Ordinal Regression Analysis of Students Satisfaction with Academic Support Services provided by Polytechnics in Ghana

Research Paper (undergraduate) from the year 2004 in the subject Sociology - Work, Education, Organisation, grade: none, language: English, abstract: The Student Satisfaction Survey was necessitated by the notion that students have needs and rights to participate in quality programmes and to receive satisfactory services. The main objective was to assess the satisfaction experienced by students in the various departments they are in contact with at the Polytechnic. The outcome variable for student satisfaction with academic support services was measured on an ordered, categorical, and four-point Likert scale- very dissatisfied, dissatisfied, satisfied, and very satisfied. Explanatory variables included two demographic variables: gender and departments, and 42 questionnaire items related to the satisfaction of department involvement, curriculum contents, support services, facilities, and extra-curricular activities at the polytechnic. Proportional stratified sampling technique was used to select the sample 0f 1044 for the survey. In order to study the effects of explanatory variables on all levels of the ordered categorical outcome, an ordinal regression was used to model students satisfaction derived from the academic support services provided by the polytechnic. The overall impression given by students was that they were highly satisfied with the services of all the departments they came in contact with at the Polytechnic. They were also highly satisfied with the quality of teaching and the helpfulness of library staff. However, they expressed low satisfaction about the non- availability of some books on their reading list in the library, lack of reading space in the library, late arrangement for examinations, and late release of examination results. Recommendations have been made for the attention of departments and the intervention of management so that suitable action plans can be developed to address issues that give cause for respondents low s