Compare means between groups SPSS

0
votes

I have this problem.

I have an SPSS sheet that looks like this (it's an analogy, so don't ask me how I have measured it). This example is about tennis players.

Player % of points won % of points won
       own service     opponent's service
1         50               10
2         80               60
3         70               40
4         80               50

Now I want to know if there's a difference between your own service, and the opponent's service, in terms of points won. (As you see, there probably is. But is it significant?)

Link naar boxplot

So Hypothesis: Own service -+-> points won

Now, Kruskal-Wallis, Independent-samples t test, One Way ANOVA, all require a grouping variable or something, but this is already implied. I could have chosen to make the data set:

Player Own service Won
1      1           0
2      1           1
3      0           1

For all games, group them on own service and see if there is a statistical difference between these groups. Perfectly doable.

The first set of data contains the same information, but presented differently. I just want to compare means between variables, only based on their own value. Can SPSS handle this type of information as well?

1
spss

1 Answers

1
votes

If the two variables are inter-related like points in a tennis match (per your example), and organized in different columns like that, a paired-samples t-test should work for you.