Chapter 3 Research design and cross-sectional design PDF

Preview

Summary

Download Chapter 3 Research design and cross-sectional design PDF

Description

Chapter 3.1: Research Designs 1. What are the three most prominent criteria with which to evaluate the quality of social research, particularly the quantitative perspective (objectivist ontology/positivist epistemology)? Provide a brief definition or explanation of each of them. - Reliability: consistency of a measure - Replicability: the ability to be replicated by others with the same results - Validity: the accuracy of a measure - measurement validity - internal validity - external validity - ecological validity 2. What other term is commonly used to refer to internal validity? Causality 3. What other term is commonly used to refer to external validity? Generalizability 4. Why is replication important? How is it related to transparency and good standards? When research is duplicated and provides the same or similar results as the original study, the findings gain increased legitimacy. If a researcher can reproduce the findings of a study, it suggests that such findings are more likely to be generalizable to a broader population. 5. What does it mean that a measurement is reliable? Or, in other words, what needs to happen for a measurement to be reliable? The consistency with which a method measures something is referred to as its reliability. The measurement is regarded as trustworthy if the same result can be consistently obtained by using the same procedures under the same conditions. 6. The experimental design is the yardstick against which all other types of research designs are assessed. Why is experimental design the “gold standard” of research methods, particularly those that follow an objectivist/positivist approach? Experimental research is commonly employed as a yardstick because it instills a high level of confidence in the robustness and reliability of causal outcomes. It is also possible to say that experimental research, real experiments, in other words, have a high level of internal validity. 7. What are the four main elements of an experiment? See Fig. 3.1 to help you answer this question. First, develop a research question and/or hypothesis Secondly, pick a research method (interviews, surveys, tests, etc.) Thirdly, gather data using the methods chosen Finally, examine the results to evaluate if the hypothesis has been proven or disproven. 8. What is the difference between a field experiment and a laboratory experiment? Field experiments: experiments that happen in real life settings

Laboratory experiments: experiments that occur in laboratories or an environment controlled by the researcher. 9. What is the difference between a quasi-experiment and a natural experiment? Quasi-experiments: experiments that have some experimental design features but cannot meet all of the internal validity standards. Natural experiments: experiments in the context that they include the notion of manipulation of a social context. Chapter 3.2 (cross-sectional design) 1. What is the structure of a cross-sectional design? Choose among these options: ■ A (preferably) large number of cases observed at a single point in time ■ A (preferably) large number of cases observed at two or more points in time 2. What can be considered a “case” in a cross-sectional design? A case in a cross-sectional design can be a variation “in respect of people, families, organizations, nation states” (Bryman 59). 3. Is there a limit to the number of variables that you can measure in a cross-sectional design? No, there are no restrictions 4. How good (i.e. strong) is the internal validity of cross-sectional designs? Explain why. It is actually weak and poor. This is attributed to the difficulty that researchers have in determining the direction of causality from the data supplied by the cross-sectional design. These designs provide correlations rather than outcomes from which causal inferences may be drawn with certainty. 5. How good (i.e. strong) is the external validity of cross-sectional designs? Explain why. The external validity of a cross-sectional design is strong and good in the case where the sample from which data is collected is chosen at random. 6. Provide an example of a cross-sectional design combined with a quantitative strategy (you can describe examples from the textbook or from other sources) Determining the percentage of patients serviced by a health clinic in a given year who have high cholesterol (Simkus). 7. Provide an example of a cross-sectional design combined with a qualitative strategy (you can describe examples from the textbook or from other sources) Beardsworth and Keil studied vegetarians' eating attitudes and practices. Both authors conducted interviews with 76 vegetarians and vegans in the East Midlands. The interviews were taped, resulting in a large archive of qualitative data....