A study on rubric development to evaluate writing skills in teaching Turkish as a foreign language

In this study, in line with the evaluation of writing skills in teaching Turkish as a foreign language, the determination of the problems and deficiencies in the scoring of the writing exams in institutions, and the lack of a standard rubric in the scoring, it is aimed to develop a valid and reliable scale suitable for A1-A2, B1-B2, C1 levels. The research was carried out on the teachers who were teaching at the Bartın University Language Education Center. The data collection tool was developed by examining the relevant literature and similar scales on the subject. In addition, the qualitative findings obtained as a result of the interview questions directed to the experts during the data collection process were used. Data were collected from teachers who are experts in their fields with a questionnaire consisting of 30 items. The obtained data were analyzed. In the study, after taking the opinions of the experts to examine the scope validity of the scale, by looking at the frequency distribution of the questionnaires, 49.75% agreed with the evaluation questions, 40.25% stated that they completely agreed, 8.46% disagreed, 1.54% did not agree at all. The reliability of the scale was made by examining the consistency of the scores given by the raters. Kendall W agreement coefficient method was used to determine the reliability. As a result of the analysis, it was found that the scales developed were .700 for A1 level, .801 for A2 level, .828 for B1-B2 levels, and .807 for C1 level. In line with the results, the consistency coefficient values approaching 1 provides evidence for the reliability of the developed rubrics. As a result of the analysis, it can be said that rubrics developed in accordance with A1-A2-, B1-B2, C1 levels are valid and reliable measurement tools and can be used in the evaluation of writing exams.