Purdue University Hypothesis Testing with an Independent Samples T Test in SPSS Project
Hypothesis Testing with an Independent Samples t test in SPSS
In the Descriptives assignment, you saw that there was a numeric difference between men and women in their responses to the question “How afraid are you of having a surgical operation?” In this assignment you will find out whether that difference was statistically significant.
Your first task is to analyze the data from the variable usrg1 in your FearXX.sav file to see if there was a statistically significant difference between men and women in their reported fear of having a surgical operation. As stated in previous assignments, do not use the surg1 variable.
In this dataset the grouping variable is “sex”. Remember from the Descriptives assignment that sex is coded 0 = Male and 1 = Female.
Your output should contain the two tables as shown in the Skill Guide for the t Test: (1) The “Group Statistics” table, which includes the means and standard deviations for males and females, and (2) the “Independent Samples Test” table which shows the results of the t-test. Upload your SPSS output file (.spv) to your submission in D2L by the due date and time listed in the Class Schedule.
Your second task is to report these results in APA style. This should take no more than two to four sentences (see Chapter 7 and the SPSS Skill Guide for the t Test). Report the exact significance level for Levene’s test (Sig.) and the t test (Sig. 2-tailed) indicated in your SPSS output. Remember that if SPSS gives you a significance level of .000 this is reported as p < .001. Report only the results of the statistical analysis. Do not discuss the reasons why you think you obtained the results you did. That information would be included in a discussion section, and not in the results. Upload your write-up in a Word document to your submission in D2L by the due date and time listed in the Class Schedule. If you prefer, you may type your write-up in the “Comments” box in the assignment function in D2L.
(1) Before doing the independent samples t test, make sure all of the cases in your FearXX.sav file are available for the analysis, not just the male or female cases. To do this, go back into the Data > Select Cases window and choose All cases.
(2) Pay particular attention to the info about Levene’s test in the Skill Guide. Levene’s test only checks the homogeneity (equality) of variance assumption. It does not test the difference between the means of your groups. Interpretation of Levene’s test determines whether you report the information in the “Equal variances assumed” row, or the information in the “Equal variances not assumed” row.
(3) Remember to italicize all statistical symbols – e.g., N, n, p, t, df, M, SD, etc. The CI abbreviation for confidence interval is a special case and is not italicized. Make sure to leave spaces on both sides of math operators (=, +, -, etc.).
(4) The independent-samples t test is technically not an appropriate procedure to use with ordinal scales of measure. However, ordinal values from Likert items such as those used in the fear study (0 = No Fear to 6 = Terror) are often treated as if they are interval or ratio. This allows you to analyze them using the t test. For basic info about Likert items see: http://en.wikipedia.org/wiki/Likert_scale.