Please watch the “one-way ANOVA” video and read the chapter’s SPSS demonstrations prior to doing the assigned exercise.
You will explore whether the amount of American adults’ television watching differs based on health. The GSS 2018 asked respondents to rate their health (ranging from excellent to poor) [HEALTH] and report their hours of TV watching on a typical day [TVHOURS].
The dependent variable, TVHOURS, is a ratio (continuous) variable. Your independent variable, HEALTH is ordinal (excellent, very good, good, fair and poor.). Thus, one-way ANOVA is the correct statistical test.
Step 1. Write the null and alternative hypothesis. [Note: In ANOVA, the alternative hypothesis is always tested non-directionally.]
Step2. Run the SPSS analysis
- From the SPSS Menu baràANALYZEàCOMPARE MEANS àONE-WAY ANOVA
- In the One-Way ANOVA dialog box, CLICKà TVHOURS into DEPENDENT LIST
- CLICKàHEALTH into the FACTOR box
- CLICKàPOST HOC, when the Post hoc dialog box opens, CLICKàBONFERRONI (NOTE: if your F-value—or ANOVA—is significant, this will provide all the possible paired comparisons to determine specific where—or between which groups—the significant differences exist.)
- CLICKàCONTINUE to return to the ANOVA dialog box
- CLICKàOPTIONS and now CLICKàDescriptives and Means Plot (Note: these two options help you more easily interpret your data.)
- CLICKà CONTINUE to return to the ANOVA dialog box.
Step 3. Interpret the Output
Begin by looking at the descriptive statistics box to understand the groups’ means and standard deviations on your dependent variable [TVHOURS]. The Means Plot offers a visually way to observe the distribution of the groups’ means. Remember that we may see differences in means that may or may not seem important, but until we look at the results of the ANOVA test, the f-value, we will not know if the differences between the means are significant. Proceed to the ANOVA box, find the F-value (or F-ratio), the df (remember, with ANOVA there are two df – one for between groups and one for within groups) – and the p-value.
If the F-value is NOT significant, proceed to Step 5. If the F-value is significant, proceed to Step 4, the post hoc comparison box.
Step 4. Examine the post hoc test results, which provide specific information on the difference of each health group to all other groups on the dependent variable [TVHOURS].
Step 5. Write up the results
- In writing up the results, be sure to provide support/evidence for your decision about rejecting or retaining the null hypothesis, including providing F-value, df, and p-value. If the F-ratio is significant, you should also explain your post-hoc paired comparison analysis.
- Create a table that reports the results of the 1-way ANOVA, including a summary of the post hoc test. Be sure to include the relevant statistics (means, SDs, the F value, the degrees of freedom, and the p value).
- Include a separate page showing work in SPSS