The StatCrunch output shows the ANOVA results for testing whether there is an association between the number of hours of TV watched per week and age group: 50 and over (AdultTV), college students (TeenTV), grade school students (ChildTV).

a. Test the hypothesis that people in different age groups spend different amounts of time, on average, watching television. Use a significance level of α = 0.05. Assume that the conditions for ANOVA are met.
b. Using the output provided, determine which sample means are significantly different from each other. Report all the given confidence intervals for the difference between means, and tell what they show. Then arrange the groups with lowest mean on the left and highest on the right, with underlines connecting groups that do not have significantly different means. Finally, write a sentence or two explaining which group means are different and how they differ.

Transcribed Image Text:

Analysls of Varlance results: Data stored in separate columns. Column means ColumnnMean Std. Ermor TeengTV25 11.88 1.5484616 ChildTV25 19.68 1.5997083 AduTV25 18.18 1.7518371 ANOVA table |source-SS | MS I F-Stat P-value Treatments 2 B54.9067 427.45334 6.3912418 0.0028 Error Tota 72 4815.44 88.88111 74 5870.3467 Tukey 95% Simultaneous Confidence Intervals TeensTV subtracted from Lower Upper ChildIV 2.2442 13.335571 Adultrv 0.7444292 11.815571 ChildTV subtracted trom Lower Upper AdunTV 7.0555706 |4.0156708