- The purpose of this study was to explore the use of a multidimensional
approach to quality assessment of master's programs in the
context of the human judgment process. Specific steps included:
(1) selecting general dimensions of quality; (2) selecting specific
measures to define dimensions; (3) use of a research model to examine
the importance of dimensions in judgment patterns of four subject
groups; and (4) consideration of the implications of such a study for
quality assessment at the master's level, traditional and non-traditional
programs and students, student services, and further research.
The literature review considered the history and development of
the master's degree, duality assessment techniques, and the iudgment
research model employed in this study. It was concluded that the
master's degree is assuming an increasingly important role in higher
education, and that a corresponding need for improved methods of
quality assessment is indicated.
In addition to the multidimensional approach adapted to the
purposes of this study, the lens model was chosen as the experimental
paradigm for examining the human judgment process in quality assessment.
Seven general quality dimensions were selected including faculty,
students, resources, environment, offerings, alumni, and overall program.
A panel of experts assisted in the selection of specific measures to
define each dimension. The final instrument included a set of instructions
followed by 35 program profiles. Profiles consisted of a
bar graph with each bar representing one of the seven quality dimensions.
Subjects reviewed each program profile and rendered a judgment
of the program's quality on a separate response sheet.
Subjects included 15 master's candidates completing degrees
through external programs, 15 master's candidates completing degrees
through resident programs, 15 faculty members serving on university
councils responsible for reviewing master's programs, and 13 academic
Subjects were tested individually over the 35 judgment tasks.
Profiles included in the tasks were developed with the following considerations
in mind: to test for averaging and consistency tendencies;
to examine influences of major variations in certain dimensions on the
judgment patterns of subjects; and to present profiles that reflected
random settings for each of the dimensions. The separate response
sheet contained 35 criterion scales ranging from one to nine with nine
representing the highest level of quality.
The subject's use of quality dimensions was determined by computing
individual product-moment correlations. These correlations
measured the extent to which judgments of quality varied systematically
with the level of each of the dimensions in the profiles. Correlations
were converted to z values and subjected to a four by seven (groups
by dimensions) repeated measurements ANOVA. A judgment analysis
procedure was employed to identify natural clusters of subjects
sharing similar dimension utilization patterns in their judgments.
Results revealed no significant interaction effects between
groups and dimensions, nor significant differences between the four
subject groups. The judgment analysis procedure did, however, reveal
two natural groups with dissimilar dimension utilization patterns.
The first group utilized all dimensions in a more balanced fashion
than did the second group. Finally, significant differences were
observed in comparing the extent to which each dimension was used
by all groups. Dimensions concerned with the quality of faculty,
alumni, resources and overall program were used significantly more
than those concerned with quality of students, environments, and