1 Reply Latest reply on Mar 8, 2015 5:25 AM by Yuriy Fal

    Z Score - Stats help

    Ryan Johnson

      I have a survey that respondents use to rate companies on various attributes.  I've created Z Scores for each attribute using the KB article.  But as I was building the sheets, I noticed something odd.  All of the companies were staying in the same Z Score rank order.  I thought maybe it was something static in my sort, but all of the mark values are indeed changing - just not changing enough to re-order the companies.  It seems odd to me that all ~30 companies would rank the same across so many different attributes, so now I'm thinking either I've done something wrong with my calculations or the way my data is distributed is completely off.  Can any Statisticians, Z Score gurus or people generally more intelligent than me tell me if I've messed up somewhere along the way?

        • 1. Re: Z Score - Stats help
          Yuriy Fal

          Hi Ryan,

           

          You'll be better using AVG([Number of Records]) --

          not SUM([Number of Records]) -- as your base calculation,

          all other calc logic being the same. That's because your dataset

          contains dirrefent number of rows for each Company.

           

          Please find the attached workbook

          (with the [Ability to discuss indications] done this way).

           

          Hope this helps.

           

          Yours,

          Yuri