Estimating High School GPA Weighting Parameters With a Graded Response Model

Abstract

The high school grade point average (GPA) is often adjusted to account for nominal indicators of course rigor, such as “honors” or “advanced placement.” Adjusted GPAs—also known as weighted GPAs—are frequently used for computing students’ rank in class and in the college admission process. Despite the high stakes attached to GPA, weighting policies vary considerably across states and high schools. Previous methods of estimating weighting parameters have used regression models with college course performance as the dependent variable. We discuss and demonstrate the suitability of the graded response model for estimating GPA weighting parameters and evaluating traditional weighting schemes. In our sample, which was limited to self‐reported performance in high school mathematics courses, we found that commonly used policies award more than twice the bonus points necessary to create parity for standard and advanced courses.

Converting cumulative grade point average to an equivalent percentage value based on a fuzzy logic

“…The minimum and maximum percentage to each grade letter in matrix M will calculated using ( 5) and ( 6). The next step is calculating the membership value to the entered cumulative GPA using (7). The novel membership function in the proposed algorithm guarantee of picking a value 1 to highest cumulative GPA and a value 0 to minimum cumulative GPA.…”

“…The meaning of the final letter grade is based on regular percentage equivalents. The final percentage is rounded to the nearest integer by normal statistical procedure before translating it to a letter grade (i.e., rounding to the next higher integer if the first digit after the decimal place is 5 or higher) [7].…”