شبكة المعلومات العربية التربوية

The reliability, invariability and effectiveness of faculty assessment tools in good diagnosis vary greatly in their design among educational institutions. Consequently, the possibility of an absolute benefit from them without reservations or restrictions is doubtful. Besides, there is a possibility of some primary sources of errors like the errors that are related to the design of evaluation tools, errors that are related to application and the choice of samples as well as errors that are related to the extent of the students’ understanding of the evaluations’ main goal and their reliability while answering them. Moreover, the variation in faculty’s evaluation timing by students during the academic semester might have a great effect on the psychometric properties of variation. It is also the responsibility of specialists in the field of measurement and evaluation to work on enriching the Arabic library, developing the measures used in the educational field and reducing the possibility of any measurement errors during the process of evaluation. Since faculty performance evaluation measures have not been tackled enough yet by research and development, and after researching data bases, it has been found that there is a great defect in dealing with this topic in a specialized manner from specialists in the field of measurement and evaluation because this measure is new and has just recently been approved by the education evaluation commission (EEC) and the national commission for academic accreditation and assessment NCAAA in Saudi Arabia. Consequently, the absolute benefits from students’ faculty evaluation results and academic leaders’ developmental decisions have been limited. This study seeks to academically participate in improving students’ faculty evaluation tools by discussing the psychometric properties of these measures according to the variables of the timing of these measures’ application and gender. Besides, it also examines the properties of building the measures’ elements through the consistence of the level of evaluation adequacy among levels of measurement invariance. This study’s community consists of all male and female Umm Al-Qura University students in the main campus at Mecca city. The sample consists of 490 male and female undergraduate students from different majors and different colleges. The reliability of the tool has been examined through the validity of its structure by using the exploratory and confirmatory factor analysis. Cronbach's alpha coefficient has shown a good internal consistency value as it reaches 0.93, and results have shown a gradual consistency of the measure’s invariance that reaches the fourth level of the used model structural invariance consistency. However, results have shown statistical significant differences in the reliability of the gender variable in favor of females. Meanwhile, there are no statistical significant differences among the students’ responses to faculty evaluation measure with regard to the variable of evaluation application timing. Consequently, it has been found that the present faculty evaluation measure needs restructuring in order to keep the main structure of building measures with regard to invariance symmetry and preserving statements invariance consistency when the sample’s properties vary. Besides, it should be taken into consideration that many studies assured that students’ responses might have other motives that are not related to evaluation objectivity. (Published abstract)