Create SPSS Research Project Using ANOVA Analysis
Your goal is to create a research project from beginning to end
Guidelines regarding how your paper should be set up (these are approximations)
- Introduction/literature review (including thesis): 5 pages double spaced
- Methods (description of variables): 1 page double spaced
- Results: 1-2 pages
- Discussion and Conclusion: 2 pages
- Bibliography: last page (at least six references from academic journals)
- Total 9-10 pages, double spaced, including tables and bibliography (SPSS output is extra)
- Perform descriptive statistics on the variables in your study
- Perform one analysis (either crosstab, correlation, or comparing means (ANOVA))
- Look at the relationship
- Treat likert variables as quantitative as long as they have four categories and are conceptually continuous
- Create your own tables in the text of the document
- Weight data
- Referencing: Include at least six academic references, and use them appropriately to support your arguments. Use the ASA referencing format
- Include your SPSS printouts at the end of the assignment.
Your paper should NOT be longer than 12 pages (excluding SPSS printouts).
You must have an abstract: A one paragraph summary of your entire research paper.
This is where you sell your topic as interesting and important. This is also where most of your references will be included. Your readers must be convinced at the end of this section that you are going to put together a solid research paper. For your thesis statement, it is perfectly fine to state “The purpose of this paper is to …” You must have a thesis statement so that your readers know the purpose of your study.
You MUST include at least six references from academic journals (i.e., journal articles with volume and issue numbers). Newspaper and magazine articles do not count as academic references. You can have more than six references, but this is the minimum. Not only do you need at least six references from academic journal articles, you must use them effectively. This is important. Including references for the sake of meeting the reference criteria is one thing; using them appropriately is more important.
Follow the ASA format style (see Lecture Notes: Referencing).
There is a wealth of literature on the returns to post secondary education (Davies and
Walters, 2007; Walters, 2004; 2006)
This is where you describe your data, variables and method of analysis. Clearly indicate which variables you are using and how they are coded. For example, is the education variable is coded such that a high score implies a lower level of education. If you recode a variable you must discuss how you did that here. If you are running your analysis on a specific group of respondents (i.e., females) indicate that here as well. Given the way your variables are coded, indicate what types of analyses will be performed (e.g., cross tabs, ANOVA, or correlation). You do not need to indicate that you are using SPSS in the paper.
You MUST present your descriptive statistics and perform one statistical analysis (you will not be given extra marks if you perform more than one statistical analysis). You are limited to three possible techniques, and this decision is made by the variables that you have selected.
- Crosstab (two categorical variables)
- Correlation (two quantitative variables)
- Comparing means (one categorical and one quantitative).
You MUST create your own tables. You MUST also report the actual level of statistical significance in your tables and in the text.
- No asterisks (not statistically significant)
- One asterisk (significant at .05)
- Two asterisks (significant at .01)
- Three asterisks (significant at .001)
You should create one table for your descriptive statistics, and one table your statistical analysis. Each table must also include the sample size. In your results section simply report your results as was done in class.
Remember to treat all ordinal variables (e.g., education) and likert variables (e.g., strongly agree to strongly disagree) as quantitative variables (thus they should be used in correlations and ANOVA’s) – as long as they have at least 4 categories. Do NOT treat these variables as categorical. In other words, these variables should not be used in crosstabulations.
Please verify that you have interpreted your results correctly. For example, many variables are reverse coded (e.g., education).
In your tables report mean category for ordinal variables–not the numerical values. For example, if the mean for an ordinal variable is 2.1 and that corresponds with the category
“Agree” then you must put “Agree” as the mean category in the table, not the value 2.1.
Keep in mind that it is possible for two groups of respondents (i.e., males and females) to be in the same category but be statistically different from each other.
Take this example:
Suppose we are comparing males and females in terms of their scores on an attitudinal question where 1= agree strongly, 2=agree, 3=disagree, and 4=disagree strongly. Suppose also that the average score for males is 2.8 and the average score for females is 2.1. In this case, the average category for both males and females is the same (even if the difference is statistically significant). Thus, in your ANOVA Table you need list the average category for each group as “Agree,” rather than the numbers (2.8 and 2.1). In the text of the paper you can state that even though the average category for both groups is “agree,” males are less likely to agree with the statement than females.
If you are using a categorical variable with many categories (e.g., religion and province) recode them so that they have less than five categories. This will make the results less tedious and easier to interpret. Include your SPSS output at the very end of the paper. You MUST do this so that the TA’s can verify your findings.
Discussion and Conclusion
Discuss your findings and relate them to the broader literature. In this section you should also provide potential explanations for your findings. You need to convince your readers that this study was worthwhile and that it could potentially contribute to knowledge (even if it is just a verification study). Well written papers will likely bring additional research (references) in this section. You will need to discuss the limitations of your research here. .
Do NOT indicate that you have proven something.
Finally, include suggestions for further research in this section as well. That is, how researchers could build on your findings in the future.