Applied mathematical statistical modeling to determine the validity and reliability of independent instructional measures

NCCU Author/Contributor (non-NCCU co-authors, if there are any, appear on document)
James Osler, Professor (Creator)
Institution
North Carolina Central University (NCCU )
Web Site: www.nccu.edu/academics/library/

Abstract: Most universities and colleges in United States and elsewhere use student ratings of instruction for summative (e.g. facultypromotion, etc.) and formative (e.g. instructional improvement, etc.) evaluation purposes. This research examines thereliability and validity of the rating instrument used to evaluate teaching effectiveness and student learning outcomes.The sample consisted of 7,919 students of Historic Black College and University [HBCU] who completed the 2012 springsemester, Student Ratings of Instruction [SRI] survey (a measurement instrument used by students to assess instruction incourses in which they are currently enrolled). The application of the Cronbach's Alpha Reliability Model, the study findsstrong internal reliability [consistency] of the rating instrument used to comprehensively assess teaching quality. At thesame time, the application of Principal Component Factor Analysis with Varimax Rotation also finds strong constructvalidity of the assessment tool used to evaluate faculty.

Additional Information

Publication
i-manager’s Journal on Mathematics Vol. 2, No. 13
Language: English
Date: 2013
Keywords
Student Ratings of Instruction; Construct Validity; Reliability; Cronbach's Alpha, Principal Component Factor Analysis, Statistics.

Email this document to