The Predictive Power of Reading Comprehension, Attitude Toward Sciences, Test Technique, And Science Subject Matter Knowledge In Predicting Pisa Scientific Literacy Test Total Score

Abstract

The purpose of this study is to identify to what extent that the reading comprehension, question types, attitude toward science, test method, and subject matter knowledge predict the PISA scientific literacy scores of the students. The study uses the method of correlational research, which is a type of quantitative research method. We reached out to 321 students who were randomly chosen using the method of cluster sampling. Five separate tests were used in the study: PISA 2015 science literacy test, 12 gap-filling exercises were asked to determine the students’ level of surface reading comprehension, Five open-ended questions were asked to determine the in-depth reading comprehension level of the students, 5-point Likert scale composed of 13 items was used to determine the students’ attitude toward sciences, 20 multiple-choice questions with 5 choices, regarding four units of the eighth-grade sciences curriculum and a 5-point Likert scale composed of 10 items was used to determine to what degree students use clues to guess the correct answer when solving multiple-choice tests. The research showed that the students’ PISA scientific literacy scores, and the scores that they got in the multiple-choice and open-ended questions of PISA scientific literacy test were strongly positively correlated with the independent variables of the research (surface reading comprehension, in-depth reading comprehension, scientific subject matter knowledge, attitude toward sciences, using test techniques).The independent variables of the research set forth the 43% of the variance of the PISA scientific literacy test total scores, 46% of the variance of the success score in open-ended questions, and 34% of the variance of the success score in multiple-choice questions. The research also reached the conclusion that the students performed much better in multiple-choice questions compared to the open-ended questions about the same text (situation/problem).