Stat472 Survey design Questionnaire Design Issues Format and

Stat-472 Survey design Questionnaire Design Issues Format and Layout, Question Format, TDM

Format and Layout There are two format or layout issues: 1. The overall physical layout of the questionnaire 2. The format of questions and response • Layout: should be clear, neat and easy to follow àGive each question a number àMake a cover sheet àAdd the Time, Date of interview àQuestionnaire identification Number Good layout improves accuracy

A few cents saved in postage or printing will ultimately cost more in terms of lower validity due to a lower response rate or of confusion of interviewers and respondents. àGive respondent instructions àPrint instruction in different style from the questions àAlways end with “Thank You for your participations” àQuestionnaire should leave the respondent with a good feeling

Question Format • Researcher should decide on a format for the questions: àCircle responses, Check box, Fill dot, put an X in a blank àListing responses down page àUse arrows and instructions for contingency questions àUse Matrix questions for a series of questions

Visual aids are also helpful. For example, hand out thermometer-like drawings to respondents when asking about how warm or cool they feel

A matrix question(or grid question) is compact way to present a series of questions using the same response categories. It saves space and makes it easier for the respondent or interviewer to note answers for the same response categories

example Do you think it is too easy or too difficult to get a divorce, or it is about right? Horizontal [ ] Too Easy [ ] Too Difficult [ ] About right Vertical [ ] Too easy [ ] Too Difficult [ ] About Right

Matrix Questions Strongly Disagree DK Agree Strongly Agree The Teacher Talks too Fast [ ] [ ] [ ] I learned a lot in this class [ ] [ ] [ ] The tests are very easy [ ] [ ] [ ] The teacher tells many jokes [ ] [ ] [ ] The teacher is organized [ ] [ ] [ ]

Total Design Method (TDM) • Developed to improve mail and telephone surveys by theoretical and practical parts. • The theory says: a survey is a social interaction in which respondents act on the basis of what they expected to receive in exchange for their cooperation (they cooperate when social costs are low, the expected benefit exceeds the perceived costs and when researchers create a feeling of trust. ) • The practical parts repeats the advice given here about good question wording and questionnaire design. • A good survey design has pilot tests, tests minimize personal costs, costs and required minimal effort and time to respondent and builds a trust with the respondent