2
$\begingroup$

For example, I have 3 different rankings for car brand quality: enter image description here

I don't have too much experience in this area, so this problme may seems rookie for most you here. I want to know if there is any good practice, algorithim related on comparing multiple rankings and extract insight from them?

Specifically:

1. How should I put these rankings into same scale to compare them?

One ranking is from good to bad, while the other is from bad to good, and in different scales.

2. Is there any good way (algo) to spot the similarities/dis-similarities across the rankings?

For example, Lexus has a really good ranking at JD Power and Consumer Reports, and Land Rover ranks low on both.

3. Some other practices that I'm not aware of.

For example, if the first 2 ranking system agree on a certain brand, then may be it's possible that the 3rd ranking system is biased at this brand.

Thanks a lot!

Improve this question
$\endgroup$
1
  • 1
    $\begingroup$ I would recommend you to dig among questions here with word ranking or tag [ranking]. You might borrow some initial insights. $\endgroup$ Commented Dec 21, 2016 at 8:35

1 Answer 1

Reset to default
1
$\begingroup$

You can treat the percentage and rating data as ranks. In R, you would do this:

rank(...)

This should give transform your variables into ranks, which you can then compare. If one of your brands is consistency outperform other brands, you'd expect higher ranks.

You can test your ranks with Friedman's test. Your test will try to address the difference in distribution between power, consumer and complaints. If you can reject your null hypothesis, your data show statistically difference between the three variables.

Improve this answer
$\endgroup$
0

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.