Quality Metrics in Higher Education
Syed Saad Andaleeb

There has been extensive and continuing discussion in Bangladesh about problems in the country’s higher education system, both public and private. Claims abound about the lack of teachers and their training, poor curriculum, high costs, low salaries, lack of commitment and sincerity both in administration and among teachers, lack of educational infrastructure, ambiguous role of the UGC, absence or unavailability of quality administrator(s), and unbounded politicization to depict the efficacy and functionality of the higher education institutions (HEIs).
Development agencies depict quality in the HEIs as low, particularly in some of the newer private universities and in the affiliated colleges. Difficulties in recruiting, retaining and offering adequate professional development support to academic staff, as well as lack of internal and external quality assurance measures perpetuate quality problems.

While a plethora of opinions exist, it is difficult to find hard metrics that assess the quality of higher education reflected in a stream of continuous assessments. In other words there is no way to assess quality trends in the system. Measuring quality, however, is a challenging task and ought to be a continuing exercise to guide the nation’s HEIs.
Students are an important source of quality assessments. I recently asked 486 university students from eight HEIs—four in the public sector (1-4 in the figure) and four in the private sector (5-8)—how they evaluated their teachers and courses. These institutions represent the better known HEIs in the country. While a plethora of measures were used in the overall study, only two measures are examined here: “overall quality of the course” and “overall quality of the instructor” using a seven-point scale.

The average ratings across the eight HEIs for “course quality” are, overall, higher than what experts would ascribe; the lowest rating being 5.17 and the highest rating being 5.79 on a 7-point scale.
The average ratings on “instructor quality” could be better but are not disquieting; the lowest being 4.66 and the highest being 5.52 on the 7-point scale. Course quality obtained a consistently higher rating than instructor quality, but the gap narrowed for the private universities.

Interestingly, there was no statistical difference on the means for “course quality” across the eight institutions selected. This finding goes against some claims that public universities are better, although assessments for a wider set of universities might prove different.

There was a single statistical difference on instructor quality between a public and a private university. For the rest of the universities, there was no difference. The higher rating was for a private university.

Digging deeper, we looked at the extent of stability of the ratings, i.e., agreement among the respondents, reflected in the standard deviation: a higher value means greater instability or disagreement on the ratings. The highest value was 1.7 for course quality and 2.03 for teacher quality, both values reflecting a lack of stability in the assessment of course content and instructor quality. The gap narrows for the private HEIs. Such perceived quality variations in students’ perceptions can be a cause for concern and ought to be in the purview of the university administrators.

Instructors in the HEIs ought to note one thing: they were always rated lower than the courses they taught and the variation of the ratings were higher for the instructors than for the courses they taught. This means that the quality of the instructors is not consistent and needs to be brought into better alignment.
We also looked at another variable: grade expectations. The findings were quite revealing. In the private universities (5-8), this expectation was really high -- 79-85% of the students expected A-grades in a randomly selected course; students in only one