A DISCRIMINATION INDEX FOR ITEMS IN INSTRUMENTS USING GROUP RESPONSES

Item discrimination for instruments used to measure characteristics by means of group responses is stressed. It is argued that a percentage of the total sum of squares which is due to groups can appropriately be used as an index of item discrimination.     

The Sixteen Personality Factor Questionnaire as a Predictor of Counselor Effectiveness

This study examined the usefulness of the Sixteen Personality Factor Questionnaire in differentiating characteristics of counselors who were rated effective and ineffective. Eight of the factors achieved significance in distinguishing between these groups. Only two factors were found to differentiate male and female counselors (p < .10).     

EFFECTS OF VIOLATING ITEM CONSTRUCTION PRINCIPLES1

Two forms of a social studies achievement test were constructed with half the items for each form containing a cue, grammar, or length fault. Faults were found to make the items easier, which was supported by confidence intervals for the differences. However, validity coefficients with achievement and intelligence criteria, as well as the reliability coefficients, were virtually unchanged. The results agreed with those of Dunn and Goldstein (1959), even though the methodology differed. A suggested measure of test-wiseness for groups is presented.     

THE RELATION OF SCALE FINENESS TO GRADE ACCURACY

The use of broad categories in grading is likely to increase the relative amount of error present in the measures on which the grades are based. This is true regardless of the degree of inaccuracy of those measures. An example to illustrate this generalization and an explanation of its basis in psychometric theory are presented.