Efficiency of the educational robot sets: A meta analysis studycovering the last 10 years theses

It is seen that the technologies that seem beyond imagination that we see in science fiction novels and movies are now in our lives. Robots are programmable devices consisting of electronic and mechanical units that can sense their environment with the help of sensors (İşman et al., 2016). With the development of technology, these devices are used in many areas such as industry, space education. The use of robots in education was first seen in the early 1980s (Papert, 1980). Investments are made in these sets, which are used in many areas in education, and therefore the effects on learning need to be investigated from multiple perspectives. The aim of this study is to examine the general effect of educational robot sets on students' problem solving, academic achievement and scientific process skills. Meta-analysis method, which is one of the literature review methods, was used in the study. The meta-analysis method is a method of creating a general result from the quantitative data obtained after the studies conducted in the similar field are grouped according to the determined criteria (Dinçer, 2021). The criteria determined as inclusion criteria in this research are as follows: The study should be between 2010-2021, the master's thesis and doctoral thesis written in Turkish, the educational robot sets were used only in the field of education, the study was experimental-semi-experimental. Effect sizes were calculated in the analysis of the data. While calculating the effect size, the conversion of statistical values such as arithmetic mean, standard deviation, t, F or r into a standard measurement value with certain formulas was used. According to the inclusion criteria, a total of 32 studies, including 6 doctoral theses and 26 master's theses, were included in the study. When the data are examined, 48.8% of the studies included in the research. It is seen that academic success is discussed at the rate of 19.5%, scientific process skills at the rate of 19.5% and problem-solving skills at the rate of 31.7%. In the context of academic achievement, according to Cohen et al.'s (2007) classification of impact levels, 0% (n=0) of the included studies were weak, 30% (n=6) small, 25% (n=5) moderate, and % It was determined that 45 percent (n=9) had a strong effect size. On the other hand, when the effect on scientific process skills is examined, 12.5% (n=1) of the studies are at a weak level, 37.5% (n=3) are at a low level, 12.5% (n=1) are at a moderate level and 37% are at a low level. It is seen that ,5 (n=3) have a strong effect size. When the effect on problem solving skills was examined, 7.69% (n=1) of the studies were at weak level, 7.69% (n=1) were at low level, 38.46% (n=5) were at moderate level, and 46.15% were at moderate level. ratio (n=6) was found to have a strong effect size.