Stage 11 of Semester Paper
Results Section k
Skip Navigation Links
Syllabus & ReadingsExpand Syllabus & Readings
Semester PaperExpand Semester Paper
Class CalendarExpand Class Calendar
AssignmentsExpand Assignments
HandoutsExpand Handouts
Instructor's Schedule
DataExpand Data
Instructional Videos
Grades
FAQsExpand FAQs
Links
Getting Help
Sociology Home
Soci 220 Home
 

Instructions for Stage 11 of the Semester Paper
Draft of the Results Section
For this stage, you must submit a draft of your results section. You already should have completed prior assignments that should have provided you with proper SPSS output for the "crosstab" tables you are completing for your analysis. You also should have completed prior assignments in Stage 7 where you completed blank "2x2" tables containing your independent, dependent, and control variables. For this stage, you will now need to fill in your tables and write an explanation about what they indicate. Please read the example results section from a prior student paper before continuing.

Click here to download the paper example.

Results Section

This section of your research paper should be 2 to 4 pages long (double spaced, standard 12 point font), including your three tables - in other words, you do not need 2 to 4 pages of narrative, just 2 to 4 pages total where the narrative is interspersed with your tables.

You should:

1. Fill in all three tables by placing the count and the row percentage from your spss output into each of the 4 cells of the table. Place the row percentages first and place the counts in parentheses afterward. Make sure that you look at the example and fill in your table in the same manner.

2. Make certain that the percentage in each row sums to 100%

3. Provide a brief summary where you compare the percentages across one of the columns as we discussed in class.

4. Make certain that your draft looks something like the example I provided in the prior student paper.

5. Because this is an introductory class, we will not be applying the concept of "statistical significance" or "confidence intervals" because we only briefly covered these concepts in class discussions. Instead, we will use the following [somewhat arbitrary] convention. (this convention is also not contained in the prior student paper, but yours should use this):

a) if the difference between groups (the two cells in either the first column or the two cells in the second column) is 1% or less, say that there is no difference between the categories/groups.

b) if the difference is greater than 1% and less than 8%, say that there is a moderate difference.

c) if the difference is 8% or more, say that there is a large difference.

As discussed in class, if the independent variable is on the rows, you should compare the cells in either column 1 or column 2. Do not repeat this comparison by also comparing the cells in the other column and make certain that you don't compare cells in the same row.

6. Discuss whether your analysis supports or does not support your hypothesis. Although not technically correct, for this introductory class, you can use the conventions above to speak about the amount of support for your hypothesis. If the difference falls under category a  (step 5a above - 1% or less) say that you found no support for you hypothesis, if it falls under category b say that you found some support, and if the difference falls under category c say that you found strong support for your hypothesis. If the difference is in an opposite direction than you anticipated, than obviously there is no support for your hypothesis, but there may be support for a competing one. Do not use the term "significant" when speaking about the differences between your categories because we are not using significance tests in this paper (unless you use apply the option below).

Optional - If you have taken a statistics course or another methods course and have learned to use a "chi-square" test of significance, you are free to do so and discuss what this means in your paper. This is entirely optional and not required for this class. If you choose to do this, you should substitute the appropriate language for testing the differences between two groups rather than using the [arbitrary] conventions above.