Chapter 7 Designing Surveys and Data Collection Instruments © 2008 Thomson, a part of the Thomson Corporation. Thomson, the Star logo, and Atomic Dog are trademarks used herein under license. All rights reserved. Survey research • Typically used in descriptive/conclusive research • Answers questions such as Who? Where? When? How? Why? © 2008 Thomson, a part of the Thomson Corporation. Thomson, the Star logo, and Atomic Dog are trademarks used herein under license. All rights reserved. Three Data Collection Methods in Survey Research • Questionnaires: by mail, email, internet • Advantages • i. Inexpensive • ii. Can be confidential or anonymous • iii. Easy to score most items • iv. Standardized items and procedures • • b. • v. Response rate may be small • vi. Cannot probe or explain items • vii. Only used by people who can read Disadvantages • Interviews • Telephones © 2008 Thomson, a part of the Thomson Corporation. Thomson, the Star logo, and Atomic Dog are trademarks used herein under license. All rights reserved. Imagine this scenario! • If a survey is done for Albertson’s stores and the data show that 36% of the respondents rate Albertson’s checkout as the best in town, this 36% figure should be an accurate reflection of feelings among the population at large. True or False or Maybe © 2008 Thomson, a part of the Thomson Corporation. Thomson, the Star logo, and Atomic Dog are trademarks used herein under license. All rights reserved. Types of Survey Error Total survey error Random sampling error Measurement error Processing error Systematic error Surrogate information error Interviewer error Response error Nonresponse bias Instrument bias Sample design error Selection/ error Frame/ coverage error Population specification error © 2008 Thomson, a part of the Thomson Corporation. Thomson, the Star logo, and Atomic Dog are trademarks used herein under license. All rights reserved. Sources of Survey Error • 1. Sampling error: The result of surveying only some, and not all, elements of the survey population • 2. Coverage error: The result of not allowing all members of the survey population have on equal or known nonzero chance of being sampled for participation in the survey • 3. Measurement Error: The result of poor question wording or questions being presented in such a way that inaccurate and interpretable answers are obtained • 4. Nonresponse error/bias: The result of people who respond to a survey being different from sampled individuals who do not respond, in a way relevant to the study © 2008 Thomson, a part of the Thomson Corporation. Thomson, the Star logo, and Atomic Dog are trademarks used herein under license. All rights reserved. Types of Errors (contd.,): Measurement Errors Surrogate Information Error - A discrepancy between the information actually required and the information being sought by the researcher. Interviewer Error - Interactions between the interviewer and respondent may influence answers. Instrument Bias - Problems with the measurement instrument or questionnaire. Processing Error Mistakes when entering information from the survey documents. © 2008 Thomson, a part of the Thomson Corporation. Thomson, the Star logo, and Atomic Dog are trademarks used herein under license. All rights reserved. Instrument Bias • Q: What is your income? Response bias © 2008 Thomson, a part of the Thomson Corporation. Thomson, the Star logo, and Atomic Dog are trademarks used herein under license. All rights reserved. Mail Survey • In December 2004, a mail survey was sent out to 1000 managers of industrial equipment to gauge their interest in a new product. 40 questionnaires were returned indicating a response rate of 4%. 90% of these responses were favorable toward the new equipment. So, the marketing managers have decided to launch the new product. • What are your thoughts? © 2008 Thomson, a part of the Thomson Corporation. Thomson, the Star logo, and Atomic Dog are trademarks used herein under license. All rights reserved. Question • In the past three months, how many times have you eaten at fast food places? • In the past three years, how many times have you been involved in automobile accidents? © 2008 Thomson, a part of the Thomson Corporation. Thomson, the Star logo, and Atomic Dog are trademarks used herein under license. All rights reserved. Types of Errors: Measurement Errors • Non-response bias Are those who did respond different from those who did not respond? • Response bias - A tendency for people to answer a particular question in a certain way deliberate falsification unconscious misrepresentation © 2008 Thomson, a part of the Thomson Corporation. Thomson, the Star logo, and Atomic Dog are trademarks used herein under license. All rights reserved. Function and Importance of Questionnaires • A questionnaire is a formalized method for collecting data from respondents. • questionnaires measure: past purchase and usage behavior attitudes and opinions, intentions, awareness and knowledge, ownership variety of respondent characteristics • To avoid measurement error, questions must pose realistic alternatives or trade-offs. © 2008 Thomson, a part of the Thomson Corporation. Thomson, the Star logo, and Atomic Dog are trademarks used herein under license. All rights reserved. Questionnaire Components A. identification data, such as: respondent's name address phone number B. request for cooperation C. instructions D. information sought (longest portion of the questionnaire) E. classification data (characteristics of the respondent, primarily "geodemographic" data) © 2008 Thomson, a part of the Thomson Corporation. Thomson, the Star logo, and Atomic Dog are trademarks used herein under license. All rights reserved. D: The blueprint • Blueprint Table:. • This table serves as an organizer that frames the major content categories, attitudes, beliefs to be assessed. The proportion of the tasks or items that will be included on the instrument or overall performance should correspond roughly with how important the domain is relative to other domains. © 2008 Thomson, a part of the Thomson Corporation. Thomson, the Star logo, and Atomic Dog are trademarks used herein under license. All rights reserved. Example: Life Satisfaction survey (100 items) Number of items Content Base Category Emotions such as: Guilt, Anger, Depression and Hopelessness, and Anxiety 30 Cognitive Functioning 25 Decision Making 15 The Capacity for Pleasure 10 Movement Toward Goals 10 Self Care and Daily Functioning 5 © 2008 Thomson, a part of the Thomson Corporation. Thomson, the Star logo, and Atomic Dog are trademarks used herein under license. All rights reserved. © 2008 Thomson, a part of the Thomson Corporation. Thomson, the Star logo, and Atomic Dog are trademarks used herein under license. All rights reserved. Questionnaire Design Steps/Best practices in questionnaire design: review preliminar y considera tions decide question content decide response format decide question wording decide question sequence decide physical characteri stics pretest and revise make final draft © 2008 Thomson, a part of the Thomson Corporation. Thomson, the Star logo, and Atomic Dog are trademarks used herein under license. All rights reserved. Questionnaire Design (cont.) Preliminary Considerations • review issues already decided by research design: detailed listing of information needs sources of data definition of the target population detailed sampling plan specification of scales and communication media visualization of the research findings • link between data-collection and information needs Each question should relate to a specific information need. © 2008 Thomson, a part of the Thomson Corporation. Thomson, the Star logo, and Atomic Dog are trademarks used herein under license. All rights reserved. Questionnaire Design (cont.) Question content Consider ways question will not be answered accurately or at all: 1. respondent unable to provide the data 2. respondent uninformed 3. respondent forgetful 4. respondent misremembers 5. respondent unwilling to respond accurately © 2008 Thomson, a part of the Thomson Corporation. Thomson, the Star logo, and Atomic Dog are trademarks used herein under license. All rights reserved. Response Format 1. open-ended questions – free-response, usually in respondent's own words 2. multiple-response questions – requires choice from an explicit list of options 3. dichotomous questions – offers choice of two responses and often a third neutral alternative © 2008 Thomson, a part of the Thomson Corporation. Thomson, the Star logo, and Atomic Dog are trademarks used herein under license. All rights reserved. Question Types What do you feel about buying Tide Detergent? ___________________________________________________________________________ ___________________________________________________________________________ ___________________________________________________________________________ Do you own a(n) __________________ at home? Personal Computer VCR Entertainment Center Yes Yes Yes No No No What is your age? (Please check one) Under 18 r 18 - 24 r 25 - 34 r 35- 44 r 45 - 54 r 55 - 64 r 65 and over r © 2008 Thomson, a part of the Thomson Corporation. Thomson, the Star logo, and Atomic Dog are trademarks used herein under license. All rights reserved. Decide on Question Wording • use simple language • use unambiguous words does the word truly convey what the researchers intended? can respondents extrapolate any alternative meaning? if so, does context help make the intended meaning clear? is there any word with similar pronunciation or spelling that could be confused with it? could we use a simpler word or phrase instead? © 2008 Thomson, a part of the Thomson Corporation. Thomson, the Star logo, and Atomic Dog are trademarks used herein under license. All rights reserved. Decide on Question Wording (cont.) • avoid leading questions • avoid biasing questions • avoid implicit alternatives • avoid implicit assumptions • avoid estimates • avoid double-barreled questions • consider frame of reference © 2008 Thomson, a part of the Thomson Corporation. Thomson, the Star logo, and Atomic Dog are trademarks used herein under license. All rights reserved. Decide on Question Sequence • sequence questions to retain respondent interest without introducing bias • use intriguing, readily understood opening question • ask general questions first • place more difficult questions in the middle of the sequence • position sensitive, threatening, and demographic questions at the end • arrange questions in logical order © 2008 Thomson, a part of the Thomson Corporation. Thomson, the Star logo, and Atomic Dog are trademarks used herein under license. All rights reserved. Questionnaire Design Steps/Best practices in questionnaire design: review preliminar y considera tions decide question content decide response format decide question wording decide question sequence decide physical characteri stics/ layout pretest and revise make final draft © 2008 Thomson, a part of the Thomson Corporation. Thomson, the Star logo, and Atomic Dog are trademarks used herein under license. All rights reserved. Issues to Address When Evaluating the Questionnaire Layout •Is each Question Necessary/ Is the Questionnaire Too Long? • Will the Questions Provide the Desired Information to Accomplish the Research Objectives? • Look at the Appearance of Mail and Self-Administered Questionnaires: - Avoid a Cluttered Look - Allow Plenty of Space for Open-Ended Responses -Instructions Printed Within The Questionnaire Should Be In Capital Letters (or set apart in some manner) PleaseTurn Page © 2008 Thomson, a part of the Thomson Corporation. Thomson, the Star logo, and Atomic Dog are trademarks used herein under license. All rights reserved. Survey implementation • First Contact: Prenotice Letter/Incentive • Second Contact: The Questionnaire Mailout • Third Contact: The Postcard Thank You/Reminder • Fourth Contact: The First Replacement questionnaire • Fifth Contact: Invoking Special Procedures • © 2008 Thomson, a part of the Thomson Corporation. Thomson, the Star logo, and Atomic Dog are trademarks used herein under license. All rights reserved. Observational Forms (alternative to surveys) • make explicit the types of observations to be made • make explicit how observations are to be measured • design should flow logically from the listing of information needs: who is to be observed? what is to be observed? when is observation to be made? where should observations be made? © 2008 Thomson, a part of the Thomson Corporation. Thomson, the Star logo, and Atomic Dog are trademarks used herein under license. All rights reserved. Acceptability Totally unacceptable, Unacceptable, Slightly unacceptable, Neutral, Slightly acceptable, Acceptable, Perfectly acceptable Agreement Strongly disagree, Disagree, Somewhat disagree, Neither agree or disagree, Somewhat agree, Agree, Strongly agree Amount of Use Never use, Almost never, Occasionally/Sometimes, Almost every time, Frequently use Appropriateness Absolutely inappropriate, Inappropriate, Slightly inappropriate, Neutral, Slightly appropriate, Appropriate, Absolutely appropriate Awareness Not at all aware, Slightly aware, Somewhat aware, Moderately aware, Extremely aware Beliefs Very untrue of what I believe, Untrue of what I believe, Somewhat untrue of what I believe, Neutral, Somewhat true of what I believe, True of what I believe, Very true of what I believe Concern Not at all concerned, Slightly concerned, Somewhat concerned, Moderately concerned, Extremely concerned Familiarity Not at all familiar, Slightly familiar, Somewhat familiar, Moderately familiar, Extremely familiar Frequency Never, Rarely, Sometimes, Often, Always Importance Not at all importance, Slightly important, Somewhat important, Very important, Extremely important Likelihood Not at all likely, Slightly likely, Somewhat likely, Moderately likely, Very likely Priority Not a priority, Low priority, Medium priority, High priority, Essential Probability Not probable, Somewhat improbable, Neutral, Somewhat probable, Very probable Quality Very poor, Poor, Fair, Good, Excellent Reflective Very untrue of me, Untrue of me, Somewhat untrue of me, Neutral, Somewhat true of me, True of me, Very true of me Satisfaction Completely dissatisfied, Mostly dissatisfied, Somewhat dissatisfied, Neither satisfied or dissatisfied, Somewhat satisfied, Mostly satisfied, Completely satisfied Satisfaction Not at all satisfied, Slightly satisfied, Moderately satisfied, Very satisfied, Extremely satisfied © 2008 Thomson, a part of the Thomson Corporation. Thomson, the Star logo, and Atomic Dog are trademarks used herein under license. All rights reserved.