Dataset on Islamic school teachers’ organizational commitment as factors affecting job satisfaction and job performance

This dataset aims to examine the effect of Organizational Commitment (OC) factors on Job Performance (JP) and Job Satisfaction (JS) at Islamic schools in Indonesia. The data propose that OC factors (Affective Commitment (AC), Normative Commitment (NC), and Continuance Commitment (CC)) have significant influences on JS and JP. Through the survey approach, 387 responses were obtained from Islamic schoolteachers in Indonesia. Face and content validity were initiated after the adaptation of the survey instrument. Further, a Partial Least Squares Structural Equation Model (PLS-SEM) was used to test the reliability and validity of the measurement model. For this purpose, the statistical process presents reflective indicator loadings, internal consistency reliability, and convergent and discriminant validity. Future researchers may reuse this dataset as a potential instrument to measure factors affecting job satisfaction and job performance within the context of education.


Specifications
Education Specific subject area Islamic education, OC, JS, and JP Type of data

Value of the Data
• The dataset presents a validation process of a survey of OC factors affecting JS and JP in the context of Islamic educational institutions. • The data are useful for educational regulators in facilitating proper policies regarding OC to improve their teachers JS and JP. • The accessible dataset could contribute to future researchers interested in doing research on similar topics

Data Description
This dataset proposes that Organizational Commitment (OC) factors that consist of Affective Commitment (AC), Normative Commitment (NC), and Continuance Commitment (CC) have significant influences on Job Performance, (JP), Job Satisfaction (JS). OC is defined as a degree to which teachers associates with their organization, the relative strength of their school involvement. AC is described as teachers' inner attachment to the school they teach, CC refers to the perception of costs, related to leaving the organization, and NC stands for the sense of responsibility possessed by the participants for the schools where they work. JP represents a term that refers to the quality of work of teachers in their profession. Meanwhile, JS is a pleasurable emotional state that results from the teaching experience that the participants have. The dataset includes two sections; demographic information and main survey. The demographic questions consist of location, gender, teaching experience, and school level ( Table 1 ). The main survey has three exogenous and two endogenous constructs ( Fig. 1 ). Three exogenous constructs included in OC measured from 1 = very disagree to 5 = very agree are AC (6 items), NC (6 items), and CC (6 items), adapted from previous academic research [1 , 2] . Meanwhile, two endogenous constructs are JS (3 items, 5 = mostly true; 1 = mostly false) and JP (3 items, 1 = very poor; 5 = very good) [2 , 3] . Table 2 informs the Mean, Standard Deviation, Skewness and Kurtosis of the data. Table 3 provides the information of the three assessments of measurement model (reflective indicator loadings, internal consistency reliability, and convergent validity). Tables 4  and 5 show the discriminant validity through the assessment of Fornell-Larcker criterion and cross-loading. Fig. 2 exhibits the measurement model of the dataset. The raw dataset and instrument are accessible on https://data.mendeley.com/datasets/dzy22g86tt/2 .

Experimental Design, Materials and Methods
For the scale development, a 2-phase procedure was implemented. In phase 1, the adaptation and translation were conducted. The adaptation of previous sources of literature was made [1][2][3] followed by the translation of the scale. The scale was translated using a back-translation method from English to Indonesian and Indonesia to English that involved 2 translation experts. In phase 2, face and content validity were conducted with two sessions of discussions with 5 users and 5 experts to evaluate the scale for the appropriateness of context and setting. We collected the data from September to December 2020 through a simple random sampling. We randomly selected a subset of participants from the population who are Islamic school teachers in two Indonesian provinces, Yogjakarta and Jambi. Every teacher has an equal opportunity of being opted as the participant. After converting the data in Microsoft Excel, the normality assessment was done by calculating Skewness and Kurtosis in SPSS 23, in which the values should be between -2 to + 2 [4] . All Skewness and Kurtosis values are in the range of the threshold; Skewness (AC4, -1.039 to NC1, -176) and Kurtosis (AC4, 2.415 to NC1,-766) ( Table 1 ). Further, the analysis of the data was conducted through the measurement model; four assessments (reflective indicator loadings, internal consistency reliability, convergent, and discriminant validity) were reported within the approach of PL S-SEM in SmartPL S 3.2. The reflective indicator loading should be .708 or higher. Table 2 performs all loading values that fulfill the threshold (.775-.930). Cronbach's alpha and Composite Reliability (CR) of greater than .700 should be applied for the internal consistency [5 , 6] . The Cronbach's alpha values of this dataset range from .736 to .906; similarly, the CR values are between .850 and .925 [7 , 8] . Convergent validity was reported through Average Variance Extracted (AVE); the values are recommended to be .500 or higher [8] . The AVE values range from .652 to .841 ( Table 2 ). By using the Fornell-Larcker and crossloading, the discriminant validity was evaluated. The AVE values of a construct should be less than the shared variance for other constructs for the Fornell-Larcker. The values of every construct are less than it's shared variance ( Table 4 ). When loading on a construct is greater than those of other constructs; cross-loading values, the discriminant validity is reported. All indicators' values (bold) of every construct were above the values of all their cross-loadings ( Table 5 ). Two items were dropped due to low loading values (AC5 and AC6). The model consists of five constructs with 22 indicators ( Fig. 2 ).

Ethics Statement
Informed consent was obtained for the data collection and the participation was voluntary. The survey was anonymous that did not include any personal information of the participants.

Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships which have or could be perceived to have influenced the work reported in this article.