Total pages: analysis 2-3 pages, discussion 1-2 pages, conclusion 1-2 pages) total
1. Analysis—2-3 (maybe 4) pages
A. Quantitative Data
Your data analysis consists of several parts:
1. Summary statistics (measures of central tendency or proportions/percentages) for all of the key quantitative variables with a very brief discussion.
2. Tables describing relationships among your key variables. Remember to provide a title for each table and to number and label your tables. Tables may also be appropriate for some qualitative data. You should not feel obligated to construct tables or figures for all your variables, but you should show at least one key relationship. You must present at least one univariate and at least one bivariate/multivariate table and your discussion should refer to the data in the tables. *** It’s useful to play around with the data a bit before making your tables. You need to know what “story” you are telling before you go to all the effort of writing up (and formatting) your results. ***
3. Analyze the responses to open-ended question(s). For the open-ended items or other qualitative data, look for patterns in the data that suggest some coding scheme. Once you have identified one or more patterns in responses, discuss trends among the responses and more specifically, discuss if the patterns found in the qualitative data have any bearing upon any hypotheses you have made.
4. Brief discussion of missing data on key variables in your data. Examine the pattern of missing data (“don’t knows” or “refusals”) in your data. What variables have the most missing data (if any)? What variables have the least? Why do you think you got the missing data that you did (e.g., problems with question wording or respondents’ unwillingness to sensitive questions)? How might this have biased your results?
OR
B. Qualitative Data
Analyze the data by looking for patterns in the data that suggest a useful coding scheme. Refer to the Schutt Chapter 11 to review some considerations when dealing with qualitative data. As you present your analysis, if you find that you have organized the coded data in a way that provides counts or frequencies or is essentially quantitative, then you should provide statistics and tables for these concepts/variables. For the concepts that are purely qualitative, you need to discuss how the “raw” data fits into your coding scheme providing examples from the interviews or observations and counter-examples of data that did not fit into the codes or categories and why you made those decisions. Remember that you know your data better than anyone else so you want to share information in a meaningful and logical way that tells a “research story.”
1. Be sure to describe your sample in terms of demographic characteristics (you do not need to provide statistics) and provide a table that helps to summarize and describe the data
2. In detail, analyze your qualitative data and the “coding scheme” that you used to measure the indicators of the key concepts in your study. Be sure to discuss both data that was considered indicative of the key concepts including any categories and/or levels of your measures as well as data that did not fit into the key concepts you have identified.
3. Considering your data, note any absences of data that would be indicative of your key concepts. Discuss why these absences might exist, with special attention to any methodological concerns resulting from how the interview or observation was conducted which may have resulted in “missing” data that may have actually been present, but unobserved for whatever reason (time of day, how questions asked, discomfort with topics and themes, etc).
2. Discussion- 1-2 pages
Your discussion consists of three parts.
1) Description and Interpretation of Results: What does your data show? Can your data be summarized in a table or graph? Does your data support your hypotheses (if deductive approach)? Do they suggest a hypothetical relationship between key concepts that can be tested in future research (if inductive approach)? Do they suggest an entirely different kind of relationship that you had not previously considered?
2) Generalizability of Results:
Does your sample seem representative of some broader population? If not, what groups/types of observations are over or under represented? If your sample is biased (and it likely will be biased), discuss how you think this might have occurred. How did you sample your respondents (and where)? What biases might this have introduced?
Did everyone you asked to fill out your surveys actually agree to do it? If not, did you see any systematic differences among the people who did or did not agree to complete a survey?
3) Reflections on Data Analysis: How might someone skeptical of your findings try to discount your results? Be an informed critic and identify the weakest part of your argument. For example, do the data support a number of different arguments, not just yours but others? Is there a possibility that your results may be spurious? Then, in spite of these issues, discuss why you believe your interpretation of the data is the best. If you cannot do this, then you have not spent enough time with the data.
3. Conclusions 1-2 pages
Write a few paragraphs summarizing your findings. What have you learned about your research question from your analyses? Do your results confirm or contradict the findings presented in the literature? Based on your findings and conclusions, end with one or two suggestions for areas future research.