0

I would like to ask some fairly basic questions so please bear with me.

I have a group audience who rated video A and video B. I would like to know if the variance (or should I be looking at Standard deviation instead) between the ratings for the two videos is present and if it is statistically significant.

Also, any online resources to perform the calculation would be much appreciated!

Thanks in advance!

Saleem Ahmed
  • 1
  • Welcome to the site. What do you mean by 'variance between the ratings for the two videos'? – Juho Kokkala Jan 14 '19 at 07:01
  • Thanks for reply. To elaborate, there 30 participants who are all watching Video A and giving it a score. Then after an intervention, the same group watch video B and each give it a score. My premise is that after the intervention, the variability is less. – Saleem Ahmed Jan 14 '19 at 08:45

2 Answers2

1

Welcome to the site. First, it doesn't matter if you look at sd or variance, one is just the square of the other.

Second, variance is always "present" so I am not sure what you mean.

Third, please edit your question, adding the contents of your comment, so it is clearer.

Fourth, your design is quite poor. You won't be able to tell whether the intervention had any effect as anything you find could be a result of seeing the first video, or of seeing any video, or of the second video, rather than the intervention. You should have some people who get the intervention and some who don't and you might consider randomizing the order the videos are shown in, as well.

Finally, you can use the F test to compare two variances, but if you follow my advice, above, you will have more groups and you won't just be comparing the variances (for which, see this thread, but making a model that the variance is related to the intervention and maybe the order. One way to do this is to transform the DV to be between 0 and 1 and then use beta regression, which models both mean and variance. See "A Better Lemon Squeezer". Since your DV is a score, it surely has bounds.

Peter Flom
  • 94,055
  • 35
  • 143
  • 276
0

Not sure I understand, but if you are asking how to compare two standard deviations (from video A and B), to check if there is significant difference, then an $F$-test can do this. This is very basic stuff and you can find calculators online trivially.

user2974951
  • 5,700
  • 2
  • 14
  • 27