تصميم اختبار لقياس مهارة فهم التعبيرات غير اللفظية وتفسيرها للانفعالات الرئيسية الستة وتقنينه وفقا لنموذج راش

This empirical study attempts to design a scale to measure the ability to comprehend and interpret the non-verbal expressions of the six primary global emotions using Rasch model analysis. More precisely, the study attempts to construct the scale, test its psychometric properties and the extent to which teachers, administrators and technical staff in Southeastern Governorate’s schools in Oman could comprehend and interpret these six emotional non-verbal expressions. The study population consists of 5028 teachers and 977 administrative and technical staff, in which a sample of 361 was randomly selected, and the instrument was distributed. The test was designed using Google Forms following global test models, with testing focusing on three channels of non-verbal expressions; sound, fixed image, and animated image. Six professors examined the scale's content validity in the related domains. The degree of agreement among these experts was calculated using the Lawshe equation, where item(s) that failed to achieve the pre-specified threshold was automatically excluded. Rasch model analysis was then used to calibrate the items and examine the scale's psychometric properties and the extent to which the constructed scale achieved the Rasch unidimensional assumption and local independence. The result of the analysis suggested that the test fulfilled the requirement of unidimensionality and local independence. It was found that the psychometric test properties were within the acceptable calibration range since the value of both items and person reliabilities were 0.98 and 0.91, respectively. These findings indicate that the test is highly reliable and valid in the measurement of comprehension and interpretation skills of non-verbal global emotions. Furthermore, the differential item functioning (DIF) analysis suggested that the items were unbiased across gender and groups except for four items. The results also showed perfect matching between the items and persons since the value of infit and outfit Mean-Square (MNSQ) closed to 1 and the value of infit and outfit Z-Standardized (ZSTD) which measuring the calibration error was closed to zero which is the ideal situation of matching in Rasch Model. The study recommended that the Rasch Model analysis be used in designing and standardizing psychological scales due to its significance, creating an equivalent scale for comparison, and re-applying this test to other samples for confirmation.