Measuring mathematical problem-solving ability: a validity and reliability testing scheme for developing quality tests

Abstract

Problem-solving skills are important because they help describe students' thinking patterns and influence how they deal with descriptive problems in everyday life. Therefore, in addition to developing students' mathematical problem-solving skills, education practitioners also need to develop an appropriate measurement instrument. The purpose of this study was to produce a valid and reliable test instrument for measuring the problem-solving skills of high school students. The research approach was Research and Development, which adapted the Borg & Gall development model. Seven research steps were taken to develop a test instrument. There were 84 eleventh-grade students at a private high school in Malang Regency as the test sample. Qualitative data was obtained from experts for validity, then the data was analyzed qualitatively and descriptively. Quantitative data was obtained from limited trials and field trials. The trial results data was then analyzed using the SPSS application to determine its empirical validity and reliability. Next, item analysis was conducted using Microsoft Excel to determine the level of difficulty and the discriminating power of each item. The results of this study indicate that the developed problem-solving ability test instrument meets the criteria of validity, reliability, has a varied level of difficulty, and has good discriminating power.