A number of studies have attempted to identify the components of the c
linical teacher role by examining learners' numerical ratings of items
on researcher-generated lists. Some of these studies have also compar
ed different groups' perceptions of clinical teaching, but have not di
rectly compared the perceptions of first- and third-year residents. Th
is study addressed two questions: (1) What do residents consider impor
tant components of the clinical teacher role? (2) Do first- and third-
year residents perceive this role similarly? Method. A content analysi
s was performed on the comments written on evaluation forms by 268 res
idents about 490 clinical teachers over a five-year period (1980-81 th
rough 1984-85) at a large family practice residency. Of 5,664 forms co
mpleted by the residents, 2,388 (42%) contained written comments; comm
ents were on 1,024 (46%) of the first-year residents' forms, 701 (41%)
of the second-year residents' forms, and 663 (39%) of the third-year
residents' forms. Themes in these comments were coded into a coding di
ctionary of 157 categories, within 37 clusters, within four roles. Res
ults. The ten highest-ranked categories (Global; Teaching: General; Kn
owledgeable; Gives Resident Responsibility; Supportive; Miscellaneous;
Interested in Teaching; Clinical Competence; Makes Effort to Teach; a
nd Gives Resident Opportunity to Do Procedures) accounted for 41% of t
he themes coded. The first- and third-year residents differed in the c
lusters they used to describe their clinical teachers on evaluation fo
rms (chi2 = 149.86, df = 36, p < .0001). Conclusion. The results sugge
st that content analysis can be used to validly and reliably study res
idents' written evaluative comments about their teachers. This study c
ontributes to the definition of the clinical teacher role, showing the
relative importances of its components, and also supports Stritter's
Learning Vector theory, finding the anticipated differences between th
e comments made by first- and third-year residents.