Methodological and Statistical Problems in Graduate Theses in the Field of Science Education

This study attempted to determine the proficiency levels and drawbacks of the purpose, significance, problem statements and/or hypothesis, premises-limitations and methodology parts of the graduate theses in science education between 2005 and 2009. To this end, stratifying their weights on university basis, the researchers selected 40 theses among a total of 125 master’s and PhD theses submitted to the elementary education and science departments of 17 universities. In order to assess the theses, the study uses the “Thesis Evaluation Form” developed by the researchers. The percentage of the average goodness-of-fit among the assessments of three experts was calculated to be .84 for the first evaluation part of the form, and .91 for the second part. As a result of the analyses, it was found that the theses had shortcomings in the study’s significance, model, universesample selection, validity and reliability processes of the measurement instruments, and statistical method-data Purpose and significance: Given that graduate theses in a particular field guide future theses and studies in the same field, it could prove helpful for future studies and theses to reveal the drawbacks and errors in such theses. Therefore, the present study aims to “determine the proficiency levels and drawbacks of the purpose, significance, problem statements, premises-limitations and methodology parts of the graduate theses in science education between 2005 and 2009”. Methods: To the purpose in question, the study uses a descriptive method. Subsequently, taking into account the weighted total numbers of theses for the universities, 40 theses were selected for analysis among a total of 125 theses of 17 universities obtained from the National Theses Database of YOK [the Council of Higher Education]. As the data collection instrument, the study uses the “Thesis Evaluation Form” developed by the researchers. Results: For examining the theses included in the scope of the research, every thesis was assessed by three researchers according to the sections in the thesis evaluation form. Following the evaluation, the goodness of fit among the researchers was calculated, and researchers then re-interpreted their examination into theses with low level of fit. As a result of the examinations, it was generally found out that  Premises parts of the theses generally use out-of-the-scope statements weakening the internal validity of the research  Although the research design definitions include experimental studies, the studies are conducted on a semi-experimental basis  Theses of experimental and semi-experimental nature chose to select a universe-sample  Dependent, independent and control variables of the theses are not clearly stated  There exist certain shortcomings in the validity and reliability processes of the data collection instrument development stages  There are also shortcomings and lack of goodness-of-fit statistics in the selection of parametricnonparametric statistical techniques in the process of statistical method and data analysis. Discussion and conclusion: The research found problems in the processes of premises-limitations, research design, universe-sample selection, expression of the variables, data collection instrument and statistical method-data analysis. Certain theses use unnecessarily too many premises. Karasar (2006)