Developing a Turkish learning motivation scale for foreign students in Turkey

The purpose of this research is to develop a motivational scale for foreign students studying Turkish in Turkey during the process of acquiring Turkish as a second language. The research method used is a quantitative research method with a survey research design. 418 people filled out an attitude scale created by sharing a Google Form link with international students who have studied or have studied Turkish in Turkey. Expert studies are used to test the validity of the content. The construct validity test was carried out by dividing the data into two groups. The first group was tested using the exploratory factor analysis (EFA) test, and the second group was tested using confirmatory factor analysis (CFA). The data used for the DFA test showed an abnormal distribution, while the data used for the EFA test had a normal distribution. As a result, by calculating the z-score, three data outliers were found, which caused the data to decrease from 209 to 206. For the next test, the KMO-Bartlett test (p.05), only the data from the first group will be tested with EFA. The score result of the KMO test is .822 and the significance value for the Bartlett test is .000. The test results showed that the study groups came from the same population. The AFA test was carried out using the varimax vertical rotation method, and the load value was .45. In addition, the principal components method is used in the extraction process with eigenvalues greater than 1. From the EFA test, it can be seen that linearity is obtained after 3 points, so it can be concluded that there are 3-factor structures in the collected scale model. The first factor named "related" consists of 8 questions. The second factor named "attractiveness of learning Turkish" consists of 7 questions. The third factor named "difficulties of learning Turkish" is represented by 4 questions. The suitability of the model that appears as a result of the DFA test with the AFA theory can be seen from the acceptable values of Chi-quadrat/DF, RMSEA, RMR, NFI, CFI, IFI, GFI, and AGFI. The reliability of the attitude scale in measuring target indicators is known by calculating the Cronbach's alpha coefficient, and the results show that the three variables are acceptable. Cronbach's alpha values starting from the first factor, as mentioned successively, are 0.837, 0.749, and 0.725. Keywords: Attitude Scale, Scale Development, Second Language Acquisition, Foreign Student, Scale Validity, Scale Reliability.