Thursday, October 9, 2014
Chapter 6: Survey Research
-the most widely used collection data technique used in sociology.
Research Questions Appropriate for a Survey
-based on the positivist approach and the people who get asked are the respondents
-the questions are strongest when the answers given are measurable variables
we are warned because few are fully aware of the causal factors and that a researcher
wants to discover a respondents subjective understanding or informal theory. A
limitation: what a person says and what hate data says may differ. * view study
The Logic of Survey Research
What is a survey?
-they measure many variables, test many hypotheses, and infer temporal order from
questions about past behaviour, experiences or characteristics.
-it is then measured with statistical techniques
-it if often called correlational because they use questions as control variables to
approx the rigorous test for causality that experiementers achieve with physical
control over temporal order and alternative explanations.
Steps in Conducting a Survey - FIGURE 6.1 ON PAGE 137
-they use a deductive approach
Thursday, October 9, 2014
-Interview Schedule: set of questions read to the respondent by the interviewer who
records the responses.
Constructing the Questionnaire
Principles of Good Question Writing
-usually are integrated as a whole
-three principles are effective: keep it clear, simple and the respondents perspective in
-Look for questions that give the researcher reliable and valid measures. It will help
the respondents feel that their answers are meaningful and that they actually
understand. Extra care if they are heterogenous.
-although they face a dilemma: they want the respondent to hear exactly the same
questions and be clear, relevant and meaningful to all the respondents.
12 principles of Question Writing for a Survey
1. Avoid Jurgon, slang and abbreviations: target the grammar and vocabulary of those
2. Avoid ambiguity, confusion, and vagueness: a researcher may make implicit
assumptions. the use of indeﬁnite words or responses to categories.
3. Avoid emotional language: use neutral language and avoid baggage.
4. Avoid prestige bias: those who have high social status can colour how people
5. Avoid double-barrelled questions: consists of two or more questions joined together
6. Do not confuse beliefs with reality: do not confuse what a respondent believes with
what you believe: ex; a respondent may think a relationship exists but it is not a
empirical measure of variables in a relationship. * view concrete example.
7. Avoid leading questions or (loaded): is a question that leads the respondents to
choose one response over another because of its wording. positive or negative
8. Avoid asking questions that are beyond the respondents abilities: hence, do not ask
questions about past details and ask about questions that are up to date so they do
not use a full-ﬁller question*