Participatory Educational Research
Yazarlar: Ertuğrul ERGÜN, Ali AYDIN
Konular:Eğitim, Bilimsel Disiplinler
DOI:10.17275/per.15.spi.2.13
Anahtar Kelimeler:Test analysis,Item analysis,Assesment and evaluation,Multiple choice tests
Özet: One of the most widely used assessment technique in educational institutions are the multiple-choice tests. Several analyses have to be made in order to determine the validity and reliability of these multiple-choice tests and items in the test. In order to make some comments about multiple choice tests, test’s average, test’s reliability, mean difficulty, standard deviance, measures of central tendency, measures of central distribution should be computed. And also to make some comments about multiple choice tests’ items, Item Difficulty index, Item Discrimination Index, item variance and standard deviance, item reliability index should be computed. These computations are time-consuming and hard to do by hand. Also even if data may be entered in a spreadsheet, formulas can be hard for a teacher to form in the software. To make comments about the produced values is also a hard point for educators. As a result, teachers in educational systems don’t/can’t do evaluations about the assessments they applied. In this study, a software has been developed for the statistical evaluation of multiple-choice tests’ results. With this software, test and item analysis of the multiple-choice exam can be done and also statistical results can be presented to the user by colorized graphics. Examinees’ scores, frequency table and analyses about the test (range, mean, median, Kr20, test’s mean difficulty, standard deviance, variance, coefficient of variation, and coefficient of skewness), every item’s Item Difficulty index, Item Discrimination Index, item variance and standard deviance, item reliability index, Point-Biserial Correlations are the main outputs of the software. Also distracters in choices can be seen easily in the graphics section. Also there is an info box in the developed software. The info box shows several information about the computed properties and their values. This box can be helpful for users who have limited information about these statistics