Page 166 - BUKU SYNOPSIS

Figure 17. The Item Fit Probability Curve of the Preliminary Field Testing

Figure 17 depicts a constant stretch for each scale, demonstrating that

the established rating scale is suitable for evaluating students and is

comprehended by raters. Also, figure illustrates that each rating scale has

their own peak. Overall, it indicates that the scoring rubric is approved for

use with the argumentative texts of students.

Additionally, the category statistic table of MFRM reveals that the

values of Rasch-Thurstone Thresholds are ranging from low, -2.54, .09,

and 2.43 respectively for scale 1 to 4 of the rating scale. The values indicate

that the raters understood well the description of each scale.

The preliminary test aims to determine the consistency of raters'

scores. This evaluation rubric includes two raters. A good assessment

instrument should be able to instruct the rater so that whoever is

conducting the evaluation can do so objectively and consistently. In

summation, Rater 1's and Rater 2's assessment results are the same or not

significantly different. Thus, the scores generated by the raters can be

161