Sampling frame generally high in coverage error12.Be familiar with the rules for good survey questions. See lecture slides.a.1. Create an outline of topics, based on research objectivesb.2. Where appropriate use or adopt questions from previous research c.3. Pay particular attention to languaged.4. Questions vs statementse.5. Open-ended v closed-ended questionsf.6. Ask questions that R is competent to anserg.7. Ask question that R is willing to answerh.8. Short items are goodi.9. Avoid double negative and negatie items
j.10.Make items clear; avoid or define ambiguous termsk.11. Avoid baised termsl.12. Tap into variationm. 13. Use approximately (or some qualifier) when appropriaten.14. When using ordinal levels measures: i.consider rangesii.equal distance but for anchorso.15. Include major categories of your population unless theoretically or empirically justifiedp.16. If using semantic differential make sure ends are polar opposites q.17. Not a testr.18. Avoid double barreled questionss.19. Ask relevant and appropriate questions 13.What are the strengths and weakness of survey research?a.Strengths:i.Generalization to populations1.Provides accurate estimates of population characteristic ii.Versatility 1.Can cover a wide range of topics2.Self-reports may provide only viable measures of attitudes, beliefs and valuesiii.Efficiency1.Numerous research questions can be addressed in the same survey2.May yield unanticipated findings 3.Secondary analysis of surveys is cost efficientb.Weakness i.Provides relatively weak evidence about casual relationshipsii.Sampling issuesiii.Fails to capture context of social lifeiv.Measurement issues 1.Measures what people say, not what they do2.Susceptible to reactive measurement effects3.Does not provide a good understanding of the context of behavior 14.What are the strengths and weaknesses of experimental research?15.Be familiar with the different threats to internal validity that can plague experiments.a.Ideal for understanding the cause of human behavior (high internal validity) but limited to what can be manipulated (low external validity)a.Selection biasb.Experimental mortality c.Historical eventsd.Maturation e.(Repeated) testingf.Instrumentation g.Statistical regression to the meanh.Experimenter bias
You've reached the end of your free preview.
Want to read all 8 pages?
- Spring '19
- researcher, internal validity, stratified random sampling