The effect of TPS and PBL learning models to the analytical ability of students in biology classroom

This research was a quasi-experimental study conducted from April to June 2016, involving students of XI Science Class at State Senior High School (SSHS) 13 of Banjarmasin as the study population. In total, there were four classes of XI Science in this school with the ability of students in biology learning between one class and another was at the same level. Therefore, a simple random sampling technique was exemplified in the selection of the samples. The XI Science 1 and the XI Science 3 classes were chosen as the samples of this study. The number of students in class XI Science 1 was 24, and the number of students in XI Science 3 was also 24 with a range of ages from 16 to 17 years old. The independent variable of this study was the learning model where students of XI Science 1 class received the PBL learning, while students of XI Science 3 class were treated with the TPS learning. The PBL syntax applied in this study was: 1) Orienting students to the problem; 2) Organizing students to learn 3) Guiding individual and group investigations, 4). Developing and presenting the work. 5). Analyzing and evaluating the problem-solving process. On the other hand, the syntax of TPS applied in this study was: 1) Thinking, 2) Pairing, and 3) Sharing. The research was conducted on the coordination system material, which consists of three subtopics, namely the nerve system, sensory system and hormonal system.

The dependent variable of this study was students' analytical thinking. The students' analytical thinking was measured using a test instrument consisting of ten essay items. The question of the essay consists of questions with cognitive levels of C3 to C5. The scoring of students' answers implemented a rubric that referred to Hart instruments with a score scale of 0 to 4. Before being used as a research instrument, the essay questions were tested first. Question trial data were analyzed using the Pearson's Product-Moment Correlation test to determine the validity of each item and the Cronbach's alpha test to determine the instrument's reliability. The results of the analysis of the items inform that all items were valid and the question instrument had a reliability value of 0.712 so that the instrument was categorized as reliable.

The test instruments were used at the beginning and at the end of the study. The collected data were then analyzed using a one-way analysis of covariance (ANCOVA) test. Before the ANCOVA test was carried out, the data were analyzed for normality and homogeneity. The Shapiro-Wilk test was selected as a normality test, while the Levene test was used to measure the data homogeneity. In addition, a linearity test between pretest and posttest data was also conducted. If these assumptions were met, the data would still be analyzed using ANCOVA, but if the data did not meet the prerequisites, then the data analysis was transferred to the nonparametric test. The nonparametric test chosen in this study was the Quade’s rank analysis of covariance test. The level of significance of this research data analysis was 5%.