Methods for improving test efficiency and accuracy in a computer adaptive test (CAT)

1. A method for using pretest items in a computer adaptive testing platform to calculate interim scores while administering a computer implemented test through a graphical user interface, the method comprising:

determining a first computer implemented test from a test script administration component, the computer implemented test having a first plurality of test items including:

an operational item; and

a pretest item having one or more pretest items parameters, wherein the one or more pretest item parameters include statistical pretest item response data acquired from other examinees;

calculating, with a latent construct estimator, a first interim score based on the response data acquired from other examinees to the operational item and the pretest item;

determining a second computer implemented test having a second plurality of test items, wherein the second computer implemented test comprises a second pretest item and a second operational item, and

wherein the second operational item is chosen based on the first interim score;

displaying, on the graphical user interface, the second computer implemented test;

obtaining, from the graphical user interface, second responses to the second operational item and the second pretest item;

calculating, with the latent construct estimator, a second interim score based on the second responses to the operational item, wherein the calculation of the second interim score includes applying a weight to one or more contributions of the one or more pretest items, and wherein the second interim score is calculated by:

{circumflex over (θ)} i,j *=(1−ω)*{circumflex over (θ)} i,j +ω*{circumflex over (θ)} i,j ′,(0<ω<1),

where w is the weight applied to the second pretest item, j is a test item administered to the current examinee i with true ability {circumflex over (θ)} i , {circumflex over (θ)} i,j is an unweighted interim score using operational items only, and {circumflex over (θ)}′ i,j is an unweighted interim score using both the operational and pretest items administered so far;

and

refining the test script administration component to select a particular test item based on the second interim score.

2. The method of claim 1 further comprising:

estimating the one or more pretest item parameters for use with a set of calibrated parameters for the operational item.

3. The method of claim 1 wherein the one or more pretest item parameters are:

equal to the one or more pretest item parameters true value(s); and

not updated during administration of the computer implemented test.

4. The method of claim 1 further comprising:

setting the one or more pretest item parameters to an average of a set of calibrated parameters for the operational item.

5. The method of claim 4 further comprising:

updating the one or more pretest item parameters based on corresponding pretest item responses from a threshold number of additional examinees.

6. The method of claim 1 wherein the one or more pretest item parameters comprise maximum likelihood estimators based on the second responses.

7. A method for use of pretest items in a test to calculate interim scores comprising:

providing a computer implemented test, generated with a computer adaptive testing component, having a plurality of test items including:

an operational item; and

a pretest item having one or more pretest item parameters, wherein the one or more pretest item parameters include statistical pretest item response data acquired from other examinees;

calculating, with a latent construct estimator, a first interim score based on the response data acquired from other examinees to the operational item and the pretest item;

determining a second computer implemented test having a second plurality of test items, wherein the second computer implemented test comprises a second pretest item and a second operational item, and

wherein the second operational item is chosen based on the first interim score;

displaying, on a graphical user interface, the second computer implemented test;

obtaining, from the graphical user interface, second responses to the operational items and the second pretest item;

calculating, with the latent construct estimator, a second interim score based on the second responses to the operational item, wherein the calculation of the second interim score includes applying a weight to one or more contributions of the one or more pretest items, and wherein the second interim score is calculated by:

{circumflex over (θ)} i,j *=(1−ω)*{circumflex over (θ)} i,j +ω*{circumflex over (θ)} i,j ′,(0<ω<1),

where w is the weight applied to the second pretest item, j is a test item administered to the current examinee i with true ability {circumflex over (θ)} i , {circumflex over (θ)} i,j is an unweighted interim score using operational items only, and {circumflex over (θ)}′ i,j is an unweighted interim score using both the operational and pretest items administered so far;

determining a particular test item based on the second interim score; and

obtaining, from the graphical user interface, an examinee response to the particular test item.

8. The method of claim 7 further comprising:

randomly selecting and administering the pretest item.

9. The method of claim 7 further comprising:

updating the second interim score for the computer implemented test based on the second responses to the second pretest item.

10. The method of claim 1 , wherein the pretest item parameters are not updated during administration of the computer implemented test.

11. The method of claim 1 , wherein the pretest item parameters are initially equal to an average of operational item parameters.

12. A computer adaptive testing (CAT) platform to calculate interim scores while administering a computer implemented test through a graphical user interface, the CAT platform comprising a computer configured to:

determine a first computer implemented test from a test script administration component, the computer implemented test having a first plurality of test items including:

an operational item; and

a pretest item having one or more pretest items parameters, wherein the pretest item parameters include statistical pretest item response data acquired from other examinees;

calculate, with a latent construct estimator of the computer, a first interim score based on the response data acquired from other examinees to the operational item and the pretest item;

determine a second computer implemented test having a second plurality of test items, wherein the second computer implemented test comprises a second pretest item and a second operational item, and

wherein the second operational item is chosen based on the first interim score;

display, on the graphical user interface, the second computer implemented test;

obtain, from the graphical user interface, second responses to the second operational item and the second pretest item;

calculate, with the latent construct estimator of the computer, a second interim score based on the second responses to the operational item, wherein the calculation of the second interim score includes applying a weight to one or more contributions of the one or more pretest items, and wherein the second interim score is calculated by:

{circumflex over (θ)} i,j *=(1−ω)*{circumflex over (θ)} i,j +ω*{circumflex over (θ)} i,j ′,(0<ω<1),

where w is the weight applied to the second pretest item, j is a test item administered to the current examinee i with true ability {circumflex over (θ)} i , {circumflex over (θ)} i,j is an unweighted interim score using operational items only, and {circumflex over (θ)}′ i,j is an unweighted interim score using both the operational and pretest items administered so far; and

refine the test script administration component to select a particular test item based on the second interim score.