essentials of marketing research chapter 7: survey research
TRANSCRIPT
Essentials of Marketing Research
Chapter 7:
Survey Research
SURVEYSSURVEYS
SURVEYS ASK RESPONDENTS FOR INFORMATION USING VERBAL OR WRITTEN QUESTIONING
RESPONDENTS = A REPRESENTATIVE SAMPLE OF PEOPLE
Advantages of Surveys
• Standardization• Ease of Administration• Ability to tap the “unseen”• Suitability for statistical analysis• Sensitivity to subgroup differences• Quick, Inexpensive• Efficient, Accurate• Flexible
PROBLEMSPROBLEMS
• POOR DESIGN
• IMPROPER EXECUTION
• Research Manager’s Task =
Total Error Minimization
Total Error
Systematic Error (bias)
Random SamplingError
Tree Diagram of Total Survey ErrorTree Diagram of Total Survey Error
RANDOM SAMPLING ERRORRANDOM SAMPLING ERROR
A STATISTICAL FLUCTUATION THAT OCCURS BECAUSE OF CHANCE VARIATION IN THE ELEMENTS SELECTED FOR THE SAMPLE
SYSTEMATIC ERRORSYSTEMATIC ERROR
SYSTEMATIC ERROR RESULTS FROM SOME IMPERFECT ASPECT OF THE RESEARCH DESIGN OR FROM A MISTAKE IN THE EXECUTION OF THE RESEARCH
Systematic Error (bias)or
Non-sampling bias
AdministrativeError
RespondentError
Tree Diagram of Total Survey ErrorTree Diagram of Total Survey Error
SAMPLE BIASSAMPLE BIAS
SAMPLE BIAS - WHEN THE RESULTS OF A SAMPLE SHOW A PERSISTENT TENDENCY TO DEVIATE IN ONE DIRECTION FROM THE TRUE VALUE OF THE POPULATION PARAMETER
RespondentError
Non-responseError
ResponseBias
Tree Diagram of Total Survey ErrorTree Diagram of Total Survey Error
RESPONDENT ERRORRESPONDENT ERROR
A CLASSIFICATION OF SAMPLE BIAS RESULTING FROM SOME RESPONDENT ACTION OR INACTION
• NONRESPONSE BIAS
• RESPONSE BIAS
NONRESPONSE ERRORNONRESPONSE ERROR
• NONRESPONDENTS - PEOPLE WHO REFUSE TO COOPERATE
• NOT-AT-HOMES
• SELF-SELECTION BIAS• OVER REPRESENTS EXTREME POSITIONS
• UNDER REPRESENTED INDIFFERENT
ResponseBias
UnconsciousMisrepresentation
DeliberateFalsification
Tree Diagram of Total Survey ErrorTree Diagram of Total Survey Error
RESPONSE BIASRESPONSE BIAS
A BIAS THAT OCCURS WHEN RESPONDENTS TEND TO ANSWER QUESTIONS WITH A CERTAIN SLANT THAT CONSCIOUSLY OR UNCONSCIOUSLY MISREPRESENT THE TRUTH
Acquiescence bias (agree with all Qs)
Extremity bias (always extreme response)
Interviewer bias (interviewer presence)
Auspices bias (halo affect)
Social desirability bias
Tree Diagram of Total Survey ErrorTree Diagram of Total Survey Error
Systematic Error (bias)
AdministrativeError
RespondentError
Tree Diagram of Total Survey ErrorTree Diagram of Total Survey Error
ADMINISTRATIVE ERRORADMINISTRATIVE ERROR
• IMPROPER ADMINISTRATION OF THE RESEARCH TASK
• BLUNDERS• CONFUSION
• NEGLECT
• OMISSION
Data Processing Error
Sample Selection Error
Interviewer Error
Interviewer Cheating
Tree Diagram of Total Survey ErrorTree Diagram of Total Survey Error
COMMUNICATING WITH COMMUNICATING WITH RESPONDENTSRESPONDENTS
• PERSONAL INTERVIEWS• DOOR-TO-DOOR
• SHOPPING MALL INTERCEPTS
• TELEPHONE INTERVIEWS
• SELF-ADMINISTERED QUESTIONNAIRES
PERSONAL INTERVIEWSPERSONAL INTERVIEWS
Speed of Data Collection Moderate to fast
Geographical Flexibility Limited to moderate
Respondent Cooperation Excellent
Versatility of Questioning Quite versatile
Door-to-Door Personal InterviewDoor-to-Door Personal Interview
Questionnaire Length Long
Item Non-response Low
Possibility of Respondent LowestMisunderstanding
Degree of Interviewer HighInfluence of Answer
Supervision of Interviewers Moderate
Door-to-Door Personal InterviewDoor-to-Door Personal Interview
Anonymity of Respondent Low
Ease of Call Back or Follow-up Difficult
Cost Highest
Special Features Visual materials may beshown or demonstrated;extended probing possible
Door-to-Door Personal InterviewDoor-to-Door Personal Interview
Speed of Data Collection Fast
Geographical Flexibility Confined, urban bias
Respondent Cooperation Moderate to low
Versatility of Questioning Extremely versatile
Questionnaire Length Moderate to Long
Mall Intercept Personal InterviewMall Intercept Personal Interview
Item Non-response Medium
Possibility of Respondent LowestMisunderstanding
Degree of Interviewer HighestInfluence of Answers
Supervision of Interviewers Moderate to high
Mall Intercept Personal InterviewMall Intercept Personal Interview
Anonymity of Respondent Low
Ease of Call Back or DifficultFollow-up
Cost Moderate to high
Special Features Taste test, viewing of TV Commercials possible
Mall Intercept Personal InterviewMall Intercept Personal Interview
MAIL SURVEYSMAIL SURVEYS
Speed of Data Collection Researcher has no controlover return of questionnaire; slow
Geographical Flexibility High
Respondent Cooperation Moderate--poorly designedquestionnaire will have low
response rate
Versatility of Questioning Highly standardized format
MAIL SURVEYSMAIL SURVEYS
Questionnaire Length Varies depending on incentive
Item Non-response High
Possibility of Respondent Highest--no interviewerMisunderstanding present for clarification
Degree of Interviewer None--interviewer absentInfluence of Answer
Supervision of Interviewers Not applicable
MAIL SURVEYSMAIL SURVEYS
Anonymity of Respondent High
Ease of Call Back or Follow-up Easy, but takes time
Cost Lowest
MAIL SURVEYSMAIL SURVEYS
• Write a “Sales Oriented” Cover Letter• Money Helps• - As a token of appreciation• - For a charity• Stimulate Respondents’ Interest with Interesting Questions• Follow Up• Keying questionnaires with codes• Advanced Notification• Sponsorship by a Well-known and Prestigious Institution
How to Increase Response Rates for How to Increase Response Rates for Mail SurveysMail Surveys
Ways to Increase Mail Survey Response
• Foot-in-the-door• Personalization• Anonymity• Response deadline• Incentives
• Length of survey• Survey size,
reproduction and color• Type of postage
– outgoing
– return envelopes
SELF-ADMINISTERED SELF-ADMINISTERED QUESTIONNAIRESQUESTIONNAIRES
• PLACE OF BUSINESS
• DROP-OFF
• COMPUTERIZED
• INTERNET
• OTHER VARIATIONS
Telephone SurveysTelephone Surveys
Speed of Data Collection Very fast
Geographical Flexibility High
Respondent Cooperation Good
Versatility of Questioning Moderate
Telephone SurveysTelephone Surveys
Questionnaire Length ModerateItem Non-response MediumPossibility of Respondent Average MisunderstandingDegree of Interviewer Moderate Influence on AnswerSupervision of Interviewers High, especially
with central locationWATS interviewing
WATS: Wide-Area Telecommunications Service
Telephone SurveysTelephone Surveys
Anonymity of Respondent Moderate
Ease of Call Back or Follow-up Easy
Cost Low to moderate
Special Features Fieldwork and supervisionof data collection aresimplified; quite adaptableto computer technology
Telephone SurveysTelephone Surveys
TIME PERIOD FOR SURVEYSTIME PERIOD FOR SURVEYS
• CROSS-SECTIONAL
• LONGITUDINAL
Comparison Of Data Collection Methods
Personal Telephone Mail
Data collection costs High Medium Low
Data collection time Medium Low High
Sample size for givenbudget
Small Medium Large
Data quantity perrespondent
High Medium Low
Reaches widelydispersed sample
No Maybe Yes
Reaches speciallocations
Yes Maybe No
Comparison Of Data Collection Methods
Personal Telephone MailInteraction withrespondents
Yes Yes No
Degree ofinterviewer bias
High Medium None
Severity ofnonresponse bias
Low Low High
Presentation ofvisual stimuli
Yes No Maybe
Fieldworkertraining required
Yes Yes No
Factors Determining Choice of Survey Method
• Researcher’s Resources and Objectives– Time horizon– Budget– Desired quality of data collected
• Generalizability vs. Completeness
Factors Determining Choice of Survey Method
• Respondent Characteristics– Incidence rate (% in target population)– Willingness to participate (suspicion, privacy,
interest)– Ability to participate (time, qualified)– Diversity of respondents (problems with
selection or response probabilities)
Factors Determining Choice of Survey Method
• Question Characteristics– Complexity of tasks (or props needed)– Amount of information per respondent (and
type of information needed)– Topic sensitivity (e.g., hygiene, charity
contributions, illicit behaviors)
Potential Survey Topics
• Attitudes– attitude comes before behavior– three parts:
• what a person knows about a topic– awareness (aided and unaided recall)
• how a person feels about a topic– positive/negative and intensity of feeling
• likelihood the person will take action based on the attitude
– past, present and future behavior
Potential Survey Topics
• Image– ask what characteristics are most important to
the subject– comparison across groups to reveal differences
• Decisions– What was the process used to make the decision?– What information sources were used?– What criteria was used for evaluation?
Potential Survey Topics
• Needs (or Wants)– Needs, desires or preferences– Find out relative importance of needs.
• Behavior– what respondents did or did not do– the timing of the behavior– the persistency of the behavior
Potential Survey Topics
• Lifestyles– lifestyle patterns identified by activities,
interests, opinions and possessions of the respondents
• Affiliations– family groups– reference groups
Potential Survey Topics
• Demographics– Variables such as age, gender, income, marital
status, education, employment, etc.– demographic groups often differ significantly
on issues– demographics can be used to identify market
segments
Survey Trends
• Over sampling of some populations
• Privacy issues
• New Technology– fax– e-mail– internet– virtual reality