SPSS Mediation

Assignment 1: Mediation in SPSS
Earlier this week, you practiced testing for mediation and, ideally, used the Collaboration Lab to ask, answer, and otherwise address any questions you had regarding mediation. In this Assignment, you apply what you learned to answer a social research question using multiple regression, including testing for possible mediation. Remember that you must still address all the assumptions for multiple regression.
To prepare
Review the datasets provided.
Construct a research question based on one of those datasets.

By Day 7
The Assignment
Based on the research question you created, complete the following tasks:
Fit a multiple regression model, testing whether a mediating variable partly or completely mediates the effect of an initial causal variable on an outcome variable. Think about whether or not the model will meet assumptions.
Fit the model, testing for mediation between two key variables.
Analyze the output, determining whether mediation was significant and how to interpret that result.
Reflect on possible implications of social change.
Write a 1- to 2-paragraph analysis in APA format that includes your data output and addresses each of the tasks listed above. See pages 683 and 684 in your Warner textbook for an excellent APA-compliant write-up of a mediation analysis.

……………………………………….

Name of Stats Test

Introduction

In a paragraph identify what statistical test you are doing, what data set you used, and what you are attempting to test using the variables you chose.

Research Question

Write your one sentence research question here.  This should not be written as a question you can answer with a yes or no.  Also, you will either use difference, effect, or relationship depending on the type of statistical test you are doing (see underlined word in RQ and hypotheses).  You can also see that the independent variables (blue) are listed first and the dependent variable last (green)—do not use colors in your paper.  Here is an example:

RQ:  What is the relationship between the number of candy bars eaten in a day, hours spent exercising each week, and BMI?

Hypotheses

Null Hypothesis (H0):  Write your null hypothesis here.  You need to use the terminology “no statistically significant” in the null (see words in bold).  You can also see that the independent variables (blue) are listed first and the dependent variable last (green)—do not use colors in your paper.  Here is an example:  There is no statistically significant relationship between the number of candy bars eaten in a day, hours spent exercising each week, and BMI.

Alternative Hypothesis (HA):  Write your alternative hypothesis here.  You need to use the terminology “statistically significant” in the alternative (see words in bold).  You can also see that the independent variables (blue) are listed first and the dependent variable last (green)—do not use colors in your paper.  Here is an example:  There is a statistically significant relationship between the number of candy bars eaten in a day, hours spent exercising each week, and BMI.

Variables

Independent Variables (IV):  List your independent variables here as well as how they are coded (values) in your dataset.  For example:

Gender—Male=0; Female=1

Age—Actual age in years at time of data collection

Dependent Variables (DV):  List your dependent variables here as well as how they are coded (values) in your dataset. For example:

Weight—Self-reported weight in pounds at time of data collection

Ever been a smoker—0=no; 1=yes

Results

Write your results section here.  This should only be 1-3 paragraphs long—be succinct—but focus the results write up towards your null hypothesis and research question.  Make sure that you address if the assumptions of the statistical test are met or not and what you did if they were not met and why.  At the end of the results you should include if you are rejecting the null hypothesis and accepting the alternative (or retaining the null if that is appropriate) and why.  Please see the examples posted in the announcements for an idea of what you results should look like and also look at the example below. ↓

I highly recommend that you provide an APA formatted table to illustrate the results that you are discussing.  It is difficult for the reader to follow the results if you just list a lot of numbers so use a table if the results are more complex than you can explain well in sentence format.  However, make sure that the table is completely in APA format (see examples starting on p. 129 of the 6^th ed. APA manual) as the tables that are generated by SPSS are not in APA format.   The table below is APA formatted.

Here is an example of an APA formatted results section:

An ANCOVA was conducted to determine the effect of two different therapy approaches on stalking-type behavior after controlling for pre-therapy stalking-type behavior. There was homogeneity of regression slopes as the interaction term was not statistically significant, F(1,46) = .882, p = .353. The Levene’s test of homogeneity of covariances (F(1, 48) = 7.19, p = .01), was statistically significant and therefore the assumption was violated. The main effect of therapy approach was statistically significant, F(1, 47) = 5.49, p = .023, partial η² = .11, indicating that the hours spent in stalking-type behavior was lower after using the cognitive-behavioral therapy approach (M = 55.30, SE = 1.87) as compared to using the psychodynamic therapy approach (M = 61.50, SE = 1.87), as indicated in Table 1.

Table 1.

Therapy Approach Differences in Stalking-Type Behavior

Column	Data	Data
Male	5.45	985
Female	6.24	265
Animal	8.96	35
Mineral	0.89	18

 

The covariate was statistically significant, F(1, 47) = 50.46, p < .05, partial η² = .52, indicating that amount of time spent in stalking-type behavior before therapy had a statistically significant effect on amount of time spent in stalking-type behavior after therapy.  Therefore the null hypothesis is rejected and the alternative accepted.  This means that the amount of time spent stalking before therapy should be considered when determining effectiveness of therapy on stalking behavior and the cognitive-behavioral therapy approach may be more effective in treating stalking behavior.

***Make sure to copy and paste your SPSS output (everything produced by SPSS at the end of the paper AFTER the reference list (not in the results section).  There is a place at the end of the paper for you to paste this.

Conclusion

Please provide a short (1 paragraph) summary conclusion.

 

 

References

(Please note that the following references are just examples of APA formatting)

Alexander, G., & Bonaparte, N. (2008). My way or the highway that I built. Ancient Dictators, 25(7), 14-31. doi:10.8220/CTCE.52.1.23-91

Babar, E. (2007). The art of being a French elephant. Adventurous Cartoon Animals, 19, 4319-4392. Retrieved from http://www.elephants104.ace.org

Bumstead, D. (2009). The essentials: Sandwiches and sleep. Journals of Famous Loafers, 5, 565-582. doi:12.2847/CEDG.39.2.51-71

Hansel, G., & Gretel, D. (1973). Candied houses and unfriendly occupants. Thousand Oaks, CA: Fairy Tale Publishing.

Hera, J. (2008). Why Paris was wrong. Journal of Greek Goddess Sore Spots, 20(4), 19-21. doi: 15.555/GGE.64.1.76-82

Laureate Education, Inc. (Producer). (2007). How to cite a video: The city is always Baltimore [DVD]. Baltimore, MD: Author.

Laureate Education, Inc. (Producer). (2010). Name of program [Video webcast]. Retrieved from http://www.courseurl.com

Sinatra, F. (2008). Zing! Went the strings of my heart. Making Good Songs Great, 18(3), 31-22. Retrieved from http://articlesextollingrecordingsofyore.192/fs.com

 

Appendix:  SPSS Output

Your SPSS Output file should be copied (Ctrl +A) and pasted here (Ctrl +V).  Do not change your page layout to landscape.  It is ok if it goes off the page on the right side.