2022 Introduction to Statistics in Research Mitchell 2nd ed

I N T R O T O R E S E A R C H : D A T A V I S U A L I Z A T I O N & C O M M O N S T A T T E S T S

You can calculate the Mann Whitney U and compare it to a critical value from a table and/or you can use the probability to determine whether to reject the null hypothesis. The sample data set is two groups,

training method A and training method B. Essentially training method A is group 1 and training method B is group 2. There are 11 participants in each group. After the salesman training, a month of sales data was collected.

Null Hypothesis:

• Hypothesis: H o there is no difference • There is no difference in sales results produced by the training methods

Alternative hypothesis:

•

Hypothesis: H a there is a difference

• There is a difference in sales results produced by the training methods. • This is a two-tailed test • Statistical Test: Mann-Whitney U ( assumptions under the parametric t -test was not met, so run Mann-Whitney U )

Significance level α = .05 (two tailed test

But first, are the two groups the same shape?

Table 113: Sample data for Mann Whitney U test (SPSS ver 27)

Table 114 Histogram of Frequency for Training Methods

Table 115: Histogram of frequency

No, they are not the same shape. If they had been the same shape, we could compare the two medians. Instead, we will compare the means rank.

Let’s look at the means rank for this dataset: Essentially, you rank on the sales from one to twenty two. For example, rank 1 is $950 (in Training Method 2 (B) and rank 22 is $1860 (in Training Method 1 (A). So, essentially if the groups do not have a similar shape, SPSS (or even Excel) will use the means rank to calculate the statistic U . A tie is assigned the average of the two corresponding ranks (see $1500 and rank of 12.5).

132