8.1 Item-level Absolute Fit measures

Chen proposed three measures for assessing item-level absolute fit.

  • Fit a model to a set of data

  • Based on the model parameters, simulate item responses of a large number of students

  • Denote the responses of students to item j in the original sample by Yj and in the simulated sample by Y~j

  • Calculate different item fit statistics:

    • proportion correct pj. This measure compares the observed proportion correct and the model-predicted for item j. pj=|P(Yj=1)P(Y~j=1)|
    • transformed correlation rjj: rjj=|Z[Corr(Yj,Yj)]Z[Corr(Y~j,Y~j)]|
    • log odds ratio ljj: ljj=|logN11N00N01N10logN~11N~00N~01N~10|
  • Perform hypothesis tests: we can estimate their standard errors and z-scores can be obtained by dividing the statistics by their corresponding standard errors.

z[pj]=pjSE[pj]N(0,1)

z[rjj]=rjjSE[rjj]N(0,1)

z[ljj]=ljjSE[ljj]N(0,1)

Exercise

If a test has 10 items, how many z[pj], z[rjj] and z[ljj] do we have?