jordan 1999 - how to read a journal article in social psychology - eng

8/8/2019 Jordan 1999 - How to Read a Journal Article in Social Psychology - EnG

http://slidepdf.com/reader/full/jordan-1999-how-to-read-a-journal-article-in-social-psychology-eng 1/5

".I.OGKINSDE

Roy F. Baumeister (coord.) (1999).

T H E SELF I N S O C I A L P SYCHOLOGY

(Colectia "Key Readings in Social

PsychoIogy")PhiladeIphia: PsychologyPress.: H O U J t o

R e a da J o u r n a l R r f i c l e i o S o c i a l P s y c h o l o g y

C h r i s t i a n H . Jordan and Mark R Z a n n a • u n i v e r s i t y

How to Read a JournalArticle in Social Psychology

When approaching a journal article for the first time, an d often on subsequent occasions,most people try to digest it as they would an y piece of prose. They start at the beginning

an d read word fo r word, unt i l eventually they arrive at the end, perhaps a little bewildered,but with a vague sense of relief. This is not an altogether terrible strategy; journal articlesdo have a logical structure tha t lends itself to this sort of reading. There are, however, moreefficient approaches-approaches that enable you, a student of social psychology, to cutthrough peripheral details, avoid sophisticated statistics w ith which you m ay not be famil-iar, and focus on the central ideas in an article. Arming yourself with a little foreknowl-edge of what is contained in journal articles, as well as some practical advice on how toread them, should help yo u read journal articles more effectively. If this sounds tempting,read on .

Journa l articles offer a window into th e i nne r workings of social psychology. They docu-ment how social psychologists formulate hypotheses, design empirical studies, analyzeth e observations (h ey collect, an d interpret: heir results. Journal art ic les also serve an in-va l ua b l e archival funct ion: They contain th e f u l l store of c om m on and cumulat ive knowl-edge of social psychology. Having documentation of past research allows researchers tobuild on past findings and advance our understanding of social behavior, without pursuing

avenues of investigation t ha i have already been explored, Perhaps mo st importantly, aresearch s tudy is never complete u n t i l iis results hav e been shiircd wi th others, colleaguesan d students a l i K c . . h i u i n a ) ar i idnt ,

;w . H pr im a ry m e a ns of communicating research find-

ings . As such, (hey ca n be ge nu i ne l y exciting and interesting to read.That last c la im ma y have c a ugh t you off guard. F or beginn ing readers, jour nal art ic les

ma y seem a nything bu t interesting an d exciting. They may, on the contrary, appear daunt-ing and esoteric, laden with jargon an d obscured by menacing sta t ist ics. Recognizing thisfact, we hope to arm you, throug h this paper, with th e basic informat ion yo u will need toread j o u rn a l articles with a greater sense of comfort an d perspective.

Social psychologists study many fascinating topics, ranging from prejudice and dis-crimination, to culture, persuasion, liking and love, conformity and obedience, aggres-

46 1



462 • The Self inSocial Psychology Appendix 463

sion, and the self. In our daily lives, these are issues w e often struggle to un derstand. Socialpsychologists present systematic observations of, as well as a w ealth of ideas about, suchissues in journal articles. It would be a shame if the fascination an d intrigue these topics

have were lost in their t ranslation into jou rnal publications. W e don' t think they are , and bythe end of this paper, hopeful ly yo u w o n ' t either.

Journal articles come in a variety of forms, including research reports, review articles,an d theoretical art ic les. Put briefly, a research report'is a formal presentat ion of an originalresearch study, or series of studies. A review article is an evaluative survey of previouslypublished work, u sually organized by a guiding theory or point of view. Th e a u thor of areview article summarizes previous investigations of acircumscribed problem, commentson what progress has been made toward i ts resolution, and suggests areas of the problem

that require fur ther study. A theoretical'articlealso evaluates past research, but focuses onthe development of theories used to explain emp irical findings. Here, the author may presenta new theory to explain a set of findings, or may compare an d contrast a set of competingtheories, suggesting why one theory might be the superior one.

This paper focuses primarily on how to read research reports, fo r several reasons. First,the bulk of published literature in social psychology consists of research reports. Second,

the summaries presented in review articles, and the ideas set forth in theoretical articles,are built on findings presented in research reports. To get a deep understanding of howresearch is done in social psychology, fluency in reading original research reports is essen-tial. Moreover, theoretical articles frequently report newstudies th at pit one theory againstanother, or test a novel prediction derived from a new theory. In order to appraise thevalidity of such theoretical contentions, a grounded understanding of basic f indings isinvaluable. Finally, most research reports are written in a standard format that is likelyunfamiliar to new readers. Th e format of review an d theoretical articles is less standard-ized, and more l ike that of textbooks and other scholarly writ ings, with which most readersare familiar. This is not to suggest th at such articles are easier to read an d comprehend thanresearch reports; they can be quite challenging indeed. It is simply the case that, because

more rules apply to the writing of research reports, more guidelines can be offered on howto read them.

Th e A n a t o m y o f Re se ar ch Re p or t s

Most research reports in social psychology, and in psychology in general , are writ ten in astandard format prescribed by the American Psychological Association (1994). This is agreat boon to both readers andwriters. It allows writers to present their ideas and f indingsin a clear, systematic manner. Consequently, as a reader, once yo u understand this format,you will not be on completely foreign ground when you approach a new research report—regardless of its specific content. You will know where in the paper particular informationis found, making it easier to locate. No matter what your reasons fo r reading a research

report , a firm understanding of the format in whic h the y ar e written will ease your task. W ediscuss the format of research reports next, with some practical suggestion s on how to readthem. Later, we discuss how this format reflects the process of scientific investigation,illustrating ho w research reports have a coherent narrative structure.

TITLE A ND A B S T R A C T .

Though you can ' t j udge a book by its cover, you can learn a lo t about a research reportsimply by reading its title. The title presents a concise statement of the theoretical issuesinvestigated, and/or the variables that were studied. For example, the following title w astaken almost at random from a prestigious journal in social psychology: "Sad an d guil ty?Affective influences on the explanation of conflict in close relationship s" (Forgas, 1994, p.

56). Just by reading th e title, it can be inferred that th e study investigated ho w e m ot iona lsta tes change the way p eople explain confl ic t in c lose re la t ionships. I t a lso suggests that

when feeling sad, people accept more personal blame for such conflicts (i.e., feel moreguilty).

Th e abstract is also an invaluable source of informat ion. It is a brief synops is of thestudy, and packs a lot of informat ion into 150 words or less. The abstract conta ins informa-tion a bou t th e problem tha t w as investigated, how it was investigated, th e major f indings ofth e study, and hints a t the theoretical and practical imp lications of the f indings. Thus, theabstract is a useful summary of the research that provides th e gist of the inves t igat ion.

Reading this outline first can be very helpful, because it tells yo u whe re the report is going,and gives you a useful framework for organizing information contained in the article.

The t i t le and abstract of a research report are l ike a m ovie preview. A m ovie previewhighlights the important aspects of a movie 's plot , and provides just enoug h informationfor one to decide whether to watch th e whole movie . Just so with titles an d abstracts; theyhighlight the key features of a research report to allow yo u to decide if you want to read th ewhole paper. And just as with movie prev iews, they do not give the wh ole story. Readingjus t th e title and abstract is never enough to fully understand a research report .

INTRODUCTION

A research report has four main sections: introduction, method, results, and discussion.Though it is not explicitly labeled, the introduction begins the main body of a researchreport. Here, the researchers set the stage for the study. They present the p roblem underinvestigation, and sta te w hy i t was important to study. By providing a b rief review of pastresearch and theory relevant to the central issue of investigation, the researchers place the

study in an historical context an d suggest how the study advances knowledge of the prob-lem. Beginning with broad theoretical an d practical considerations, th e researchers delin-eate the rationale thafrled them to the specific set of hypotheses tested in the study. Theyalso describe how they decided on their research strategy (e.g., why they chose an experi-

ment or a correlational study).The introduction generally begins w ith a broad consideration of the problem investi-

gated. Here, the researchers wa nt to illustrate that the problem they studied is a real prob-le m about which people should care. If the researchers are studying prejudice, they m aycite statistics that suggest discrimination is prevalent, or describe specific cases of dis-crimination. Such information helps i l lustra te why the research is both p ractically andtheoretically meaningful, and why you should bother reading about it. Such discussionsar e often quite interesting an d useful . They ca n help yo u decide fo r yourself if the researchhas merit . But they may not be essentia l for understanding the study at han d. Read theintroduction carefully, but choose judiciously w hat to focus on and remember. To under-stand a study, what yo u really need to understand is wha t th e researchers' hypotheses were,

and how t he y we re de r ive d from theory, informal observation, or intui t ion. Othe r back-ground informat ion may be int r iguing, but may not be cri t ical to understand w h a t th eresearchers did and why they did it.

While reading th e introduction, try answering these questions: What problem w as stud-ied, and why? How does this study relate to, and go beyond, past investigations of theproblem? H ow did the researchers derive their hypotheses? What questions do the researchershope to answer with this study?

METHOD

In the method section, the researchers translate their hypotheses into a set of specific,testable questions. Here, the researchers introduce the main characters of the study—the



464 • The Self inSocial Psychology

subjects or participants—describing their characteristics (gender, age, etc.) and how manyof them w ere involved. Then, they describe the materia ls (or apparatus), such as any ques-tionnaires or special equipment, used in the study. Finally, they describe chronologically

/ the procedures of the study; t ha t is, how the study was conducted. Often, an overview of

the research design wil l begin the method section. This overview provides a broad outlineof the design, a lert ing you to w h a t yo u should at tend.

The method is presen ted in great detail so that other researche rs can recreate the study toconfirm (o r que s t ion) its results. This degree of detail is no rm a l ly no t necessary to under-stand a study, so don' t get bogged down t rying to memorize the part iculars of the proce-dures. Focus on how the independent variables were manipulated (o r measured) and how

th e dependent variables were measured.Measuring variables adequately is not always an easy matter. Man y of the varia blespsychologists are interested in cannot be directly observed, so they must be inferred from

participants' behavior. Happiness, for example, cannot be directly observed. Thus, research-ers interested in how being happy influences people 's judgmen ts must infer happiness (orits absence) from their behavior—perhaps by asking people ho w happy they are, an d j udg-in g their degree of hap piness from their responses; perhaps by studying people 's facialexpressions fo r signs of happiness, such as smiling. Think about th e measures researchers

use while reading the m ethod section. Do they adequately ref lect or capture the conceptsthey are meant to measure? If a measur e seems odd, consider carefully how the researche rsjustify its use. % f

Oftentimes in social psychology, getting there is half the fun. In other words, how aresult is obtained can be just as interesting as the result itself. Social psychologists oftenstrive to have participants behave in a natural, spontaneous mann er, while controlling eno ughof their environment to pinpoint the causes of their behavior. Sometimes, th e major contri-

bution of a research report is i ts presentat ion of a novel method o f investigation. Whe n thisis the case, the method will be discussed in some detail in the introduction.

Participants in social psychology studies ar e intelligent an d inquisitive! people who areresponsive to what happens around them. Because of this, they are not always initially toldth e true purpose of a study. If they were told, they might not act naturally. Thus, research-ers frequently need to be creative, presenting a credible rationale for complying with pro-cedures, w ithout revealing the study's purpose. This ra t ionale is know n as a cover story,

an d is often an elaborate scenario. While reading the method section, try putting yourselfin the shoes of a part ic ipant in the study, and ask yourself if the instructions given topart ic ipants seem sensible , realistic , and engaging. Imagining w hat i t was like to be in thestudy will a lso help you remember the study's procedure, and aid you in interpreting the

study's results.While reading th e method section, try answering these questions: H ow were the hypoth-

eses translated into testable questions? H ow were the variables of interest manipulatedand/or measured? Did the measures used adequately reflect the variables of interest? For

example, is self-reported incom e an adequate measure of social c lass? Wh y o r w h y no t?

RESULTS •

Th e results section describes how the observations collected were analyzed to determinewhether the original hypotheses were supported. Here, the data (observations of behavior)are described, and statistica l tests are presented. Beca use of this, the results section is oftenint imidating to readers w ho have l i t t le or no tra ining in sta t ist ics. Wading thro ugh complexand u nfamiliar statistical analyses is understandably confusing an d frustrating. As a result ,many students are tempted to skip over reading this section. We advise you not to do so.Empirical findings are the foundation of any science an d results sections are where such

findings are presented.

Appendix • 465

Take heart. Even th e most prestigious researchers were once in your shoes an d sympa-thize with you. Thoug h space in psychology journals is l imited, researchers try to strike abalance between th e need to be clear and the need to be brief in describing their results. Inan influent ia l paper on how to write good research reports, Bern (1987) offered this a dv ic eto researchers:

No matter ho w technical or abstruse your art ic le is in its part iculars, inte l l igentnonpsychologists with no expertise in statistics or experimental design shouldbe able to comprehend the broad outl ines of what you did and wh y. They shouldunderstand in general terms what was learned, (p. 74)

Generally speaking, social psychologists try to practice this advice.Most statistical analyses presented in research reports test specific hypotheses. Often,

each analysis presented is preceded by a reminder of the hypothesis it is meant to test.After an analysis is presented, researchers usually provide a narrat ive description of theresult in plain English. When th e hypothesis tested by a statistical analysis is not explicitlystated, you can usually determine th e hypothesis that w as tested by reading this narrat ivedescription of the result , and referring back to the introduction to locate an hypothesis thatcorresponds to that result . After even th e most complex sta t istical analysis, there wil l be awritten description of what the result means conceptually. Turn your attention to thesedescriptions. Focus on the conceptual meaning of research f indings, not on the mechanicsof how they were obtained (unless you're comfortable with statistics).

Aside from statistical tests an d narrative descriptions of results, results sections alsofrequently contain tables an d graphs. These ar e efficient summaries of data . Even if youare not familiar with statistics, look closely at tables and graphs, and pay attention to the

means or correlations presented in them. Researchers always include written descriptionsof the pertinent aspects of tables and graphs. While reading these descriptions, check the

tables an d graphs to make sure what the researchers say accurately reflects their data. Ifthey say there was a difference between tw o groups on a particular dependent measure,look at the means in th e table that correspond to those tw o groups, and see if the means dodiffer as described. Occasionally, results seem to become stronger in their narrat ive de -scription than an examination of the data w ould warrant .

Statistics canbe misused. When they are, results aredifficult to interpret . Havin g saidthis, a lack of sta t ist ical knowledge sho uld not make you overly cautious while readingresults sections. Th ough not a perfect antidote , journal art ic les undergo extensive reviewby professional researchers before publication. Thus, most misapplications of statistics ar ecaugh t and corrected before an article is published. So, if you are unfam iliar with statistics,you can be reasonably confident that findings ar e accurately reported.

While reading th e results section, try answering these questions: Did the researchersprovide evidence that an y independent variable manipulations were effective? Fo r example,if test ing fo r behavioral di fferences be twe e n happy and sad pa r t i c ipa n t s ,did (h e research-ers demonstrate that one group was in fact happier than the other? What w ere the majorfindings of the study? Were th e researchers' original hypotheses supported by their obser-vations? If not, look in the discussion section for how the researchers explain th e f indingsthat were obtained.

DISCUSSION

Th e discussion section frequently opens with a summary of wha t th e study found, and anevaluation of whether the findings supported the original hypotheses. Here, the researchersevaluate the theoretical andpractical implications of their results. This can be particularlyinterest ing w hen the results did not w ork out exactly as the researchers antic ipated. Wh en



468 •' The Self in S o c i a l P s y c h o l o g y

Once you have read through a research report , and un derstand the researchers' f indingsand their explanations of them, the story does not end there. There is more than one inter-pretation for any set of findings . Different researchers often e xp la in th e same set of facts in

different wa ys .Let 's t ake a mo me n t l o o i - . < v l a nasty r um or , Th e r u m o r is th i s: Researchers present thei r

s tudies in a dispass ionate M a nne r , intending only to inform readers of thei r f i nd i ngs an d

their interpretat ion of t hos e f ind ings. In t ruth, researchers aim not only to inform readers,bu t also \opersuade them (Sternberg, 1995). Researchers w a n t to convince you their ideasare right . There is never on ly one explanation for a set of findings. Certainly, some expla-nations are better than o thers; some f i t the available data better, are more parsimonious, orrequire fewer questionable assumptions. The point here is that researchers are very pas-sionate about their ideas, an d w a n t you to believe them. It's up to you to decide if you w a n tto buy their ideas or not .

Let's compare social psych ologists to salesclerks. Both social psychologists and sales-clerks w ant to sell you som ething; either th eir ideas, or their wares. You need to decide ifyo u want to buy what they' re sel l ing or not—and there are potentia l ly negative conse-quences for e i ther decision. I f you le t a sales c lerk dazzle you with a sales pitch, withoutthinking about i t carefully, you might end up buyi ng a substandard product that you don ' treally need. After having done this a few times, people tend to become cynical, steelingthemselves against any and all sales pitches. This too is dangerous. If you are overly criti-cal of sales pitches, yo u could end up foregoing genuinely useful products. Thus, by anal-ogy, when you are too critical in your reading of research reports, you might dismiss, outof hand, some genuinely useful ideas—ideas that can help shed l ight on why people be-have the way they do.

This discussion raises the important question of how critical one should be while read-

in g a research report. In part, this will depend on why one is reading the report. If you arereading it simply to learn what th e researchers have to say about a particular issue, fo rexample, then there is usually no need to be overly critical. If you want to use the research ias a basis for planning a new study, then you should be more critical. As you develop anunderstanding of psychological theory and research methods, you will also develop anability to criticize research on many different levels. An d any piece of research can becriticized at some level. As Jacob Cohen put it, "A successful piece of research doesn'tconclusively settle an issue, it jus t makes some theoretical proposition to some degreemore likely" (1990, p. 1311). Thus, as a consumer of research reports, you have to strike adelicate balance between being overly critical an d overly accepting.

While reading a research report, at least initially, try to suspend you r disbelief . Try tounderstand the researchers' story; that is, t ry to understand the fac ts—the findings and howthey were obtained—and the suggested explanation of those facts—the researchers' inter-pretation of the findings and what they mean. Take the research to task only after you feelyou understan d what the authors are trying to say.

Research reports serve no t on ly an i m por t a n t archival funct ion, doc um e nt ing re sea rc hand i t s f ind ings, bu t also an inva luable s t i m u l u s f u n c t i o n . They ca n excite other researchersto join th e investigation of a pa r t i c u l a r issue, or to npp l y ne w m e thods or theory to a differ-

ent, perhaps novel, issue. I t is this st imulus function that Ell iot Aronson, an eminent socialpsychologist , referred to w hen he admitted that , in publishing a study, he hopes his col-leagues wil l "look at it, be stimulated by it, be provoked by it, annoyed by it, and then goahead and do it better.... That ' s the excit ing thing abou t science; i t progresses by peopletaking off on one another's work" (1995, p. 5). Science is indeed a cumulative enterprise,and each new study builds on w hat has (or, sometimes, has not) gone before i t . In this way,research articles keep social psychology vibrant.

A study ca n inspire ne w research in a num be r of different ways, such as: (1) it can leadone to conduct a better test of the hypotheses, t rying to ru le out a l ternative explanations of

A p p e n d i x • 469

the f indings; (2) i t can lead one to explore the l imits of the f indings, to see how widelyapplicable they are , perhaps exploring si tuations to which they do not apply; (3) i t can leadone to test th e implications of the findings , fur ther ing scientific investigation of the phe-nom e non ; (4) it can inspire one to a pp ly the findings, or a novel methodology, to a differentarea of inves t igat ion; and (5) it can provoke one to test th e findings in the c on te x t of aspecific real world problem, to see if they ca n shed l ight on it. All of these ar e e xc e l l e n textensions of the origin al research, and there are , undou btedly, other w ays that researchiindings ca n spur ne w investigations.

The problem with being too critical, too soon, while reading research reports is that th eonly fur ther research one may be wil l ing to a t temp t is research of the f irst type: Redoing astudy better. Sometimes this is desirable , part icu larly in the early stages of investigating apart icular issue, w h e n th e findings are novel and perhaps unexpected. But redoing a rea-sonably compell ing study, without extending it in any way, does little to advance our un-derstanding of hu man behavior. Although the new stu dy might be "better," i t wil l not be"perfect," so // 'would have to be run again, and again, l ikely never reaching a stage whereit is beyond cri t ic ism. At some point , researchers have to decide that the evidence is com-pell ing enough to w arrant investigation of the last four types. It is these types of studiesthat most advance ou r knowle dge of social behavior. As you read more research reports,yo u will become more comfortable deciding when a study is "good enough" to move be-yond i t . This is a somewhat subjective judgment, and should be made carefully.

Whe n social psychologists write up a research report fo r publication, it is because theybelieve they have something new and excit ing to communicate about social behavior. Mostresearch reports that are submitted for publication are rejected. Thus, the reports that areeventua l ly published are deemed pert inent not only by the researchers who w rote them, butalso by the reviewers and editors of the journ als in wh ich they are publishe d. These people,

at least, believe th e research reports they write an d publish have something important an dinteresting to say. Sometimes, you' l l disagree; not all journal articles ar e created equal,after all. But we recommend that you, at least initially, give these well-meaning socialpsychologists the benefit of the doubt. Look for what they're excited about. Try to under-stand th e authors' story, and see where it leads you.

Author Notes

Preparation of this paper was facilitated by a Natural Sciences and Engineering ResearchCouncil of Canada doctoral fe l lowship to Christian H . Jordan. Thanks to Ro y Baumeister,Arie Kruglanski, Ziva Kunda, John Levine, Geoff MacDonald, Richard Moreland, Ia nNewby-Clark, Steve Spencer, and Adam Zanna for their insightful comments on, and ap-praisals of, various drafts of this paper. Thanks also to Arie Kruglanski and four anony-m o u s editors of vo lum e s in the series, Ke y R e a d i n g s in Social Psycholog y fo r their helpfu l

cr i t iques of an initial out l ine of this paper. Correspondence concern ing this art ic le shou ldbe addressed to Christ ian H. Jordan, Department of Psychology, Un iversi ty of Waterloo,W a t e r l o o , O n t a r i o , C a n a d a N2 L 3G 1 . E le c t ron ic m a i l c a n b e s e n t [email protected] .



470NThe Self inSoc ia l Psycho logy

REFERENCES

Am e r i c a n P s y c ho l og i c a l As s oc i a t i on (1994). Publication

manual'(4th ed.). W a s h in g to n , D.C.

Aronson, E. (1995). R e s e a r c h in soc ia l psychology,ns a l eap

• of faith. In E. A;:o/:.son (Ed.) , Readings ai%>ttt'fatfiMciftl

anifnal (li\\ ed. , pp . 3-9). Ne w Y ork : W . H. F i i -om a n an d

C om pa ny .

Bern, D. J . (1987). W r i l i n g th e e m p i r i c a l j o u r n a l a r t i c l e . In

M. P. Zanna & J . M. Darley (Eds.), The compleat academic:

A practical guide for the beginning social scientist (pp.

171-201). Ne w Y ork : R a nd o m Hous e .

Bowe r i ng , G. (1988). Errata. Re d Deer, A l t a . : Re d Deer Col-

lege Press. C o h e n , J. (1990). T hi ngs I ha ve l e a rne d (so f a r ) .

American Psychnlngi.it, 45, 1304-1312.

F;org,is, I. P. (1994). Sad and gui l ty? Affcc l isc- i n f l ue nc e s on

th e e x p l a n a t i o n of c o n f l i c t in c lose r e l a t i o n s h i p s . Journal

of Personality and Social Psychology , 66, 56-68.

St e rnbe rg , R. J . (1995). Th e psychologist's companion: Aguide to scientific writing or students and researcliers (3rd

ed.). Cambridge: Cambridge Univers i ty Press.

[NSBE!

Roy F. Baum eister (coord.) (1999).

T H E SELF I N S O C I A L P S Y C H O L O G Y

(Colectia "Key Readings in Social

PsychoIogy")PhiIadelphia: Psychology Press.

jordan 1999 - how to read a journal article in social psychology - eng

Documents