Question

Topic: Research/Metrics

Strength Of Correlation In Excel

Posted by Anonymous on 3/13/2013 at 11:31 AM ET 250 Points
Hi, I'd like to do an correlation study in excel to see how strong the relationship is between:

buying ad's off (users paying $1.99 to turn off ads from their mobile gaming app) vs buying coins (users purchasing coins in the game)

How do I set it up in excel?

Will the it something like this:
Y-axis | X-axis
User 1 $1.99 (if they pay to turn off ads) (amt spent buying coins)
User 2 " "
...
...
...

Or do I have to score the data via a Yes/No?
To continue reading this question and the solution, sign up ... it's free!
SUBSCRIBE
Or, SIGN IN

RESPONSES

  • Posted by Moriarty on 3/13/2013 at 12:07 PM Member
    I'm no analytics expert - this short (12min) video will give you an idea of how Pivot Tables work, and I'm pretty sure that a few hours puzzling over your computer will yield you some sensible results. And of course, some serious grounding in the use of a tool beloved of the Analytics guys.

    https://www.youtube.com/watch?v=UE1kaiyb6UY&feature=player_embedded

    Anyway, there's lots more to have a look at if you don't get it the first time.

    Hope this helps. M

  • Posted on 3/13/2013 at 5:38 PM Moderator
    I don't understand what you're trying to correlate with what. What is the question you're trying to answer. Just use ordinary English; don't try to put it in statistics terms.
  • Posted on 3/14/2013 at 7:18 PM Member
    Try using the excel data analysis pack (assuming you have the add-in installed). Set your data up in either columns or rows and indicate that (as well as whether you have labels for your columns or rows) after you select from the Data ribbon "data analysis/correlation." Highlight your input and designate where you'd like the output. Push OK and read your outcome. Pretty simple.
  • Posted by koen.h.pauwels on 3/18/2013 at 6:48 AM Accepted
    You have to set it up as a yes/no (1 = yes, 2 = no) with each variable in its own column. Each row is a user. Next, you simply type in the correlation formula: =correl(column1, column2) and you get the correlation coefficient.

    Of course this assumes your true question is indeed to "see how strong the relationship is between
    buying ad's off vs buying coins". A negative (positive) correlation coeffiicent implies a negative (positive) relation between the 2 variables. As Michael said, do tell us if you have another purpose in mind

BACK TO TOP

Post a Comment

Related Discussions

Community Info

Top 25 Experts
(Research/Metrics)