|Abstract or Summary
- This study was designed to determine instructor performance
appraisal methods being used in community colleges (in Florida, Iowa,
New York, Texas, and Washington) and to identify factors responsible
for variations in performance appraisal procedures. Questions about
instructor performance appraisal which were examined included: who
does it, for what purposes, what methods are used, and what is
evaluated. Hypotheses which were tested stated that significant differences
do not occur (1) among the states, (2) between the responses
of administrators and instructors, (3) between the responses from
multi-campus and single-campus colleges, and (4) among the various
interaction combinations of state, personnel, and campus factors.
Data were obtained with a mailed questionnaire sent to administrators
and instructors selected at random from faculties in participating
colleges. A total of 160 questionnaires were used, 32 from
each state. The data were statistically analyzed through the multianalysis
of variance technique.
Conclusions were that:
1. The instructor's immediate supervisor is primarily
responsible for appraisal, but contributions are provided by
students and other administrators.
2. Important purposes for appraisal include: improvement
of instruction, to encourage retention of effective personnel,
and articulation of goals and objectives which affect supervisor
3. Rating scales are the most extensively used method
4. Criteria for measuring instructor performance
emphasized: classroom interaction with students; attitude,
judgement, initiative, leadership; classroom management;
instructional planning; commitment to institutional goals; interaction with faculty; and improvement since last
5. Significant differences occurred among responses
from the five states for 25 of the 60 items on the questionnaire.
6. Significant differences between administrator and
instructor responses occurred for 19 of the 60 items.
7. Multi-campus and single-campus colleges were essentially
not different (the null hypothesis was rejected in four
8. The null hypothesis stating that there are no significant
interaction differences for the mean scores of the variables in
the study was rejected in 23 instances, but no pattern of rejection