Abstract

This research aimed to investigate the components and indicators of digital leadership among school administrators under the Primary Educational Service Area Offices in the three southern border provinces of Thailand. The study employed a quantitative research methodology. The sample consisted of 200 school administrators under the Primary Educational Service Area Offices in the three southern border provinces, selected through simple random sampling using a lottery method without replacement. The research instrument was a five-point Likert scale rating questionnaire. Content validity was examined by seven experts, yielding a Content Validity Index (CVI) ranging from .86 to 1.00. The internal consistency reliability, calculated using Cronbach’s alpha coefficient, was .988. Data were analyzed using descriptive statistics, Exploratory Factor Analysis (EFA), and Confirmatory Factor Analysis (CFA). The research findings revealed that: 1. The results of the Exploratory Factor Analysis (EFA) indicated that the data were suitable for factor analysis, with a Kaiser-Meyer-Olkin (KMO) value of .959 and Bartlett’s Test of Sphericity significant at the .001 level, demonstrating sufficient correlations among variables. From the initial conceptual framework consisting of 6 components, 18 indicators, and 90 items, factor extraction was conducted using Principal Axis Factoring with Promax rotation. Based on the criteria of Eigenvalues ≥ 1.00 and factor loadings ≥ .50, the results yielded 4 appropriate components, 12 indicators, and 39 items. The four components were: (1) Digital Vision, (2) Digital Culture Development, (3) Digital Literacy, and (4) Communication. These four components collectively explained 71.062% of the total variance, with factor loadings ranging from .501 to .769. 2. The results of the Confirmatory Factor Analysis (CFA) confirmed the indicators of digital leadership among school administrators. The measurement model demonstrated a good fit with the empirical data. The goodness-of-fit indices indicated that the model demonstrated an acceptable fit to the data, with χ² = 116.532, df = 44, χ²/df = 2.648, CFI = 0.976, TLI = 0.963, and RMSEA = 0.085. The standardized factor loadings ranged from 0.798 to 0.957 and were statistically significant at the 0.001 level. The Composite Reliability (CR) values ranged from 0.902 to 0.949, and the Average Variance Extracted (AVE) ranged from 0.755 to 0.862. These results indicate that the indicators possess strong convergent validity and high reliability.

This work is licensed under a Creative Commons Attribution 4.0 License.