NotesFAQContact Us
Search Tips
ERIC Number: ED312961
Record Type: Non-Journal
Publication Date: 1989-Oct
Pages: 31
Abstractor: N/A
Reference Count: N/A
Conjoint Analysis: A Study of the Effects of Using Person Variables.
Fraas, John W.; Newman, Isadore
Three statistical techniques--conjoint analysis, a multiple linear regression model, and a multiple linear regression model with a surrogate person variable--were used to estimate the relative importance of five university attributes for students in the process of selecting a college. The five attributes include: availability and variety of financial aid, dorm life (living conditions and food quality), quality of education (quality of teaching, career relevance of the curriculum, and overall institutional reputation), student-faculty relationships and interaction (availability of faculty to students, faculty promotion of student development, and degree of faculty advice given to students on personal as well as academic matters), and campus social life. Comparison of the results of the three techniques showed that all three produced identical estimates of the relative importance of the five attributes, but the multiple linear regression model with a surrogate person variable produced the highest correlation between observed and predicted ratings for the hypothetical universities not included in the estimating procedures. The questionnaire is appended. Contains 16 references. (MSE)
Publication Type: Reports - Research; Speeches/Meeting Papers; Tests/Questionnaires
Education Level: N/A
Audience: N/A
Language: English
Sponsor: N/A
Authoring Institution: N/A