Independent Samples t Tests SPSS
Read This Article Wuensch, K. L., & Moore, C. H. (2004). Effects of physical attractiveness on evaluations of a male employee’s allegation of sexual harassment by his female employer. Journal of Social Psychology, 144, 207-217.
Download the Data http://core.ecu.edu/psyc/wuenschk/SPSS/Reverse-Pattr-Guilt.sav These data are from the research reported in the article. Bring the data into SPSS.
The Variables The grouping variable is P_Attr 1 indicates the plaintiff was attractive 2 indicates that he was not attractive The test variable is Guilt 1 = definitely not guilty 9 = definitely guilty
The Values Are Already Labeled
The Analysis Analyze, Compare Means, Independent Samples T Test
The Group Statistics
The t Test
Download Files to Estimate Cohen’s d Go to http://core.ecu.edu/psyc/wuenschk/SPSS/SPSS-Programs.htm Download two files, .sav and .sps
Enter Values in .sav File Enter “Equal Variances Assumed” statistics.
Run the Syntax File Just click on RUN, ALL.
Look Back at the .sav File
Write the Summary Statement When the plaintiff was physically attractive, the jurors were significantly more certain of that the defendant was guilty (M = 6.21, s = 1.81, n = 160) than when the plaintiff was not physically attractive (M = 5.52, s = 2.17, n = 164), t(314.4) = 3.08, p = .002, d = .34, 95% CI [.12, .56].
Equal Variances Assumed or Not I followed Zimmerman’s advice and reported the “equal variances not assumed” t, because the sample sizes were not equal. But you must always use the “equal variances assumed” t and df when using the program that estimates Cohen’s d.
SD or s For many years, “SD” was the APA-approved symbol for standard deviation. In the most recent Publication Manual they added “s” as appropriate to symbolize the sample standard deviation. Only time will tell whether authors shift from “SD” to “s.”