The authors propose using stopping rules that will terminate the test when a certain amount of measurement precision is reached. To be specific, four stopping rules are introduced：The former three (D-rule, T-rule, D-rule) are used to quantify the measurement precision of the ability vector and the last one (K-rule) is for quantifying the amount of available information carried by each item.
Three simulation studies were implemented。
The simulation results showed that all four stopping rules successfully terminated the test when the mean squared error of ability estimation is within a desired range, regardless of examinees’ true abilities.
1 The values for the 4 stopping rules seem conditional dependent, no criteria?
2 How can we sure the four methods are achieve its accuracy, may use tradition methods as a comparison give more strong support
3 Will the value of covariance matrix in the simulation affect the results?
4 The K-rule seems perform better and consistent among all the condition, the following 4-rules comparison may be interesting