0
$\begingroup$

It's been awhile since I've taken statistics, so forgive the simple question.

My company is attempting to evaluate the effectiveness of a training we've developed. We have teams spread throughout the country and want to compare two independent groups who go through the training. The 2015 group was the original trainees, the 2016 went through the modified training.

Our data set looks like this:

Groups / 2015 Size / 2015 Avg Score / 2016 size / 2016 Score

Group 1 / 35 / 105.5 / 42 / 105.1

Group 2 / 14 / 95.6 / 12 / 99

...

I thought at first I could use an ANOVA calculation to understand the differences or a simple difference of means - but I don't think either would account for the difference in sizes from year to year of the groups. If anyone could give me a nudge I would really appreciate it.

Improve this question
$\endgroup$
1
  • $\begingroup$ Was there random assignment to modified training? Or is there reasonable concern that more highly skilled people (or something else related to outcomes) may have systematically selected one of the training regimes? $\endgroup$
    – Matthew Gunn
    Commented Oct 28, 2016 at 22:17

1 Answer 1

Reset to default
0
$\begingroup$

I think a t-test for independent groups is in order here.

Different group sizes can be addressed with a Pooled variance. It is a weighted average of the group-specific variances, with greater weight given to whichever group has the larger sample size.

A Pooled variance estimate is used if one is willing to assume a constant variance across the groups. If there is some doubt about the constant variance assumption, then use the Pooled Variance calculation for unequal variance.

See the following Wikipedia pages for the exact formulas: Independent two-sample t-test

Improve this answer
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.