Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

Sign In Apply for Membership

Categories

PT Score vs. Predicted on 7Sage

sarahblairsarahblair Core Member
in General 604 karma

Hey guys,

So I don't know if this is stupid and I'm just confused on the issue but are the estimated stats that 7sage has on the homepage curved differently than the regular PT scores we receive? I'm talking about the bar graph that you input your amount of incorrect answers on ____ amount of recent PT's. For example, I got 21 wrong on a PT and 7Sage estimated that to be a 162 point something on the most recent 10 PT's but my actual score was a 159. Just confused about the discrepancy. Generally, I thought that I could get roughly 25 wrong to get a 160 when using their estimated predictor but is this not the case?

Thanks!

Show Related Discussions

Comments

  • andre3000andre3000 Alum Member
    137 karma

    I'm not an expert, but could it be the difference between flex & the old version?

  • tahurrrrrtahurrrrr Member
    1106 karma

    I think that estimation is just based on the average of each test's actual curve results. Each test is curved differently, so you can't definitively say -25=160.

    For example one PT a -25 might be 160 but on a different PT it's a 158.

  • sarahblairsarahblair Core Member
    604 karma

    @tahurrrrr right, thank you! So, I guess what I'm asking is I don't get why I got a 159 when all of the 7sage predicted results are from 161-164 when getting a -21on the 10 most recent PT's (this one was within that range). Like the 10% is a 161, 20% 162 and 164, and 50% 163. 160 isn't even listed let alone a 159.

  • sarahblairsarahblair Core Member
    604 karma

    @tahurrrrr the reasoning above the graph says for my example inputs: getting 22 incorrect answers on the 10 most recent PT's would result in a scaled score of __. Do you know if our raw PT scores just aren't scaled?

  • fin..501fin..501 Member
    125 karma

    @sarahblair that specific test must just be an outlier. Honestly, that tool isn't really that useful. Each individual test has its own raw score -> result calculation, and that one must be especially unforgiving, perhaps because most takers found it slightly easier than average. A common metric is the "curve" which is the number of wrong questions it takes to get 170. common curves range from -9 to -11 where -11 is a very difficult test, so you can get more wrong without losing points, and -9 is an easier test, so you can't get as many wrong.

  • sarahblairsarahblair Core Member
    604 karma

    @"fin.cobb" I see. Thank you so much!!!

Sign In or Register to comment.