A meta-analytical investigation of the influence of computer assisted instruction on achievement in science

Using computers in science education has become popular in Turkey as well as in other countries. Since science has many theoretical and abstract concepts, which are difficult to understand by students, students need some visual materials in order to learn these theoretical and abstract concepts. The importance of computer assisted instruction (CAI) has increased in schools. CAI is a method that uses computers in a learning media and strengthens students’ motivation and educational processes. It gives opportunities to both students and teachers to learn and teach more quickly and to combine active learning with computer technology (Akçay et. al., 2006). Collette and Collette (1989) explained that using a computer increases motivation and desire during lectures and laboratory work in the process of learning (Akçay et. al., 2006). There are a lot of important reasons to use a computer in science education. They can provide text, graphs, audio, video, pictures, animation and simulation in the same media to students. Simulations foster learning and help students to see different aspects of a subject and generalize about it (Akpinar & Ergin, 2007). Some studies showed that CAI was more effective than the other methods in increasing students’ interest in science lessons (Geban, Askar & Özkan, 1992; Hounshell & Hill, 1989).

Many primary studies that investigate phenomenon directly have been carried out in determining the effect of CAI on students (Özmen, 2008; Akçay et. al., 2006; Tas, Köse & Çepni, 2006; Karamustafaoglu, Aydin & Özmen, 2005). Evaluation of these studies plays an important role at certain characteristics by gathering them together for determination of the effectiveness of CAI. Meta-analysis is an effective review method used for evaluating similar studies.

Meta-analysis refers to the critical review and integration of the findings of separate studies. In a meta-analysis, the researcher compares outcomes across several studies using quantitative methods. The goal is to summarize the findings and characteristics of different studies (Göçmen, 2005). The method focuses on a common problem or topic and pools findings of several studies in an effort to draw inferences as to the meaning of a collective body of research (Hannafin et al., 1996). Basically meta-analysis helps researchers to translate results from different studies to a common metric and statistically explore the relations between the characteristics and findings of the studies.

During the past three decades, a large number of meta-analyses have systematically examined the effects of technology on student outcomes. Kulik and his associates have reported several studies focused on the effectiveness of computer based education in elementary and secondary schools, colleges and adult education (Kulik & Kulik, 1986; Kulik & Kulik, 1991; Kulik, Kulik & Schwalb, 1986; Kulik, Kulik & Bangert-Drowns, 1985; Kulik, Kulik & Cohen, 1980). In these studies, positive outcomes were found on students in favour of computer assisted instruction.

In the last decade, meta-analysis studies on CAI have increased in various areas. The first study during this period examined the effectiveness of CAI on the academic achievement of secondary students (Christmann, Lucking, & Badgett, 1997). In another study, Christmann, Badgett and Lucking (1997) focused on microcomputer-based CAI within differing subject areas. Whitley (1997) investigated gender differences in computer-related attitudes and behavior. Moreover, Cavanaugh (2001) examined the effectiveness of interactive distance education. Recent meta-analyses investigated the effectiveness of CAI programs in supporting beginning readers (Blok, Oostdam, Otter & Overmaat, 2002) and the effects of CAI versus traditional instruction on students’ achievement in Taiwan (Liao, 2007).

Even though there are a lot of studies on the effect of CAI on students’ academic achievement, there are few in science education (Bayraktar, 2001-2002; Christmann & Badgett, 1999). Bayraktar (2001-2002) investigated how effective CAI was on student achievement in secondary and college science education when compared to traditional instruction. She found the overall effect size as 0.273 from 42 studies yielding 108 effect sizes between the years 1970 and 1999. The results of the study also indicated that some study characteristics, such as student-to-computer ratio, CAI mode, and duration of treatment were significantly related to the effectiveness of CAI. Christmann and Badgett (1999) examined the effects of CAI on students’ achievement in differing science and demographic areas. They combined 11 studies on CAI in science. Schroeder et. al., (2007) studied a meta-analysis of U.S. research published from 1980 to 2004 on the effect of specific science teaching strategies on student achievement. The major implication of their research is that they have generated empirical evidence supporting the effectiveness of alternative teaching strategies in science. Moreover, Tekbiyik, Birinci Konur and Pirasa (2008), investigated effects of computer assisted instruction on students’ attitudes towards science courses in Turkey meta-analytically. In the study the arithmetic mean for 23 ESs included 17 studies was found to be 0.68.

Although, studies were started towards computer assisted education in Turkey in 1984, the first great project, called the Project of Computer Assisted Education, was put into practice by the National Ministry of Education. 7,541 teachers were trained in computer assisted education at the project.

During the 1990–1991 academic years, the National Ministry of Education bought 12,000 computers and attempted CAI in elementary and secondary schools (Alyaz & Gürsoy, 2002). The use of educational professional software for CAI started during the 1990s in Turkey when big software production companies brought their programs to Turkey, and the computerization process was accelerated (Özmen, 2008).

Schools and teachers have been constantly supported until the 2000s to encourage using CAI. Using computers has become widespread in schools and research on the effectiveness of CAI has increased after 2000. Several studies have been conducted in order to determine the effects of computer assisted instruction in science education, like all other subject areas. There have not been any studies providing insight on the effectiveness of CAI and evaluating the process of its development in Turkey yet.

Aim of the Study

The aim of the study is to meta-analytically determine the overall effectiveness of CAI on Turkish students’ academic achievement in science education from the year 2001 to 2007. In order to reach this aim following research questions were formulated:

1. What is the effectiveness of CAI on students’ academic achievement in science in terms of effect sizes (ESs) calculated from primary studies?

2. Are there any significant relationships between the effectiveness of CAI and the variables such as grade level, subject area and instruction methods of comparison group, publication year, type of publication and sample size?