0
$\begingroup$

I collect data about a videogame (think about an ego shooter with kills / deaths etc)

Obviously the higher "kills" is and the lower "deaths" is, the better the players result is.

If I track a lot of different games - these values should be normally distributed - right?

So then I want to implement a major change - like using a different weapon or strategy or something like that - , and after some games evaluate, wheter the results are significantly higher, or not.

Can this be done with the student's t-test? Or would you recommend a different test? How can this be done in R? And how many games would be necessary to get a good result?

Thanks in advance

$\endgroup$
0
0
$\begingroup$

Re: "these values should be normally distributed - right?" Not necessarily.

Yes, this can be done in R (easily).

Comparing results with two different weapons would be a two sample t-test. You would be comparing the mean kills with weapon A vs. the mean kills with weapon B.

If the values aren't normally distributed, you can use non-parametric tests.

You could use ANOVA to see if there's a difference in the means of 3 or more groups at the same time. Assumptions need to be met for the results to be valid. For example, you have a dataset with weapon equals A, B, or C, and you have the number of kill. You could use ANOVA to see if the weapon type significantly (statistically) impacts the mean.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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