Textbook Notes (280,000)
CA (170,000)
U of G (10,000)
SOAN (400)
SOAN 2120 (100)
Chapter 6

SOAN 2120 Chapter Notes - Chapter 6: Social Desirability Bias, Jargon, Design Issues

Sociology and Anthropology
Course Code
SOAN 2120
William Walters

This preview shows pages 1-2. to view the full 7 pages of the document.
Chapter 6 - Survey Research
*See Figure 6.1 (page 137) - Steps in the Process of Survey Research
Constructing the Questionnaire
Principles of Good Question Writing
-three principles for effective survey questions:
-keep it clear
-keep it simple
-keep the respondent’s perspective in mind
-good survey questions give researcher valid and reliable measures
-help respondents feel they understand question and their responses are meaningful
1. Avoid jargon, slang and abbreviations
-jargon and technical terms are field specific terms that laymen may not understand
-slang is jargon of a subculture
-unless specialized population is being observed
2. Avoid ambiguity, confusion and vagueness
3. Avoid emotional language
-strong emotional connotations can colour how respondents hear and answer survey questions
4. Avoid prestige bias
-titles or positions in society can carry prestige or status
-avoid associating a statement with a prestigious person nor group
5. Avoid double-barreled questions
-double-barreled question: two or more questions joined together (makes a respondent’s answer
6. Do not confuse beliefs with reality
-don’t confuse the respondent’s beliefs with what you, a researcher, measures
7. Avoid leading questions
-leading (or loaded) question: one that leads respondent to choose one response over another by its
-can be stated to get either negative or positive answers
8. Avoid asking questions that are beyond respondents’ capabilities
-asking something few respondents know frustrated respondents and produces poor-quality responses
9. Avoid false premises
-don’t begin question with a premise with which respondents may not agree and then ask about choices
regarding it
10. Avoid asking about intentions in the distant future
-avoid asking about hypothetical circumstances far in the future
-responses poor predictors of behaviour removed far from current situation
11. Avoid double negatives
-double negatives are confusing

Only pages 1-2 are available for preview. Some parts have been intentionally blurred.

12. Avoid overlapping or unbalanced response categories
-mutually exclusive: response categories don’t overlap
-exhaustive: every respondent has a choice
-balance: ex. by offering polar opposites
Types of Questions and Response Categories
Threatening Questions
-survey researchers sometimes ask about sensitive issues or issues respondents believe threaten their
presentation of self
-ex. drug use, mental health problems, deviant behaviour
-might be reluctant to answer these truthfully or at all
-be extra cautious about asking these
-respondents want to present positive images of themselves
-might be ashamed to answer these questions
-may underreport or self-censor answers
-ex. likely to underreport having illness or disability, engaging in deviant/illegal behaviour, revealing
financial status
-several techniques to increase truthful answers to threatening questions
-ex. should ask only after a warm-up (after rapport has been built with respondent and tell
respondent they would like honest answers)
-can phrase question in “enhanced way”
-can embed threatening response with more serious activities to make their behaviour seem less
Socially Desirable Questions
-social desirability bias: when respondents distort answers to make their reports conform to social
-tend to overreport being cultured, giving money to charity, having good marriage, loving their
children, etc
-questionnaire writers try to reduce social desirability bias by phrasing questions to make norm violation
appear less objectionable and present wider range of “acceptable” behaviour
-can also offer multiple response categories to give respondents “face-saving” alternatives
Knowledge Questions
-can be threatening because people don’t want to appear ignorant
-can be phrased in a way to make respondents feel more comfortable
-ex. “How much, if anything, have you heard about…”
Skip or Contingency Questions
-contingency question (or “screen” or “skip” questions: two- (or more) part question
-first part of question determines which of two different questions a respondent next receives
-select respondents for whom a second question is relevant
Open vs. Closed Questions
-open-ended question: unstructured, free response, can give any answer
-close-ended: structured, fixed response, choose from fixed responses to choose
-one or the other is most appropriate in different situations
You're Reading a Preview

Unlock to view full version