Testing

5 posts / 0 new
Last post
Super Admin
Testing

Lawshe (1975) provided a table of critical values for the CVR by which a test evaluator could determine, for a pool of SMEs of a given size, the size of a calculated CVR necessary to exceed chance expectation. This table had been calculated for Lawshe by his friend, Lowell Schipper. Close examination of this published table revealed an anomaly. In Schipper's table, the critical value for the CVR increases monotonically from the case of 40 SMEs (minimum value = .29) to the case of 9 SMEs (minimum value = .78) only to unexpectedly drop at the case of 8 SMEs (minimum value = .75) before hitting its ceiling value at the case of 7 SMEs (minimum value = .99). However, it is important to understand when applying the formula to 8 raters, the result from 7 Essential and 1 other rating yields a CVR of .75. If .75 was not the critical value, then you would need 8 of 8 raters of Essential that would yield a CVR of 1.00. In that case, to be consistent with the ascending order of CVRs the value for 8 raters would have to be 1.00. That would violate the same principle because you would have the "perfect" value required for 8 raters, but not for ratings at other numbers of raters at either higher or lower than 8 raters. Whether this departure from the table's otherwise monotonic progression was due to a calculation error on Schipper's part or an error in typing or type setting is unclear.

Dislike
0
Super Admin

testing

Super Admin

Testing1

Super Admin

sdwqwdw

Super Admin

test