Path: > Advice > Format for the Paper
Advice on Statistics Research Paper: Format for Writing the Paper


Perhaps you like the paper-writing phase of research; maybe you dread it. The difference usually hinges on whether you regard yourself as a "good writer"--as determined by grades earned on countless other writing assignments. My experience with student research papers suggests that reporting the results of quantitative research is very different from other types of writing. Students who do well in creative writing may find this form of exposition more challenging; others rarely applauded for clever turns of phrase may receive compliments on their clarity of expression. Writing a research report can be a challenge for students who excel at writing essays and an opportunity to shine for those who do not ordinarily "write well." You can improve your writing performance by paying close attention to these suggestions for reporting your research.

The watchword for this type of writing is structure. The format of your paper should reveal the structure of your thinking. Devices such as paragraphing, headings, indentation, and enumeration actually help your reader see the major points you want to make. If you tend to string sentences together without organizing your thoughts into paragraphs, you are not helping him or her make sense of your writing. As a rule of thumb, if you type a full page (double spaced) without indenting for a new paragraph, you almos tcertainly have run one thought into another and have missed an opportunity to differentiate your ideas.

Headings can convey the major topics discussed in your paper. A research report (see the Lacy article on analysis of variance) typically contains four basic components:

1. Statement of the problem that gave rise to the research

2. Discussion of how the research was designed to clarify the problem

3. Analysis of the data produced by the research

4. Summary and conclusion of the study

Although you could include those sections in your report without separate headings, the underlying logic of your paper will be readily apparent with headings that identify its basic components: (1) the problem, (2) research design, (3) data analysis, (4) summary and conclusion.  

The 310 Statistics research paper is intended to demonstrate your competence in applying statistics to political and social analysis. The paper should be no more than 5 typewritten pages (double-spaced, not counting tables or graphs). It should be similar to articles in the "Research Notes" section of the Journal of Politics, source of the Lacy reading on Analysis of Variance.. Virtually all these articles are structured (some explicitly, other implicitly) according to the outline below. To help us evaluate your papers for the 25 points that they can earn, please follow the outline explicitly in writing your papers. They will be scored as indicated under the four section headings:

The Problem (worth 3 points)

Begin by stating briefly the intellectual concern with the topic, indicating why it is worthy of study. For example, does the topic reflect an established interest (e.g., explaining voting turnout), or does it pertain to a relatively new area (e.g., the changing role of Hispanics in politics)? To emphasize the ongoing nature of research, each paper should cite at least one previous study or publication relevant to your research. (The subject index to Sociological Abstracts, which contains articles in political science, is a good source for publications. USPD: United States Political Science Documents is another good source, and it also contains abstracts of the articles cited. Both sources are in the Reference Room.)

You can either cite your references in footnotes (giving author, title, and publication particulars), or you can cite the author and date in parentheses within the text. For example, (Tufte, 1974: 314) -- and then give the complete citation under "References" at the end of the paper:

Tufte, Edward R. (1974) Data Analysis for Politics and Policy. Englewood Cliffs, New Jersey: Prentice-Hall.

Research Design and Hypotheses (worth 7 points)

This section should translate the intellectual concerns expressed above into your research. Indicate here the nature and source of your data (i.e., state the data set that you are using in your analysis), the operational measures of your theoretical concepts, and any controls for other factors affecting your dependent variable. For example, do you expect the hypothesized relationship to hold across sex and race (for individual-level data) or across types of political systems (for national-level data)? You must also formalize your hypotheses in this section.

By formalize, I mean physically distinguish your hypotheses from the rest of the text in two ways: (1) labeling them as H1, H2, etc., and (2) underlining them. For example, you might say, "This leads to our first hypothesis:

H1: The greater the GNP per capita, the higher the literacy rate."

Hypotheses should be bold assertions of expectations that lend themselves to falsification. They gain in credibility as they survive attempts to test them -- i.e., to falsify them. (Admittedly, it is intellectually more satisfying to propose hypotheses that are supported rather than falsified through data analysis. Whether your hypotheses are supported or falsified will have no effect on the paper's grade.) Whenever possible, formulate directional hypotheses, which invite falsification more readily than non-directional hypotheses. (We will discuss the difference between the two soon.)

Also pay attention to the linkage between the concepts in your theory and in the way you operationalize those concepts in formulating your hypotheses. Be careful not to throw away data by collapsing variables to do crosstabulations when they might more properly be analyzed instead through correlational and regression analysis. For example, the "thermometer" variables in the VOTE88 data are expressed from 0 to 100, while those in VOTE96 are collapsed into a few ordinal categories. So these VOTE88 variables make better quantitative dependent variables in regression analysis than the recoded variables in VOTE96.

Data Analysis (worth 10 points)

Report here the results of your statistical tests. Refer explicitly to the hypotheses being tested by number: H1, H2, and so on. In most cases, your data should report tabulations of statistics. If you use ordinal or continuous data, your statistics will involve correlation coefficients, regression coefficients, or results of t-tests or F-tests. Do not simply accept and report the format of SPSS computer printout. That's not very classy. Instead, reformat the data into tables like those in the Journal of Politics or someother professional journal. Take some care in reporting your tables. Provide informative titles. Be sure to include the Ns on which any percentages are based. (We will deduct points if Ns are not included.)

Statistical tables should contain all the information that the reader needs to analyze the test. Your job as writer is to point out the key features of the analysis, not to repeat all the numbers in the tables. The data are in the table; the text should be used to summarize its particulars. Example: "All but one of the correlations in Table 1 are in the expected direction and are statistically significant." Quote actual numbers only to emphasize special points: "Note that the correlation of .25 between GNP per capita and death in foreign wars is substantially lower than that of .50 between GNP and deaths in domestic violence."

Please report correlations and slopes (if you employ regression analysis) only to the second decimal point. Do not slavishly reproduce them to the ultimate decimal point from the SPSS output. If your analysis involves plots, you may use the PLOT printout if you label it properly and mount it on a page in your paper with aesthetic feeling. Where possible, avoid reference to variables by their SPSS labels (e.g., PCTBLACK, CLINTON), for these labels convey little meaning to an outside reader, for whom this paper should be written. Instead, refer to them in more descriptive terms: "percent black" and "vote for Clinton in 1992." This makes for more pleasant reading.

Summary and Conclusion (worth 5 points)

This section should return you to the problem raised at the beginning of the paper. It provides the link between your narrow data analysis and the broader intellectual concerns with which you began. You might start by summarizing the results of your statistical tests and determining whether your research supported or contradicted prevailing theory. If your hypotheses are supported, how powerful is the theory? That is, how much variance are you explaining in the dependent variable? If your research fails to support the theory tested, what are the possible sources of failure? The theory itself? The presence of confounding variables? The inadequacy of the data or the way the variables were measured? The basic research design? If you see weaknesses in your research, here is the place to comment and perhaps make suggestions about future research.

About the Data That you Will Analyze

I strongly advise against trying to collect your own data to write this paper. Data collection is a time-consuming and often frustrating activity. I would prefer that you spend your time in doing data analysis rather than data collection for this paper. You have several data sets from which to choose. You can review the available data sets through DOIT. You will probably consider different data sets before settling on one for your analysis. You should explore the variables in which you are interested by running FREQUENCIES for discrete variables and DESCRIPTIVES for continuous variables. (You can tell which are discrete and continous by the MIN and MAX values on the printouts from the DOIT procedure.) Use CROSSTABS for discrete variables and PLOT for continuous variables to get a feel for the data. You will soon learn more powerful statistical techniques to employ in your analysis, which will make writing the paper more interesting. Guide to useful SPSS commands for use in your research