The purpose of this study was to build a Random Forest supervised machine learning model in order to predict musical rater-type classifications based upon a Rasch analysis of raters' differential severity/leniency related to item use. Raw scores (N = 1,704) from 142 raters across nine high school solo and ensemble festivals (grades 9-12) were…

Five methods of equating minimum-competency tests were compared using the Florida Statewide Student Assessment Test, Part II, for 1984 and 1986. Four of five methods yielded essentially comparable results for the highest scoring 84% of the students. Different lengths of anchor items were compared, using the concurrent item response theory equating…

Two data collection designs, counterbalanced and spiraling (Angoff's Design I and Angoff's Design II) were compared using item response theory and equipercentile equating methodology in the vertical equating of 2 mathematics achievement tests using 1,000 eleventh graders and 1,000 twelfth graders. The greater stability of Design II is discussed.…

Three administrations of two forms of a biology achievement test were conducted with high school sophomores and seniors. The 58 common items between the two tests were used for score equating purposes. (TJH)

Measurement scales developed using Thurstone and item-response theory (IRT) methods of scaling achievement tests for the same single-level data were compared for approximately 4,000 high school students taking the Iowa Tests of Educational Development in 1975. Results of both approaches indicate that variability increases as grade level increases.…

Item-preequating and random groups designs were used to equate forms of the American College Testing Assessment Mathematics Test for over 36,000 students. Equipercentile and three-parameter logistic model item response theory (IRT) procedures were used for both designs. The pretest methods did not compare well with the random groups method. (SLD)

Calculating the reliability of a testlet-based test is demonstrated using data from 1,812 males and 2,216 females taking the Scholastic Aptitude Test verbal section and 3,866 examinees taking another reading test. Traditional reliabilities calculated on reading comprehension tests constructed of four testlets provided substantial overestimates.…

The use of restricted item response models to test hypotheses regarding item difficulty ordering and slope uniformity was demonstrated in a study in which 597 algebra students were asked to solve word problems reflecting various types of cognitive processing. Benefits and limitations of the procedures are discussed. (SLD)

Recent findings on differential item functioning (DIF) for minority examinees (Asian Americans, Hispanics, and Blacks) taking the Scholastic Aptitude Test (SAT) are presented, and the standardization approach to assessing DIF is described. Item characteristics related to DIF that generalize across ethnic groups are discussed. (SLD)