# Stats Guys: Can You Stat Test Coefficients of Variation

1. ## Stats Guys: Can You Stat Test Coefficients of Variation

I am looking at survey data of users of a specific product type. Within the data set, I am looking at overall rating scores, which is derived from a 5 point scale that is converted to a ratio. From this I have generated a CV for two different group, Brand A brand used most often users and Brand A not brand used most often users. I get CVs of 18% and 24% respectively.

What I would like to know is if there is a method to stat test these numbers to determine a significance difference.

2. ## Re: Stats Guys: Can You Stat Test Coefficients of Variation

It is difficult to know exactly what you are looking for, since I am not familiar with the term "coefficient of variation". Perhaps if you can explain exactly what a coefficient of variation is, and how you obtain it from the raw data, then, if any of us here is statistically savvy enough, we may be able to suggest a test.

I might also suggest some time here (http://www.real-statistics.com/hypothesis-testing/ ) refreshing your memory of how to structure some basic hypothesis testing. You can also look at the various tests/distributions he discusses. As I recall, it seems that most Intro to Stats texts/courses introduce three main "tests": z test, t test, and chisquare tests. Which test is most appropriate will depend on the details of the sample/population you are studying and the information you have.

3. ## Re: Stats Guys: Can You Stat Test Coefficients of Variation

The coefficient of variation is the standard deviation divided by the mean.

4. ## Re: Stats Guys: Can You Stat Test Coefficients of Variation

So it sounds like you have mean and standard deviation for each sample set, and you are trying to test if the two sample sets are really different. My Intro to Stats text suggests using the t-test for this to test if the difference between the means m1-m2=0.

Review the different two sample T tests here: http://www.real-statistics.com/students-t-distribution/
This one is probably also useful: http://stattrek.com/sampling/differe...px?Tutorial=AP

5. ## Re: Stats Guys: Can You Stat Test Coefficients of Variation

While I do have what you are talking about, this is a focused look at the variance. I can tell if the means are significantly different, that is no problem. What I am trying to look at is if the variance between the two measures and groups is significant.

6. ## Re: Stats Guys: Can You Stat Test Coefficients of Variation

My Intro to Stat's text suggest using an F-test to compare two variances. Similar to: http://www.real-statistics.com/chi-s...ing-variances/

7. ## Re: Stats Guys: Can You Stat Test Coefficients of Variation

I will look into chi-square. Thank you.

There are currently 1 users browsing this thread. (0 members and 1 guests)

#### Posting Permissions

• You may not post new threads
• You may not post replies
• You may not post attachments
• You may not edit your posts

Search Engine Friendly URLs by vBSEO 3.6.0 RC 1