docunent resume ed 323 658 ea 022 242 author anzalone, stephen
TRANSCRIPT
DOCUNENT RESUME
ED 323 658 EA 022 242
AUTHOR Anzalone, StephenTITLE Using Instructional Hardware for Primary Education in
Developing Countries: A Review of the Literature.Education Development Discussion Papers Series.
INSTITUTION Harvard Univ., Cambridge, MA. Inst. for InternationalDevelopment.
SPONS AGENCY Agency for International Development (IDCA),Washington, DC. Bureau of Science and Technology.
PUB DATE Mar 88
CONTRACT USAID-DPE-5824-A-5076NOTE 88p.; A product of the Basic Research and
Implementation in DevelopinG Education SystemsProject. A seven-page table contains small type.
PUB TYPE Information Analyses (070)
EDRS PRICE MF01/PC04 Plus Postage.
DESCRIPTORS Audiovisual Aids; Autoinstructional Aids; Computers;Cost Effectiveness; Educational Equipment;*Educational Nedia; Educational Radio; *EducationalTechnology; *Electromechanical Aids;Electromechanical Technology; Elementary SecondaryEducation; Family School Relationship; ForeignCountries; Mass Media; School Community Relationship;Television; Videodisks
ABSTRACTResearch on the utilization of educational media, or
electronic classroom instructional aids, in primary schools indeveloping countries is reviewed in this paper. Five kinds ofhardware--radio, television, computers, videodiscs, and hand-heldelectronic devices--are compared according to cost and learningeffectiveness. A model of learning production is used to define theeducational system in terms of its interaction with family andsociety, and to illustrate the relationships among the components ofeducational management, which include content, instructionalmaterials, classroom management, physical facilities, and learningtechnology. Radio was found to be the most cost effective means ofimproving both scope and quality cf instruction. Effectiveness isenhanced if the technologies are used in combination with curriculumdevelopment, instructional resources, administrator support, andteacher training. A comparative table, describing seven educationalinterventions, a diagram of learning production, and an extensivebibliography are included. (LNI)
Reproductions supplied by EDRS are the best that can be madefrom the original document.
*/
Education Development Discussion Papers
Mfflfflffl M M M MMffiffl M IE MEE El Mfflffl MM MM USING INSTRUCTIONAL HARDWARE MM FOR PRIMARY EDUCATION Mi
1MIN DEVELOPING COUNTRIES: M
M A Reviev of the Literature Mi
Le by Mil
IM STEPHEN ANZALONE Mi
1M Mi
1MInstitute for International Research im
6715 Whittier Avenue WI
IM McLean, VA 22101 El19 USA El19 ElLe M A Rc. ri 'i l'i El1M ElIMMffifflfflfflfflfflffiMfflffifflMfflfflfflfflfflffifflffl
BRIDGESBasic Research and Implementation in DevelopinG Education Systems
U $ DEPARTMENT OF EDUCATIONDeice or Education& Research and Improvemnt
EDUCATIONAL RESOURCES INFORMATIONCENTER (ERIC)
Xis document has been reproduced asreceived from the person or orgenizationoriginating it
r Minor changes hays been made to improvereproduction duality
Points of view or opinions SteMd in Mrs docu-
e('mint do not necessarily represent officialOE RI position or policy
%
CI
"PERMISSION TO REPRODUCE THISMATERIAL HAS BEEN GRANTED SY
TO THE EDUCATIONAL RESOURCESINFORMATION CENTER (ERIC)."
*Note: The formal publication of this paper is out of print -
CIyou are receiving a xerox copy.,
.
.le
tiaI.
vki A project of the Harvard kistitute for Internattonal Development.r
to Harvard Graduate of Education.and the Office otEstduatioh. for Science and Technology.
.
United States Agency for Internattonal Developmentv
.
ABSTRACT
This papist reviews research on "educational media," that is,the electronic hardware that constitutes an alternative toteachers or printed lessons. Five kinds of hardware areincluded: radio, television, computers, videodiscs, and hand-heldelectronic devices.
Anzalone compares the various hardware in terms of cost andeffect on learning. However, adequate evaluat.Lon research hasbeen done only with respect to radio and instructionaltelevision, and in these cases detailed information is presented.
Using the BRIDGES model of learning production, Anzalone'sdiagram defines an education system by its interaction withfamily and society. Family and society determine much of whattakes place in schools, but they are in turn effected by theoutcome of that schooling. The diagram also shows thateducational management and planning (as done by teachers,principals, supervisors, etc.) is comprised of interrelatedfactors that include: content, instructional materials, classroommanagement, physical facilities, and learning technology.Anzalone shows how the choice of learning technology influencesthese other factors.
Several clear-cut conclusions can be drawn from this review.The first is that television is not a cost-effective alternative,in either rich or poor countries, in comparison with conventionaltechnologies based on classroom teachers. Countries that beganwith instructional television in primary schools found them toodifficult and costly to manage, and have abandoned them. On theother hand, ziglio_im_am_stagaimiLmumm_sajammo
I 9
Anzalone notes that radio should not be seen primarily as a meansto replace teachers. On the contrary, the combination of radioinstruction--at per student costs close to that of a textbook--with teacher training leads to significant gains in learning.
Also, none of the other kinds of instructional hardware haveyet been shown to be cost-effective in the production of learningin primary schools. There is little evidence to justify theenthusiasm for widespread introduction of microcomputers inclassrooms. Research in the United States suggests that whilecomputers help in some cases, particularly in practice and drillon repetitive subjects, there is no evidence that they are acost-effective substitute for teachers.
Vid=discs are described as a sophisticated technology withgreat potential that is beyond the reach of schools even in therich countries.
The review reports one systematic study on hand-heldelectronisLdevices, which are small microprocessor driven devices
3
that teach basic spelling and arithmetic to young learners.These devices, which are relatively inexpensive if they can beused over a period of time, were shown to be effective in onecountry. No follow-up research has been done.
The last and most important conclusion to be drawn from thisreview is that learning technologies based on hardware are bythemselves sterile. Unless they go hand in hand withimprovements in content, with increased use of instructionalresources, and with greater administrative support, they havelittle chance of contributing to either the quality or thequantity of primary education.
A companion to this paper, Using Soft Technolocies ofLearning for Primary Education in DeveloRing Countries, bySivsailam Thiagarajan and Aida L. Pasigna, reviews research onthose instructional technologies (such as new forms of classroomorganization) that do not require investments in hardware. Thatpaper and this one should be read together, as the "hard-soft"distinction between instructional technologies is arbitrary:together the papers provide a comprehensive coverage of what isknown about the effectiveness and costs of many kinds ofinstructional technologies.
4
TABLE OF CONTENTS
I. INTRODUCTION 4
II. HOW THE LITERATURE SEARCH WAS CONDUCTED 5
III. SELECTION OF INSTRUCTIONAL HARDWARE 6
IV. A MODEL OF LEARNING 8
V. ISSUES
1. 0Within School" Relationehips 82. School - Family Relationships 103. School - Society Relationthips 11
1. The Medium vs. Method Issue2. Equivalence of costs and outcomes
VI. COMPARING EDUCATIONAL OUTCOMES
12
1213
15
VII. SELECTED INSTRUCTIONAL HARDWARE 16
1. Radio 172. Television 313. Computers 404. Videodiscs 525. Hand-held Electronic Devices 55
VIII. CONCLUSIONS AND OBSERVATIONS 60
IX. AGENDA FOR FUTURE RESEARCH AND IMPLEMENTATION 62
X. BIBLIOGRAPHY 65
4
I. INTRODUCTION
The use of learning technologies may provide developingnations with the means for extending access and improv.mg thequality of education at substantially lower costs thantraditional means. By learning technologies we mean procedurosand techniques based upon theories which help to design,organize, deliver, and manage educational information and thelearning process. These procedures and techniques may or may notbe mediated through devices and instruments.
In a companion review, "Literature Review on the SoftTechnologies of Learning," we examine the procedures andtechniques--what sight be called the "soft" aspect--of learningtechnology. But for the present review, we want to focus on thedevices aspect of learning technology, or the electronic mediaused to automate, enhance, or extend parts of the learningprocess. We refer to this "hard" aspect of learning technologyas, instructional/ hardware.
Since the purpose of this review is to determine how and towhat extent instructional hardware is or might be significantlyinvolved in primary education, we do not address distanceeducation projects at the post-primary level or primary schoolequivalency programs for older students--areas whereinstructional hardware has been used with considerable success.
We define instructional hardware as electronic presenters ofinstructional sequences that are significant enough in scope andcurricular importance to constitute an alternative to lessonsusually provided by teachers or textbooks and similar printedmaterials. In some cases, the wider concept of educational mediacan and will be used interchangeably with instructional hardware.
We look specifically at the use of radio, television,computers, videodiscs, and hand-held electronic devices tosupport primary school education in developing countries.
Our reasons for focusing on these five possibilities arediscussed below.
We recognize that--since, for example, we learn fromtelevision programs and not from televisionlooking atinstructional hardware per se is problematic. The literature isoften vague about the content or method that is being delivered,either because it is believed that these are less important thanthe attributes of the medium itself or because of anunderstandable and often premature desire to generalize about thepotential educational value of the medium--especially for guidingpolicy or investment decisions. We elaborate later in the reviewon the confounding influence of method on the medium.
6
5
We begin the Laview by discussing how we conducted ourliterature search and our selection of instructional hardware tobe examined. We then orient our review within the framework ofthe BRIDGES learning model. Our discussion highlights some majorissues involved in comparing and generalizing from research oninstructional hardware in developing countries. Finally, wereview evidence and draw conclusions about each type ofinstructional hardware with respect to whether their introductionwould permit an education system to reduce costs whilemaintaining coverage and quality.
II. HOW THE LITERATURE SEARCH WAS CONDUCTED
Our review is based upon a search of a disparate body ofliterature, mostly written in the U.S., that includes books,articles, unpublished research reports, project evaluations, andbibliographies. We conducted on-line searches of the DIALOGdatabases (including ERIC, Dissertation Abstracts International,PsycINFO, and SOCIAL SCISEARCH) and obtained abstracts ofrelevant literature. Microfiche versions of several keydocuments were obtained. In general, we did not find thesedatabases to be a good source of information on instructionalhardware in developing countries.
Advice and documents were received from specialistsincluding John Mayo, Jamesine Friend, Jean Meadowcroft, KlausGelder Michael Laflin, Clifton Chadwick, Judy Brace, Peter King,Maurice Imhoof, Tom Eisemon, and Noel McGinn. Jim Hoxeng andJulianne Gilmore of the Office of Education, Bureau for Scienceand Technology of the U.S. Agency for International Development,helped us obtain several important documents. We also obtainedleads (and confirmation of a lack of Third World literature oneducational applications of computers! from participants in theUNESCO/Stanford symposium on "Computers in Education: The Rolefor International Research" held in Palo Altc in March, 1986.
The literature we found is identified in the bibliography.Cooper (1984) warns of two threats to validity associated withgathering data for literature reviews: (a) not locating allstudies pertinent to the topic of interest, and (b) elements orindividuals in the retrieved studies not representing allindividuals or elements in the target population. We tried tominimize the first threat by circulation of this draft to otherinvestigators in the field to determine what may have beenmissing. The second threat is not so easily dealt with. Not allcharacteristics of primary school pupils in developing countriesare represented by the studies reviewed, and it is difficult tosee how they could be.
6
III. SELECTION OF INSTRUCTIONAL HARDWARE
All of the instructional hardware considered in this review,are examples of what Schramm (1973) called "big media." In makingthis definition, Schramm considers complexity of design andproduction, the size of the audience, possible educationalsignificance, and the complexity of support necessary for successin school settings. "Big media" originated in systems forcommunications, entertainmont, advertising, computation, and datastorage and retrieval--not as educational tools.
This review looks at various kinds of instructional hardwarecapable of providing a significant equence of instruction andused as an alternative or supplement to lessons provided byteachers or textbooks and similar printed materials. Thisexcludes various audiovisual devices or objects a teacher mightuse in a lesson to aid comprehension, stimulete imagination, orotherwise occupy a student's time. We include educational filmsand filmstrips in this category, since their role in primaryschool instruction has usually been to enrich or supplementregular lessons.
We consider kinds of instructional hardware that are ormight one day be feasible for primary schools in developingcountries. This is a subjective determination. It isconditioned by the way the instructional hardware is used, anddoes not suggest that all instructional hardware will be feasiblein all schools in all countries. The following kinds ofinstructional hardware are examined:
o Radio. Radio is the instructional hardwaremost widely used in primary schools indeveloping countries. Radio includes bothcommercial and public broadcast channels.Audiocassettes provide instruction similar tolessons broadcast by rtdio. So far, they donot appear to have been widely used forprimary education in developing countries.The same holds true for audiovisualcombinations like audiovision or radiovision.
o Television. Television has been much lesswidely used than radio for primary educationin developing countries. There are, however,several important national experiments witheducational television. We limit ourselves
7
to instructional television related to theschool curriculum and not children's "openbroadcast" programs with incidental ordeliberate educational content.Videocassettes are now becoming commonthroughout the Third World. They freeeducational television from rigid broadcastschedules and thereby make television a moreconvenient (although more expensive) mediumfor classroom use.
=miters. The use of computers in primaryducation in developing countries is not yetwidespread. Growing interest in computers,recognition of their impact on soclety,falling prices of hardware, and the"demonstration effect" of their increasinguse in developed countries are likely tocause this to change, especially in well-equipped urban schools.
o Videodiscs. Videodiscs are just now on thehorizon for use in education in countrieslike the U.S. Their use in some primaryschools in developing countries is admittedlyspeculative but not inconceivable at thistime. Developments with mass storage CD ROM(compact disc read-only memory) technologycould make this a possibility in some placesand could woll be offered in the schoollibrary of the future.
o Band-held Electronic Devices. Hand-heldelectronic learning devices includemicroprocessor-driven drill and practice aidslike the Speak & Spell and pocketcalculators. Despite their relatively lowcost and ease of operation, their use ininstruction in developing countries has sofar received little attention.
There are other kinds of instructional hardware that havebeen used effectively, often in combination with other media, inclassrooms and distance teaching programs both in developed anddeveloping countries. Bates (1982) provides a good discussion ofthe range of delivery media and comparative advantages ofdifferent systems. Some instructional hardware that might proveuseful for edu:ation in developed countries has had little or noapplication at the primary school level in developing countries.We do not consider several varieties of instructional hardwarethat have been or might soon be used in developed countries or
9
8
for distance education at post-primary levels in developingcountries, including teleconferencing, telephones, two-way radio,videotext, teletext, interactive cable, computer-augmentedtelevision, electronic blackboards, still-picture television, andslow-scan television. Similarly, the use of satellites is notconsidered, since we consider satellites as carriers rather thanas presenters of instruction.
IV. A MODEL OF LEARNING
The purpose of this review is to determine how and to whatextent learning technology is or might be useful in primaryeducation in developing countries. We ask whether the learningtechnology cum instructional hardware variables really need to beincluded in a model of learning separately from, say, theinstructional content and materials variables, and whether thealternative or supplementary instruction provided by thesesystems adds to the explanative or predictive power of the model.
The BRIDGES theoretical model of learning production (seediagram on the next page) situates the educational system in aposition where it can prevent "social origin" from deterwining"social destiny." At the perimeter of the educational system isthe set of management and planning functions that steers themystem. Within the perimeter are schools. Within schools areblocks of variables (including learning technology) that may beinvolved in the production of learning. These variables interactwith each other and with inputs--skills, aspirations, andknowledge brought to sdhool from the family and community andpolicies and resources supplied by society. The production oflearning results in values, skills, and knowledge returning tothe family and society. These outcomes are produced with varyingdegrees of effectiveness and efficiency.
1. "Nithin_fichml!AilatimishignBoth intuition and the literature suggest that the most
important relationship between the use of instructional hardwareand other "within school" variables is the relationship withinstructional management. Indeed, the literature is rich withexamples of the insufficient attention paid to this relationshipand the disastrous consequences of this neglect.
The two most important aspects of the relationship betweenuse of instructional hardware and instructional management areteacher acceptance of the technology and teacher competence inits use.
Teacher acceptance seems to depend to some extent on thetype of, and the purpose to which instructional hardware is being
1 0
FIGURE 5.1 CONCEPTUAL MODEL OP LEARNING PRODUCTIO0
EDUCATIONAL GYSTEM
MANAGEMENT AND PLANNING
"SCHOOL" ORLEA RN DA CENTERft.. . 1 CONTEKT 1
I WC TRUC 1,..-"---1% sis
1 TIONAL i1%MANNA/V.0r
%. ,*
1..
o %
%
; I WO' IOC- 1
I%
1. I OPAL I .OATEItt AL %
51 pgys I CAL \....... PAC! L I 71111SI I
.. ,'....'
I
I
VALuES,SK:LLS, ENOKJEAC
SKILLSIASPINATToNs. KHOOL
... ...I .
I LwAprömq 1,reatooLoo :
,....-%
........M.IM. Yr
SKILLS, VALUES, KNOWLINICA
PoLicies ANO RiSouncRS
11
FAMILY
SOCIAL cONTEXT
COMMUNITY. N.so "" ft
soct eTY
SOCIAL CONTEAT
BCONOMT
POLITY
1 2
1109
applied. The introduction of new technology in the production oflearning, as in the productiou of other social products, oftenanticipates a substitution of capital for labor in order toachieve a net reduction in cost. The substitution ofinstructional hardware for labor in educational settings provesto be economical when more highly trained teachers can bereplaced by less trained (and lower paid) teachers, or when thereis less frequent contact between teacher and student (Jamison andOrivel, 1982). In many instances, the relationship betweenteachers and the use of alternative instructional hardware may befundamentally antagonistic. Our review treats the substitutionof capital for labor as a policy issue rather than as a classroommanagement issue.
We believe that the most important question about teacheracceptance is the following:
Does different instructional hardware meetwith and sustain different levels of teacheracceptance?
The second aspect of the relationship between instructionalmanagement and the use of instructional hardware has to do witi:what is necessary to operate the system. This involves both theteacher's existing knowledge and skills as well as whetsveradditional training or orientation is required to operateinstructional hardware. There is considerable variation here,.Some teachers have been oriented to a new curriculum delivered byradio in as little as three hours. However, to prepare them fora more complex intervention, say, to teach LOGO, considerablymore training ts required. The most important questions are:
o What knowledge and skills are required ofteachers to use ihstructional hardware and,perhaps, to learn a new instructional role?
o Do knowledge and skills gained by the teacherfrom working with instructional hardwaretransfer to areas where such hardware is notbeing used?
A second important "within school" relationship is betweenthe use of instructional hardware and content.
The questions which should be addressed are:
o How readily are the various school subjects taughtby different kinds of instructional hardware?
o Does this vary with grade level?
o Does use of instructional hardware permit (or
13
10
dictate) adding or deleting content in thecurriculum or acquiring skills within asubject that are either now or possible at ane arlier age?
The third "within school" relationship is between the use ofinstructional hardware and instructional materials. Theimportant questions we consider are the following:
o How effectively and how cheaply cane lectronic messages substitute for printedmessages?
o What visual supports are necessary for audiomedia and vice versa?
The final "within school" relationship is betweeninstmctional hardware and the physical facilities of the school.Clearly/ the condition of school facilities, especially lack ofelectricity, would rule out the use of some instructionalhardware and constrain the use of others For many people,policy makers, and teachers, poor facilities often renderincongruous the introduction of alternative instructionalhardware. The face validity of the priority of improving basicfacilities over the introduction of instructional hardware ispowerful. Investments in improved school facilities are likelyto be made for reasons apart from school achievement and apartfrom accommodating or facilitating the use of alternativeinstructional hardware. The important research question appearsto be how alternative instructional hardware can or cannot bemade to operate in the typically abject conditions of remoterural or crowded urban schools.
2. fighook:rjaily_RaatjanakimThe BRIDGES model (page 6) highlights the reci;rocal process
by which skills, aspirations, and knowledge are brought to schoolby the student and values, skills, and knowledge return to thefamily and community. With respect to the first of these, themajor questions are: Do factors beyond a student's contzto, suchas socioeconomic background, ethnicity, gender, mother tongue, orgeographic location of his/her home affect entry into school,continuation of schooling, and what is learned while in chool?To what extent are these changed or reinforced by the use ofalternative instructional hardware in instruction?
Issues related to how values, skills, and knowledge returnto the family and community are not generally treated in the -
literature. It is difficult enough to measure the gross effectsof either schooling or the content of particular school subjectson family or community life. It would be extremely difficult todisentangle or even detect the effects of what may be very
11
limited contact with instruction provided by any kind ofinstructional hardware.
3. figliggiallipaA student needs certain skills, values, and knowledge to
participate in adult life. These skills are to a large extentmanifest while in school and, after leaving school, affect andare affected by social and economic life.
It is the latter part of this process that is of mostinterest to policy makers. It is the former part, however, thatreceives attention in the literature. Apart from the occasionalspeculation that exposure to media in schools contributes to aweakening of traditional rural values, causing a drift to thecities, there appear to be few attempts to examine the oxternalefficiency" of instruction provided by instructional hardware.As was argued above, if long-term effects resulting from what isoften episodic contact with alternative instructional hardware dooccur, they would be extremely difficult to detect. The problemof assessing the external efficiency of learning associated withalternative instructional hardware is particularly germane in thecase of the newer devices, where the introduction of computersinto education is often preparation for the economic and socialpossibilities engendered by the rapidly developing globalinformation society. These claims are only beginning to beexamined in countries like the United States (Carnoy et al.,1986).
The use of other instructional hardware has largely beenpredicated on the more modest claims that they affect learning orinternal efficiency within schools. The variables mostfrequently treated in the literature are achievement in differentsubjects, attendance, repetition of grades, and continuation inschool. Often, there are measures of attitude toward a schoolsubject or the instructional hardware itself. The adequa(4 ofthese measures and the difficulties in comparing them with theeffects of conventional teaching is discussed in the followingsection. The notion of internal efficiency could be treated as a"within school" consideration. We treat it in this section inorder to keep discussion of outcomes in the same place and toemphasize that interncl efficiency is something wider thanclassroom proceee, and something closely related to policies andresources.
The policiss and resources that determine the use ofinstructional hardwars in schools are also of fundamentalimportance. An undevstanding of how policies --or perhaps moreaccurately, actions of policy makers--affect the selection,introduction, and continued use of instructional hardware isessential and largely lacking in the literature. Perhaps thesingle most important question here is:
1
12
o Has the introduction of instructional hardware found asustained or expanded place in national educationalpolicies?
The fact that particular resources are needed forinstructional hardware is of indisputable importance. This facthas received considerable attention, and this literature isconsidered throughout the review.
At the level of policies and resources, two questions areimport:Ant:
o Has the introduction of an alternativeinstructional hardware permitted asubstitution of lower skilled (and lowerpaid) instructional personnel or a reducedrole for the teacher in the improvement ofinstruction or extension of schooling?
o Where introduction of instructional hardware does notpermit the substitution of capital for labor, what arethe additional educational costs and theirimplications?
V. ISSUES
There are two issues of special importance that make itdifficult to determine whether instructional hardware cancontribute significantly to the learning in schools: the mediumversus method issue and the equivalence of costs and outcomesissue.
1. ThsAsslimaLyirimiktjualumAdvocates of the use of media in education often point to
various attributes of the media as being particularlyfacilitative of learning. They cite such things as the immediacyand liveliness of radio, the ability of television to command andhold attention and to provide camplex visual stimuli, and thecapability of computers to offer immediate and individualizedfeedback. Explanations for poor reoults from particularinstructional media often cite the failure of the instructionaldesigners to make good use of the features of the medium.
There are those who dispute the claim that various featuresof media affect learning. Clark (1983, p.445) argues thatcurrent summaries and meta-analyses of media and learning studiessuggest that "media do not influence learning under any
13
conditions." He contends that in the instances where largechanges in student achievement have been associated with the useof media, such as (and this is his example) with educationaltelevision in El Salvador, the changes are attributable toclurriculum reform and not the medium. Clark insists that betterperformance associated with use of media is explained by theconfounding effects of instructional method, and to a lesserextent, novelty effects.
Others take exception with Clark's hypothesis. Petrovichand Tennyson (1984) argue that the meta-analyses Clark usesinvolved classroom studies without adequate controls to test hishypothesis, that Clark did not adequately take intc account howvarious components of the huuan information processing systemaffect learning, and that recent evidence from computer-assistedinstruction studies do not support his view.
The medium versus method issue is clearly relevant to theliterature on instructional hardware in developing countries.Howver, it is often difficult to determine what ^astructionalmethod is being used in a particular media application andhazardous to generalize about what methods work best with whatmedia. In an application where the method is well-defined, suchas the interactive radio method, there is likely to be wideagreement with Clark, that the outcomes are attributable to useof the interactive method specifically rather than to the mediumof radio generally. For policy makers in developing countriesthe medium versus method issue may not be critical. Even if itcan be shown that it is an instructional method rather than amedium that is responsible for gains in student achievemnt, theuse of media may often be the only means to deliver the methodefficiently.
2. Equivalence of Comte and OutcomesCritics of the use of instructional hardware in developing
countries raise several related questions about how costs andoutcomes are analyzed and inUtrpreted. Carnoy and Levin (1975)argue that evaluators' "modes of analysis" have possibly beencontaminated because of the advocacy positions taken with respectto the media being evaluated and because of their closeassociations with the agencies that have sponsored both theimplementation and evaluation of projects being examined. Emery(1985) takes a similar view.
Carnoy and Levin (p.387) identify two types of bias in theevaluations of various applications of instructional hardware ineducation in developing countries: (a) the use of "deficientdata" when they favor instructional technology over traditionalalternatives, a phenomenon they refer to as "benefit-of-the-doubt" bias, and (b) "narrowing of the scope of analysis to thoseitems on the agenda of the sponsoring agency while ignoring other
1 7
14
effects."
Carnoy and Lavin (p.388) submit that cost analyses ofapplications of instructional technology are faulty because "inalmost every case the reported costs of instructionaltechnologies are understated relative to their true costs."According to them, this results from basing cost estimations upon"drawing board" approaches, which, they contend, are invariablylower than actual costs. Moreover, they suggest that costestimations tend to ignore costs of such things as "contributedfacilities" and technical assistance costs associated withdevelopment of the instruction to be provided.
The second bias identified by'Carnoy and Levin (p.391) isthe use of "a narrow Aasure of educational attainment" whichtends to "overstate the attractiveness" of the technology beingevaluated. Thdy argue that the outcomes of education go farbeyond performance on measures of reading and mathematics. Thisis, they note:
an especially serious problem in a cost-effectivenessstudy because the evaluation instruments are invariablydesigned to compare the instructional approaches on thebasis of what the instructional technology is designedto achieve rather than the far larger outcomes that thetraditional systems of instruction focus on. (p. 391)
We believe that these contentions should be taken asimportant cautions in interpreting the literature on the use ofinstructional hardware to support primary education in developingcountries. However, two points should be made. First, costanalyses that have appeared subsequent to the time period withwhich Carnoy and Levin were concerned appear to approachestimation of "true costs" by taking into account actual costsand fixed costs for instructional development (see, for example,Jamison, (lees, and Wells, 1978; Jamison and McAnany, 1978;Leslie and Jamison, 1980). Second, although few would disagreethat educational outcomes include more than performance onachievement tests in reading and mathematics, ths importanceassigned to these attainments by educational systems all over theworld makes it difficult to accept the idea that emphasis onthese outcomes in evaluations is a self-serving "narrowing" ofthe educational process by the advocates of instructionaltechnology. In recent evaluations a the use of instructionalhardware, there does appear to be sensitivity to the issue ofwhether an equivalent range of educational outcomes is beingcompared (see, for example, Friend and Kozlow, 1985).
15
VI. COMPAING EDUCATIONAL OUTCOMES
Our attempt to determine whether and how instructionalhardware can be significantly involved in the production oflearning in primary schools in developing countries requires thatwe evaluate and summarize information about educational outcomes.There are enormous difficulties in producing a synthesis ofresearch findings fram studies involving the use of differentmedia since applications have been undertaken in differentcountries, at different points in the educational process, andfor different educational purposes.
In the past several years, research reviews have madeincreasing use of quantitative meta-analytical procedures forevaluating and summarizing research on educational outcomes. One
of the most widely used methods of meta-analysis is thestatistical measure called effect size. This is a comparison ofdifferences between an experimental and control group. Theeffect size is simply the difference between the means of theexperimental and control group divided by the standard deviationof the control group (Glass, 1977; Glass, McGaw, and Smith,1981).
Walberg (1984b, p.216) describes the advantages of usingeffect sizes:
Effect sizes permit a rough calibratior of comparisonsacross tests, contexts, subjects, and othercharacteristics of studies. The estimates, however,are affected by the variances in the groups, thereliabilities of the outcomes, the match of curriculumwith outcome measures, and a host of other factors,whose influences, in some cases, can be estimatedspecifically or generally. Although effect sizes aresubject to distortions, they are the only explicitmeans of comparing the sizes of effects in primaryresearch that employs various outcome measures onnonuniform groups. They are likely to be necessaryuntil an advanced theory and science of educationalmeasurement develops ratio measures that are directlycomparable across studies and populations.
Effect sizes are expressed in standard deviation units.Although this is often not readily understood by persons without
a background in research, effect sizes can be expressed in termsof changes in percentiles. For example, the meta-analysis of 32research studies on the effects of computer-assisted instructionon achievement of elementary school students in the United Statesconducted by Kulik, Kulik, and Bangert -Drowns (1985), found anaverage effect size of .47. This maans that the use of computer-assisted instruction had the effect of raising the mean of pupil
19
16
achievement from the 50th percentile, which would be averageperformance without computer-assisted instruction, to the 68thpercentile of that group. In interpreting effect sizes, Walberg(1984b, p.222) reports that from samples of recent educationalresearch, an effect size of .20 (equivalent to raising the meanfrom the 50th to the 58th percentile) is about average and one of.45 is "considered large and exceeds about 84% of those typicallyfound in educational research." Research syntheses undertaken byWalberg (1984a, p.23) show, for example, that IQ has a strongeffect on learning with an effect size of .71. Similarly, ofvarious methods used in instruction, reinforcement or reward forcorrect performance had the largest overall impact, with aneffect size of 1.17 (equivalent to raising means from the 50th tothe 88th percentile).
We have attempted in this review to use effect sizes toevaluate and compare achievement gains reported in the literatureon the use of instructional hardware in primary education indeveloping countr1es. In the present draft, we are limited tothe interactive radio vise studies, where evaluations providenecessary information.1
For purposes of formulating educational policy andallocating resources within educational systems in developingcountries, the important question to be answered is: How largean effect is required to make use of instructional hardwarecompelling? Is an effect size of .45 (or, raising the mean fromthe 50th to about the 68th percentile) --judged to be large indeveloped countrieslarge enough to motivate policy makers in adeveloping country to undertake a major investment andreorganization of school instruction?
VII. SELECTED INSTRUCTIONAL HARDWARE
The possibilities for using instructional hardware toimprove or extend education in developing countries have beenconsistently acknowledged since the 1960s. Perhaps the landmarkendeavor of the early era was the four part series NewEducational Media in Action (1967), edited by Wilbur Schramm.This collection included three volumes of case studies and asynthesis entitled The New Media: Memo to Educational Planners(Schramm et al., 1967). These books emerged from a projectundertaken by UNESCO's International Institute for EducationalPlanning under a contract from the U.S. Agency for InternationalDevelopment. The series includes twenty three cases studies from
1This was made possible through discussions and sharing ofinformation with Jamesine Friend, Peter Spain, and Julianne Gilmore.
2e
seventeen countries and covers a wide range of educational
applications for tha use of radio and television. The wide
analytical sweep of these documents, their great readability,
their generally upbeat appraisal of the educational possibilit.
for the "new" media have set the tone for international
discussions for years to come.
At least twenty years have passed since radio and televisi
emerged as possible transformers of education in developing
countries. At a time when we are still trying to assess the
effects of what has been called the "first electronic
revolution," by considering whether instructional hardware,
especially radio, will or should play a wider role in primary
education, the "second electronic revolution" (computers,
videodiscs, and similar devices) is already arriving at the door
of schools in thesis countries.
We look now at the literature on specific instruccional
hardware: radio, television, computers, videodiscs, and hand-held
learning devices. We look retrospectively at developing
countries' experience with radio and television and
futuristically at the possibilities for computers, videodiscs,
and hand-held learning aids. Our treatment of radio 3s the most
detailed at this point.
1. WigRadio signals have come to reach increasingly large numbers
of citizens in developing countries. Between 1965 and 1982, the
number of radio receivers per thousand inhabitants has, for
example, grown from 12 to 131 in Indonesia and !rom 11 to 177 in
Sierra Leone (UNESCO, 1904). McAnany (1973) was correct when,
more than a decade ago, he described radio as the "most universal
mass medium of communication."
With improvements in the processes of manufacturing
receivers and developments in the us4 of transistors and
batteries, radio became affordable,portable, and practical--
even in remote regions. Radio has, since the 1960s, played a
varied and important role in social and economic development
efforts. There is extensive literature describing applications
of radio, often in combination with other media, to agricultural
extension and farmer education, adult literacy and basic
education, health, nutrition, population education, and family
planning. (See, for example, Clearinghouse on Development
Communication, 1978; 1982.)
Radio has also been used in various ways in classrooms in
developing countries, including primary school instruction. In
the 1960s, Thailand emerged as an example of how the primary
school curriculum could be enriched at relatively low cost
through radio broadcasts. In 1967, about 800,000 pupils were
18
receiving radio lessons in several subjects (UNESCO, 1967). Theexperience in Thailand is summarized on page 23, a.c. We discussvarious aspects of Thailand's experience later in this sectionand in connection with the questions raised related to theBRIDGES model.
In 1972, UNESCO's International Commission on theDevelopment of Education noted this about radio:
People often seem to be deterred by the reputedlygreater efficiency of other media which, however, havethe major defect, compared with radio, of being unableto hope for such widespread distributionor anythinglike it--for a long time to come. The very low costand adequate reliability in all climates of miniaturetransistor radios mean that radio broadcasting shouldmore and more be recognized as a particularly suitablemedium for educational purposes (Faure et al., 1972,p.122).
Radio's potential for delivering education in school wasnot, and perhaps is still not, apparent to all educators. As theFaure Commission noted, the attractiveness of television largelypre-empted consideration of radio as an educational tool in manyplaces. Moreover, radio is not now widely used in classrooms inthe early industrialized societies, thus precluding any kind of"emulation effect" to stimulate applications in developingcountries.
The early promise of radio related closely to some of itsattributes as a communication medium that offered potentialadvantages for instruction. These included radio's livelinessand immediacy and its apparent ability to require concentrationof attention and mental effort on the part of students. Schrammet al. (1967) cited four other potential advantages:
Radio can spread and equalize learningopportunities while raising educationalquality
o The old media such as textbooks could sharethe substance of teaching; the new media likeradio can share the dynamics of teaching
o Radio can provide demonstrations to schoolsthat could not otherwise afford them
Radio's potential was seen by Jamison and McAnany (1978,p.12) to fall into three cetegories: "improving educationalquality and relevance; lowering educational costs (or the rate ofincrease of costs); and improving access to education,particularly in rural areas."
22
19
These were formulations of educational potential and not anassessment of effects observed through research. The earlyempirical "basis" for the effectiveness of radio is outlined inJamison and McAnany (1978). They cite the survey by Chu andSchramm (1967), which concludes that under proper conditions
students can learn from any of the media that were currentlyavailable and that not only is a visual channel not alwaysnecessary for learning, sometimes it can interfere with it.
These conclusions were shared by Forsythe (1970). Jamison,
Suppes, and Wells (1974, p.33-34) reached similar conclubions butadmit that the empirical basis for these conclusions was small:
Radio has been used extensively for formal classroominstruction in the United States (more in the past than
in the present) and elsewhere. There exist, however,only a limited number of mood evaluations of theeffectiveness:: of instructional radio. Theseevaluations indicate that instructional radio can beused to teach most subjects as effectively as a liveclassroom teacher or ITV.
The radio initiatives and the case studies reported duringthe 19705 were helpful in moving from an assessment of potential
to an assessment of effects, even if the information collected onlearning outcomes was often sparse or potentially unreliable.Two rich case studies came from Mexico: the Tarahumara RadioSchools (Schmelkes de Sotelo, 1972; 1977) and MexiJan
Radioprimaria (Spain, 1977). Insightful commentaries on these
case studies followed in Jamison and McAnany (1978). To
facilitate our present analysis, these case studies have beenbroken down into summary form on the following pages.
Both of the Mexican radio initiatives were atteLpts todevelop low cost extensions of primary education in underserved
areas. Both used radio for direct rather than supplementalteaching and for several subjects. Both report achievement data
that, if not clearly indicative of radio's effectiveness, at
least point in that direction. Each project failed in one of its
principal objectives: the Tarahumara schools failed to promote
much equity among a disadvantaged sub-population and
Radioprimaria essentially failed in its objective to use radio as
a substitute for hiring more teachers in a school. Both projects
demonstrated that receiving radio signals and keeping radio setsfunctioning properly were matters not to be taken lightly by
educational planners. Despite the promise shown by these two
efforts, both were abandoned and seem to have fallen victim to
the prevailing disenchantment with formal education witnessed
during the 1970s.
Two similar initiatives undertaken during the 1970s are
23
I5S=.2.1221
Italia Oroect 1n theolond
csolcsibilm
Thelland
To Offir Instruction in
Objects Woe schools had
limited abilities 'a ovoids
T. twirl dselrsd sestet
volume
Primer, and woomftry school
and teachers
teacher impacted to grew.
stuants fer broadcasts.
leed students rewrote
review meteriets and
conduct plet-broacest
activion
oatalater:als Uszl
Scripts and wee printed
and circulated to owners
Lit:ninon'
In cogsrliene Women radio
S W couret twants, in social
togies. Marengo in
favor of tee ratio *twines
sere Ow in ride 6. and
We me Inbred* 8. In
ayella differwas hi Hoer ofpone wawa mere highly
signitleamt in written toots
e nd mete a denting end
inging. In MOW, *WIo we MO IgnIfIcant elf.
erre= sport hem writing
sialltIes et Irma 6 (In few
of antra tudents) and et
Crow ? (n fever of rune
scubas)
Mica UMW=
so infliesetlen
Iwn IMPOIOIMPO to oresametionel rano for orhuryelirmian Mos* the
w arisro am. imb WM,
Ulba fil0=21
fedi. 113718: 206 scowls (220 elementary end 46
setordery
ill,: SOO schools
legit 1.000 Wore
teat 3,080 Wools. 800.001 tudents
VIS3 MI
lasslanwou..ILLam
362 bars or year
Lfflailt-dIrausrairo
Theintal lei
Insfainrcagme
30 Simone teachers program
twice mach tort of the school
year
sunlit aerial all POW*
teachers to use the Inelish
podia broadcasts
wattle mamlieralmt 12Stial
It yes difficult to fit a
tisht broadcast scheatte to
riddle varyine time-taro
of schools
Iniscoal.111.1Lim
so Informolen reported
en the oriole Indicators
of Internet Offshoot
Nen free student
achievement
MIA
tan" et M.. MIbrim. SWIM MP. 1,0
larlralibus
mato eam 1.8; 3.4
Seslo4 $i.t bridle 141 34
Ingstleb se Ileserd unarm Crean 3, 6 6 ?
ma foie II. ond III of toorilery seheet;
A oelldrars Itema hour progress a einem
o f adatetlenel meterfild ud 014411411MMOOt
broOltalit every dry of the embeet nor
/011011111011.11123111
Vented emerang to subject
rmatle Wiper wed in sesta staples
Utile emikede weld for Inaleh ard swig
aggbiLliageong
N o infeemetion
MOW Oeitraticinat talm4n
kcMiisuIity
MANI oldie teachers
unfavorable et first.
Attitudes believed to
hew hawed during
wend ere third
yore Ott deteriorated
* gain because of
growing difficulty in
receiving signet frem
Sanaa. PulAta endpurls ttitudes nos
witty,
flnaretal we foreign achange isommed
develeport of pewee *silty ore isesnello
of the system. llooptien of stoats Wow
herder es we COIONVIIII tetter bowtromasetIng. Maintenance 0 sets I problem
for WM Wore. C0.4100Patieft of teachers
d ee often o wages
Man
e a Information
apreal SapinilttoNO 14, the educational brombeets hoe boon loos
than 0.81 of the Ministry WOW. The cest par pupil 0 serving the
20Z of Tar puplle We received the brodcasts was to awn 0311 of
the total apansitture per pupil en olutotion. This treats Wet ts a
coat 0 Wad i cents per studoi.bour
tasina
24
113=1011.111
Tarawa* Radio Scheele
csoscatalm
Malin (Siam torlawars,
mato region in ; 'Ocala
scow.)
IIIIMILLIX-110111-1111111)
SANWA, eshimAimg to muneansaties awe it was
possibl- to satabileh an
n offislan sok*
Oreviding instruction of
sufficient astir, to ensure
contimation in railer scials
Ira utIsi
Originally intended for the
Tantalisers India' but in.
eluded white children se well
Ono or tin nowalllere teachers
intorectins with redis tanner,
tat frequently provided in-
struction in lieu of radi
Officiel sada tostbats dig-
tributed free to pupils
lifjabieggi
Fourth redo WOW in 1971
w enn G00% en third wade
own In arienntels and %DaniaallOssfid to SSA% for moilsfrom CISO41 C private schools
in aria City en the awe toots
Radio sagas bean in VOSS.
ladle an no t..er uNd lifter
the 1171-74 school your
MLR
tale
IMELIIIILLIE1111111
1937-72
1 a I 1 a 11101:
Vedas 14 amorally
fastigiate se
t IMAM slasersouWenn: rItalina.
erithatie/gesear7,
history, geogrepa.
' claw, civics, ert,
snd Calms
11011111-ILLIMILIES1111
Opals* ielthough the
majority of Terahmere
fcallos spook no Sanishi
Luslicicuatallr
iii resorted
ftisc_becatimailthad
tel reception snd difficulty
obtaining repoirs ad bat-
teries
gvaiustion found only 7 of
24 wheels visited actually
eating use of radial
1.210111111fariet
lama 1917-71, Interwar
dramas wore 013 (notions
am ad 75 for rural
schools); athletes OU0Orat
S O may es 1311 of dropouts
resulted from aka elsola
D ropouts increase et higher
grudge, suggesting higher
opartunity costs
ft date em failure in school
111131M-All
11111 polls In SA schools
(11171)
puhes Orevieina Oats
ahoolkes de Setele (1971;
1917); esisan & wont,(MI)
LIOOLUMIMOLV-IMMIll
tS Mate broodiest*,
folload bY 0 shoals of
anallsog throughout the
la, doily
IMUALL1011-1111bil
Almon tiles* Milosof motorist free
@Motel Magas by rale manor
Tads, ablest eater
wooneine er ottoman's
tt NNW aurae dfd
net effect pail
g alavant
Ulf NCO11101.111fill
not notated
S opa apensien of system Wand admin..
'strata cepaity of the systa resulted
little again' er of Wont*
Eaten possaned little aperients sn ioIndio
LULU
Studonts ethnic Sect-
grand as Silnificono
(pc.a) predictor of
overall ecologic per-
foram via white
pupils outperforming
Toraloplara pails (but
not In orftfolltia):
them' stiffosisosecruse Masa first
ond emend 'rode
unworn mere likely
to dropout ond ore
unierropromented in
grates 3 ad 4
total instructional and misigoltrinin emits (ISO-
M) per stoat as slat SIC This amieres to
an eatfastod per pupil plenty aponilionnt of 0.2
innate* In 1971. '140ft AMONWOrostioste
tor 1971 on oversee causi par sOMPONS Met of
MAO (er 0.331 per WO Wad Man 11
SOIMOINCO
fraternal Efficiacx
little herd ate.
[valuation concludes
thet pails who com-
plete fourth wade
ere wowed tO
cOntinme in rog-
via schools
Evaluation suggests
thet Instruction
net suited to the
mini ud sanomic
allay ad natal
thet dl Is Ormilootell
de nit use idiot
Is lama In
sena, some Ors*
atm use *sheet
kmadas toplat their pospien
ad thee namefind wort In tans
ad cities
CIithe WOO of radio to
mesa pillar, a-
untie% pool dy to
offsets tO promote
the use of midis
for emommity ed.
~WI inapirod by
Paulo Peeks
Insmainue
sexism teditykrimerie
limicanige
mr:ies (lesion areund tan
Luis hstesi 260 eiles
nerthwest of te capital)
latiinabeilgraBolla
Te illerenee etTiehin4Y ifrural wheels by tu 4ng
six grade wheel with four
teeshere
V. resell school dreamsOmer laplannesd)
aural children
(no ethnic distinct's''ers
reported)
ritismzukt
little informeitiono reported
that enty 102 of teachers felt
they we up any authority bs .
cause of radio
204.1244.104..14231
Mimeographed biweekly guides for
teachers; free textbsoks for
pupils
In 1975 pupils received
supplementary materials
MagninAchievement tests in arithmetic
Ind Spanish given three mewls
apart to radio end non-redie
sixth waders; gmin seem in
were en Spanish fer radio
pupils (3.7) significantly
better than for neared% pupils
(2.3); on aritemetie radio
pupils ales outperformed
normal. polls 3.3 to 2.9 gains
en amen tut net significant
&stemma Ifficfmmv
Nsbe
Sadie
atm jaimapjagel
1,70.72
Igrs West*
Iacta&edge 64o itemmilh.
arithadtie history.
ilmegrephy; aloe @4WD
physical adlosetion,
%mere stmer.
prettiest activitiom
lacers.11.1raussiise
Swish
Lugum/rigicium
me orientation provided;
high rote ef teacher
tUrreVer
BelatistrisimaLluset
Peer reeeptien in eeny areas
Only 111 of 44 schools
found to have audible.
functioning radio
Jnternmt Midi=
es conclusions can be drawn
Newt effects of radio en
repetition er drogue
teacher one pil attend .
ans9 opegred 9oed in
radio schools
In 1972, 36 et 44 redie
schools were already
complete schools orid used
radio se supplement. net
to improve efficieneY
Intel-rifts with parents indicate widespread perception
test priory education hem little utility in rural areas
Interview Jib representatives of local industry pointed
to dimoulty in finding lurk with only Primm77 119/1071 eau'
cation
A 44 WWI 0 47 grodUstes of primary schools in ere. }pond
352 et graduate, working me 3$11 continuing studies but re
seeparisies of mei* one renredio schools were made
Mairritlmilsaugairm
Sestina in sumer if 2441.411 lain, radio free 40 in 1070.71air im 160244. Tbis ineremeed to 4 in 44 tut me this
eime mils mos wed may Is MO rub ohms
idx111:111
2075 pil') In 65 schools (1975)
Itudies Providing Pate
Spain (1977); eseAsen & mcAnent (1976)
611:1116aL1111111IMALIMICa
live er eix fourtean minute broadcasts
each day, semt et them multigrade forest
Idetructionst method
little informetion, but indications that
bereedcests followed Comely lessons in
official towitbeeks
in 1975. lessen fermate said to rely lets on
direct teacher talk. sere dramatizations end
concrete mummies provided
ashr-lirasszel
spew to lune meat
school 40Wcation; re:
of teachers welting in
the pros live in the
city and commute to
to rurel schools
mato, nonSeemint Issue%
Inadequate mechanisms for
mensimmers and supervision
of schools
Max
Accootabititx
442 of teachers felt
Wine radio as an
aid superior to
direct teaching
stone; 441 said it
was not
Me COnetUilienit con be drawn about
any differential effects of using
radio with particular grooms of
pupils
large proportion of schools that already
had sixth grads suggests stoups less on
need of new technology benefitting more
Jamison & OcArurn (197S) compute en swaps
ennust cost of 1116.44 per student ( s.odia
perhour) belied upon participation of 2004
students
2 6
Wows
lattrentar WWI catiallt
meopfon asolio natamoucs Om li 1.000 4000 balls 1 onset of protect.
P reimtrISI,1 te 10.0011 pits in pertictootIns
w heats harem nabs, (istonars in
other seluois and a nano)
i11121Cia01110
e loweges
Cengs indite Os. Juin
ormolus*
112212o2i2.122.221talothe
lallaem samotionat basiity
repining misting
striation with reels Ur
striation
(nnegmt me to devise
Wuretrete wag of
o ffeetive ad afforlabl rale
biotin/Mien Wet wad be wed
In era develepAng seuntry)
liELJ22211.12222=1limps Prwiffinallaci
1173.1,11
Methenoties Vedas 1.4
ihapes. Sedrie, an/ Primal (Min
fielenk Wrie, awe (1910):
Friona (1100); Mill* lel* and fowls
tunitalls maws and Ramo (1024)
LonhatormiLlimaif Waste iffebbol, S days por wok
Ielleved by WM Waage a post.
Weedoist activities
lioarast.lzatnation Itairist(222LINsbt
era ond teem children OpenishProject Wid famietlon for the interettivr
MO esde. fliotwes insiuded estivs and
trauma' pall ressanws, imsealate
fOinfeallbant, diatfiebtoal preetios vith *te$ .
mentor etnature labia S Wow twin
pr.sarted per lessen): lessens wed famence
essuplee, *entrails* oticalmisrc, timer of
'WNW Iletfoserf, 4111 an0o1ne ItninSfv '
animation of (awe
1222212..22i2lucky Item."
amlie aflame Me Owl? lessons, throe hew tralt..ng we-
teachers wan peaulareatest eien east yeer.
activitiesfecomers oom bade
NI no regular super.
vision
guar materials age 55A5? fagna2.3.23ball
Wimasets owl Mr VW 1, *Attire& siassrama (MG
Iwo 24 mild inwansiv. of Mei)
notollsalm. IM Mina Ws el problase
Maw applamtely neteriale
Isu0.c.122122201
Nether, kad en oversee
11 yeas of eacation,
anchor's linewliage of
o ath bawd to be wok
n ame flanoomenc Icsoml
IfftI LoUraiLlillram Mtn
W ait Gloom outperfornal
control iiMMOI In al grades.
ifferansee owe ummewilly
strong in great ene. affect
altos ers as fellows
erode lo 1.32; Weft 2: .21;
Weft 31 .47; grads 4: .02
1222024 A11r1cci
se InforsotIon
fbe we a ratio did not
beve strong positive
offsets on stanares.
dropout. a repetition
A savoy of palleii
peting tosfaleml
(Mew ene fowl.
Ion) nowt1 tease.
ors found :Vie
lessee help chit+
ron loan and ere
helpful for thar
w ives. tbey 11104
0 woman
Met MH lessons
wad fat be oh
rand t Me Mt.form abilities of
children
Wien radio alldron outperformed rural
podia Millen. se significant Alfieri
wan ettrilatebio le wolf
Jesuang elates owe 011.11/1111N1 te 1211,000 wars ane intludina
W aludwe hail sest allegation, Miaow and *Mew (1,211)
miebiato Mat to be 113.0$ W etwilwit per hew. Mullins(
casts West of adding ate we MIAOW to e system) we
C1.04
2 7fit Obe lenelesOlen of Aill assistance,
Mermin$ ass under toy te word smarefibla smeress of
gbe progoes end for govawins lessee for &Ode S. nowt maw beaded off la ebe wheinetion
Ihe twobegtes I. 1001
ILIMAICLISC
terve Smile lenposoe Arts Project
ilICLCO21110
Coco Orme districts
selected ts posit s netiensi
sample repreemntetivo of differentnetienel limey groups
I. tomb english es a mend
Impose sere effectively
(Project psis mos to compere
offsets of reMe Instruction
eith senventienel instruction)
laciabiastLin
Rural children:
eisht different mother tempos
P reVidllettbor tenontree:lotions es needed, provide
ewe corrections to students.
cerduct postbroMpost activities
9111C-MILIALILISI
P upil morteheete bard in
beeklets
Smile pupils outperformed
nen-radio pupils on ell tests .
miegrImpie newly elways stet .
lotto:Hy significant. largest
Offerers.* marred in listenins
stills. effect sites ss follows:
Grade! Grade? Grods3
listening .$ f 1.17 .37
feeding .32 .29 .41
Mitine .22 .rtIpositing
(Word Totai) .23 .1,
(018111111118) 39 .49
(Groomer) .31 .24
laiteralltrnim
N o date reported on imams of
redie ebsients In school after
mode 3 but roselneee of radio
polls to continue in English
lames. inetrustion after
grade 3 Indleeted ip 90X of
umber.
The Weiser, of education of
Imps decided net I. seMinue
kayugusua otter Me
ladie
(wilts palish in actionl
LunittaLartinil
Instil& erodes 1.3
imammailratairabe
english
maier Omeretionst Immo
eerie reception *probiemetic .
in leZ of schseis and yew*
in 32. One district ei .
eresseddifliculty obtaining
printed motorists en tme
Interne( efficiency
Owes tiess es mush content
presented in Wolf hour blocks
of instruction compered with
conventienel instruction
less disruptive pupil
behavior observed in radio
classes
tapestry end transfers Into
radio Messes wound to
Wye sem diffiguity
ftetchins upP
mg pipits In 21 Wool.
(including control row)
;tulles Providing Oats
*Mord et el. (MG).
Christensen (INS). end
Friend 4 lesmeror (1101
Laceshamernstleura
minutes. S defo PRP .11.9
)nstructionsi Method
Version of the ititereiltivo
radio sedsi used open-ended
pupii sempetemies. Wise
peretetpottat, Immeiets
reinfensmeent, distributed
pretties; form:tile animation
Inaktf.-IsralLeal
teachers evereSo *INWOO 32. with Ie Mr,
of Mut:Definer/
veers oeperiense
Plater Nerweement togs
Mum LUSE
SOSO teachers/besemeeters indicated
rod% me gem We.
ful; 6411 teachers
theught enroll Inc
lish llhlity of radio
pils better then
ocher pupils
Intor.yesr teacher turnover
occurred in 5 large minority
of classes
ifed32
teciorecenesIc factors related to
performance were net smoured
the enly variable found to effect Pail
perforeenee wee *district.. The
underlyins variable within district
sneered to be *veers of teechhie
imporIenre
estivate of future eremel pee pupil recurrent costs projected for
loplomentatIon in ell priory schools In Cams provided bo friend 4
Comerse. INS. Mame radios lest 5 fors. 3 sets of Notting.. Pr
Coot redie per year um of esisting radio Onennels. Cost per pupil
(including transmission costs. tescher trainins and menual,
batteries, radio and seintenence. and Me Prier= mt./ 2.34
Eau= 28
Intervention ftuse gdarisd
gal, !kills Pilot Project Radio In 1963, 50 schools in S provinces
(MatAgwetics Component)
igssiCialnia0
Thailand
Rationale for Using Mediij
oargve quality of Instruction,
especially where towhees are
endertreined
-Reduce regional inequities in
edUcetienel opportunities
lacathailitig0
Scheet children throe/Sh-
out t4e country
'sag:Lisa:
Teachers empected te observe ad
help students during radio lessons
and in s few cases emy spend time
teaching another Class. Teacher
lead post-broadcast activities
Other Materials Used
Teecher1s guide; student's
notebooks; student workkooks
in grade I
Elfectivenesg
Radio students outperformed
control students in sll regions
tor terve years. In a 1961
caparison of achievement of
mod graders, differences in
ecielevement in the Ifortheest hod
an effect size of .36 and .24 in
the Control *loin (Sengkeit wee).
This compered with .36 fee a
smiler comparison in Nicaragua
gAternal Efficiency
No informaion
Although it opposes thst
originml seml of expending to
all 10,000 schools hos net yet
been achieved, project is
ongoing
Tears Nine ileviewed
1961 - 1963
1 - 3
Thei lenpaga
NatgagetiCS
terwmase of instruction
Thai
Teacher Preparation
Twe-dgy workshop
Woe Ocerational Issued
No information
latanolAWsignei
Sale evidence that
radio helped lareve
discipline and conctu-
ality an pert of
towhees and students
Mil
Itudies Providing Oats
Oetgo (1961); San-jen (1962);
Friend, Gelds end Searle (1979)
IMALEclarocr-aLlisam
Daily 23 minute breedgest and IS
minute post-breedeset activity
Jnetructional method
The NiCaraguan Radio math lessons
were adapted for Thailand
Inez Alaimo
se information
Major management Issued
Ile information
AcCeatagiiity
Ouestienneires
indicated high
degree of en-
thusiasm on pert
of fesChers and
headmasters
In a 195? comparison between second
grade children in the more odvanta:IA
Central Plain region end tne
advantaged *atheist region, achieve-
ment levels rose in both regions and
achievement disperities between two
regions was nearly overcome
Estimates of future annual per pupil recurrent costs for implementation
in 10,693 schools (1,041,733 ;Wits) grOvidSN by 9ev,(00414mt
aumusisisaimi (Spring 1963). Naaullse ratio, Lost 5 years,
342 of schools hove electricity and batteries lost for 100 hours.
Cost per spit (includine nen-reusable workbooks and teacher
Nonmais and excluding teacher orientation) 6.44.
9
Laticemsji.
Radivessisted Community limit
&Nation te00440)
LE1101130111.213
Inicsi limpublic Province
si Sermhons in the Southwest
Mogion
To pnovide priory edication
to children in rut* areas
whore there ere nel schools
IscuLtstialice
Children 7 - 14 whe do not
Moil scam to schools
Ingstigt
Dees Ofiesprofessionsl moni-
tors to take cherge of make-
shift radio °schools."
All instruction is provided
by the radio
Othp Materials Used
Worksheets
faistinam
!valuation activities ere on-
going ond being taken OYOr by
host tawdry. Evidence so
far sunsets that when cowprod to conventionsl schools
in the same Mien, the RAPIXI
instruction in meth is swirl-
or end in roodint newly es
pod
Toe early to evoluota. First
cohort will finish grade 4 in
November, 1166. Plans hove
been mode to allow students to
tObe Ithellengse mime which
would permit placement in
conventional schools
latairsaialframitlaidetal
Geed possibility that the
project will be opPOd* to Include tIto Northwest
Region and adapted for othor
countries
ftS1.1.10;maim
Radio Sy 1906, 61 centers with 1117 children
1111:1-11I0Lie.10116
19e3 1964
folualtisla1 to 4 dithmestits and
topple with swoons of
spiel science, notural
science, civic education, end
phyaical education upon into
the loosens.
"amuses of instr!ction
Spud sh
Teacher Preparation
An orientation course
listing a few days
major Ostracises( Issues
Maps have boon encauntorod
in deliverino and collecting
workshoots. lsoistical dif-
ficulties ift providing reou.
tor supervision of rsdio
schools
jnternel Efficiency
Absenteeism of 636 in some
plops. Neese.% evIdence
that children mey be loom-
ing st least es well es
students in regular schools
but st half the price is
highly indicative of
internal efficancy
MilAnnual test ow pupil
Nos Opt half the cost of
a year of conventionsl
primary school
inetruction
S2121611
39
;tidies Providthajtots
walker (1166), Monsoon, Koslow and Olson
(1966), Sanpufnetty (1141), Prim!! e Koslow
(1916), Friend (1M), Nelwig I Friend (19e)
New. (personal communisation)
linthilassurudinum
Ono hour per dey
petructionai Ptthod
Derivation of the interactive radio instruc-
tien model. Math lessons were adopted frao
the Nicaraguan Radio Math Pro:set
alter_ jirazzal *cc,otab,Ltiy
monitors are expected
to be literate with 6 -
9 years of schooling;
most hed a inimum of
3rd grads
major Panimement Issue!'
Coordination of the functions in the Santo
Domineo office (technical, curriculum
design, preporation of materials) and those
of the Sarshans office (supervision of the
comounities, training of the monitors) has
posed problems
gauf ty
(a) Project has provids) education and
promoted social intopration of childron
of Neitian immigrants. (b) Evidence that
sose students lOaylt regular Schools for
RAD1CO schools suggests success in not
becoming on inequitable inferior track
of rural education
20
reported in the literature: the Paraguay Rural Radio EducationProject and the Shuar Radio Schools of Ecuador. (Information onthese initiatives was not complete enough to allow us to preparedetailed summaries.)
The Shuar Radio Schools aimed at providing more relevant aruaculturally sensitive education to the school children of theShuar Indians of Ecuador. (The information on these snhoola is
drawn from the Clearinghouse on Development Communication's[1982] Project Profiles, p.39-40). The few Shuars who had beentrained as teachers were used as radio teachers, and other Shuarswere used as paraprofessionals. Enrollments in the Shuar RadioSchools gl.-ew from 486 (1972-73) to 3100 (1978). The RadioSchools demonstrated high pass rates, which declined somewhatwith expansion of the -system from 97.4% (1972-73) to 84.5% (1976-
77). The Radio Schools also apparently reduced the drop-out ratefrom 30% in the conventional schools that were replaced to a
level described as "minimal."
As of 1986, according to the Ministry of Education inEcuador, the Shuar Radio Schools continue to operate.
The Rural Radio Education Project in Paraguay was anunsuccessful attempt in the 19700 to app:y radio to the task ofdelivering primary education (to adults as well aa children) in
rural areas. According to the Clearinghouse on DevelopmentCommunication (Project Profiles, 1982), the project, like theShuar Radio Schools, took a bilingual approach and attempted todevelop a route for rural children and adults to obtain a
complete primary education. The project was terminated when AIDfunding ended.
The major breakthrough in the use of radio for primaryeducation also took place during the 1970s. This was the RadioMathematics Project in Nicaragua, sponsored by USAID andundertaken by Stanford University's Institute for MathematicalStudies in the Social Sciences. This effort laid the groundworkfor later applications of radio that would come to be called"interactive radio."
The model for radio instruction developed in Nicaragua (thisand the other interactive radio cases are outlined on thefollowing pages) applied empirically-validated techniques for
developing and testing radio lessons. Radio was used for directinstruction and not curriculum enrichment. What was distinctiveabout the approach in Nicaragua was the conversational formfollowed by the radio teacher, who would elicit several oralresponses (often 100 per half hour lesson) from students in theradio classroom. Although the instruction provided by the radiowas essentially one-way, the frequent student responses andfeedback provided in anticipation of these responses by the radioteacher is arguably a form of interaction. Paisley and Chen
3 1
21
(1983, p.9) define interactivity as the ratio of user activity tosystem activity. By this criterion, the interactive model doesappear to distinguish itself from other applications of radio toprimary education in developing countries.
The Nicaraguan effort and the subsequent Radio Language ArtsProject in Kenya, the Thai.Basic Skills Pilot Project, and theRadio Language and Radio-Assisted Community Basic Education(RADECO) Project in the Dominican Republic constitute aconcentrated effort to explore the effects of a singleinstructional medium foJused on the early years of primary schoolacross several subjects and countries. The cases, which aresummarized on the preceding pages, have been rich in informationand provide the best empirical evidence to date of radio'seffectiveness as an instructional tool in primary education indeveloping countries. (The Nicaragua experience is described inSuppes, Searle, and Friend, 1978; Friend, Searle, and Suppes,1980; Friend, 1985; the Kenya experience is described in Oxfordet al., 1986; Christensen, 1985; Friend and Kemmerer, 1985; theThailand experience is described in Galda, 1985; Friend, Galda,and Searle, 1979; and Sang-jan, 1982; the RADECO experience isdescribed in Hanssen, Kozlow, and Olsen, 1983; Helwig and Friend,1985; Friend, 1984; Sanguinetty, 1985; Walker, 1986; theinteractive model is also discussed in Block, 1983; Searle,1983.)
Radio lessons for primary school science following theinteractive radio model are to be developed and tested in PapuaNew Guinea during the next few years. There will also bereplications or adaptations of the interactive radio model inother countries during the next few years under AID's RadioLearning Project.
If the interacti-e radio experience is well-documented, thesame cannot be said for educational radio as a whole indeveloping countries. Hawkridge and Robinson (1982) mentionthat, "Almost every country claims to use educational radio oreducational television or both." UNESCO (1984), in reportingstatistical data collected from member states, indicates thenumber of hours of radio broadcasting "devoted to a specificcurriculum." However, it is not reportei how much of this is foreducation at different levels. So, for the majority ofdeveloping countries, we really do not know how radio is beingused in primary ducation and with what consequences.
Summary of ResultsIn this section, we summarize the outcomes of the radio case
studies under the headings of instructional management, content,instructional materials, operational outcomes, equity,effectiveness, internal efficiency, external efficiency,sustainability, and costs.
32
22
Ingtructional Manaqemant. Our interest here is in hovradio, when used as an alterrative to lessons usually provided bythe teacher, fits into the classroom. Numerous issues could beaddressed in this connection, but we limit our attention toquestions of teacher acceptance of radio use and what new skillsare necessary for teachers to use radio.
From our case summaries, it appears that teachers generallyaccepted the use of radio in the classroom. /n Mexico, almosthalf the Radioprimaria teadhers expressed the belief that usingradio ac an aid to instruction was superior to direct teachingalone. Teachers' acceptance of radio was also positive inNicaragua, Kenya, and Thailand.
The use of radio vilquired teachers to play a newinstructional role. In all the cases we studied, responsibilityfor direct instruction was taken from the classroom teacher andassigned to the radio teacher. The classroom teacher wasexpected to prepare children for broadcasts, distributeaccompanying materials, and provide help during the lessons. Inmany cases, there were postbroadcast activities for the teacherto conduct.
Where we have information, ve find that teachers wereprepared to undertake these tasks in anywhere from three hours(Nicaragua) to three days (Thailand and the Dominican Republic).We found that this preparation worked even thouvh the educationalbackground of teachers usually was weak (10 to 11 years ofschooling in Kenya and Nicaragua or a very weak 3 to 9 years inthe Dominican Republic).
A final question related to instructional management: Didthe use of radio in one subject lead to improved teaching inother subjects? This has not received a good deal of explicitattention in the literature. It is often unclear what isexpected te transfer- -possibly a modeling of more active methodsand the use of reinforcement.
We do find some evidence of teachers expressing the viewthat the use of radio has helped them become better teachers.Perhaps a more important question is whether it is reallydesirable for a teacher to model classroom instruction on radioinstruction. Friend (1980, p.xviii) puts it this way:
Do teachers learn new pedagogical techniques fromradio? (We have some anecdotal and observationalevidence that they do, but the techniques they learnare sometimes imitations of techniques that are well-suited to radio but not ideal for human teachers.)
Content. Radio has been used to teach almost all school
33
23
subjects offered at the primary level. From the different caseswe looked at, which permit a limited number of comparisons, wemight tentatively conclude that radio seems to be most successfulin teaching mathematics. The evidence from Kenya stronglysuggests, perhaps not surprisingly, that radio is a good meansfor teaching language-learning and listening skills. So far,teaching reading by radio seems more difficult than teachingmathematics. No conclusions can be drawn about subjects likescience and social science without tests similar to the ones inNicaragua and Kenya.
With respect to using radio at different grale levels, mostof the information we found caue from applications at the earlyprimary level. We do not know whether radio, especially theinteractive model, would be equally effective after the fourthgrade.
IllesiOactignal_lis. A question of great practicalsignificance but one that is not easily answered is whether radiolessons can effectively substitute for more expensive printedmaterials like textbooks. Rarely do circumstances permit validcomparisons to be made between the use of radio and textbooks.The one comparison we did find was in Nicaragua (Jamison, Searle,Gelder and Meyneman, 1981). Three groups (in different schools)of first graders were compared: one that received radio lessons,one that received textbooks, and one that served as a controlgroup. Although both the radio and textbook interventions showedpositive effects on achievement and helped close rural-urbandisparities in achievement, radio had a stronger impact. Thetextbook group scored about .33 of a standard deviation above thecontrol group; the radio group scored 1.5 standard deviationsabove the control group. We wonder why more comparisons of thiskind have not been attempted.
It should be noted that the use of radio is not necessarilya substitute for using printed materials. We found no instanceswhere radir was used without supporting printed materials. InNicaragua, however, substantial cost reductions were obtained,after grade one, by doing without student worksheets (exerciseswere copied into notebooks).
Operational Outcomes. Many operational problems associatedwith using radio were noted. We found that obtaining batteriesand keeping Tadios in good repair posed problems in most of thecases we examined. Reception of clear signals was difficult inmany schools in Kenya, Mexico, and in the early radio project inThailaad. MUltiple grade classrooms posed problems in Nicaragua.
Clearly, sullh difficulties can lead to a substantial loss ofinstruction during the school year. What cannot be answered iswhat level of instructional loss schools are likely to tolerate.We can only speculate that for radio instruction to be viewed as
,34
24
an acceptable and reliable deliverer of instruction, theinstructional loss coming about because of poor physicalfacilities and other operating conditions must be considerablyless than what is routinely lost through such things as teacherabsenteeism. A teacher missing school for tweuty days a yearmight well be regarded as normal. Conversely, missing ten radiolessons per year may be regarded as making radxo unreliable. TheRadio Math Project in Nicregua, however, showed that it waspossible to address the issue of missed lessons (because of notreceiving broadcasts or absenteeism) by using redundancythroughout the sequence of radio lessons.
Zauity. In all the cases we examined, radio was used(althouah not exclusively) to reach children in rural areas whereeducational opportunities were jrlged to be inferior to those inurban areas. In two cases, the Auer and Tarahumara RadioScnools, radio was used to reach disadvantaged sub-populations.The Shuar schools appear to have succeeded in making educationmore relevant to the community. The Tarahumara schools failed toreduce the gap in achilvement between Indian and non-Indianstudents. In the Dominican Republic, RADECO reached ruralstudents who for economic and geographical reasons were unable toattend conventional schools.
In Thailand and Nicaragua, we found, as might be expected,that urban radio pupils outperformed their rural counterparts.In Thailand, existing disparities in achievement levels inmathematics were narrowed as a result of radio instruction.
The available evidence (Kenya and Nicaragua) suggests thatboys and girls learn equally well from radio.
rails we see in some cases that the use of radio in primaryeducation contributed to improved equity and that in other casesit did not, we found little information about radio'srelationship to specific aspects of family or community life.
Ufectivenese. There appears little doubt that radio, whenused properly, can be an effective means for improving studentachievement in a variety of school subjects.
Thailand's experience in the 1960s demonstrated that radiocould be used effectively to enrich the school curriculum. InQomparison with control students, radio students showed positivedifferences for achievement in music, mixed results for socialstudies, and no real differences in English.
In the two Mexican studies, we also found positive learningeffects, although it was not entirely certain that these could be
attributable to the radio lessons.
For the interactive .:4clio interventio s, the impact of radio
35
on student achievement is impressive. From effect sizes reportedin our project summary sheets for the grade/subject/skillstested, we compare average effect sizes from each of theseinterventions in the table below.
Intervention Average Effect Size Raises Mean From...
Nicaragua .52 50th to 70th percentile
Kenya .45 50th to 67th percentile
ThailandUrban
Rural
.24 50th to 59th percentile
.58 50th to 72nd percentile
We find a large avlrage eftect size of .52 across fourgrades for mathematics instruction in Nicaragua. If we did notinclude the results from grade 4, when the school year coincided'4ith the onset of the revolution, this would rise to .69. ForAmway we also find a high average effect size across threegrades in listening, reading, speaking, and writing skills. InThailand, effect sizes range from a modest .24 for mathematicsinstruction among second graders in the area of Bangkok to .58 inthe rural Northeast. (It should be kept in mind that differencesin effect size refer to the size of gains and not to absolutelevels of achievement.) We workre not able to calculate effectsizes for RADECO in the Dominican Republic. The fact that thereare no differences between the hour long daily radio interventionin an out-of-school setting and a conventional day of schoolinstruction is indicative of success.
We found little or no information about radio'seffectiveness on outcomes other than achievement.
Internal Efficiency. We are interested in whether radiocontributes to internal efficiency (apart from studentachievement) in the delivery of primary education in developingcountries. Our judgments regarding internal efficiency are basedupon changes in indicators such as school dropout rates,repetition rates, and the number of children being reached by theeducational system or the quality of education being deliveredfor a given amount of resources.
There is evidence that radio contributes to internalefficiency. Thz use of radio in Kenya made it possible to.
36
26
deliver three times as much content in a half hour block of timeas was delivered through conventional instruction (Oxford et al.,1986). Teacher and student attendance appeared to be improved inthe Mexican Radioprimaria classes (Spain 1977). Teacher andstudent punctuality were positively affected by radio in Thailand(Galda, 1985).
Except for the positive experience of the Shuar RadioSchools in Ecuador (Clearinghouse on Development Communications,1982), we were unable to find any evidence that radio reduceddropout rates.
The Mexican Radioprimaria's attempt to achiev, efficiency byusing radio to "complete" a sdhool by offering six grades ofinstruction with four teachers was unsnccessful. Most schoolsthat ended up using the radio lessons already were completeschools (Spain, 1977).
We found no evidence that radio reduced repetition rates.This was especially disappointing in the case of Nicaragua, whereas Jamison and McAnany (1978, p.109) observed:
...it is perhaps well worth noting that, even thoughthe cost per student is relatively high, adoption ofthe RMP (Radio Mathematics Project) would almost surelyreduce costs per graduate because of the effect it hason re4ucing repetition rates.
Unfortunately, this hoped-for improvement was not realized.It was learned in Nicaragua that repetition is only partlyrelated to student performance (George, 1980; Galda and Gonzalez,1980; Jamison, 1980).
Perhaps the best example of radio's potential for increasinginternal efficiency in primary education is the CN1M If RADECO inthe Dominican Republic. The evidence currently availablesuggests students perform at least as well on math and readingtests as students do in conventional schools in the region, butwith about half the time spent in learnina at about half the cost(Friend and Koslow, 1985; Sanguinetty, 1985). Although wo arenot comparing full or identical sets of educational outcomes--indeed, Friend and Koslow (1985) note that no comparisons weremade in science and social science and that conventional sdhoolsaim at a wider range of outcomes than RADECO schoolsthe MECOresults aro nonetheless impressive.
Zmternal Efficiency. We found almost no information thatallows us to evaluate radio's contribution to the externalefficiency of primary education in developing countries. This isnot surprising, since, in most countries, little empiricalattention is given to the relationship of primary education tothe employment or the continuation in education of primary school
37
tc-
27
dropouts. Few of the radio interventions that we examined lastedlong enough or had the necessary resources to look at the longerterm success in school or in employment of students who had beestaught by radio. We mentioned earlier that it would be very hardto presume that exposure to a half hour broadcast in what mighthave been one sdhool subject during the course of a few years islikely to have a discernible effect later in life. However, theRADECO experience in the Dominican Republic, where radio carriesvirtually the entire instructional load, may offer a goodopportunity for examining the external efficiency of radioinstruction.
Sustainabilitv. We found that among the cases we examined,there was mixed success for radio instruction moving from anexperimental or demonstration phase to a sustained or expandedrole in national educational policies. One of the most promisingdemonstrations of instructional radio, the Nicaraguan RadioMathematics Project, was abandoned following the overthrow ofSomoza. The Radio Language Arts Project in Kenya wasdiscontinued after a successful pilot phase. Radio efforts inMexico were abandoned as was one in Paraguay (although apparentlyfor reasons unrelated to radio). On the other hand, theDominican Republic is apparently moving forward to extend RADECO.The Shuar Radio Schools, as best we can tell, are continuing toflourish. Thailand, supplemental radio has enjoyed sustainedapplication Zor more than two decades, and the use of directinstruction radio lessons adapted from Nicaragua may one day beused on a large scale.
Costs. What does it cost developing countries to use radioto improve the quality of instruction in one or more subjects?We believe that the evidence shows that it costs either a littleor a lot. Whether radio is expensive or cheap depends to a largeextent on whether one assigns fixed capital costs for developingradio lessons to the annual por student costs, to the number ofstudents served by radio instruction, to other assumptions madein the calculation, or to how incremental, annual per studentcosts relate to the country's annual overall expenditures forprimary education.
The costs of radio appear to be most attractive for adeveloping country when a successful pilot program has beencompleted and the costs of instructional development have beenpaid by a donor. No new lessons need to be developed, and thereare no new transmission facilities required to broadcast acrossthe country. There are large numbers of students to be served
and the fixed capital costs for instructional development (paidby the donor) will not be added to the calculation of averageannual per student cost in order to arrive at an estimation of"true c:sts" to society. Subject to the assumptions mentioned inour project summary sheets, annual ar student costs were foundto be $.34 and $.44 for Kenya (Fr.Lend and Kemmerer, 1985) and
38
28
Thailand, respectively (Development Communications Report, 1985).Again, these are based on a projection of large numbers ofstudents to be served (about one million).
The annual per student costs will not be so low when thereare fewer students, when the capital costs of instructionaldevelopment are added, or when new broadcasting capacity must be
developed to add new grades or reach new regions. Jamison andMcAnany (1978) estimate the costs of the Nicaraguan Radio MathProject, which they regard as expensive, to be $3.05 per student
per year. This figure was based on 120,000 students being servedand includes a per-student aliocation of fixed capital costs forinstructional development. This is what instructional radiocould cost a country unable to secure financial assistance from a
donor to cover capital costs. When per-student allocations offixed costs are not included, the actual costs to the countryapproach the marginal cost (the cost of adding one more studen*),
which, for Nicaragua, was $1.64 per student (Jamison and McAnany,
1978).
The costs of the other cases we examined varied widely.Thailand during the 1960s, with 800,000 students being servad,showed annual per student costs of $.44 (Jamison and McAnany,
1978). In the Mexican cases, where radio was being used forsmall numbers of students (1,000 to 1,800), the annual perstudent costs for the Tarahumara Radio Schools were $52.86, andfor Radioprimaria, $16.44 (Jamison and McAnany, 1978).
Although annual per student costs are useful for purposes ofanalysis and comparison they must also be viewad in the aggregatesense relative to yearly educational budgets. For example, acost of $.44 per student might appear to be very low. But whenthis is multiplied by a million students, officials in ministriesof education and finance must face adding $440,000 to the annualrecurrent educational budget for improvement in one subject.This may not appear attractive if the alternative is,
say, making a fixed investment in improving textbooks and passingon recurrent costs to parents.
22ncluilanaThe evidence suggests that radio can and has played a
positive role in the production of learning in primary schools in
developing countries. We reach this conclusion on the basis ofeight cases from seven countries, in which one model of radioinstruction is overrepresented.
We find that radio has not been fully exploited as aninstrument of mass communications in the delivery of primaryinstruction on a national scale, except in the case of Thailand.In most of the cases reviewed, radio has been used with audiencesof about 1,000 - 4,000 students. Similarly, we find that radio
(1 9
29
has been used mostly to improve the quality of instruction in oneor more subjects and not, except in the Dominican Republic, as ameans to extend educational opportunities to those not srvd byschoola. The evidence suggests a generally high degree ofteacher acceptance and that teachers can learn to work with radioinstruction with relatively little training. W found that radiolessons wore almost always supported by printed materials.Receiving broadcast signals, keeping radios in repair andsupplied with batteries, and getting suCh things av printedsupporting materials to schools on time represented considerableoperational gnd management challenges.
Because of lack of evidence, we were unable to conclude thatradio has had an effect on the external efficiency of primaryeducation in developing countries. The evidence that isavailable uggests that radio, apart from its effects in raisingstudent achievement, has had mixed results in affecting otheraspects of the internal efficiency of education.
We learned that radio instruction in a school subject can bedelivered to children in schools for about the cost of atextbook, between about $.40 and $3 per student per year,depending on assumptions and conditions.
At least for the interactive radio cases, the evidencesuggests that radio can have a strong impact on studentachievement. In view of this evidence, we must ask why morecountries have not adopted th interactive model. Questions thatmust be asked should include the following:
o Has the case for textbook availability beenmade so persuasively as to overwhelm thelikelihood of consideration of otherinstructional possibilities?
o Does a general resistance to the use ofinstructional hardware in education indeveloping countries preclude considerationof radio?
o Are the transmission and reception of radiosignals regarded as simply too unreliable forusing radio to deliver instruction?
o Are the recurrent costs of using radio todeliver primary instruction perceived as toohigh?
o Is the effort to reorganize instruction andmanage a radio delivery system regarded astoo burdensome?
4 9
3 0
The use of radio has been endorsed widel at all levels
of the international community as offering an - opriate,
effective, and affordable means for improving ,ducation in
dtmeloping countries. To a largo extent, its potential has been
demonstrated. Yet it appears that e :iloping countries have not
made as much use of radio in primary e_ucation as one might
expect. It is important that we learn why.
Radio as a Policy Option. We believe that the literature
just reviewed points to several key questions that educational
policy makers in developing countries should ask when considering
the possible use of radio to extend or imprcov primer- education.
They are the following:
o Radio offers both the advantages anddisadvantages of what might be described as a
"centralized" curriculum. Broadcasts comefrom a fixed point at a fixed time; thecontent, language, and "culture" of th:,
lessons are not subject to much variation orcontrol at the local level. Therefore, thequestion to be asked is : Do nationalconditions permit using a centralizedcurriculum such as could be offered by radio?
o Are there adequate facilities and resourcesto ensure satisfactory transmission andreception of broadcasts in the areas to be
served? Would this be true beyond the stageof what might be a preliminary pilot project?
Does local broadcasting capacity permitadding new grades and subjects?
o Is thane likely to be msistance from thestakeholders in the "curriculum status quo"--textbook writers and publishers, localcurriculum specialists, teacher trainers, orexpatriate educational advisers? How mightthis resistance be overcome?
o While there is evidence that high qualityradio lessons are often more effective thanconventional instruction, mediocre radiolessons might be less effective than mediocreconventional instruction. Therefore, thequestion to be asked is: Is there sufficienthigh calibre staff to develop, produce, anddeliver high quality radio lessons? Wouldthere be a willingness, as there was inThailand, to offset some of the fixed capitalfor developing high quality radio lessons by
41
a
31
adapting lessons that have proved to beuccessful in other countries?
o What will be the "add on" costs to tta annualbudgot and how will they be financed?
2. TelevisionWhen developing countries have turned to instructional
hardware to lead the way in major efforts to expand or improveeducation, they have turned to television. Although radio hasbeen recognized as the most universal means of mass communicationavailable to developing countries, we observed in the precedingsection that most nations have o far failed to fully exploitradio's potential for reaching large numbers of children. Thisis not so with television. Where experiments with radiotypically involved 1,000 to 4,000 students, experiments withtelevision have reached school audiences of up to a half million.
Tho television experiments of the sixties and seventies wereclearly ambitious, and often linked to a zomplete reform of partor all of an educational system. The appeal of television andthe great expectations that were placed upon it reflected more ofan appreciation of television's growing economic and social powerin industrialized countries than in any demonstratedinstructional capabilities. Indeed, Schramm's (1962) comparisonsof instructional television with conventional teaching shouldhardly have aroused great expectations. Of 421 comparisons, C3favored television, 50 favored conventional teaching, and 308showed no differences. The proposition that television can be asgood as conventional teaching would be compelling if the strategycalled for replacing teachers. But in most of the experiments ofthe sixties and seventies, television was used for only part ofthe instructional day, to aid teachers rather than replace them.Before making an investment in a complex and costly undertakinglike television, one would want evidence to suggest thattelevision could be a good deal more effective than conventionalteaching. W. look at the use of instructional television inseven countries--American Samoa, El Salvador, Ivory Coast, Niger,Colombia, India, and Nigeria.
haszicam_11112AOne of the most ambitious attempts to make use of television
to improve school instruction took place during the sixties andseventies in American Samoa. (Unless otherwise indicat,d,information in this section is drawn from Schramm et al., 1981.)Television was chosen as an effectivn and economical meails toforge a complete modernization of primary and secondary 4chooleducation. The hope was that through a transformation of theeducational system with instructional television as the prtncipaldelivery vehicle, the academic performance of Samoan students
4 2
32
could be brought to levels equal to those of students in goodschools on the U.S. mainland.
The experiment in American Samoa spanned more than fifteen
years. It included a reworking of the curriculum. Lessons weredeveloped to cover about 35% of all classroom time. Lessons,
which were produced in English, lasted about 8 minutes in theearly grades and 25 minutes in the upper grades. Televisionlessons were accompanied by a packet of curriculum materials,including lesson plans, instructions for the teachers, and
supplemental readings.
Advisers from the U.S. took charge of almost all aspects of
the educational program. It was recognized late in theexperiment that the failure to involve Samoan teachers in theplanning and development of instruction and the great haste inmoving to cover the entire primary-secondary school curriculum
were serious errors.
It was planned that the experiment would not be evaluateduntil the first cohort had received a complete twelve years ofinstruction under the new system. Baseline information was not
collected. About six years into the project, standardized tests
were administered to students in various subjects and at variousgrade levels. At best, the achievement tests yielded a mixed andinconsistent picture of the effects of student learning on
achievement.
Emery (1985, p.498) takes exception with what often has beena generally positive portrayal of the results obtained inAmerican Samoa. She amserts that:
Results showed that after seven years of expensiveconcentrated ETV under the most optimal conditions,grades 3-12 achieved 59 per cent of United Statesnorms. This was a drop from the 60 per cent they hadbeen achieving in 1935 when there was only the most'primitive' oral society teaching by untrained native
teachers...
The data scattered through the 'bold experiment' leaveslittle doubt about two matters (M. Emery, 1984).First, for every subject for which any reasonablebaseline data could be established, ETV failed and it
failed in direct correlation with grade, or increasingabstraction and conceptualization of knowledge.Second, the senior author (Schramm)...appears to havedone everything possible to put the results in the most
favorable light. As an example, we have from hisTables 21 and 23 (pages 115 and 116) the fact thatthere was a 9 per cent decline in oral English scoresin Grade 7 after three years with ETV. After
4 3
33
discussion of the possible reasons for this result,none of which touched upon the failure of televlsion asa teacher, the team concluded that (p.127) 'there isevery reason, at least, to think that the standard ofEnglish usage improved notably during those years'(i.e. 1964-70). Than, when we examine Table A.II wefind in their own figures for the teaching of Englishthat face-to-face teaching is from 16-19 times moreeffective than television.
The costs of instructional television in American Samoa werehigh. In 1972, with school enrollments of 8,100, the per studentcost (including.a share of annualized capital costs) was about$166 per year.
A survey conducted in 1972 found a high degree of acceptanceof instructional television among elementary school teachers andadministrators, especially for teaching oral English. At thehigh school level, the use of television was strongly criticizedby teachers and etudents.
In 1970, political winds changed. The incoming Governor toAmerican Samoa described the experiment with instructionaltelevision as E.1 "utter and complete failure" (quoted in Schrammet al., 1981, p.80). Opposition to television led to an eventualreduction in its use as teachers were given the authority todecide when and where to use it. In 1974-75, power shortageswere encountered throughout the year, and television was notreceived during the final month of school. By this time, onlyone channel was being used instead of six (Masland and Masland,1976). In 1975, the use of television was eliminated entirely atthe secondary level.
El SalvadorThe use of instructional television was also the centerpiece
of a major education reform in El Salvador during the sixties andseventies. (Unless otherwise specified, information in thissection is drawn from Mayo et al., 1976.) El Salvador embarkedupon a reform of instruction for grades 7-9 in order to widenaccess to secondary education. Television was used to provideinstruction in mathematics, English, Spanish, social sciences,and natural sciences. The use of television was embedded in alarger curriculum reform which included retraining of teachers,revision of instructional materials, and the building of newfacilities.
The reform was very successful in widening the path tosecondary school. In 1968, there were 19,104 students enrolledin grades 7-9; this increased to 65,390 in 1973. The governmentplaced hopes on increasing the number of persons who would leaveschool with a ninth grade education to take training for middlelevel technical jobs.
44
34
The costs of instructional television were considerablylower in El Salvador than in American Samoa. The per studentcost, based upon enrollments of 48,000 students and assuming 170hours of programs, were calculated to be $24.35 per year in 1972.Carnoy (1976) estimated costs to be about $26 per studeht. He
points out that this would come to about $780 per classroom of 30students. This was almo6t 60% of the annual salary of a juniorhigh school teacher in 1970 ($1,400).
The results obtained in El Salvador were mixed. Mayo et al.
compared the performance of three cohorts of students (those whostarted seventh grade in 1969, 1970, and 191). In the final two
years, comparisons were made between students receivinginstructional television and those not receiving instructionaltelevision but benefiting from other aspects of the educational
reform. The ITV students fared better than non-ITV students in
the seventh grade. In eighth and ninth grades, Mayo et rl.
conclude that the "advantages of ITV were not apparent."
The ITV students outperformed their non-televisioncounterparts on tests of general ability. On achievement tests,between seventh and ninth grades, ITV students showed bettergains in mathematics; gains in reading were about equal; results
in social sciences were mixed; and in science, non-ITV studentsshowed better gains than students learning science through
television.
It was discovered that boys gained more than girls, and ingeneral, television did not reduce the effects of studentbackground variables (father's education, family wealth, andurbanization) on achievement. Over 85% of students were found to
be continuing their education after the ninth grade, but half ofthem were enrolled in courses leading to university studies and
not to preparation for middle-level technical jobs, as had been
hoped for by government planners. University enrollments were
not widened. Mayo et al. concluded at the time that students'
"high educational aspirations are not realistic in terms of thetangible opportunities available to them in the near future."
By the end of 1972, some disenchantment with television wasgrowing on the part of teachers. Mayo et al. note that themajority still believed that educational television could be an
important teaching tool. But in later years, opposition to
television grew, and the cost of television became an issue whenteachers went on strike for higher wages.
Ivory CoastDuring the seventies, the Ivory Coast also undertook an
ambitious reform of primary education with television playing a
leading role. (Unless otherwise indicated, information from this
section is drawn from Kaye, 1976.) The strategy in the Ivory
45
35
Coast was to make use of television to increase the quality ofinstruction, which would allow for introduction of a policy of
. automatic promotion and great reductions in dropouts. Thus, evenwith the add-on costs of television and improved teachertraining, cycle costs per graduate would be reduced.
The project began in 1971 with 21,000 first graders. Theplan called for adding one grade per year. Over a fifteen yearperiod, enrollments in primary schools were to be tripled, and by1986, the entire 6-12 year age group would be enrolled in ITVschools. A thorough revision of the curriculum and upgrading ofteacher qualifications was planned to accompany the use oftelevision. There was also an ITV component for out-of-schooladult groups.
By 1979-80, 15,635 classes were receiving instructionaltelevision, representing about 84% of all students (Hawkridge andRobinson, 1982) or in excess of a half million students. -.essonswere broadcast covering French, reading, writing, mathematics,hygiene, moral education, civics, environmental studies, andphysical education.
Carnoy (1976) estimates a cost of $13 per student per yearfor 1976, based upon enrollments of 336,000 students.
W. did not find published data with respect to studentachievement. Hawkridge and Robinson (1982, p.163-164) refer to aseries of studies undertaken by a team of researchers from theUniversity of Liege in Belgium and observe that:
Interpretations of the results of testing the studentsdiffer. The Belgians found that many students did notmaster the first grade curriculum; this was true intelevision schools and those without television. Itcaused problems in second grade in television schools,where the curriculum was based on mastery having beenachieved in the previous year. Students in televisionschools whe repeated first grade did better the secondtime but only rarely did they achieve mastery. Theteam noted that students had difficulty in particularwith graphical representations. On the other hand, itseems that television teaching was most effective inteaching spoken French, with the children o televisionschools improving their proficiency in the languagemarkedly more than those in other schools. Lesssatisfactory progress was recorded in written French,and changes are to be made in the methods of teachingthe written language. Simiilarly,*changes have alreadybeen made in the content and pedagogy of the modernmathematics programmes, which were initially radicallynew to both teachers and students...
46
3 6
Hawkridge and Robinson also report that the informationcollected by the Belgian team shows that the use of televisionhelped equalize educational opportunities in rural areas, thatthe "range of differences in standards between the best andpoorest television schools is small when measured by studentachievement on tests. Some of the rural schools are in factamong the best now." (p.164).
Kaye (1976) notes that the use of television in the Ivor,Coast brought about a reduction in rates of repetition and drop-out, and unit costs per graduate were lower than under thetraditional system. Unfortunately, according to Kaye, only aquarter of graduating students would find places in secondaryschools.
Despite the comparatively modest annual per student costs,the Ivory Coast apparently found television too expensive and nolonger uses it in schools.
NismrNiger turned to television during the sixties as a means to
help overcome an acute shortage of trained primary schoolteachers and to inject modern methods of thought, xpression, andteaching into the school setting. (Unless otherwise indicated,information in this section is drawn from the Clearinghouse onDevelopment Communication, 1982.)
Lessons were broadcast daily in French and arithmetic.Classroom monitors, who lacked the qualisications of trainedteachers, were supported by short training courses and in-serviceeducation via radio. About 700 children were reached during apilot phase; this rose to 9,000 in 1975.
The use of television appeared to have a strong impact onreducing school dropout rates. During the sixties, dropout ratesin Niger approached 4041 674 of the original 716 students whobegan Tele-Niger classes in 1966 completed the four-year cycle in1970 (this represents a dropout rate of less than 6%).Attendance was also found to be considerably better than intraditional classes.
There seems to be little readily available information onstudent achievement. The Clearinghouse on DevelormentCommunication (p.64) notes:
The problems identified in various studies relate tothe use of French in the broadcasts and to the qualityof education received by students in the televisedclassroom. In particular, TV students performed lesswell than students in the traditional classroom instandardized math and grammar tests. To help assuagethis problem, a fifth year was added to the televised
4 7
37
curriculum in 1970.
We were unable to locate cost information relevant to yearswhen coverage of Tele-Niger expanded beyond its experimentalphase. Carnoy (1976) estimated costs for 1969 based upon 800students taking part in the pilot phase. This came to $1156 peryear. Needless to say, this is not a good stimate of whattelevision would cost in later years when the system expanded.
From what we can tell, television is no longer being usedfor primary school education in Niger. Apparently, opposition tothe use of television was encountered from teachers' unions andfrom within the ministry of education itself.
=ShiaIn the sixties, Colombia put television to the task of
improving primary school instruction and providing in-serviceeducation to teachers. (Unless indicated otherwise, informationin this section is drawn from Schramm, 1967.) Television wasused to provide instruction in five grades. Each grade receivedninety minutes of instruction in different combinations ofsubjects, which included mathematics, social studies, language,natural science, and music.
In 1965, television reached about 250,000 pupils. Thisresulted in an annual per student cost of only $4 (Carnoy, 1976).
There were mixed results with respect to studentachievement. At the end of the first year's use of television,groups of television and non-television students were tested,5,000 in all. Eight comparisons were made. In three of them,the television students significantly outperformed non-televisionstudents (grade 2 language, grade 5 mathematics, and grade 4natural science). In five comparisons, there were no significantdifferences.
We found no information about whether television continuesto be used in Colombia. Lyle (1982, p.292) suggests that the useof television may have been curtailed.
The massive Colombian project reported in the 1967series was attempting to upgrade rural primary schoolsthrough direct television instruction. It reliedheavily on the use of American Peace Corps Volunteersfor utilization supervision. For this the project wasfortunate to have been the focus of the largest singlecadre of volunteers in the entire Peace Corps. Buteven at the time of the case study, the Peace Corps waswinding down.
To the extent that the project has redirected its focusfrom rural to urban schools, the move may well have
48
3 8
been a necessary and wise one, simply because of theinability to continue providing adequate utilizationsupervision in the rural areas.
IndiaIndia also has had experience with the use of television in
primary education. This came as part of the SatelliteInstructional TV Experiment (SITE) and is described in Shukla(1979). Television was used to bring a daily 22 minute programof general enrichment to rural schools in six states. Lessonswere aimed at grades one to five. The children were fromdisadvantaged homes, often thtJ first generation in tho family toattend school. The objectives of the programs were to help thechildren learn community living skills, develop awareness of themodernization of society, instill habits of hygiene, promoteaesthetic sensitivity, and improve basic concepts and skills inmathzmatics, language, and technology. There was also a seriesof science programs for 9-12 year old children.
An evaluaiion of SITE showed mixed results. There were nosignificant differences between SITE and non-SITE children withrespect to achievement in school subjects or in schoolattendance. It appeared that centralized development andproduction of program material failed to come to grips with thecultural diversity of the audience. The amount of informationand pacing of messagea proved difficult for the children tounderstand, especially in science. There did appear to beevidence that the language development of some children waspositively affected. Although it appeared that teachers were notgenerally unfavorable to the use of television, they were notpersuaded that television instruction was superior to face-to-face instruction.
We found ro information on the current status of SITE inprimary education in India or on costs.
There was a six year attempt (1959-65) to use instructionaltelevision in Nigeria. Different regions made use of televisionat different educational levels. In Lagos and Kaduna, televisionwas tried at the primary level. According to Schramm (1967),Nigeria experienced almost every difficulty imaginable:jurisdictional conflicts between the Ministries Education andInformation, difficulties in transmission and reception, powerfailures, uncertainties about broadcast schedules, oppositionfrom teachers, and lack of trained personnel of all kinds.Perhaps the most telling point was a discovery that three-fourthsof the television sets found in schools were inoperative.
Estimated annual costs (for Lagos) were $6 per student forabout 26 hours of instruction, based upon 17,200 students(Carnoy, 1975). We found no information about the educational
4 9
39
effectiveness of the television lessons.
ConclugionsTaken together, the diverse experience of these seven
countries with the use of instructional television for primaryeducation has been disappointing. The overall impression is thatthe use of television can be expensive and that it contributeslittle (in some cases, nothing) to student achievement.Television often arouses strong opposition on the part ofteachers and sometimes resistance on the part of students. Thereis a striking lack of success in upgrading telovisior1 use from anexperimental phase to a permanent feature of national education.
However, care must be taken when trying to draw conclusionsabout using television as a medium to deliver primary schooleducation in developing countries. It may well be the case,bearing in mind the earlier discussion of the medium versusmethod issue, that unlike radio, a satisfactory model forinstructional television programs for the primary schoolclassroom has not been found.
Television gs a Policy Option. Educational policy makers indeveloping countries who may wish to use television to deliverprimary educatton should ask the same questions asked for radioand some other questions as well. These would include thefollowing:
o Has consideration of the possibilitiesoffered by television followed an adequateconsideration of the possibilities offered byradio?
o Given the generally disappointing resultswith instructional television in otherdeveloping countries, what suggests thatthings could be different in the presentcontext?
o Would time and resources be sufficient toallow an experimental phase of activities todevelop and test a model of televisionprogramming that "works" in the primaryschool classroom in a developing country?
o Could the chances of success in usingtelevision be improved by strategies (andpossibly a combined use of related technologysuch as videocassette recorders) that wouldallow greater flexibility and teacherinvolvement in using television in theclassroom?
50
40
3. C9MORISERPeople must be properly prepared to live in a society wheremost will encounter computers at workplaces, in private life, andin any other activity; we draw the attention of decision makersto their responsibilities in meeting effectively the newchallenges to equity at national and international levels.
In recent years, more and more countries have beenmaking decisions on a national level to ensure theintroduction of computers in education on a largescale, and more decisions are in preparation in manyother countries, both developing and industrialised.Educational authorities all over the world should takeinto account existing scientific evidence and applyscientific monitoring and objective evaluation to arapidly changing domain... (Stanford SymposiumStatement, Stanford/UNESCO Symposium On Computers andEducation: Which Role for International Research, 10-14March 1986)
Discussion about the application of instructional technologyin developing countries is increasingly becoming discussion aboutcomputers. Most often the interest in introducing computers ineducation, as the Stanford Symposium Statement suggests, stemsfrom perceived social and economic imperatives rather than from adssire to use technology to improve or extend education.Although the generally declining costs of microcomputers in the19805 have made introduction of computers a possibility for moreand more countries, the question of affordability still loomslargo. Carnoy et al. (1986) point out that per student costs percontact hour for direct computer access are higher than costsmany developing countries now incur for education per student perweek. Nevertheless, many developing countries are nowconsidering ways to begin introducing computers into theireducetional systems.
The literature on the use of computers in education indeveloping countries is scant and largely addresses problemsdeveloping countries are likely to face based on the experienceof the industrialized countries. Our discussion of the possibleuse of computers at the primary school level is especiallyanticipatory.
The evidence now available suggests that the introduction ofcomputers in education in both developed and developing countrieshas taken place mostly at post-primary levels. At the primarylevel, UNESCO (1986) reports "general measures" for introductionof computers in nine countries: Australia, Canada, France,Hungary, Korea, Mexico, Spain, the United Kingdom, and the UnitedStates. "Experimental measures" aro reported for Argentina,
51
41
Belgium, Brazil, China, Colombia, Japan, Kenya, the Netherlands,New Zealand, and Senegal. However, UNESCO admits that the paceof change taking place in most countries makes it difficult tokeep an up-to-date inventory. Assessing the extensiveness ofcomputer use in developing countries is also complicated by thedifficulty of obtaining infctmation at a subnational level--where J.ndividual or groups of schools may be introducingcomputers without government assistance. These efforts often goundocumented.
Sducational UsesThe possible educational uses of computers in developing
countries can be anticipated by taking the experience ofindustrialized countries as a starting point. These uses, whichin concept or practice are not always distinct, include thefollowing:
Computer Literacy. There are numerous definitions ofcomputer literacy. The most common element is, perhaps, the goalof preparing people to live in the "information society" or to"demystify" the computer. India and Mexico are two developingcountries now embarking on national programs of computer literacy(Carroy and Loop, 1986).
The applicability of the notion of computer literacy fordeveloping countries is considered in Papagiannis et al. (1987).They contend that the idea of computer literacy must be broadenedand that its educational agenda must be linked to the widercontext of national development. They suggest three maindimensions of computer literacy:
(a) computer literacy as employment oreparation--The authorssee some value for employment-oriented technical training incomputer-related skills but urge that these programs be keptsmall in size and confined to the upper secondary schoollevel;
(b) computer literacy as informatics--The authors contendthat an understanding of the social, economic, political,and cultural dimensions of information technology will beimportant for national development and should aim at closinghe erp between the rich and the poor. They recommendprograms of adult education that address the issues ofinformation technology and its role in development andrecommend providing opportunities for greater access tomicrocomputers and other information technology. The issuesof information technology for national development mightalso be addressed in social studies courses in schools--even those where computers have not been intr!duced;
(c) computer literacy as productivity enhancement--Theauthors point out that "computer literacy as productivity
52
4 2
enhancement relates to learning to usr computers to improvenon-computer related activities," (p.78). They suggest that
the use of productivity software will and should follow aless "personal" course than it has in industrializedcountries. They urge that computer literacy as productivityenhancement should be oriented toward the goal of nationaldevelopment and might take place within demonstrationprojects in health, nutrition, and community development.The aim would be to promote greater local control and self-reliance in the development process.
Direct Instruction. Computers have been used as a means to
provide direct instruction in school subjects like mathematics,language, and scievce. Distinctions--often value laden--are made
between instructional strategies used in computer instruction:
drill and practice, tutorials, games, simulations, and problem-
solving. Friend (1985) argues that those categories are not
always distinct. She notes that the common distinction between
tutorial programs and drill and practice is not always clear,since tutorial programs often contain practice components andthat well-designed drill and practice programs allow students toabstract to general principles from the exercises given.
In 1985, about 77 percent of the time spent in computer usein elementary schools (below grade 4) in the United States wasfor drill and practice in reading, language arts, and arithmetic
(Becker, 1987b).
proaramminq. Computers have also been used in schools for
the purpose of teaching programming. This is especially true atthe secondary level. In the United States, for example, about
42* of the computer time in high schools is devoted toinstruction in programming (Becker, 1987b). The intention here
is often two-fold. Advocates of teaching programming argue that
this will (a) lead to a better or more rapid development ofhigher order cognitive skills, or (b) that programming skillswill be useful in finding employment. At the primary level,there is sometimes an effort to teach what Friend (1935) refers
to as the "beginner's programming language." These includelanguages such as BASIC, LOGO, and Smell Talk.
Lockheed and Mandinach (1986) tie a declining interest inpost-secondary computer-related careers now being witnessed inthe United States to poorly taught pre-college instruction in
programming. With respect to the external efficiency of recent
efforts to teach programming, they cite a study by Linn (1985,p.22) that points to the conclusion that:
Programming instruction at the pre-college level wasinadequate in scope, depth, and choice of language. In
most cases, students enrolled in college courses had tounlearn misconceptions and poor techniques before they
s 3
43
could profit from college instruction.
amputers as Toole'. There appears to be an emergingconsensus on the value of teaching students to use computers asproductivity tools either within existing school subjects or inspecial courses. This involves instruction in the use ofspreadsheets, word processors, outliners, and database managers.
Compared to computer-assisted direct instruction andprogramming, using computers as tools appears to be gainingrapidly in relative importance in schools (Hunter, 1987). Themerits of the "tools" approach are considered in Lockheed andMandinach (1986). They cite the enormous difficulties inintegrating computers into the curriculum. They suggest an"applications-based computer course" as ar, alternative to fullcurriculum integration and programming courses. Such a coursewould emphasize generalizable skills through instruction in useof the various applications packages mentioned above and would belinked to the regular school curriculum through collaborativeteaching.
Papagiannis et al. (1987, p.61) believe that using computersas a tool may be a fruitful approach for developing countries.They contend:
that the link between these (enduring) goals (ofeducation) and what can be practiced using the variousdata-base management programs, word and concept-processing (outlining) programs, spreadsheeting,quantitative and qualitative data-management programs,within the various subject-area specialties, areprecisely the skills we urge our teachers to preparestudents in. In other words, our traditionaleducational aims of improving problem-solvinq, problem-posing and critical understanding through the use ofproductivity tools and creative programs in thesubject-areas could become a reality in practice...
Besides the above categories of classroom use, computers mayalso be used in education f.Jr various diagnostic, monitoring, andtesting purposes (computer-managed ins6ruction) as well as forschcl administration tasks.
H-rdware Configarations and AccessThere are essentially two options that developing countries
might follow with regard to selection of computer hardware foreducation. One uses a minicomputer to route instrur.tion toindividual student terminals. These systems tend to use completecurriculum packages and offer features for diagnosis, evaluation,and monitoring of student progress. Such a system has made itpossible to run more powerful sor-ware and offers the potential
54
44
advantage of greater ease in classroom management. With aminicomputer system, software does not need to be passed aroundand generally only a single system manager, rather than a host ofteachers, must be trained. (AID's Lrarning Technologies Projectis currently evaluating the Control Data Corporation/WICATminlcomputer application in a primary school in Grenada.)
More common, however, will be a configuration usingmicrocomputers. These may be installed in different classroomstnroughout the school or placed together (and sometimes linkedtogether) in a single room. UNESCO's (1986) survey of 43countries found the trend to be toward the one-room "computerlaboratory."
Schools may begin acquiring microcomputers without making aslarge an initial investment as with a minicomputer system.Microcomputers offer more flexibility with regard to how theymight be used in classroom activities and in software selection.
Access and EquityBoth the availability of hardware in relation to the number
of students in a school and a school's curriculum decisionsaffect the amount of access a student has to the computer. Evenin the industrialized countries, there is a great deal ofvariation to be found with rempect to access. In the UnitedStates, by spring ',985, five out of six primary schools hadacquired computers for instruction; at the secondary level, themajority of U.S. high schools had fifteen or more computers forinstruction (Becker, 1987b). In the space of two years (1985-87), the computer to student ratio in elementary schools isestimated to have gone from about 60 to 1 to about 40 to 1(Becker, 1987b).
Other industrialized nations have considerably fewercomputers available in schools. UNESCO's (1986) survey foundthat for 30 countries having a national policy for the use ofcomputers in education, the average number of computers perschool was five.
In developing countries, it is very likely that schoolsinterested in introducing computers will attempt to get alongwith even fewer machines, perhaps one per school. /t isdifficult to predict the possible effects or even how one mightorganize instruction in such a "lone or sparse micro" scenario.
The use of computers in educAktion raises questions thataccess to computers may be affectdd by race, ethnicity, gender,and social class (see Carnoy et al., 1986; Lockheed andMandinach, 1986; and Papagiannis et al., 1987, for a discussionof these issues).
45
With respect to possible gender bias in access to computersin education, Papagiannis et al. (1987, p.50) put the issue asfollows:
While access to computers at the primary level showedlittle or no gender bias, at the secondary level,access and use by male students was much higher thanfor females. This parallels the predominant use ofmicrocomputers in math and science, where gender biashas long been entrenched.
The effects of gender on computer access may disappear whencontextual variables are taken into consideration. Becker(1987a, p.14) notes the following:
Boys and higher-ability students also use schoolcomputers disproportionately, although not in allrespects. The sex discrepancy is greatest wherecomputers are least connected to school curricula-4nbefore- and after-school al:tivities, in game playi4.4 insecondary schools, and in elective programmingactivities in elementary school. Parity between boysand girls is most often founl in computer-assistedinstruction activities, in writing and word processing,in elementary-level learning games, and in programmingactivities in secondary school. Although peopleusually perceive girls as being underrepresented inprogramming activities, on national basis thedifferences are small in hi,,11 schools, especially wherecurricular activities exist for entry into programmingclasses...
Williams, (1987) study of computer access and studentattitudes in Trinidad and Tobago confirms the importance of theconnection to the school curriculum when looking at the questionof gender anti 4ccess to computers. She found that within schoolswhere computers are being used in instruction, females hadgreater access than males, but that outside of school. thereverse was true.
A second access/equity issue has to do with how (zomputer usewithin classrooms is affected by characteristics of students suchas race and social class. It is often pointed out that studentsfrom more well-to-do families are more likely to receive exposureto more "cognitively exciting" software like LOGO and that lesswell-to-do students are likely to be exposed to the "drudgery" ofdrill and practice software. But the charges of inequity in thiscomparison are not clear-cut since the 'Inefits of computer-assisted instruction for remediation in basic skills have beendemonstrated and the benefits of LOGO for the development ofthinking skills have not.
5'6
46
How these issues of access/equity suggested by theexperience of industrialized nations play out in the context ofdeveloping countries cannot be predicted. The access/equity7uestions within and between computer-using classrooms are,Jertainly likely to be of less concern to policy makers than thequestions of disparities between the "have's and have not's,"that is, between public and private schools, between urban andrural schools, and with respect to a choice between wider accessto computer literacy at lower grade levels end a more focusedemployment-oriented concentration at higher levels.
Likely EffectivenessThere have been inmerous summaries and interpretations of
the literature on the effectiveness of computer-assistedinstruction (Carnoy et al., 1986; Papagiannis et al., 1987; Burnsand Bozewan, 1981; Vinsonhaler and Bass, 1972; Forman, 1982;Rogosta, Holland, and Jamison, 1982; Hartley, 1978).
For our purposes, a useful benchmark was esta aishedKulik, Kulik, and Bangert-Drowns' (1985) meta-analysis of 32research studies on the effects of computer-assisted instruztionon the achievement of elementary school pupils in various schoolsubjects. They report an average effect size of .47 (consideredto be high) for the use of computer-assisted instruction. Othermeta-analyses have shown effect sizes of similar magnitude(Niemiec and Walberg, 1985).
One problem with the meta-analyses is that the studios uponwhich they are based are becoming dated. Becker (1987a, p.5-6)observes that 99% of the studies drawn upon in the meta-analysesuse "computer systems, software, and social organization ofimplementations that characterize, at most, one or two percent ofall the instruction in traditional academic subject matter usingschool computers today." Only a few of the studies in the meta-analyses reflect the u-e of microcomputers. However, a review ofresearch on computer etfectiveness in instruction (Roblyer,Castine, and King, forthcoming) that takes into account morerecent studies found an average effect size of .53 from studieson the use of computers at the elementary school level.
The syntheses of studies on the effectiveness of computersin education are of limited value to educational policy makers indeveloping countries, where the educational circumstances are fardifferent from those in more developed countries and whereempirical investigation into the effectiveness of computers ineducation is only beginning. Nevertheless, the evidence nowavailable does provide justification for the expectation thatinstruction delivered by computers will positively affect studentachievement. This is a useful starting and reference point forthe more situation-specific investigation that must take place indeveloping countries.
5 7
47
Again it must be stressed that achievement in traditionalacademic subjects is only one of many possible outcomes ofinterest to educators in developing countries. Attention must begiven to assessing the effects of computer use on motivationaland other attitudinal changes toward self, school, science, andtechnology. Similarly, the acquisition of even basic levelemployment-oriented computer skills is likely to be seen as moreuseful in developing countries than in industrialized countries.Especially in situations where computer literacy is the goal, itwill be important and by no means easy to find appropriate waysto evaluate the effectiveness of these interventions.
ConstraintsWe find in the literature mention of four major constraintsto the widespread use of computers in education in developingcountries. These are costs, software suitability, operatingrequirements, and teacher training.
COsts. Friend (1985b) maintains that the cost of purchasingand operating computers is the main reason why computers have notbeen more widely used in developing countries. Carnoy and Loop(1986, p.9) note the paucity of studies of costs, cost-benefit,and cost-effectiveness. They suggest that it is difficult toassess the cost for using computers in developing countriesbecause:
At prebent, most computers are used in environmentswhere maintenance and use infrastructures are well-developed and there is a market for such services(hence they can be priced). In many developingcountries this is certainly not the case (p. 9).
Carnoy and Loop also mention that no cost-benefit studies onthe use of computer-assisted instruction have been undertaken.Cost effectiveness studies that have been undertaken suggest thatwhile it appears that computer-assistedinstruction can becompared to other technologies, the results so far are too sparseto draw conclusions about the cost-effectiveness of computerscompared with other technologies.
Levin, Glass, and Meister (1984) estimate the costs ofproviding computer-assisted instruction through a fully-equippedcomputer laboratory in the United States. The costs to a schoolwith 736 students providing 23 sessions (10 minutes) per day on32 terminals would be about $119 par student per year at 1980prices. The authors point out that even with falling prices forcomputer equipment, the per student costs of computer-assistedinstruction in a fully-equipped computer laboratory are notlikely to be reduced, since the biggest component of these costsis for salaries of persons running the laboratory.
48
Such a relationship is likely to be different in developingcountries where costs of hardware are likely to be larger inrelation to salaries. In general, however, without knowing whata country hopes to accomplish with computers in the classroom,the hardware, software, and personnel available within schools,the number of students who can be served by the computers, thespecific conditions under which students will work withcomputers, and the effects of possible cost reductions throughusing computers after school hours for other purposes, it isimpossible at this time to speak with assurance about the likelyeducational costs of computers in developing countries.
Software Suitability. Concerns about the transfer ofsoftware to developing countries are widespread in theliterature. UNESCO (1986) reports that the problem consists ofthe suitability of imported software for the curriculum, thesuitability of software in various linguistic/cultural settings,and the portability of software across the hardware and operatingsystems used in different countries.
UNESCO maintains that governments pay insufficient attentionto the question of software in their plans. While many countriesappear to be producing software of their own, UNESCO notes thateven a software exporting country like Australia depends for themost part on imported software because domestic production isinsufficient. Of equal concern, available software is often notadequately evaluated and validated. UNESCO (1986, p.16) notesthat in the United States, there are 20,000 educational programson the market, and it is estimated that only about 10% of thesehave been assessed.
Hebenstreit (1986) cites the problem posed by copyrightarrangements for software. If developing countries are not ableto make legal copies of software packages and must purchase asmany copies as *here are computers, the possibilities for usingcomputers in the classroom will be severely constrained.
In educational policy discussions that will take place indeveloping countries during the coming years, the issue ofsoftware suitability could raise the question of educationalrelevance more forcefully than anything experienced since ThirdWorld countries began refashioning curricula and curriculummaterials following independence. In most instances, developingcountries intending to use computers in education will be forcedto rely on imported software. Policy makers considering theinsertion of imported software into an existing curriculum mustask:
o Where in the curriculum is this appropriate?
o How much of an existing curriculum might be
49
replaced or supplemented by importedsoftware?
Do cultural or linguistic mismatches betweenimported software and local circumstancesoffend national or regional pride? Do theyalso affect learning?
Freeman's (1987) study of a computer classroom in Grenada isan early attempt to explore some of the problems Third Worldchildren encounter while working with imported educationalsoftware.
Oueratina Reauirements. The problems posed by heat,humidity, dust, access to necessary sources for maintenance andrepair, and access to satisfactory eleCtrical current indeveloping countries are obvious to most people and need not beelaborated on here. Phillips (1987) and Freeman (1987) describehow computer-assisted instruction in Grenada is affected bysystem reliability and poor electrical supply.
Teacher Training. There is wide consensus that teachers areusually not adequately prepared to operate effectively withcomputers in inatruction (Hebonstreit, 1986; Carnoy et al., 1986;UNESCO, 1986; Lockheed and Mandinach, 1986). Carnoy and Loop(1986, p.6) describe the situation as follows:
Few countries seem to have taken the necessary steps toprepare teachers to use computers, even when hardwareIs installed in schools. There is also littleagreement on how to prepare teachers beyond short termcourses (6-15 days) that merely help understand how touse computers in the classroom. The problems ofimplementing even this type of training are apparentlygreat. Those countries most committed to computertraining for teachers (Sweden, UK, France, Australia,and Canada) have reached only 25% of their teachingforce. In most other countries, less than 5 percent ofteachers have taken such courses (2 percent in LatinAmerica). Even though some countries have recentlylaunched national teacher training programmes (India,Chile, Korea, Cuba and Mexico), most are not devotingthe necessary resources to training, but rather topurchasing hardware...
Idkelv _Future DirectionsThe widespread introduction of microcomputers into schools
in the United States began prior to a clear demonstration oftheir effectiveness as educational tools, and, in many cases, inthe absence of systematic planning on how they would be used. Wesee ei similar pattern emerging in developing countries.
60
50
Developing countries, especially in Latin America and theCaril-bean, are becoming increasingly interested in the use ofcomputers in education. This trend cannot yet be fullydocumented from the literature, and until we are able to do so,we offer it (we hope) as "reliable hearsay" derived from numerousconversations with educators, government officials, researchers,computer specialists, and manufacturers.
We believe that computers are an object and instrument ofinstruction unlike other instructional hardware. That is to say,few countries find it important to offer education abouttelevision and radio. The importance of learning about computersmay start a trend in which students will increasingly learnthrough computers.
The importance of learning about or through computers islargely a concern for relevance: education must be relevant toan emerging world economy where computers will play an importantrole--an economy in which few countries wish to fall furtherbehind. A concern for keeping up in the world economy,described by some as the pursuit of equity at the internationallevel, is likely tc. force acceptance of inequities at thenational level. Developing countries may judge that it isnecessary to give some children experience with computers, evenin the face of great uncertainties that this experience can beeasily generalized throughout the entire national educationalsystem. Although this consideration makes it more likely thatthe use of computers will be concentrated at post-primary levels,applications at the primary level can also be expected toincrease. This will be pushed in part by government authorities'intentions to keep up with private schools.
Primary school children in developing countries will havedifferential access' to computers based upon the financial meansof their countries and families. Some children will have morefavorable opportunities because they attend private schools orbecause they have a computer at home. Others are likely to havemore limited access to whatever machines are available in theschools they attend. Perhaps this might be only one machine perschool, where a microcomputer serves a multipurpose function forschool administration, computer literacy, direct instruction(possibly concentrating on either remedial work or acceleration),and service as part of a school "library." How effectively onemicrocomputer, or at best a very few, might contribute to primaryeducation in such a "lone or sparse micro" situation remains tobe studied.
Finally, we see that children in developing countries arelikely to encounter possibilities for computer-assisted educationthrough private learning centers. As costs of hardware andsoftware continue to fall, we expect to see a proliferation of
61
51private learning centers offering a variety of services to thepublic: computer sales and service, management consulting, dataanalysis, training in the use of computers,
and computer-assistedinstructioa in subjects taught in schools. These may be wellcapitalized venturrs possibly supported through franchises withmxnufacturers of computers from abroad or less capitalized localcottage industiies.
Private learning centers will compete for private fundsparents now spend on outside tutoring for their children. Suchtutoring is often conducted prior to a student's sitting forexaminations required to enter or graduate from secondary school.It will be interesting to see how effectivelyprivate learningcenters are able to use an imported and possibly eclecticcurriculum of off-the-shelf software in subjects like language,mathematics, and science to improve student performance onnational examinations in these subjects.
ConclusionsComputers have become, as many suggest, a "culturalphenomenon." Apart from the obvious functions they perform inthe workplace, the phenomenon of computers is generating newvocabulary, symbols, images, and metaphors that are rapidlybecoming part of both everyday culture and educated discoursethroughout the world. The computer phenomenon is increasinglymaking itself felt in developing countries.
It is too early to draw any conclusions about how computerswill contribute to the production of learning in schools indeveloping countries. It is obvious, however, that it is now toolate for discussions about whether the use of computers foreducation in developing countries is a good idea or not. Manycountries have already decided to have a look for themselves;others are soon to follow. It is important that we learn whatthey learn.
Commters as a Policy Option. Educational policy makers'indeveloping countries who may be considering the introduction ofcomputers into education should ask the following questions:o What can computers do for instruction thatconventional formats can't do?
o What specific educational uses of the computer are tobe attempted: direct instruction, computer literacy,programming, or learning to use computers as tools?o Are one or more of these uses feasible andaffordable at the primary school level orshould they be postponed to post-primaryschool education?
62
52
o What software will adequately deliver theintended instruction?
o What hardware can be obtained and servicedlocally to run the software selected?
o Will technical and financial resources besufficient for training teachers to work withcomputers?
4. VideodiscsVideodisc technology has been heralded as one of the richest
possibilities for the delivery of instruction. It wedstelevision's capacity to provide audio and visual material(including still photographs, motion, graphics, and text) to thecomputer's capacity for interactivity. It provides unmatchedpossibilities for the simulation of reality within a controlledinstructional setting. The relatively high costs of thistechnology have so far precluded wide application in schools,even in affluent societies. Continuing technologicaldevelopments seem likely to cause this to change. Although themost immediate applications in developing countries are likely tobe in the area of higher education and technical training forspecialized personnel, it is not inconceivable that, as withcomputers, primary school students in some places will encountervideodisc technology as part of libraries of the future, inprivate or public learning centers, in well-equipped schools, orat home.
Optical laser discs currently being used can hold over100,000 still frames of visual material (or about 25-30 minutesof motion). An audio track of less than one hour would accompany100,000 frames. The information stored on discs is read by avideodisc player and presented on a television monitor. Donahueand Donahue (1983) describe four levels of videodiscinstructional hardware:
Level 1: The videodisc system consists of avideodisc player and a television monitor andpossibly a remote control device. Itfunctions in a fixed sequence "slide show"mode. The user controls point of entry.Instruction is described as linear;
o Level 2: The videodisc player has a computercomponent which allows some branching,greater speed, and more user control;
o Level 3: The videodisc player is coupled toa microcomputer, which allows for more
53
complex programming and interactivity;
o Level 4: The system consists of two videodiscplayers, permitting more complex branching andalmost instantaneous feedback. Such systems areused for simulated instruction such as pilottraining.
The cost of an interactive "level 3" sybtem is likely to runbetween $2500 - $3000. Estimates of costs for producingvideodiscs vary. Donahue and Donahue (1983) estimate that afirst copy of a "professionally produced one-hour instructionaldisc" is likely to cost more than $100,000. Costs ol additionalcopies of the disc are around $20.
Interest in the use of videodiscs has heightened with theemergence of what is referred to as CD-ROM (compact disc-readonly memory), CD-I (compact disc-interactive), and CD-V (compactdiscvideo). Information on these 5 1/4" discs is stored indigital form and in great volume (perhaps around 550 megabytesand going higher). The CD-ROM standard is being applied tostoring and retrieving large databases and library materials.Many people expect that the development of a multi-purposecompact disc player will make it possible to read all kinds ofinformation, from computer data to film and music, 'ith onedevice.
The interactive CD-I technology is likely to become ofconsiderable interest to educators. The 5 1/4" CD-I discs willhold about 650 megabytes of information (storage of up to 1.3gigabytes is foreseen for the future). Current capabilities makeit possible to provide a 2-hour sequence of interactiveinstruction with high fidelity audio. The high resolution videowill rely on still pictures, text, and graphics. A CD-I discwill be "mastered" for about $5000, and copies will cost about$30.
What makes CD-I technology so interesting is the fact thatit will bring down dramatically the costs of an interactivevideodisc system. The Sony and Philips Corporations expect tohave an inexpensive CD-/ player on the market by the end of 1987that will make interactive videodisc technology a possibility forhomes and schools. The CD-I player will connect to a televisionset and interactivity will be achieved through a remote sensingdevice (or other peripherals). The built-in computer will notonly drive the system but offer some computing capability aswell. The player is expected to sell for about $500.
One of the current limitations of CD-I is that it does noteasily accommodate digitized full-motion video. This will becomea possibility soon, however. Recently, General Electric/RCAannounced a prototype machine called DVI (digital video
R4
54
interactive) that does permit the use of moving video ininteractive instruction (Ofiesh, 1987).
MUch of the attention given to videodiscs so far hasconcentrated on the hardware. Halloway (1982, p.133) warns that"less is known about teaching and learning using the videodiscthan is known about the engineering of the technology." Hanafin(1985, p.236) remarks that "little empirical research of any typehas been published to support the differential effectiveness, theeffects on learning, or the validation of the basic assumptionsof interactive video."
A considerable amount of research needs to be done on howbest to use the various capabilities of interactive video fordesigning effective instruction. Halloway (1982, p.135) argues:
The multimedia aspect is more than having everything(motion, sound, graphics) available at once, thoughthat is convenient. The advantage for instructionaldesign is having the choice, in being able to pick thebest mode for the learning goals. Research is somehelp in this selection. For instance, adding sound topictures is not necessarily an improvement and may evendetract from learning. 'Real' pictures rarely increaselearning; research suggests line drawings often workbetter. In these and other 'areas the computer alonemay be quite sufficient to assure a very high degree oflearning without the videodisc' (Bork, 1980). Thebroad selection of media in combination can provide thebest fit between learner and lesson.
For developing countries, the instructional design issuesare not likely to be of great interest in the immediate future.Few are in a position to marshall the financial or humanresources to engage in professional instructional design ofvideodisc instruction. What will be of interest is thedevelopment of effective software to run on the low costinteractive systems. Such software could be affordable if it is"generic" enough to be useful in several countries, therebyreducing costs. Instruction in mathematics is something thatmight lend itself to this kind of treatment.
Hanafin (1985) believes that rlteractive video does notdiffer gr-Atly from "allied technology" such as computers fromeither lea....ing or cognitive perspectives. In a parallel sense,we anticipate that the application of videodisc technology ineducation in developing countries will not differ significantlyfrom the application of computer technology where a high premiumis assigned to using computers for direct instructin. Low costinteractive video systems are likely to find use in homes(possibly replacing the videocassette recorder), especially asmotion pictures and recorded music become available on compact
P5
55
discs and systems come to serve both educational andentertainment functions. Interactive video is also likely tofind instructional use alongside microcomputers in privatelearning centers.
yisltssuisaiuuLisaisy_salan. At the moment, educationalpolicy makers in developing countries are not pressed to come toterms with the possibilities offered by videodisc technology.They may wait until low cost videodisc players have provedthemselves, software becomes widely available, and experimentalwork has been conducted that demonstrates where and ',low the useof videodiscs could be cost effective.
Given the high fixed capital costs associated with producinghigh quality instruction on videodiscs, the use of videodiscs ineducation in developing countries is likely to occur throughreliance on imported software. The quality and relevance of thissoftware will be an issue. Ways must be found to help educatorsin developing countries be intelligent consumers of videodiscsoftware (and computer software as well) prior to the day theyare able to produce their own.
5. Nand-held Electronic DevicesWe consider two varieties of hand-held electronic devices:
microprocessor-driven learning aids and calculators. It appeaxsthat neither is widely used in instruction in developingcountries at this time. However, the instructional possibilitiesthey offer and their low cost, ease of use, and ability to run onbatteries (sometimes solar cells) recommend them forconsideration as appropriate tools for supporting primary schoolinstruction.
Electronic Learning AidsOne can currently buy numerous electronic games that offer
educational content. Some of these operate with microprocessorssimilar to ones used in computers. These devices diaer fromcomputers in that they are much simpler to use, are lessexpensive, are not usually programmable, and deliver onlyinstructional routines resident in the device or that might beadded with insertable modules.
The most well-known of the electronic learning aids is theseries produced by Texas Instruments, Inc.: the Speak & Spell,Speak & Read, and Speak & Math. These "talking" devices utilizea keyboard, synthetic speech, and an eight-character alphanumericdisplay to offer interactive drill and practice in language andarithmetic skills. The aids weigh just over one pound and sellfor about $40 in retail stores in the United States. Otherelectronic aids to appear recently include Coleco's "TalkingTeacher" and Video Technology's "Electronic Lesson One."
66
56
The first known effort to examine the use of electroniclearning aids in classrooms in a developing country is describedin Anzalone and McLaughlin (1984). This was a pilot study thatinvolved the use of the Speak & Read and Speak & Math in primaryschools in the Kingdom cf Lesotho.
The purpose of the study was to determine whether the use ofexisting models of the Speak & Read and Speak & Math was feasiblein a developing country like Lesotho, how the aids fit into thestructure and processes of the classroom, and whether pupils'mastery of English and arithmetic was strengthened by use of theaids. The study ran for twelve weeks in five primary schools.The language aids were used in standards 3 and 4, and the mathaids were used in standard 6. Pupils worked with the aids ingroups of four for about one hour per day, two to five times perweek, during the usual classroom time for English or arithmetic.
The study reported the following results:Technical Feasibility. The aids were introduced into
primary schools without great cost and preparation. The aidsstood up well physically. Only 3 of 62 aids used requiredreplacement, and none were lost because of poor housekeeping onthe part of teachers. Pupils adapted easily to the syntheticspeech contained in the aids.
Human Use Effects. Interviews conducted with teachers,administrators, and pupils showed a widespread belief that theaids were useful. The result of pupil observation showsmaintenance of a high degree of involvement during periods whenthe aids were used. Both the interviews and observationdemonstrated that it was feasible to use the aids in groups offour.
Learning Effects. The results of weekly tests of pupils'English word recognition and arithmetic demonstrated positivelearning effects associated with use of the aids. For bothlanguage and arithmetic, pupils using the aids made larger gairwin achievement than those not using the aids. DI the cases ofthe language aid, the difference was statistically significant at
the .001 level. For the arithmetic aid, the difference on thefinal test was not statistically significant but was so (at the.03 level) on the test given the week before. The aids had agreater impact on learning among the loss able 'pupils. Nosignificant differences in achievement were discovered betweenboys and girls using the aids.
The study concluded that it is feasible to use electronicaids to supplement primary instruction in a developing countrylike Lesotho. The study outlines a scenario whereby aids wouldbe rotated among grades and schools. Under these and otherassumptions, the cost of a 60 hour supplement of electronic drill
57
and practice (excluding batteries) was estimated to cost $.94.Anzalone and McLaughlin also concluded that more information wasneeded on the use of the aids over longer periods of time, inother country settings, and on the range and magnitude ofpossible learning effects.
We have found no other attempts to look at the use of thistechnology within the context of a developing country. Althoughthe learning aids such as those used in Lesotho continue to besold and new ones have appeared, there appears to be littleinterest on the part of manufacturers to improve theircapabilities. Similarly, unlike microcomputers, the prices ofelectronic learning aids have not fallen during the past fewyears.
CalculatorsHand-held calculators, now often solar powered and sslling
for $5 or less, are perhaps the most affordable educationally-related electronic product available today. They are to be foundalmost everywhere in the Third World. The one place that theyare noticeably absent is the classroom. In educational systemswith which we are familiar, parents and teachers tend to beopposed to the use of calculators because of the belief thatusing calculators is likely to lead to inadequate mastery ofnumber facts and development of computational skill.
The hand-held calculator does not strictly fall under ourdefinition of instructional hardware because of our insistencethat the device must be capable of offering instruction thatcould serve as an alternative to a lesson usually provided by ateacher or a textbook. The instructional use of calculators inthe classroom would almost certainly require direct instructionfrom the teacher or from exercises in worksheets or another formof printed text.
We were unable to find any published studies of the use ofcalculators in schools in developing countries. There have beenmany studies undertaken in North America and Europe. A review ofthe literature by Suydam (1983) looked at over 175 studies, ofwhich at least 75 explored whether the use of calculators in theclassroom would have a deleterious effect on achievement inmathematics. In all but a few instances, achievement test scoreswere as high or higher when calculators were used in instruction(but not on the test) than when they were not used. Suydam foundevidence that the use of calculators can help students learnnumber facts and can help in expanding a student's repertoire ofproblem-solving strategies.
A review of thirteen students by Sigg (1982) concluded thatcalculators were not detrimental to the development ofcomputational skills. Seven of the thirteen studies showed gains
CS
./111
58
in computational skills when calculators were used ininstruction; three showed improvements in students' attitudestoward mathematics. Less able students appeared to gain morefrom using calculators in instruction. The evidence regardingimprovements in problem-solving performance was inconclusive.
Etlinger (1974) sees two roles for the use of calculators inmathematics instruction; the functional and pedagogical. Thefunctional role foresees calculators being used to perform allmental calculations in solving problems. The pedagogical roleforesees calculators being used as a tool to facilitate learning.The first role is likely to be unacceptable in most developingcountries, and the second one is likely to be suspect.
If we ignore the possibilities offered by hand-heldcalculators we may miss an opportunity to improve the externalefficiency of education in those settings where the majority ofchildren leave school du:ring the first few grades. In thesesituations, number facts are often poorly acquired andcomputational skill is weak. PrGficiency in the use ofcalculators would give those childien who receive no furthereducation a means to solve many of the arithmetic problems theywill encounter on the job and in daily life. Although this wouldcreate a dependency on an external device, there is no socialstigma associated with using calculators for purposes of dailycomputation. And just as those of us who wear eyeglasses findthis dependency preferable to not seeing properly, it is likelythat being able to solve problems by using an inexpensiveelectronic device is preferable to not being able to solve theproblems at all.
Perhaps "calculator literacy" should be taught in additionto basic mathematics. The use of calculators might kindleinterest in and facilitate learning of more sophisticatedmathematics than is now generally achieved in most of the primarschools in developing countries.
Hand-held Electronic Devices as a Policy Option. Furtherexperimental work with electronic learning aids in classrooms indeveloping countries is necessary before recommending toeducational policy makers that they be used. There appears to belittle interest in shaping this technology to make it apractical, affordable, and effective alternative within primaryschool instruction in developing countries.
Similarly, educational policy makers are likely to have adifficult time exploiting the educational potential ofinexpensive calculators. It may be difficult or impossible toconvince teachers and parents that learning to use calculators isnot detrimental to the development of computational skill or todemonstrate the advantages of "calculator literacy." Moreover,the use of calculators would have to be contained in an
69
59
instructinnal program. The development costs and the costs andadministrative burden for training teachers and producing anddistributing the materials required for such a program are likelyto detract from the attractiveness of using this technology forinstructional purposes in the classroom.
79
60
VIII. CONCLUSIONS AND OBSERVATIONS
Our purpose in this review was to determine how and to whatextent learning technology in the form of instructional hardwareis or might be significantly involved in the production oflearning in primary school classrooms in developing countries.We have looked in some detail at the past experience of severalcountries and have tried to look ahead at the likely directionsand possible consequences of the introduction of the devices ofthe "second electronic revolution." We reached numerousconclusions throughout the review about how the use ofinstructional hardware relates or might relate to the productionof learning at various points in the process. We will close withsome general conclusions and observations.
We defined instructional hardware as electronic presentersof instructional sequences that are significant enough in scopeand curricular content to constitute an alternative to lessonsusually provided by teachers or textbooks and similar printedmaterials. We find that the experience just reviewed points tothe importance of the word alternative.
We found that instructional hardware has not been used in away that would constitute a complete alternative to formal schoolinstruction delivered by teachers. The one possible exception tothis is RADECO, in the Dominican Republic, where learning byradio takes place in makeshift schools supervised byparaprofessionals. Even here, however, radio instruction islimited to one hour per day and is not meant to constitute acomplete alternative to formal primary schools. The strategy forextending primary education by using instructional hardware as alams costly alternative to formal schools remains largelyuntried.
We did find that instructional hardware, especiallytelevision, has been used as a primary means of delivery insystems-wide attempts to upgrade the quality of primary schoolinstruction. In several cases, this was in response to largeshortages of trained teachers. Here, the use of instructionalhardware was not intended to constitute an alternative toconventional instruction but as a component of a strategy thatoften included improvement of conventional instruction throughsuch means as teacher training. It appears that as levels ofteacher training rise or the numbers of certified teachers in thesystem grow, teachers resist (usually successfully) the use oftelevision as a primary means of delivering school instruction.Few developing countries face the situation they did in thesixties where there were often massive shortages of trainedteachers, low numbers of children enrolled in school, and verylittle perceived relevance or quality in the curriculum. The
71
61
conditions favorable for a strategy where instructional hardwareis used as a catalyst in multi-faceted reform of all or part ofan entire educational system have largely disappeared.
Instructional hardware appears to be most commonly used todeliver instruction that is not an alternative to, but rather anr'ternative within, regular school instruction. Within thecontext of regular school instruction, learning technologydelivered through instructional hardware makes it possible toprovide productive periods of instructional time superior to whatroutinely takes place in the usual primary school classroom in adeveloping country. These periods (where they have beendelivered with instructional hardware) typically vary in durationfrom an hour per week to an hour per day. Often these periodshave led to large gains in student achievement during the courseof a school year. Our companion review, "Review on the SoftTechnologies of Learning," suggests that these productive periodsof instructional time result from application of methods andtechniques emanating from our knowledge of the learning process.In the present review, we have found that these enriched segmentsof instruction can be delivered with varying degrees ofefficiency with the us* of instructional hardware.
The idea of productive periods of instructional time,delivered as an alternative within regular school instruction, isacknowledged in the BRIDGES model of learning production. Thelearning technology cluster of variables is conceived as beingseparate from (but related to) other "within school" variables.Learning technology should arguably be conceived as intrinsicallyinvolved in, and not separate from, such things as classroommanagement and instructional materials. The evidence, however,suggests the existence and potential effectiveness of a "withinschool" alternative provided by the application of learningtechnology during part of the school day.
In the future, whether learning technology delivered throughinstructional hardware will be used less for quality enhancementwithin the context of regular school instruction and more as analternative to conventional instruction will probably depend onhow well resources for education keep up with demographics. Upto now, using instructional hardware as an alternative toconventional school instruction has not been compelling for mostdeveloping countries. In the future, however, instructiondelivered thrcugh instructional hardware may be an alternative toproviding nothing, and the use of instructional hardware couldwell be not just compelling but inevitable.
6 2
IX. AGENDA FOR FUTURE RESEARCH AND IMPLEMENTATION
The preceding analysis, conclusions, and observationspoint to recommendations for an agenda for future research andimplementation for using instructional hardware to improve andextend primary education in developing countries. It is a jointagenda for investigators and policy makers--a short list of whatwe believe to be the most promising avenues of activity.
We would first note that the use of instructional hardwareis likely to be more attractive at post-primary levels.Interventions that rely on child/machine interaction are likelyto be seen by many as incompatible with primary school'straditional mission of socialization of the child--no matter howeffective the intervention proves to be. Nevertheless, insituations where there is acceptance that productive periods ofinstructional time mediated by the use of instructional hardwareare either desirable, inevitable, or both, we contend thefollowing.
Of the instructional hardware we reviewed, radio continuesto offer the most promise of being an effective and affordabledeliverer of instruction on a mass scale. The interactive radiomodel has set a standard.against which the success of futureapplications may be judged. For the future, the priorities forresearch and implementation for the use of radio in primaryeducation in developing countries are these four:
o More needs to be learned about the experienceof all those countries reportedly using radiofor primary school education but whoseexperience does not figure into theliterature. How extensive is thisexperience? What are the successful models?Do they Ethers common characteristics? Whatlessons can be drawn so as to allow greatergeneralization about the use of radio inprimary school education in developingcountries?
o We need to have a better understanding of thebarriers to acceptance of the interactiveradio model. At the moment, we can onlyspeculate why more countries have not decidedto adopt the model. In view of theinvestments already made and the results that
73
63
have been achieved, speculation is nosubstitute for empirical investigation.
o The RADECO model for extending primaryeducation through radio seems especiallypromising and likely to be relevant forcountries that come to experience difficultyin stretching resources to keep pace withpopulation growth. Assistance should beprovided to countries prepared to adopt oradapt the RADECO model and share thisexperience with the international community.
o It would be useful to investigate theintroduction and implementation of theinteractive model following a strategy thatis less concerned with demonstrating radio'seffectiveness (this has already beendemonstrated) and more concerned withdemonstrating the development of localexpertise, commitment to quality,identification with the lessons produced, andsustainable institutional capacity for futureimplementation.
The other area on the joint agenda for research andimplementation with respect to instructional hardware and primaryeducation in developing countries has to do with futureapplications of computers and associated technologies. Althoughthe industrialized nat4,ons are selling an increasingly largenumber of computers to developing countries, very littleassistance is being provided to investigate how computers mightbe used most productively in classrooms. We recommendexploration in four areas:
o We must look for viable and affordable usesof the computer in situations where only oneor a few microcomputers can be introduced.What difference does it make to have acomputer present in the school? Caneffective activities for individuals, groups,or the entire class be designed so as torevolve around the use of a singlemicrocomputer?
o We need to discover the most effectiveapproaches that integrate computer literacyand direct instruction in traditional schoolsubjects.
o In countries where successful national orregional curriculum development activities
74
6 4
(such as in the Caribbean) heive producedmaterials that are widely used, it would beuseful to explore the development of related,high quality educational software. It wouldbe unfortunate if decisions about thepotential effectiveness of computers ineducation in developing countries are madeentirely on the basis of software producedfor markets in industrialized societies.
o Some attention should be given to the effectsof using computers for primary educationindirectly through teacher education. Theuse of computers, especially if linked tovideodiscs, would be feasible, manageable,and affordable for many individual or groupsof teachers' colleges. Such applicationswould allow the modeling of effective andineffective teaching behaviors and a goodmeans to teach methods for diagnosis of, pupildifficulties in different subjects.
75
65
X. BIBLIOGRAPHY
Agence de Cooperation Culturelle et Technique (1985). Niger -television scolaire en merge. Direct. (September, pp 15-49).
Aghi, M. (1979). The effectiveness of science educationprogramming. Journal of Communication. (Vol. 29, F. 14, pp.104-5).
Anzalone, S. (1987). The effect of PLATO instruction on basicskills in Jamaica. McLean, VA: Institute for InternationalResearch - Learning Technologies Project.
Anzalone, S., and Conrad, K. (1984). A review of evaluations.Microcomputers for basic and vocational skills. Amherst, MA:University of Massachusetts.
Anzalone, S., and McLaughlin, S. (1984). Electronic aids in adeveloping country: Improving basic skills in Lesotho. Amherst,MA: University of Massachusetts.
Arnove, R.F. (Ed.). (1976). Educational television: A policycritique and guide for developing countries. New York: Praeger.
Ball, J. (1971). A radio series for primary schools in Africa.Educational broadcasting international. (Vol. 5, 2. pp. 79-82).
Bates, T. (1982). Options for delivery media. In H. Perraton,(Ed.). Alternative routes to formal education. Baltimore: TheJohns Hopkins University.
Bates, T., and Robinson, J. (1977). Evaluating educationaltelevision and radio. London: The Open University Press.
Becker, H.J. (1987a). The impact of computer use on children'slearning: What research has shown and what it has not.Baltimore: The Johns Hopkins University.
Becker, H.J. (1987b). What we know about computer use and itseffectivenes in schools: Existing knowledge and new researchdirections. Cupertino, California: Apple Computer, Inc.
Bienfait, A. (1964-71). Recherches peaagogiques. La televisionscolaire du Niger. Paris: Ministere de la Cooperation. (Vol.
8).
76
66
Block, C. (1985). Interactive radio and educational development.Development Communication Report, 49.
Brusling, C. (1982). Rise and fall of educational technology inSweden. Prospects. (Vol. 12, 3).
Bryant, J., and Anderson, D. (1983). Children's understanding oftelevision. New York: Academic Press.
Bunnel, D. (June 1986). The challenge of CD-ROM. PC WORLDMagazine.
Burns, P.K., and Bozeman, W.C. (1981). Computer-assistedinstruction and mathematics achievement: Is there a relationship?Educational Technology, 21. (pp. 32-39).
Carnoy, M. (1976). The economic costs and returns to educationaltelevision. In R.F. Arnove, (Ed.). Educational television: Apolicy critique and guide for developing countries. New York:Praeger. (pp. 22-73).
Carnoy, M.; Daly, H.; DeVillar, R.; Earthman, E.; Herrmann, F.;Loop, L.; Misra, M., and Starbuck, J. (1986). Education andcomputers: Vision and reality in the mid-1980s. Stanford, CA:Stanford University School of Education.
Carnoy, M., and Levin, H.M. (1975). Evaluation of educationalmedia: Some issues. /nstructional Science. Amsterdam:Elsevier Scientific Publishing Co.
Carnoy, M., and Loop, L. (1986). Computers and education: Whichrole for international research? Paris: UNESCO.
Chadwick, C. (1982). An overview of educational technology inLatin America. Prospects. (Vol. XII, 3).
Chu, G., and Alfian, C. (1967). Programming for development inIndonesia. Journal of Communication. (Vol. 30, No.4, pp. 50-7).
Chu, G., and Schramm, W. (1967). Learning from television: Whatthe research says. National Association of EducationalBroadcasters.
Clark, R. (1983). Reconsidering research on learning from media.Review of Educational Research. (Vol. 53,4, pp. 445-459).
Clark, R. (Winter 1984). Clark's reply to Petkovich andTennyson. Educational communication and technology - A journalof theory research and development.
7 7
67
Clearinghouse on Development Communication (1978). A sourcebookon radio's role in development. Washington, DC.
Clearinghouse on Development Communication (1982). Projectprofiles. Washington, D.C.
Collins, A.W.; Sobol, B.L.; and Westby, S. (1981). Effects ofadult commentary on children's comprehension and inferences abouttelevised aggressive portrayal. Child Developzent, 52. (pp. 158-63).
Collins, A.W., and Wiens, M. (Intelevision viewing: DescriptionM. Pressly and J. Levin, (Eds.).Educational, clinical and socialSpringer-Verlag.
press). Cognitive processes inand strategic implications. InCognitive strategy training:applications. New York:
Cooper, H.M. (1984). The integrative research review: Asystematic approach. Beverly Hills: Sage Publications.
Christensen, P. (1985). The radio language arts project:Adapting the radio mathematics model. Development CommunicationReport, 49.
Winne, I. (1985). El aprendizaje del numero y la computadora.Buenos Aires: Laboratorio de Psicologia Evolutiva.
Donahue, T., znd Donahue M. (1983). Understanding interactivevideo. Training and Development Journal.
Eicher, J.C., and Orivel, F. (1980). Cost analysis of primaryeducation by television in the Ivory Coast. The economics of neweducational media. Paris: UNESCO.
Eicher, J.C.; Hawkridge, D.; McAnany, E.; Mariet, F.; andOrivel F. (1982). The economics of new educational media:Overview and synthesis. Paris: UNESCO.
Emery, M. (1985). Another exciting revolution? A rejoinder toKupisiewicz and White. Prospects. (Vol. XV, 4, pp. 493-509).
Etlinger, L. (1974). The electronic calculator: A new trend inschool mathematics. Educational Technology. (14/12 pp. 43-5).
Faure, E. et al., (1972). Learring to be. Paris and London:UNESCO-Harrap.
Forman, D. (January 1982). Search of the li'erature. TheComputing Teacher.
78
68
Forsythe, R. (1970). Instructional radio: A position paper.Stanford, CA: Stanford University.
Freeman, C. (1987). Computer classroom thnography: CrochuPrimary School, Grenada. McLean, VA: Institute forInternational Research - Learning Technologies Project.
Friedlander, B.; Wetstone, H.; and Scott, J. (1974). Suburbanpre-school children's comprehension of In age-appropriateinformational television program. Child Development. (Vol. 43,pp. 561-5).
Friend, J. (1984). (Mimeo). Evaluation of first gradeinstructional materials. In The Radio-Assisted Community BasicEducation Project, Report 2. Washington, D.C.: InterAmericaResearch Associates.
Friend, J. (1985). (Mimeo). Classroom uses of the computer: Aretrospective view with implications for developing countries.Washington, D.C.: The World Bank.
Friend, J. (1985). A model for interactive radio lessons: Theradio mathematics project. Development Communication Report, 49.
Friend, J.; Galda, K.; and Searle, B. (1979). (Mimeo). Transferof instructional technology: Radio mathematics in Nicaragua andThailand.
Friend, J., and Kemmerer, F. (1985). Strategies for and costs ofdisseminating the radio language arts project throughout Kenya.In Interactive radio in education. Washington, D.C.: Agency forInternational Development and Institute for InternationalResearch, Communications Support Project.
Friend, J., and Kozlow, S. (1985). Evaluation of second gradeinstructional materials produced by The Radio-Assisted CommunityBasic Education Project. Washington, D. C.: InterAmericaResearch Associates.
Friend, J.; Searle, B.; and Suppes, P. (Eds.). (1980). Radiomathematics in Nicaragua. Stanford, CA: Stanford University.
Galda, K. (1985). Learning math by radio. DevelopmentCommunication Report, 49.
Gilmore, J. (1986). Afterward. In M. Imhoof and P. Christensen,(Eds.). Teaching English by radio: Interactive radio in Kenya.Washington, D.C.: Academy for Educational Development.
Glass, G.V. (1977). Integrating findings: The meta-analysis of
79
69
research. Review of Research in Education, 5. (pp. 351-379).
Glass, G.V.; McGaw, B.; and Smith, M.L. (1981). Meta-analysis insocial research. Beverly Hills, CA: Sage Publications.
Grant, S. (1977). Administrative history of out-of-schooltelevision in the Ivory Coast. Washington, D.C.: Academy forEducational Development.
Green, C. (1985). Instruction by audio conference: An Alaskanexample. Development Communication Report, 49.
Halosworth, B. (1971). Radio broadcasts for schools - somelessons learnt. Educational Broadcasting International. (Vol. 5,1, pp. 56-9).
Hannafin, M. (1985). Empirical issues in the study of computer-assisted interactive video. Educational Communication andTechnology - a Journal of Theory, Research and Development.(Vol.33, 4, pp. 235-247).
Hancock, A. (1976). Planning for educational mass media.London/New York: Longman.
Hanssen, A.; Koslow, S.; and Olsen, A. (1983). RADECO: Radio-based primary education in the Dominican Republic. DevelopmentCommunication Report, 42.
Hartley, S.S. Meta-analysis of the effects of individually pacedinstruction in mathematics. (Doctoral dissertation, Universityof Colorado, 1977.) Dissertation Abstracts International, 1978,38(7-A), 4003. University Microfilms No. 77-29, 926.
Hawkridge, D., and Robinson, J. (1982). Organizing educationalbroadcasting. Paris: UNBSCO.
Hebenstreit, J. (1986). The use of informatics in education:Present situation, trends and perspectives. Paris: UNESCO,Division of Structures, Content, Methods and Techniques ofEducation.
Helwig, J., and Friend, J. (1985). Teaching where there are noschools. Development Communication Report, 49.
Hodapp, T.V. (1977). Children's ability to learn problem-solvingstrategies from television. Alberta Journal of EducationalResearch. (Vol. 23, No.3, pp. 171-7).
Holloway, R. (Spring 1982). The videodisc in education: A casefor technological gradualism. Videodisc/Videotext. (Vol. 2, No.
so
70
2, pp. 132-7).
Hunter B. (1987). Computer-based tools in learnir3 and teachinga research agenda. Cupertino, California: Apple Computer, Inc.
Imhoof, M. (1985). Interactive radio in the classroom: Tenyears of p-oven success. Development Communication Report, 49.
Imhoof, M., and Christenseli, P. (Eds.). (1986). TeachingEnglish by radio: Interactive radio in Kenya. Washington, D.C.:Academy for Educational Development.
Ingle, H. (1986). Reconsidering the use of television foreducational reform: The case of El Salvador. In R. F. Arnove,(Ed.). Educational Television. New York: Praeger. (pp. 114-39).
Jamison, D. (1982). An introduction to the methods of costanalysis. In H. Perraton, (Ed.). Alternative routes to formaleducation. Baltimore: The Johns Hopkins University.
Jamison, D.; Klees, S.; and Wells, S. (1978). The costs ofeducational media: Guidelines for planning and evaluation.Beverly Hills, CA: Sage Publications.
Jamison, D., and McAnany, E. (1978). Radio for education anddevelopment. Beverly Hills, CA: Sage Publications.
Jamison, D., and Orivel, F. (1982). The cost-effectiveness ofdistance teachings for school equivalency. In H. Perraton,(Ed.). Alternative routes to formal education. Baltimore: TheJohns Hopkins University.
Jamison, D.; Searle, B.; Galda, K.; and Heyneman S. (1981).Improving elementary mathematics education in Nicaragua: Anexperimental study of the impact of textbooks and radioachievement. Journal of Educational Psychology. (Vol. 73, pp.556-566).
Jamison, D.; Suppos, r., and Wells,of alternative instructional media:Educational Research. (Vol. 44, 1,
S. (1974). The effectivenessA survey. Review of
pp. 1-67).
Kaye, A.E. (1976). The Ivory Coast educational televisionproject. In R.F. Arnove, (Ed.). Educational television. (pp.140-79).
Kleem, S. (1977). Cost analysis of nonformal ETV systems: Acase study of the extra-scolaire system in the Ivory Coast.Washington, D.C.: Academy for Educational Research.
81
4
71
Konan, D., and N'Guessan, A. (1979). Educational reform andtechnological innovation: The Ivory Coast experiment.Educational reforms: Experiences and prospects. Paris: UNESCo.
Kulik, J.; Kulik C.; and Bangert-Drowns, R. (1985).Effectiveness of computer-based education in lementary schools.Journal of Computers in HUman Behavior. (Vol 1, pp. 59-74).
Kumar, M. S. V. (1986). Introducing technological innovationsfor education in a developing country: Implications forplanning. Doctoral dissertation. Amherst, MA: University ofMassachusetts.
Kumar, N. (1967). Educational radio in India. New Delhi: AryaBook Department.
Leslie, J., and Jamison, D. (1980). Policy implications ofinstructional technology: Cost and implications. In J. Simmons(Ed.). The education dilemma: Policy issues for developingcountries in the 19805. New York: Pergamon Press.
Levin, H.; Glass, G.; and Meister, G. (1984). Cost effectivenessof four educational interventions. Palo Alto, CA: ,StanfordUniversity School of Education.
Lewis, J. (1975). A revieo of scientific calculators for school.School Science Review. (Vol. 57, 198. pp. 177-182).
Linn, M.C. (1985). The cognitive consequences of programminginstruction in classroom,. Educational Researcher. (Vol. 14, 5,pp. 14-16, 25-29).
Lockheed, M., and Mandinach, E. (1986). Trends in educationalcomputing: Decreasing interest and the changing focus ofinstruction. Educational Researcher. (Vol. 15, 5, pp. 21-26).
Luskin, B.J. (1987). The current and future implications of CD-1. Technological Horizons in Education. (Vol. 14, No. 10, p.47).
Lyle, J. (1982). Since 1967: The original case studiesreviewed. In D. Hawkridge and J. Robinson. Organizingeducational broadcasting. Paris: UNESCO.
Masland, L., and Masland, G. (1976). Some cross-culturalimplications of educational television: the Samoan ETV project.In R. F. Arnove, (Eds.). Educational television: A critique andguide for developing countries. New York: Praeger.
s2
72
Mayer, R. (1980). Contributions of cognitive science and relatedresearch on learning to the design of computer literacycurricula. Series in Learning and Cognition. Report 81-1.Santa Barbara, CA: University of California.
Mayo, J., and Hornik, R.C. (i981). Evaluation and research inthe planning, development and support of media-based education.Paris: International Institute for Educational Planning.
Mayo, J.; Hornik, R.C.; and McAnany, E.G. (1976). Educationalreform with television: The El Salvador experience. Stanford,Cio Stanford University.
Mayo, J.; McAnany, E.G.; and Kleee, S. (1973). The Mexicantelesecundaria: A cost-effeotiveness approach. Washington:Academy for Educeational Development.
McArany, E. (1973). Radio's role in development: Fivestrategies for use. (Vol. 4, pp. 28). Washington, D.C.:Information Center on Instructional Technology.
KcAnany, E.; Oliveira, J.; Orivel, F.; and Stone, J. (1983).Distance education: Evaluating new approaches in education fordeveloping countries. Evaluation in Education. (Vol. 6, pp. 289-376).
Nguchu, R.R. (1981). The extent to which radio is used inteaching of home science in urban primary schools in Kenya.Nairobi, Kenya: Institute of Edvlation.
Niemiec, R., and Walberg, H. (1985). Computers and achievementin the elementary schenls. Journal of Educational ComputingResearch. (Vol. 14. pp. 435-439).
Ofiesh, G. (1987). Reflections c Microsoft's 2n4 CDconference. Technological Horizons in Education. (Vol. 14, No.9, p. b3).
Oxford, R.; Clark, J.; Hermansen, J.; Christensen, P.: andImhoof, M. (1986). Final report: Evaluation of the Kenya radiolanguage arta project volume 1: Narrative results. Washington,D.C.: Academy for Educational Development.
Paisley, W., and Chen, M. (1983). Children and interactivemedia: Exploring the effects of the second electronicrevolution. Stanford, CA: Stanford University Press.
Papagiannis, G.J.; Dougal, C.;(1987). Information technologytheory, research and practice.
Williamson, N.; and La Mon, R.and education: Implications forOttawa: International
R3
73Development Research Center.
Perraton, H. (Ed.). (1982). Alternative routes to formaleducation: Distance teaching for school equivalency. Baltimore:The Johns Hopkins University Press.
Perraton, H. (Ed.) (1986). Alternative forms of post-primaryeducation and training in Africa. London: Marlborough House.Porraton, H. (1982). The scope of distance teaching.Alternative routes to formal education. Baltimore: The JohnHopkins University.
Petkovich, M., and Tennyson, R. (1984). Clark's "Learning frommedia": A critique. Educational Communication and Technology -A journal of theory research and development.
Phillips, E.A. (1987). Report on Activities at the CrochuLearning Center - Term 1 1986. (Mimeo).
Polsky, R.M. (1974). Getting to Sesame Street. New York:Praeger.
Roblyer, M.D.; Castine, W.H.; and King, F.J. The effectivenessof computers in instruction: A review and synthesis of recentresearch findings. (Submitted for publication, 1987).
Romiszowski, A.J. (1984). Producing instructional systems:lesson planning for individualized and group learning activities.London: Nichols Publishing.
Saloman, G. (1977). Effects of ncouraging Israeli mothers toco-observe Sesame Street with their five-year-olds. ChildDevelopment. (Vol. 48, No. 3, pp. 1146-51).
Saloman, G.; Eglstein, S.; Finkelstein, R.; Finkelstein, I.;Mintzberg, E.; Maly., D.; and Velner, L. (1972). Educatioraleffects of Sesame Street on Israeli children.
Saloman, G., and Gardner, H. (1986). The computer as educator;lessons from television research. Educational Researcher.(January, pp. 13-19).
Sang-jan, P. (1982). Media and the teaching of mathematics inThailand. Media Asia. Bangkok: Centre for EducationalInnovation and Technology
SAnguinetty, J. A. (198b). The replication of the RADECO projectin the Dominican Republic: Recurrent cost implications.Washington, D.C.: InterAmerica Research Associates.
R4
Schmelkes, S. (1972). Estudio de evaluacion aproximativa de :
escuelas radiofonicas de la Tarahumara.Revista del Centro de
Estudios Educativos, 2, 2. (pp. 11-36).
Schmlkes, S. (1977). The radio schools of Tarahumara, Mexic
An evaluation. Radio for education and development: Case
studies. Washington, D.C.: The World Bank.
Schramm, W. (1962). What
instructional television.
years. Stanford, CA: The
Stanford University. (pp.
we know about learning from
Educational television: The next ten
Institute of Communication Research,
52-74).
Schramm, W. (1973). Big media, little media. Washington D.C.:
Academy for Educational Development.
Schramm, W.; Coombs, P.; Kahnert: F.; and Lyle, J. (1967). The
new media: Memo to educational planners. 14..ris: UNESCO.
Schramm, W. (Ed.). (1967). New educational media in action:
Case studies for planners. Paris: UNESCO/Internascional
Institute for Educational Planning.
Schramm, W.; Nelson, L.M.; and Betham, M.T. (1981). Bold
experiment: The story of educationaltelevision in American
Samoa. Stanford, CA: Stanford University.
Searle, B. (1985). Evaluations of three interactive radio
projects. Development CommunicationReport, 49.
Siann, G., and Macleod, H. (1986). Computers and cLildren of
primary school age: Issues and questions. British Journal of
Educational Technology.(Vol. 2, No. 7, pp. 133-143).
Sigg, P. (1982). The hand-held calculator:Effects on
mathematicalabilities and implications for curriculum change.
Bloomington, Indiana: University of Indiana.
Singer, D.G., and Singer, J.L. (1980a). Television viewing and
aggressive behavior in preschool children: A field study.
Fortasic Psychology, 347. (pp. 289-303).
Singer, J.L. (1980) The power and liritations of televisiOn: A
cognitive-affuctiveanalysis. In P. Tannenbaum, (Ed.). The
entertainmentfunctions of television. Hillsdale, N.J.: Erlbaum.
Singer, J.L., and Singer, D.G. (1980b). Television viewing:
Family style and aggressive behavior in preschool children.
Violence and the family. Boulder, CO: American Association for
75
the Advancement of Science. (pp. 37-65).
Singer, J.L., and Singer, D.G. (1983). Psychologiats look attelevision; cognitive, developmental, personality and socialpolicy implications. American Psychologist, 38, 7. (pp. 826-34).
Spain, P.L. (1977). The Mexican Radioprimaria project. In P.Spain et al., (Eds.). Radio for education and development: Casestudies, 1. Washington, D.C.: The World Hank.
Spain, P.L.; Jamison, D.T.; and McAnany, E.G. (Eds.). (1977).Radio for education and development: Case studies. Washington:World Bank.
Sprigle, H.A. (1971). Can poverty children live on SesameStreet? Young Children. (1Pal. XXVI,4, pp. 202-217).
Suppes, P.; Searle, B.; and Friend, J. (Eds.). (1978). The radiomathematics project: Nicaragua 1976-1977. Stanford, CA: StanfordUniversity Press.
Suydam, M.education:Ohio State
Suydam, M.Arithmetic
(1982). The use of calculators in pre-collegeFifth annuu1 state-of-the-art review. Columbus:
University.
(1983). Research report: Achieving with calculators.Teacher. (Vol. 31, 3., Nov., p. 20).
Sylla, F. S. (1985). Computers and literacy in Senegal.Doctoral Dissertation. Cambridge, MA: Massachusetts Institute ofTechnology.
Tavasoli, G. (1978). Impact of educational television ontechnical and agricultural improvements in rural areas of Iran.International Review ol Modern Sociology, 8, I. (pp. 49-63).
Tilson T., and Searle, H. (1978). An assessment of teacherattitudes. In P. Suppes, B. Searle, and J. Friend (Eds.). Theradio mathematics project: Nicaragua 1976-1977. Stanford, CA:Stanford University.
UNESCO. (1967). New educational media in action: Case studiesfor planners, I - III. Paris: UNESCO.
UNESCO. (1978). The economics of new educational media: Cost andstatus of research trends, 1. Paris: UNESCO.
UNESCO. (1980), The economics of new educational media: Cost andeffectiveness, 2. Paris: UNESCO.
76
UNESCO. (1984). Media education. Paris: UNESCo.
UNESCO. (1984). Statistical Yearbook. Paris: UNESCo.
UNESCO. (1986). Informatics and education, a first survey of thestate of the art in 43 countries. Paris: UNESCO.
Vinsonhaler, J.R., and Sass, R.K. A summary of ten major studieson CAI drill and practice. Educational Technology. (July 1972,pp. 29-32).
Walberg, H.J. (1984a). Improving the productivity of America'sschools. Educational Leadership, 41. (pp. 19-30).
Walberg, H.J. (1984b). Synthesis of research on. teaching. InM.C. Wittrock, (Ed.). Handbook of Research on Teaching.Washington D.C.: American Educational Research Association.
Walker, H. (1986). Evaluacion cualitativa del programa deradioeducacion comunitaria (RADECO) de la Republica Dominicana.Washington, D.C.: U.S. Agency for International Development.
White, R. (1976) An alternative pattern of basic education:Radio Santa Maria. Paris: UNESCO.
White, R. The use of radio in primary and secondary formaleducation: The radio Santa Maria Model in the DominicanRepublic. In P. Spain, et al. (Eds.). Radio for education anddevelopment: Case studies, 1. (pp. 172-98).
Williamsf K. (1987). Computrs in Trinidad and Tobago: A surveyof access and attitudes towards the technology in secondaryschools. Docoral dissertation. Cambridge, MA: HarvardUniversity.