09.12.2012 Views

I__. - International Military Testing Association

I__. - International Military Testing Association

I__. - International Military Testing Association

SHOW MORE
SHOW LESS

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

Test Mode Supervisory<br />

WRITTEN TESTS Situational Judgment Test<br />

(Mean of effectiveness<br />

weight for "M' responses<br />

minus effectiveness weight<br />

for "L" responses)<br />

-<br />

Job Components<br />

CommxT MOS-Specific<br />

Job Knowledge Tests Job Knowledge Tests<br />

of Comnon Tasks of MOS-Specific Tasks<br />

(Percent items correct) (Percent items correct)<br />

JOB SAMPLE TESTS Supervisory Role-Plays Hands-On Tests Hands-On Tests<br />

(Mean across role-plays of Cornnon Tasks of MOS-Specific Tasks<br />

of ratings on 3-point (Mean across tasks of (Mean across tasks of<br />

effective behavior scales) percent steps passed) percent steps passed)<br />

RATINGS Rating Scales - Army Wide Rating Scales - Army Wide Rating Scales -<br />

Supervisory Dimensions Non-Supervisory Dimensions MOS-Specific Scales<br />

(Mean across dimensions (Mean across dimensions (Mean across dimensions<br />

of supervisor ratings on of supervisor ratings on of supervisor ratings on<br />

7-point rating scales) 7-point rating scales) 7-point rating scales)<br />

Figure 1. <strong>Testing</strong> instruments providing coverage of each job component, by<br />

test mode.<br />

Table 2<br />

Statistical Characteristics of Test Instruments Across Nine MOS<br />

Supervisory Comnon MOS-Specific<br />

Mean SD Rel. Mean SD Rel. Mean SD Rel.<br />

Situational Judgment Tests 1.37 .60 .75<br />

Job Knowledge Tests 65.4 12.5 .79 64.9 13.5 .73<br />

Supervisory Role-Plays 2.26 .42 .71<br />

Hands-On Tests 72.6 15.4 .46 69.4 19.5 .44<br />

Rating Scales - Army-Wide 4.49 1.06 .50 5.13 1.13 .48<br />

Rating Scales - MOS-Specific 5.19 0.97 .43<br />

Note. Situational judgment test results ranged from -.77 to 2.57 (thus the mean score of 1.37 is roughly<br />

equivalent to a score of 4.46 on a 7-point scale, with a standard deviation of 1.26): reliability estimate<br />

is split-half on items, corrected to test length.<br />

Job knowledge tests and hands-on test scores are proportions correct: reliability estimate for job knowledge<br />

tests is the median across MOS of a split-half on odd-even items, corrected to test length; the reliability<br />

estimate for hands-on tests is the median across MOS of the split-half on task scores, corrected to number<br />

of tasks.<br />

Ratings were made on a 7-point scale, where a 1 represents poor performance; reliability estimates are onerater<br />

reliabilities across dimensions, using the median across MOS for MOS-specific ratings.<br />

Role-play ratings were made on a 3-point scale, where a 1 represents less effective supervision; reliability<br />

eStim&eS are the median one-rater reliability across items, averaged across the three role-plays.<br />

544

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!