Clinical Teaching in Physical Therapy: Student and Teacher Perceptions

Many practicing pbysical therapbtspartic@atein the most crucial phase of a student's education by seruing as Clinical Instructors. The purpases of this study were to identtfv the clinical teaching behaviors perceived as most effkctive and most hindering by students and CISand to compare the response rates of students in bachelor's and musterk degree programs. A published 58-item questionnaire was completed by 172participants from eight pbysical therapy educationprogram The results were analyzed by multivariate analysis of variance. The perceived most helpful teaching behaviorspmained to providing inforhation throughfeedback. The perceived most hindering behaviors were intimidating questioning and correcting student errors in the presence of patien&. The dt#erent student and CI ratingsfor the item 'leaz~es student alone until asked to supembe" has important ethical and educational implications.Master's and bachelor's degree students' ratings dtffered sign@cantly on four teaching behaviors. Dt#erent in.slru~tionalmethods might be necewy for educating these s&&. Oarski RW, Kulig K, Olson KE: Clinical teaching in pbysical therapy: Student and teacher perceptions. Phys Ther 70:173-178, 199Oj

Robert W Jarski Kornelia Kulig Ronald E Olson

Key Words: Education:pbysical therapist, clinical educationlteaching methods; Education, professional; Teaching.

Clinical instruction constitutes a major portion of the physical therapy curriculum. Clinical teaching involves exposing a student to conditions, usually in an active patient setting, where the probability of learning clinical information is high. The process uniquely involves a student, a Clinical Instructor, and patients. The conditions include the physical environ-

ment as well as the behaviors and attitudes of all people in the clinic where the teaching takes pkace. The learning outcomes may be intentional or unintentional, positive or negative. Because students in clinical education programs learn behaviors that influence their lifetime professional performance, improvement of physical therapy professional services depends

R Jarski, PhD, PA-C, is Associate Professor o f Health Sciences, Oakland University and Meadow Brook Health Enhancement Institute. School of Health Sciences. Rochester. MI 48309-4401. K Kulig, PhD, FT, is .4ssistant I'rofessor of Physical Therapy, Oakland University, and Staf Physical Therapist, Oakland Physical Therapy and Rehabilitation, PC, Novi, M I 48050.

to a great degree on maintaining a high quality of clinical education. Many practicing physical therapists in hospitals as well as outpatient settings are involved in the clinical phase of student education. Effective clinical instruction in these settings is believed to require a unique subset of teaching skills,'-5 and specific teaching behaviors should be identified and evaluatedl.2 for at least three reasons. The first reason is to promote and help ensure a positive and constructive learning" ex~erience(j.7so that appropriate skills, behaviors, and attitudes future professional praaice are not only learned but also assimilated. The clinical phase is where the theoretical and practical educational components are integrated into reallife situations with actual patients 1

K Olson, PhD, is Dean and Professor, School of Health Sciences, Oakland University Address all correspondence to Dr Jarski at Program in Physical Therapy, Oakland University, Kwhester, MI 48309-4401 (USA). This project was funded in part by a grant from the Oakland University Educat~onalDevelopment Fund.

7bk arlicle u~assubmitted Februun, 2.3, 1969; ulas ulith the authorsfor retlisionfor 14 uleeh; and u~asaccqied Nozlember 8, 1989.

36 / 173

Physical TherapyNolume 70, Number 3/March 1990

brought into the instructional process. Clinical Instructors serve as role models for students and should facilitate the integration of the educational components. Clinical Instructors, therefore, should exemplify the highest caliber of cognitive, interpersonal, and humanitarian qualities. Second, physical therapy education is moving toward a master's degreelevel curriculum for entry-level professional graduates.8 It is not known, however, whether students entering bachelor's and master's degree programs actually differ in their instructional needs or whether specific teaching practices are required to meet these needs. This knowledge may be especially important in clinical instruction because of its preponderance of personal interactions. Different prerequisites are usually required for admission to graduate programs. Admissions committees, therefore, might select applicants differently, and applicants might also select programs differently. 'Third, many CIS have not had formal preparation in education and have been selected, not because of their teaching abilities, but because of their professional skills.9 Because the qualities that constitute effective clinical teaching in physical therapy education are not well publicized, CIS may lack information and direction in planning their professional development activities. Currently, it is often not possible to accurately assess clinical instruction. The primary purpose of this study was to identify the teaching behaviors perceived to be the most effective and those perceived to be the most hindering by students in physical therapy clinical education programs and by CIS.The secondary purpose was to compare teaching behavior ratings of students in bachelor's and master's degree programs. Our null hypothesis was that there would be no difference between bachelor's and master's degree student ratings.

Method A list of 58 teaching behaviors thought to be effective or ineffective have been identified by Gjerde and Coble3 based primarily on the theoretical work of Stritter et al.5 Gjerde and Coble developed a questionnaire for use in family medicine.3 The questionnaire items address four areas important in clinical education: communication skills, professional skills, interpersonal skills, and andragogic (adult education) skills. With the authors' permission, we adapted the questionnaire for use in allied health education by changing some of the terminology. For example, we changed the word "resident" to "student." Two questions regarding student or CI status and professional degrees were added at the beginning of the instrument. To investigate the role of CIS in clinical practice, an additional item was added at the end of the instrument: "Is actively and regularly engaged in clinical practice." The five-point (1-5) rating scale of the original instrument was expanded to a seven-point scale to increase the numerical choices for rating behaviors. The ratings were weighted as follows: 1 = very helpful, 2 = moderately helpful, 3 = slightly helpful, 4 = neither helpful nor hindering, 5 = slightly hindering, 6 = moderately hindering, and 7 = very hindering. Results were analyzed for 1) the entire group of respondents, 2) the differences between student and CI ratings, and 3) the differences between ratings by students in bachelor's and master's degree programs. To help increase the accuracy and general applicability of the findings, two physical therapy programs were selected from different states in each of four different geographical regions of the United States: Northwest, Northeast, Southwest, and Southeast. By chance, four programs were bachelor's degree programs and four were entry-level master's degree programs. For the purposes of this study, this distribution of bachelor's and master's degree programs was retained in the sample.

Physical TherapyNolume 70, Number 3/March 1990

Questionnaires with instructions and postage-paid return envelopes were mailed with an accompanying letter to the program directors. The program directors were asked to distribute a questionnaire to each CI who was actively involved in clinical teaching and to each student who had completed at least one clinical rotation. In the questionnaire, the CIS were asked to indicate the length of time they had been involved in clinical teaching and the students were asked to indicate the duration of their clinical learning experience in their present program. A cover letter attached to each questionnaire explained the general purpose of the study and that participation was voluntary. The project was approved by the Oakland University (Rochester, Mich) Institutional Review Board. One hundred thirty-nine students and 33 CIS returned completed questionnaires. The characteristics of the subjects are shown in Table 1. The percentage of questionnaires returned by each program varied from approximately 14% to 100%. It was not possible to determine the exact percentage returned because questionnaires were distributed by the program directors. The various return rates possibly could have affected the results of the study. In an unpublished 1988 study on physician assistant students using the same survey instrument and sampling method, we tested for possible rating differences among 10 physician assistant programs with high and low return rates. Instrument ratings from the programs having the two highest return rates (94% and 60%; n = 38) were compared with the responses from the five programs having the lowest return rates (28%-33%; n = 34). A comparison of mean ratings by t tests showed no significant difference for any of the 58 questionnaire items. We concluded that the different return rates did not significantly affect the results. We expected that the different response rates would not affect the results of the present study because I ) the allied health professions have many shared characteri~ticsR.10~11 and 2) the questionnaire

items in both studies addressed educational issues generally applicable to all students in clinical education programs and were not emotional or personal items for which sampling biases would be highly probable. The perceived most helpful and most hindering teaching behaviors were identified by ranking mean ratings. To test for behavior rating differences between CIS and students and between students in master's and bachelor's degree programs, the following procedures were used. Instrument ratings were analyzed by multivariate analysis of variance to detect overall item differences. Where a significant multivariate F statistic was found, a follow-up univariate F statistic was calculated to identify individual items that differed significantly. When testing for differences between students' and CIS' ratings, a consenrative alpha level of .Ol was used to decrease the probability of detecting chance differences. When testing for differences between bachelor's and master's degree students' ratings, the conventional .05 alpha level was used.

The perceived most hindering behavior-"questions students in an intimidating manner1'-had a mean rating of 5.88 based on the combined ratings of the students and the CIS.The other 9 behaviors perceived as most hindering were rated between 5.67 and 4.44(Tab. 3). Like the behaviors rated as most helpful, the most hindering behaviors focused primarily on the teaching process. In addition to these behaviors, availability and negative interpersonal skill behaviors were perceived as hindering, especially those pertaining to behavior around patients. Five behaviors were rated differently by students and CIS.The significantly different behaviors are shown in Table 4 along with their univariate F statistics and significance levels. The greatest difference was found in the mean ratings for the behavior "leaves student alone until asked to supervise." Students, with a mean rating of 3.36, considered this behavior helpful (ie, less than 4,the neutral rating on the seven-point scale); CIS,with a mean rating of 5.09, considered this behavior hindering.

Results The 20 behaviors perceived as most helpful by both students and CIS are rank-ordered and listed in Table 2. The mean ratings were between 1.20 ("takes time for discussion and questions") and 1.57 ("demonstrates sensitivity to student needs [eg, feelings of inadequacy, frustration]"). The most highly rated perceived helpful behaviors pertain to the teaching process, such as answering and discussing questions, providing constructive feedback, and facilitating practice and problem solving. Interpersonal behaviors such as "deals with students in a friendly, outgoing manner" were the next most highly rated category. The behavior "is actively and regularly engaged in clinical practice," which was added to the revised survey instrument, ranked 18th with a mean rating of 1.55. This behavior is one of only three professional skill behaviors included among the 20 teaching behaviors rated as most helpful.

Four behaviors were perceived differently by bachelor's and master's degree students. Table 5 lists the behaviors and their univariate F statistics and significance levels. These behaviors related to student supervision, behavior around patients, and basing judgments of students on indirect evidence. Interestingly, all of the behaviors were rated significantly higher (ie, more hindering or less helpful) by the master's degree students compared with their bachelor's degree counterparts.

Discussion The primary purpose of this study was to identify the teaching behaviors perceived to be the most effective and those perceived to be the most hindering by students in physical therapy clinical education programs and by CIS.

-

Table 1. Ue.m@tion of Subjects

(N = 1 72) Variable

Total

Number of programs represented

8

Bachelor's degree

4

Master's degree

4

Number of states represented Number of student respondents by program

8 139

Bachelor's degree

82

Master's degree

57

Number of student respondents by geographic region

139

Northwest

28

Northeast

58

Southwest

37

Southeast

16

Duration of clinical instruction (mo) X

5.64

s

6.01

Mode (N = 30)

2

Number of Clinical Instructor respondents Physical therapists

33 31

Othera

1

No response

1

Duration of clinical teaching (yr)

X

7.22

s

5.27

Bimodalities (N = 4,4)

7,8

"Respondent indicated only Bachelor of Science degree.

Most Helpful Behaviors Gjerde and Coble identified 58 behaviors believed to be important in clinical instruction.3 These behaviors were classified into four slull domains: communication slulls, professional skills, interpersonal skills, and andragogic skills. The mean ratings in our study indicate that andragogic ski1l.sbehaviors pertaining to providing information through feedback or through discussion and answering questions--were perceived as most helpful. Seven (35%) of the 20 behaviors rated as most helphl pertained to providing information. There was also strong support for CIS to provide

I'hysical TherapyNolume 70, Number 3/March 1990

Table 2. Teaching Behaviors Identified as Most Helpful

Rank

Ratinga X

s

BehavloP Takes time for discussion and questions Answers questions clearly Provides constructive feedback Provides students with opportunities to practice both technical and problem-solving skills Is willingly accessible to students Discusses practical applications of knowledge and skills Shares his or her knowledge and experience Creates practice opportunities for students Asks questions that stimulate problem solving Deals with students in a friendly, outgoing manner Emphasizes problem-solving approaches rather than solutions per se Asks questions in a nonthreatening manner Demonstrates a genuine interest in the student Demonstrates enthusiasm for teaching Demonstrates sensitivity to patient needs Summarizes major points at the conclusion of the teaching session Demonstrates skills for students Is actively and regularly engaged in clinical practice Actively promotes discussion Demonstrates sensitivity to student needs (eg, feelings of inadequacy, frustration)

"Based o n con~binedratings of 58 teaching behaviors by 139 students and 33 Clinical Instructors. ( 1 = very helpful, 2 = moderately helpful, 3 = slightly helpful, 4 = neither helpful nor hindering, 5 = slightly hindering, 6 = moderately hindering, 7 = very hindering.) b ~ d a p t e dfrom Gjerde and Cable..?

practice opportunities with active participatiori in patient care. Six (30%) of the 20 behaviors perceived as most helpful (30%) were interpersonal skill behaviors such as friendliness toward students, enthusiasm for reaching, and sensitivity to patient needs. Behaviors relating to providing information and discussing questions, however, were consistently rated as more helpful. Few professional skill behaviors were rated among tlne 20 behaviors perceived as most helpful. Even though certain affective behaviors were regarded as important, teaching skills were considered more important in facilitating learning.

Our results are generally consistent with the findings in the family medicine study by Gjerde and Coble.3 A similar survey instrument was used, and the mean ratings of the 18 teaching behaviors identified as most helpful by family practice CIS and residents-in-trainingwere reported. The results of our physical therapy study differed from those of Gjerde and Coble's3 family medicine study on only two items ("willing to admit when he or she does not know" and "is well prepared for teaching sessions"). Apparently, physical therapy and family medicine respondents do not differ widely in their perceptions of effective clinical teaching. This finding would be expected because

Physical TherapyNolume 70, Number 3/March 1990

both physical therapists and family physicians are health care professionals having intensive patient care responsibilities during the clinical education phase of their respective programs. Our results are also consistent with those reponed in the ethnographic study of physical therapy CIS by Scully and Shepard,' but different from the results reponed by Moore and Perry9 and by Emery,' who found that physical therapy students rated interpersonal relations as more important than teaching behaviors. This difference in results is not explained by our study, which found only five significant dserences between student and CI ratings. One possible explanation is the overlap between the categories used in Emery's1 investigation. For example, "communication" and "teaching behaviors" cannot always be clearly differentiated. Using exact definitions of the categories might have increased the precision in categorizing the behaviors in Emery's valuable study.

Most Hindering Behaviors In our study as well as in the study by Gjerde and Cable? the same two teaching behaviors were rated as most hindering (ie, "questions students in an intimidating manner" and "corrects students' errors in front of patients"). Students and CIS did not differ in their perceptions of intimidating behavior in our survey, which called upon subjects to imagine or recall clinical teaching behaviors. Their perceptions, however, might differ in actual practice. For example, a CI might think his or her feedback is helpful, whereas the student might think the CI's teaching behavior is intimidating. Future studies should test for differences between ratings for imagined versus observed teaching behaviors. Videotaped interactions might be instrumental, and rating assessments should consider the teacher, the learner, and the patient.

-

learning might jeopardize the quality of health care or patient safety? Can students recognize when they really need supervision either for learning or for ensuring patient safety?

Table 3. Teaching Behaviors Ident@ed as Most IIindering

Rank

Rating' X

s

Behavior" Questi,ons students in an intimidating manner Corrects students' errors in front of patients Bases judgments of students on indirect evidence Fails to adhere to teaching schedule Fails to recognize extra effort Discusses medical cases in front of patients Is difficult to summon for consultation after hours Appears to discourage student-faculty relationships outside of clinical areas Gives general answers to specific questions Fails to set time limits for teaching activities

"Based on combined ratings of 58 teaching behaviors by 139 students and 33 Clinical Instructors. (1 = very helpful, 2 = moderately helpful, 3 = slightly helpful, 4 = neither helpful nor hindering, 5 = slightly hindering, 6 = moderately hindering, 7 = very hindering.) "Adaptetl from Gjerde and Coble?

Different Ratings by Students and Clinical Instructors

insights into the educational process; students may prefer to develop their own clinical skills without constant, direct supervision-the "discc~verylearning" mode. This finding could also represent students' fear of teacher criticism. That CIS considered this a hindering behavior is possibly due to ethical as well as didactic concerns. Do students know when their

Of the five behaviors rated differently by students and CIS, the greatest difference was for the item "leaves student alone until asked to supervise." Students consider this behavior helpful, whereas CIS considered it hindering. 'This finding poses interesting

-

Table 4. I3ehul1ior.s Clinical Ir2structors and Students Rated Di$erentlf Rating CI

Student

(El

(El

F

P

Leaves student alone until asked to supervise

5.09

3.36

39.44

,001

Fails to recognize extra effort

5.81

4.71

14.45

,001

Fails to set time limits for teaching activities

5.00

4.31

9.59

,002

Answers questions clearly

1.03

1.31

8.16

,005

Discusses medical cases in front of patients

5.48

4.66

7.28

,008

"a = .01.

"Adapted from Gjerde and Cable.'

40/ 177

These judgments should be the prerogative of the experienced clinical professional who is also able to use educational strategies that help students learn. It must be recognized, however, that students and teachers have opposite perceptions of this potentially critical behavior. With this difference in mind, the CI should allow discovery learning only when it has been ascertained that the activity will be safely performed by the student. Different Ratings by Bachelor's and Master's Degree Students The secondary purpose of this study was to compare teaching behavior ratings of students in bachelor's and master's degree programs. Four behaviors were rated differently by these two student groups. Differences in mean ratings show that master's degree students considered as more hindering 1) discussing medical cases in the presence of patients, 2) having their errors corrected in the presence of patients, and 3) having judgments made about them based on indirect evidence. Interestingly, bachelor's degree students rated "leaves student alone until asked to supervise" more helpful than did master's degree students. One possible explanation for these results might be that master's degree students consider graduatelevel education a more serious endeavor than undergraduate education. Further studies might help further explain these findings and add further credence to adopting the master's degree as the entry-level degree.

Conclusions and Recommendations From these data, we conclude that clinical teaching behaviors rated as most helpful pertain to the instructional process. Facilitating a favorable learning environment depends on instructor availability and positive

I'hysical TherapyNolume 70, Number 3lMarch 1990

-

Table 5. Behavior Ratings that I)l$ered Sign@cant(P Between Students in Master's and Bachelor's Degree Program7 Rating Master's Degree Students

Bachelor's Degree Students

Behavior"

(XI

(XI

F

P

Leaves st~~dent alone until asked to supervise

3.91

2.88

11.24

,001

Discusses medical cases in front of patierits

5.09

4.03

7.92

,006

Corrects students' errors in front of patients

6.1 1

5.33

6.08

,016

Bases judgments of students on indirect evidence

5.93

5.22

5.78

,019

"a = .05.

"Adapted from Gjerde and Coble.3

interpersonal relationships with the student, patients, and other personnel. Professional skills appear to play a less important role in effective clinical teaching. As physical therapy programs move toward the master's degree entry level, CIS might need to change some of their behaviors to meet the needs of' their new students. The results of this study should help CIS to I ) assess their own clinical teaching behaviors and 2) identify and practice educational skills found to be most effective.

Many graduate physical therapists are called upon to be CIS sometime during their careers, and all physical therapy practitioners should participate in educating patients. Continuing education planners are encouraged to make education-related sessions available to graduate physical therapists, and practicing physical therapists are encouraged to select continuing education courses in related disciplines. For students, mastering instructional skills should be an educational program objective. It may be achieved by learning essential instructional theory,

Physical TherapyNolume 70, Number 31March 1990

and then practicing by making classroom presentations and assisting in laboratory sessions with proper faculty guidance and supervision. This level of education is consistent with the master's degree level of preparation. Using instructional methods known to be effective should enrich the clinical education of our students and help enhance future physical therapy services for patients. References 1 Emery MJ: Effectiveness of the clinical

instructor: Students' perspecrive. Phys Ther 64:1079-1083, 1984 2 Scully RM, Shepard KF: Clinical teaching in physical therapy education: An ethnographic study. Phys Ther 63:349-358, 1983 3 Gjerde CL, Coble RJ: Resident and faculty perceptions of effective clinical teaching in family practice. J Fam Pract 14323-327, 1982 4 lrby DM: Clinical teacher effectiveness in medicine. J Med Educ 53:80%815, 1978 5 Stritter IT,Hain JD, Grimes DA: Clinical teaching reexamined. J Med Educ 50:87&882, 1975 6 DeCecco JP, Crawford WR: The Psychology of Learning and Instruction. Englewood Cliffs, NJ, Prentice-Hall Inc, 1974 7 Knowles MS: Andragogy in Action: Applying Modern Principles of Adult Learning. San Francisco, CA, Jossey-Bass Inc, Publishers, 1985 8 Tammivaara J, Yarbrough P, Shepard KF: Assessing the Quality of Physical Therapy Education Programs. Alexandria, VA, American Physical Therapy Association, 1986 9 Moore ML, Perry JF: Clinical Education in Physical Therapy: Present StatudFuture Needs. Washington, DC, American Physical Therapy Association, 1976 10 Sch& GE, Cawley JF: The Physician Assistant in a Changing Health Care Environment. Rockville, MD, Aspen Publishers Inc, 1987 11 Hislop HJ: Tenth Mary McMillan Lecture: The not-so-impossible dream. Phys Ther 55:1069-1080, 1975