Measurement of Parameters with Survey Research in Sociology

Measurement of Parameters with Survey Research in Sociology

Course Description

This is a graduate-level introduction to issues associated with survey research. The course touches on the role that theory plays in doing such research as well as the contribution of survey research to the theory. In addition, the course deals with issues of measurement, sampling, questionnaire design, interviewing, measurement error assessment, and data analysis. These issues are dealt with in several ways. First, we will examine the theoretical basis of the issues (e.g., of sampling); second, we will then deal with their application (e.g., applied sampling). In this version of the class, I am planning to put greater emphasis on application.

Assignment Questions

Answer all 3 questions; detail is preferred over brevity.

You have been given the opportunity to conduct research about intergenerational conflict across three generations. Mother-daughter conflicts have received some attention from researchers but the addition of a third party in the form of a grandmother has received little. The possibility for coalitions between grandmother and granddaughter in opposing the “mother” (the middle generation female). In addition, Simmel has suggested that triadic relationships are by their very nature conflict-prone. These claims appear most salient when the three members of the triad live in the same household. You have two opportunities in which to conduct this research. The first you have access to $2.5 million in funding. This provides you with the opportunity to conduct a large-scale study, including a national study. But this money is insufficient for a simple random sample. Instead, a more complex design ought to be considered. Develop a sample design and plan for collecting data from each member of the above-described triads. Statistical power considerations call for a sample of 600 triads. Interview costs per triad are approximately $1,000 (these involve face to face interviews). Salaries of the principal investigator and co-investigators consume about $250,000. Travel costs will vary from $300 per trip to $2000 per trip for farther away states. With these costs in mind, design a study focusing on the sampling. The second opportunity involves a much smaller budget, in the neighborhood of $100,000. There are no co-investigators and you have decided to donate your own time in order to maximize the money available for the study. You have determined that you could study up to 600 triads using the telephone. Your cost per interview is $135. Design a sample and sampling plan to fit both these opportunities. In both opportunities, you will be required to collect information on social class, race, and residence (rural urban), for example. Stratifying your sample along these lines may be a good idea. Considering the sensitivity of the topic, this may affect how you choose to collect data. Defend your decisions. PS: data on households in the Census is off-limit to researchers. [This meant to be a tricky question. A difficult triad to locate and a sensitive topic.] This may take 2500 words or 4 pages. %50 point)
Both Aneshensel and Chafetz address the issue of the relationship between theory and data. Discuss five of their differences regarding this relationship. To do so you should discuss what they mean by theory and its components, variables, and so on. 1750/2000 words or more. ( 30% point)
Estimate the sizes of a sample to test a t-test using a two-tailed test versus a single-tail test when the effect size (d) is 4.7. Why are these samples different? It amy take 500 pr few words (20%)