Study Guides (390,000)
CA (150,000)
U of G (8,000)
SOAN (200)
Midterm

SOAN 2120 Study Guide - Midterm Guide: Social Desirability Bias, Random Assignment, Design Issues


Department
Sociology and Anthropology
Course Code
SOAN 2120
Professor
David Walters
Study Guide
Midterm

This preview shows pages 1-3. to view the full 12 pages of the document.
Survey Research
- the survey asks respondents about their beliefs, opinions, characteristics, and
past or present behaviour often called correlational
- researchers usually ask about many things at home time, measure many
variables and test several hypotheses in a single survey
- the following can be asked in a survey:
o behaviour
o attitudes/beliefs/opinions
o characteristics
o expectations
o self-classification (what do you consider yourself… i.e. liberal)
o knowledge
- limitation
o it provides data only of what a person says, and this may differ from
what he or she actually does
Steps in the Process of Survey Research
Step One:
- develop hypotheses
- decide on type of survey
- write survey questions
- decide on response categories
- design layout
Step Two:
- plan how to record data
- test survey instrument
Step Three:
- decide on target population
- get sampling frame
- decide on sampling size
- select sample
Step Four:
- Locate respondents
- Conduct interviews
- Carefully record data
Step Five:
- enter data into computers
- recheck all data
- perform statistical analysis on data

Only pages 1-3 are available for preview. Some parts have been intentionally blurred.

Step Six:
- describe methods and findings in research report
- present findings to others for critique and evaluation
Principles of Good Question Writing
- keep it clear, keep it simple, keep the respondent’s perspective in mind
- 12 things to avoid when writing survey questions
o jargon, slang, abbreviations
o vagueness
o emotional language
o prestige bias
o double-barreled questions
o beliefs as real
o leading questions
o issues beyond respondent capabilities
o false premises
o distant future intentions
o double negatives
o unbalanced responses
Types of Questions and Response Categories
Threatening Questions
- researchers must ask with extreme care
- these questions are part of self-presentation and ego protection
- respondents may underreport or self-censor reports or behaviour that they
believe are a violation of social norms
- masturbation is considered ‘very uneasy’ while sports activity is the ‘least
uneasy’
Socially Desirable Questions
- social desirability bias occurs when respondents distort answers to make
their reports conform to social norms
Knowledge Questions
- many have inaccurate factual knowledge
Skip or Contingency Questions
- avoid asking questions that are irrelevant for a respondent
- contingency question: two-or more part question
- contingency questions select respondents for whom a second question is
relevant
Open Versus Closed Questions
- open-ended question: asks a question to which a respondent can give any
answer

Only pages 1-3 are available for preview. Some parts have been intentionally blurred.

- closed-ended question: both asks a question and gives the respondent fixed
responses from which to choose (i.e. good, very good, poor, very poor)
o large scale surveys have closed-ended questions because they are
quicker and easier for respondents and researchers
- Advantages and Disadvantages
Open
Closed
Pros
- unlimited number of
answers
- respondents can answer in
detail
- reveal a person’s logic
- easier and quicker for
respondents to answer
- easier to compare
- easy to code and analyze
Cons
- different degrees of detail
- responses may be
irrelevant
- coding responses is difficult
- respondents with no opinion
can answer anyways
- respondents may get
frustration
- misinterpretation of question
can go unnoticed
- Nonattitudes and Middle Positions
o Two types of errors can be made:
Accepting a middle choice or ‘no attitude’ respondent when
responds hold a nonnetrual opinion
Forcing respondents to choose a position on an issue when
they have no opinion about it
o 3 Kinds of Attitude Questions
Standard Format
Does not offer a ‘don’t know’ response – agree or
disagree
Quasi-Filter
Offers a ‘don’t know’ response
Full filter
Asks if they have an opinion if yes, explain
- Agree/Disagree, Rankings or Ratings
o Less well-educated respondents are more likely to agree with a
statement, whereas forced-choice alternatives encourage thought and
avoid the response set bias
o Better to ask respondents to choose among alternatives by ranking
instead of rating
- use a funnel sequence
o ask more general questions before specific questions
You're Reading a Preview

Unlock to view full version