MODEL DEVELOPMENT OF JOB SATISFACTION THROUGH THE ANALYSIS OF DETERMINANT VARIABLES OF JOB SATISFACTION OF SCHOOL PRINCIPAL (CASE STUDY ON A JUNIOR HIGH SCHOOL IN MEDAN)

The purpose of this study was to determine the theoretical model to describe the relationship between the causalistic latent variables that determine job satisfaction of SMP principals. This study will be conducted at the State and Private Junior High School in Medan involving 173 school principals as respondents, its execution for 2 years by a team of researchers and 26 students helper investigators. First research in 2016 focused on efforts to identify and map the job satisfaction of all SMP and the results was in the form of instruments determinant of job satisfaction are valid and reliable, which is compiled into a report on the research and used as articles to be published in international journals. In the first try out was conducted to prepare research instruments the form of multiple-choice Likert scale model, which subsequently spread to 54 principals of SMP. After calculations, Innovative Behavior Instrument for 42 items distributed gained 1 point invalid. Instruments Leadership Learning as much as 50 items of disseminated obtained 3 items that are not valid. Interpersonal communication instruments as much as 40 items disseminated items obtained 3 invalid. Work Motivation instruments as much as 50 items disseminated items obtained 6 invalid. Job Satisfaction instrument as much as 50 items of disseminated gained 1 point invalid.

To determine the reliability of the instrument is done by using the Cronbach Alpha Formula. After calculation of the 41 items of the Innovative Behavior instrument, obtained reliability coefficient of 0.895, 47 item instrument for the Instructional Leadership, obtained reliability coefficient of 0.923, 37 item instrument for the Interpersonal Communication, obtained reliability coefficient of 0.790. For 44 item instrument Work Motivation, the coefficient of reliability for 0.882, 49 item instrument for the job satisfaction, obtained reliability coefficient of 0.964. Furthermore, the second try out is done after all of the points are not valid discarded, and held mid August, 2016. After calculations, Instruments Innovative Behavior as many as 41 items of disseminated obtained 2 items that are not valid. Instruments Instructional Learning as much as 47 items of disseminated obtained 4 items that are not valid. Interpersonal Communication instruments deployed as many as 37 items obtained 4 items that are not valid. Work Motivation instruments as much as 44 items of disseminated obtained 6 items that are not valid. Job Satisfaction instrument with 49 items distributed gained 2 items that are not valid. While the results of the calculation of the reliability test, to 39 items of the Innovative Behavior instrument, obtained reliability coefficient of 0.898, 43 items instrument for the Instructional Leadership, obtained reliability coefficient of 0.932. For 33 items instrument Interpersonal Communication, the coefficient of reliability for 0.860. For 38 items instrument Work Motivation, the coefficient of reliability for 0.896, and 47 items instrument for the job satisfaction, obtained reliability coefficient of 0.966. Furthermore, the second year of research in 2017 designed a model of job satisfaction for which data is captured with a valid and reliable instrument in the first year, and then analyzed. Data analysis techniques carried out in two phases, which are descriptive and inferential. The purpose of descriptive analysis is to describe the characteristics of the data such as the average, median, mode and variance. Inferential analysis was used to test the requirements and the research hypothesis. Test Requirements analysis include: 1) data normality test, and 2) the linearity test regression. Test each variable data normality using the Kolmogorov Smirnof Test. Test linearity regression between variables was performed using statistical F test. To test the hypothesis used path analysis. And test matches theoretical models used to test the goodness of fit by using Chi Square. Thus, the product of the job satisfaction of SMP principals models seminars to get feedback on improvements, which will be published in international journals.