Clinical supervision is the process by which supervisors treat their supervisees through interpersonal relationships to help supervisees develop their knowledge of the ability to perform clinical duties to provide quality service. However, from the literature review, it was found that a tool for assessing supervisors’ clinical supervision has not been clearly developed, especially in the context of the Intensive Care Unit (ICU), which requires close supervision and monitoring of nurses to provide the highest level of safety for patients. Therefore, the research objective was to analyze the construct validity of the clinical supervision scale in the care units of tertiary hospitals in Northeast Thailand. Methods Participants were 234 nurses in intensive care units at tertiary hospitals located in Northeast of Thailand. Data were collected using a questionnaire as perceived by registered nurses based on the framework of Proctor’s model using the 18-item Clinical Supervision Scale. Content validity was explored by seven experts, and the Cronbach’s alpha coefficient for reliability was .967. EFA was conducted to identify factors affecting the function of clinical supervision. CFA was used to ascertain the model fit. Results The EFA showed a three-factor clinical supervision and CFA. The results indicated that the hypothesis model of clinical supervision showed goodness of fit with empirical data throughout the sample (Chi-square = 103.501, df = 100, p-value = .385, CFI = .999, GFI = .955, AGFI= .923, RMSEA =.012). These results were within acceptable ranges, which indicated that the construct validity of the clinical supervision scale in the intensive care unit was consistent with Proctor’s functions: promoting professional accountability and quality control (normative function), skill and knowledge development (formative function), and providing support and personal well-being (restorative function).