In: Statistics and Probability
Questions posted in pics.
A sample size N = 14 participants was asked to rate their agreement to the statement "I use Facebook to get to know people more closely" on a 5 point Likert scale (ranging from 1 meaning "I do not agree at all" to 5 meaning "I totally agree")
Ratings from men: 2,5,3,4,3,4,5
Ratings from women: 1,3,2,2,2,4,3
Check the research hypothesis that men use Facebook more often to get to know people more closely than women.
1. Select an appropriate statistical test and use the five steps of hypothesis testing.
2. Illustrate your answer with a sketch showing comparison distribution, the cutoff and the score of the sample of this distribution.
3. What could be an issue with a sample size this small?
Please rate. Thanks.
From F test : since p value = 0.758 is > alpha (0.05), thus we fail to reject Null and conclude that population variance of both the samples are equal.
2 sample t test is below : minitab output.
3. Please note that prior to proceed for 2 sample t test, we did equality of variance test by F method. and we know that F test assumes normality of the data for better results, since here sample size is very small and assumption of normality for such sample size could lead to false outcomes of the hypothesis test.
So it is better to have large sample sizes to minimum of 30.