Questionnaire Design Clinic Spring 2010 Seminar Series Survey

Questionnaire Design Clinic Spring 2010 Seminar Series Survey Research Laboratory University of Illinois www. srl. uic. edu 1

Cognitive Steps in Answering Questions 1. Understand question. 2. Search memory for information. 3. Integrate information into summary judgment. 4. Translate judgment onto response alternatives. Survey Research Laboratory 2

Survey Intro/Cover Letter Introduction should indicate: § § § who is conducting the survey the topics to be covered in the survey an assurance of confidentiality any IRB stipulations whether you offer how long it will take depends on mode, topic, population Survey Research Laboratory 3

Survey Intro/Cover Letter example The goal of this survey is to understand the patterns of interactions between the groups and individuals who will be attending …. . The results of this survey will be used in part for University of Illinois at Chicago research purposes. Ultimately, we hope to provide a visual representation of the social network structure of various interested parties in the ______ region. By completing this survey, you are agreeing to participate in the research. Survey Research Laboratory 4

Cover letter - example The Department of _____at the University of Illinois at Chicago, is conducting to understand the patterns of interactions between the groups and individuals …. … The research team is headed by… You are receiving this questionnaire because you are. . . It should take you about 10 minutes to complete the questionnaire. Please respond by August 1. Participation in this survey is voluntary, and you may choose to stop answering questions at any time. Your choice to participate or not will have no effect on your course grade or on your relation with the University of Illinois. Your response is anonymous as we will not be asking for any personally identifying information – please do not write your name anywhere on this questionnaire. We plan to use the results of this survey in part for University of Illinois at Chicago research purposes. Ultimately, we hope to provide a visual representation of the social network structure of various interested parties in the Calumet region. ……. Only aggregated results will be presented. Completing this questionnaire indicates your consent to participate in the survey. Please contact ____ if you have any questions. If you have any questions about your rights as a research participant, please contact the IRB (OPRS) at…. Thank you for your participation! Names Affiliation Email / phone number 5 Survey Research Laboratory

Ordering the Questions § First questions should be directly related to the topic as described in the introduction or advance/cover letter § Put demographic questions at the end § Leave objectionable questions (e. g. , income) for the end Survey Research Laboratory 6

Characteristics of a good question § One that yields a truthful, accurate answer § One that asks for one answer on one dimension § One that accommodates all possible contingencies of response § One that uses specific, simple language § One that produces variability in response § One that minimizes social desirability § One that is pretested Survey Research Laboratory 7

Acquiescence - example § Would you be interested in learning about courses from the past experiences of other students who have taken them? • Yes • No § Do you feel ready to apply what you have learned today? • Yes • No Survey Research Laboratory 8

Acquiescence - example § Administrators (e. g. , principal, assistant principal, dean of students ) at this school ……………… a. Collaborate with faculty and staff to make the school run effectively. • Strongly Agree, Neutral, Disagree, Strongly Disagree Survey Research Laboratory 9

Acquiescence Form A (agree I’s version): Individuals are more to blame than social conditions for crime and lawlessness in this country. Form A (Agree I’s) Agree (I): 59. 6% Form B (agree SCs version): Social conditions are more to blame than individuals for crime and lawlessness in this country. Form B (Agree SCs) Agree (SC): 56. 8% Disagree (SC): 40. 4 100 (473) Disagree (I): 43. 2 100 (472) Survey Research Laboratory 10

Acquiescence - example § Do you feel ready to apply what you have learned today? • Yes • No § Do you feel ready to apply what you have learned today, or do you not feel ready to do that? • Feel ready • Do not feel ready Survey Research Laboratory 11

Acquiescence - example § Would you be interested in learning about courses from the past experiences of other students who have taken them? • Yes • No § How interested are you in learning about courses from the past experiences of other students who have taken them? • Not at all interested, Slightly interested, Moderately interested, Very interested, Extremely interested Survey Research Laboratory 12

Characteristics of a good question example Administrators (e. g. , principal, assistant principal, dean of students ) at this school collaborate with faculty and staff to make the school run effectively. Strongly Agree, Neutral, Disagree, Strongly Disagree How well do administrators (e. g. , principal, assistant principal, dean of students ) at this school collaborate with faculty and staff to make the school run effectively? Not well at all, Slightly well, Moderately well, Very well, Extremely well Survey Research Laboratory 13

Characteristics of a good question § § One that yields a truthful, accurate answer One that asks for one answer on one dimension One that uses specific, simple language One that accommodates all possible contingencies of response § One that produces variability in response § One that minimizes social desirability § One that is pretested Survey Research Laboratory 14

Characteristics of a good question example The instructor was able to answer questions clearly and concisely in this course. Participants materials (high quality; easy to follow; helpful, etc) Survey Research Laboratory 15

Characteristics of a good question § § One that yields a truthful, accurate answer One that asks for one answer on one dimension One that uses specific, simple language One that accommodates all possible contingencies of response § One that produces variability in response § One that minimizes social desirability § One that is pretested Survey Research Laboratory 16

Characteristics of a good question example Indicate how the amount of use of the following types of classroom assessment have changed in your classroom Assessment during instruction to determine: degree of comprehension on part of an individual student…. . Survey Research Laboratory 17

Characteristics of a good question example Has your use of the following types of assessments increased, decreased or stayed the same? Assessment during instruction to find out how well a student has understood the material…. . Survey Research Laboratory 18

Characteristics of a good question § § One that yields a truthful, accurate answer One that asks for one answer on one dimension One that uses specific, simple language One that accommodates all possible contingencies of response § One that produces variability in response § One that minimizes social desirability § One that is pretested Survey Research Laboratory 19

Characteristics of a good question example How often do you go talk with your academic advisor in the Department of Communication? Multiple times a semester Once a semester For emergencies only Once a school year How often do you get in touch with your academic advisor in the Department of Communication? • • • Never Once a school year or less often Once a semester Multiple times a semester or more often In emergencies only Survey Research Laboratory 20

Characteristics of a good question § § One that yields a truthful, accurate answer One that asks for one answer on one dimension One that uses specific, simple language One that accommodates all possible contingencies of response § One that produces variability in response § One that minimizes social desirability § One that is pretested Survey Research Laboratory 21

Pretesting § Preferable to test the questionnaires with people like those in your main study population § Test in same mode to be used for the study § Consider cognitive pretesting Survey Research Laboratory 22

Other characteristics of a good question § Specify who, what, when, where and how. § Over what period of time? What’s included? § Provide cues Survey Research Laboratory 23

Other characteristics of a good question - example § Does this table of contents cover recent developments? § Does this table of contents cover recent research / topics in this field, or does it not? § How many people live in your household? Please include yourself in the count, and also those who people who live with you but are away temporarily. Don’t forget to count any babies and children. § How many years have you been enrolled in a college or University? § Not counting summers, for how many years have you been enrolled as a student in a college or university? Please count each year that you were enrolled even if these are not consecutive years. Survey Research Laboratory 24

Open vs. Closed Questions Closed questions are usually better § § Easier for the respondent Less coding later Better to have respondent do categorizing Categories help define the question Survey Research Laboratory 25

Open vs. Closed Questions example § What grade level(s) have you worked with in the past 12 months? (e. g. , 6 th, 7 th, 8 th) _________ § Which grade level(s) have you worked with in the past 12 months? (please select all that apply). q Grade 6 q Grade 7 q Grade 8 § How often do you send text messages? _________ § How often do you send text messages? You can answer in any of the three different ways below, whichever one would be most accurate for you. You can tell us how many times per DAY you send text messages, or how many times per WEEK you send text messages, or how many times per MONTH you send text messages. Survey Research Laboratory 26

Disadvantages of Closed Questions Categories may be leading to respondents May make it too easy to answer without thinking May limit spontaneity Not best when § asking for frequency of sensitive behaviors § there are numerous possible responses • role of initial question development and pretesting Survey Research Laboratory 27

Response Options § Response categories should be consistent with the question § Categories should typically be exhaustive, including every possible answer § Categories should be mutually exclusive § “Don’t know” category § usually not appropriate § knowledge questions Survey Research Laboratory 28

Response Options - example Did the facilitator have a thorough understanding of the subject matter? Very Poor, Needs Improvement, Okay, Good, Excellent How good or bad was the facilitator’s understanding of the subject matter? Extremely good, Somewhat good, Neither good nor bad, Somewhat bad , Extremely bad Survey Research Laboratory 29

Response Options - example Please indicate below which of the following organizations you interact with on a regular basis. Please check all that apply. Organization A • I regularly exchange ideas or advice with members of this organization. • I regularly collaborate on projects with members of this organization. • I have personal friends in this organization. • I provide funding or receive funding from this organization. How often do you exchange ideas or advice with members of each of the following organizations? Organization A: Never, Rarely, Sometimes, Very Often, Extremely Often Organization B: Never, Rarely, Sometimes, Very Often, Extremely Often Do you have at least one personal friend in each of the following organizations, or do you not? Organization A: Have, Do not have Organization B: Have, Do not have Survey Research Laboratory 30

Response Options § 5 -7 response options § General guidelines • number of scale points (sensitivity vs. meaning) • labeling of scale points Survey Research Laboratory 31

Response Option Labels example How would you rate the instructor’s ability to explain the material in this course? Extremely poor, Somewhat poor, Fair, Somewhat good, Extremely good (or Excellent) Survey Research Laboratory 32

Physical Format Checklist § Number all questions sequentially § Use large, clear type; don’t crowd § ‘White space: ’ Place more blank space between questions than between subcomponents of questions § List answer categories vertically instead of horizontally § Avoid double/triple ‘banking’ of response choices § Be consistent with direction of response categories § Be consistent with placement of response categories Survey Research Laboratory 33

Physical Format Checklist (continued) § Don’t split questions across pages. If necessary (e. g. , question requires 1. 5 pages), restate question and response categories on next page § Put special instructions on questionnaire as needed, next to question § Distinguish directions from questions – be consistent Survey Research Laboratory 34

Key References To Get Started Dillman, Don. Mail and Internet Surveys: The Tailored Design Method. New York: John, Wiley & Sons, Inc, 1999. Bradburn, N, Sudman, S. and Wansink, B. Asking Questions: The Definitive Guide to Questionnaire design – for Market Research, Political Polls, and Social and Health Questionnaires. San Francisco: Jossey Bass, 2004. See also past issues of Public Opinion Quarterly Survey Research Laboratory 35

Other References Schuman, H. , & Presser, S. (1981). Questions and answers in attitude surveys. New York: Academic Press. Krosnick, J. A. , Holbrook, A. L. , Berent, M. K. , Carson, R. T. , Hanemann, W. M. , Kopp, R. J. , Mitchell, R. C. , et al. (2002). The impact of "No Opinion" response options on data quality: Nonattitude reduction or an invitation to satisfice? Public Opinion Quarterly, 66, 371– 403. Saris, W. E. , Krosnick, J. A. , & Schaeffer, E. M. (2005). Comparing questions with agree/disagree response options to questions with construct-specific response options. Unpublished manuscript, Political, Social, Cultural Sciences, University of Amsterdam. Survey Research Laboratory 36