Assignment:
Research Project
Overview
A significant portion of the course grade (30%) will be based upon a semester-long research paper examining the relationship between a set variables selected from the General Social Survey (GSS). Using the GSS, students will select a question (or dependent variable) of interest, conduct a brief review of the literature to identify relevant theories and causal factors (i.e., independent variables), perform an appropriate statistical analysis, and interpret the results. The goal of this course requirement is to:
1. Learn the role of theory and hypotheses in social research
2. Learn how to operationalize and measure abstract concepts
3. Learn how sample sizes and sampling methods affect estimates
4. Become acquainted with various sources of data available on the Internet
5. Learn how to use computers to analyze quantitative and qualitative data
6. Learn how to identify and locate scholarly journal articles using the library and Internet
7. Develop writing and analytical abilities
The General Social Survey
The GSS is a survey carried out by the National Opinion Research Center (NORC) at the University of Chicago. Previously conducted every year from 1972-1994, it is now conducted every other year. As of 2014, 30 national samples with 59,599 respondents and 5,900+ variables have been collected. It is considered the gold standard of public opinion data and is widely used by a variety of groups, including survey researchers, political scientists, sociologists, and historians.
It is also one of the easiest datasets to learn how to use! Like most datasets, data can be selected and downloaded for use with a statistical application (e.g., SPSS), but with the SDA interface (https://sda.berkeley.edu/archive.htm), you can also easily do simple cross-tabulations and perform basic statistical functions such as chi-square with just a click of a button. While I encourage students to familiarize themselves with this website, please note that we will also spend time in several classes as well as on an exercise learning how to use this dataset. But before we do any analysis, we need to go back and start at the beginning stages of the research process.
Getting Started: Selecting A Variable
Most research begins with a question or hypothesis regarding the relationship between two or more variables. For example, are married people more likely to self-identify as "conservative" rather than "liberal"? Are those without a college degree more likely to report their workplace as being "stressful" compared to college graduates? Has the leisure gap between men and women increased in recent years? These are just a few examples of the kinds of topics and questions you can explore with the GSS.
Spend a few hours browsing the GSS codebook examining variables of interest; then, select a question (i.e., a variable) that you believe operationalizes a concept of interest to you. For example, the variable CAPPUN2 is based on responses to the survey question, "Are you in favor of the death penalty for persons convicted of murder?" This would be your dependent variable. Once you have selected your dependent variable, review the literature on your topic to identify relevant theories and associated causal factors (i.e., independent variables) important for your topic.
Reviewing the Literature
Once you have selected a dependent variable that reflects a concept or question of interest, start doing some research using the various databases and search strategies we learned at the University Library. Review the literature on your topic to identify the relevant theories and/or associated causal factors for your topic. Sources should include peer-reviewed journal articles (e.g., American Sociological Review), not newspaper articles and/or websites (e.g., Huffington Post).
Note that some topics may overlap several disciplines, requiring you to search literature across disciplines. For example, the effect of political ideology on attitudes towards capital punishment is a topic not only studied by sociologists but also criminologists, political scientists, and legal scholars. Accordingly, you might need to broaden your search to include journals from other disciplines (e.g., law journals).
Although you must include at least four scholarly articles in your paper and list of references, I would recommend creating a bibliography of at least 6-10 sources. This will give you some flexibility and options from which to choose when deciding what to include in your paper as well as a series of examples you can refer to when writing sections of your paper.
Construct an Annotated Bibliography
Next, create an annotated bibliography of the relevant articles, listing articles alphabetically by the first author's last name. Below each article reference, include a copy of the abstract or a brief annotation of the article (≤ 1¶). Annotations should be comparable to abstracts in length (100-250 words), summarizing the key findings as well as how they relate to other related items listed in the bibliography:
Contrary to earlier findings by Smith et al. (1985), this study of public attitudes towards capital punishment finds that men, especially those in the South, are more inclined to support capital punishment relative to women and those residing in the Midwest, West Coast, or Northeast. Based on a nationally representative sample of elected public officials at the local and state level (n = 2,038), this study finds certain demographic groups are significantly more likely to favor the death penalty, even when presented with a substantial body of empirical evidence questioning its effect on deterrence of crime. The article concludes with a discussion of whether gender and geography are in fact valid predictors of respondents' attitudes or merely reflect underlying differences in political ideological that vary by region and gender.
The purpose of annotated bibliographies is to provide a brief summary of what each item adds to the existing literature - a key preliminary step in organizing a literature review or summary of existing research. A good annotated bibliography will provide students with the raw materials necessary for the next step in the writing process and make organizing such writing significantly more manageable!
This will help you to not only a) review the literature, but also b) select additional independent variables that may affect your dependent variable.
Select Independent Variables
Next, return to the GSS codebook and select two to three variables which operationalize possible causal factors that the literature suggests may affect the dependent variable. These will be your independent variables. For example, based on the literature on attitudes towards capital punishment you might pick "gender" (SEX), "educationalattainment" (EDUC), and "political views" (POLVIEWS).
Theory and Hypotheses
Once you have done this, state your propositions and write a brief theoretical discussion of how and why you believe the independent variables affect the dependent variable (Note: you will need a brief literature review to accomplish this). Explicitly state at least one hypothesis regarding the relationship between each independent variable and the dependent variable. For example, one hypothesis might be that "women are more likely than men to support the death penalty for persons convicted of murder." Be sure to report the level of measurement of each of your variables (e.g., ordinal) as well as the response categories given to respondents for each variable (e.g., male or female for SEX). You must also include a statement of the data source used in your study (i.e., GSS) and how you have operationalized your concepts-that is, how the variables you selected capture the concepts of interest. If you are struggling with this section, consider the very articles you selected - how did the authors present their research questions or hypotheses? How did they describe the relationship between the variables or concepts of interest?
Data and Methods
Nearly all empirical sociology articles include a description of the data and methods immediately following the introduction, literature review, and stated hypotheses or research questions. After all, readers will want to know about the data you used, aspects of the variables included, and other aspects of your research methodology.
Accordingly, you should spend some time examining the data you are working with. For example, what is the GSS and why is it a good source of data for your concept/variable of interest? Why is a quantitative approach (e.g., survey research) more suitable or appropriate for this question than a qualitative one (e.g., ethnography)? What aspects of the data sampled are important to note (and why)? For example, you might find that certain questions were only asked in certain years, limiting the generalizability of your results; likewise, you might find that certain groups were undersampled, leading you to question the validity of certain response categories. For example, if your variable only included six nonwhite respondents, you might be hesitant to draw any conclusions about the effect of race on the variable selected.
Descriptive statistics of the sample and variables selected are also worth noting. For example, what was your sample size and/or years sampled? How many respondents did you have grouped to certain variables? How are your variables distributed (e.g., frequency distributions, measures of central tendency)? This information helps the reader to understand why you selected the data and variables included in your study, what your data and variables "look" like, and possible limitations or caveats to consider when examining the findings. Again, I would encourage you to go back and see how the authors of the articles you selected described their data and methodology.
Findings/Results
The last major section of the research paper should include a brief description of the findings that includes at least one table or figure illustrating the results. For example, you might include a contingency table of responses or table comparing averages across different groups. The findings should essentially explain to the reader what you found and what is noteworthy. Look at the "Results" or "Findings" sections of the articles you selected for examples of what this section might look like. I will also try to provide some time in class for us to work on these together so you can see different examples.
Summary
To summarize, you will submit a research paper that includes, in this order,
1) a descriptive title
2) a statement of your topic of interest and research question (≤1¶),
3) a defense or rationalefor your quantitative methodological approach to answering your question (≤1¶),
4) a brief discussion of the relevant literature and theories (1-2 pgs.),
5) a brief statement of propositions and their associated hypotheses (≤1¶),
6) a brief description of the dataset and variables selected (1-2¶),
7) a brief description of the results including at least one table or figure (1-2 pgs.), and
8) a reference page including at least four scholarly articles.
Please do not exceed twelve (12) pages (not including references); all papers must be typed and double spaced. Proper ASA format guidelinesMUST be followed. You must also include at least four(4) scholarly works in your literature review and reference page; at least three of these must be from peer reviewed journals such as theAmerican Sociological Review or American Journal of Sociology.
This project will require every student to revise and resubmit the paper based on comments provided by the instructor on the first draft; I am also happy to provide input or feedback on any aspect of this project during office hours or by appointment throughout the semester. I will also work with each student on the results section to assist in the selection of appropriate statistics, interpreting results, and constructing tables. There is also a direct link to Jody Caldwell, Head of Research Services at the Drew University Library, on our course web page for your assistance.