Covert bias in evaluation of physical therapists students' clinical performance.

[Haskins AR, Rose-St Prix C, Elbaum L. Covert bias in evaluation of physical therapist students' clinical performance. Phys Ther. 1997;77:155-168.]

Key Words: Bias, Clinical education, Cross-cultural education, Cultural diversity, Minority students.

Minorities are underrepresented in physical therapy in the United States, as they are in higher education and in most of the health care professions.[1] Since 1985, the number of minority students in the physical therapy program at Florida International University (Miami, Fla) has dramatically increased as a result of the program's minority student recruitment and retention efforts. Concomitant with the increased number of minority students in the program, however, the physical therapy faculty perceived an increase in the number of problems with students in the clinic during clinical affiliations. Often, a clinical instructor would criticize a student's "professional behavior" or affect. It appeared to the faculty that, most of the time, the problems were related to "personality clashes" and that often the students who were involved were minority or nontraditional students. Although there were no complaints of overt bias, we wondered whether the problems these students were experiencing in the clinic could be due to covert racial or ethnic bias.

It is known that racial and ethnic bias exists in the United States. According to Maguire, "White America has always had problems acknowledging people with marked differences-particularly skin color, language, and culture-as equals."[2](p28)

Polakow-Suransky and Ulaby[3] found that although many white students believe that they no longer discriminate against African Americans, African American students often feel that they are treated unfairly and are continually subject to discrimination from white students and professors. Furthermore, white students frequently continue to hold stereotypical views of minorities, such as, "Orientals are smart, blacks are dumb, and Hispanics are troublemakers."[3](p605)

Research on teachers at all levels has shown that "most white teachers are not prepared by background or training to work with minority students."[4](5) Instructors expect middle-class social characteristics from all students[5] and, at all grade levels, teachers are less likely to try to communicate with minority and female students.[6] In studies of elementary and secondary school children, minority and poor children are usually taught more by rote and less by questioning,[7] which may place them at a disadvantage in higher education, where teachers value discussion and debate.[8] Even in higher education, faculty can view teaching ethnic minority students as a threat,[9] and they tend to avoid the culturally different student.[7]

Faculty attitudes, behaviors, expectations, and teaching strategies may be misunderstood by minority students, resulting in diminished teacher effectiveness in the multicultural classroom.[10] Allen and Niss[11] found that minorities were treated differentially in a college classroom. Although they did not observe overtly racist behavior on the part of the professors, Allen and Niss did find that the professors "displayed subtle negative reactions to minority students (almost all of whom were black)."[11](p608) Teachers were more likely to criticize a minority student's comments, while rewarding and praising the comments of white students. Other subtle biases on the part of the teachers included acknowledging white students more often, exhibiting a warmer tone when conversing with nonminority students, and maintaining longer eye contact with white students.

Jenkins and Bainer[10] believe that subconscious faculty attitudes and behaviors communicate unease and negatively influence the academic success of ethnic and minority students. They state, ". . . much of the differential treatment that may occur in classroom and related instruction is inadvertent, and often below the level of consciousness of both faculty and students."[10](p11) Unconscious biases that categorize students can result in stereotypical treatment, lowered expectations, and inequality in educational opportunities for certain ethnic or racial groups.[12,13]

Understanding the impact of unconscious biases in physical therapy clinical education is important for the profession. Professional (entry-level) physical therapy education programs in the United States rely on clinical instructors to evaluate the performance of students during clinical rotations. May et al[14] confirmed the importance of professional behaviors in clinical success. The most frequent reason for students experiencing difficulties in the clinic is failure to meet the clinical instructor's expectations in the area of professional behaviors. Because culture can define what is considered to be "professional," it seems appropriate to consider whether cultural bias may be responsible for racial or ethnic minority students experiencing difficulty in the clinic.

If there is covert bias in the evaluation of students based on race or ethnicity, it is possible that, based on these negative experiences, minorities are being excluded from the profession if academically capable minority students do not graduate, or if minority physical therapists elect not to participate in their professional association because of negative experiences with majority therapists within the Association. Discrimination by teachers and low expectations of college faculty have been cited as factors contributing to the attrition of academically capable minority students.[4,15]

We are not aware of any published research addressing the attrition of minority students from physical therapy professional education, although there is some evidence that minorities have higher attrition rates than nonminorities in higher education and in the health care professions. Although first-year nonpersistence among college students is high irrespective of race or ethnicity,[16] the college completion rates for black, Hispanic, and Native American students are far below those of white students.[17] Two-year colleges have higher attrition rates for nonwhite students and for those from low socioeconomic backgrounds.[18] There is some evidence that, although the withdrawal rates for allied health education programs are lower than expected when compared with other academic disciplines,[19] the under-representation of minorities may be due to the greater attrition of minority students. In the health care professions, minorities receive only 7.9% of the degrees conferred,[20] and black students not of Hispanic origin have a higher attrition rate (25.4%) when compared with students from all other categories of race and ethnic origin.[21]

Because 94% of physical therapists are non-Hispanic whites,[22] it seems reasonable that most physical therapy clinical instructors, like other white teachers, may not be prepared by background or training to evaluate minority students. Bias in other health care disciplines has been demonstrated in the clinical setting. Williams and Rogers[23] reported that nursing preceptors are poorly prepared to deal with the cultural diversity of nursing students and their potential patients. Kurtz[24] reported that social work students stereotype patients on a socio-cultural basis. We wondered whether physical therapy clinical instructors are prepared to work with minority students. The purpose of this study was to determine whether there is covert bias among physical therapist's when asked to assess a physical therapist student's ability to give a verbal progress report.

Method

Study Sample

Physical therapists attending a fall conference of the Florida Physical Therapy Association (FPTA) were asked to participate in the research study as they walked past the data-collection center located adjacent to the exhibit hall entrance. Eighty-three physical therapists volunteered to participate in the study. The subjects (31 male, 52 female) had a mean age of 41.5 years (SD = 8.2, range = 26-73). Three of the participants were black, 7 participants were Hispanic, and 73 participants were white. The mean number of years as a physical therapy practitioner was 16.9 years (SD = 8.1, range = 1-50). Seven subjects had earned entry-level master's degrees in physical therapy, 69 subjects held baccalaureate degrees, 1 subject had earned an associate degree, and 3 subjects had certificates in physical therapy. For 48 subjects, the highest degree earned was the bachelor's degree; 30 subjects held a master's degree, and 4 subjects had earned a doctoral degree. Thirty-four participants described themselves as clinicians, 33 participants described themselves as administrators, and 12 participants described themselves as teachers. The overwhelming majority of the subjects (n = 79) indicated that they had received their physical therapy education in the United States. Seventy-seven of the respondents indicated that they had supervised physical therapy students in a clinical setting, whereas 4 respondents indicated that they had never supervised students in the clinic. The mean number of students supervised in the past 5 years was 12.4 students. Fifty-five of the participants indicated that they had received formal training or continuing education regarding physical therapy student supervision and instruction, and 27 participants indicated that they had not received training.

Procedure

We created a hypothetical patient, Mrs Henry, who had undergone a total hip replacement. We described her postoperative status (eg, range of motion, strength, functional status, living conditions) in a detailed case study. We then wrote a script wherein a physical therapy student, walking down the hallway of the facility, encounters the referring practitioner, who asks, "How is Mrs Henry doing?" The student's response to this query is shown in the Appendix.

We selected four female physical therapist students of differing ethnic and racial backgrounds (white, Hispanic, Asian, and black). The white student grew up in Georgia, the Hispanic student was Cuban, the Asian student was from the Philippines, and the black student was from Haiti. The Hispanic and Asian students each had a slight accent, and the black student had a marked Haitian accent.

Wanting the students to appear believable in their responses, we prevailed upon a professor in the Theater Department at the university to coach the students until each student appeared to be natural in reciting the script she had memorized. Staff from the university's Instructional Television Department videotaped each woman in the hallway of the Student Health Clinic, responding to the query, "How is Mrs Henry doing?"

A data-collection instrument was developed that included instructions, a copy of the case study, a one-page questionnaire for collecting demographic data about the subject, and a six-item rating scale to be used after viewing the videotape. The instructions indicated that the purpose of the study was to assess the interrater reliability of practitioner assessments of a physical therapist student's ability to give a verbal progress report. The instructions asked the participants to read the case study about Mrs Henry, complete part 1 of the survey (participant information), review the rating scale (part 2 of the survey), view the videotape, grade the student's performance, and return the survey instrument to the researcher.

The case study described Mrs Jane Henry, aged 67 years, on the third day following a left total hip replacement. The case study described her mental status, range of motion, muscle strength, sensation, functional status, endurance, posture, gait, skin condition, discharge status, and current treatment.

Part 1 of the survey, the participant information questionnaire, asked the participants to describe themselves. Data were collected on gender, age, race or ethnicity, highest degree earned, years as a physical therapy practitioner, and physical therapy degree. Participants were asked whether they had received their physical therapy degree in the United States; whether they had ever supervised physical therapist or physical therapist assistant students in a clinical setting; how many physical therapist or physical therapist assistant students they had supervised in the last 5 years; whether they had ever had formal training or continuing education regarding physical therapy student supervision and instruction; and whether they described themselves as administrators, clinicians, teachers, or other.

The rating scale was based on an instrument used by the physical therapy faculty to evaluate oral presentations in the classroom. Each participant was asked to rate the student's verbal progress reports on a scale of 1 to 10 (with 1 being "poor" and 10 being "excellent") in five areas: communication of appropriate information, thoroughness of report, organization of information, clarity of presentation, and maintaining interest. The participant was also asked to give the student an overall rating on a scale of 1 to 10, and space was provided for any written comments that the participant might want to make. No formal reliability data have been obtained for the instrument.

Data were collected by two of the researchers (ARM, CR) at a fall conference of the FPTA. The study, cosponsored by the FPTA Council on Cultural Diversity, was given the working title of "Interrater Reliability in Clinical Education" so as not to betray the real purpose (ie, whether there is covert bias). Revealing the true purpose of the study, it was assumed, might skew the results. Cosponsorship by the Council on Cultural Diversity was not advertised for the same reason.

Participation in the study was promoted to conference participants by inclusion in the conference program, flyers distributed at the conference registration desk, word of mouth, and a sign displayed at the data-collection center. The data-collection center, adjacent to the entrance of the exhibit hall, consisted of a table at which participants could sit to read the instructions and the case study and to complete the questionnaire and rating scale. A videotape monitor/player was positioned so as to be visible to persons seated directly in front of it, but not to passersby or to participants seated at the table. Access to the video equipment was controlled by the researcher.

After entering the data-collection center, participants were given the data-collection instrument. After reading the instructions and the case study and completing the questions on demographic data, subjects were escorted individually, in pairs, or three at a time to view one videotape, depending on the number of people waiting to view the videotape. The presentation of videotapes was systematically varied for each successive escorted individual, pair, or trio to ensure approximately equal numbers of therapists viewing each videotape. The videotape took approximately 1 minute to view. The subjects were then instructed to return to rate the student on a scale of 1 to 10 on five characteristics and to give the student an overall rating. Therapists who viewed the videotape together were not allowed to converse. Of the 83 physical therapists surveyed, 20 therapists viewed videotape 1 (the white student), 19 therapists viewed videotape 2 (the Hispanic student), 27 therapists viewed videotape 3 (the Asian student), and 17 therapists viewed videotape 4 (the black student).

Data Analysis

Data were analyzed using SPSS for MS Windows, version 6.1 (standard version).(*) We used an analysis of variance and a Scheffe test for post hoc analysis to determine whether there were any differences in the mean ratings received by any of the four students. Written comments were reviewed and grouped by similarity of content based on the judgment of the researchers. A chi-square test was used to determine whether there was a difference in the proportion of negative and positive comments made concerning the students.

Results

The white, Hispanic, and Asian students received higher ratings than the black student on clarity of presentation and overall rating. The white and Hispanic students received a higher rating than the black student on maintaining interest. The white and Asian students received a higher rating than the black student on communication of appropriate information. The white student received a higher rating than the black student on organization of information. The results of the analysis of variance for thoroughness of report were significant (P=.03), but the post hoc analysis indicated that no two groups were different at the P=.05 level (Tab. 1). Chi-square statistical analysis revealed no significant associations between the ratings and the participants' gender, race, or prior experience in supervising students.

Table 1.
Ratings of Students' Presentation by Students' Race/Ethnicity

                 Rating
                      Group 1   Group 2    Group 3  Group 4
                       (White)  (Hispanic)  (Asian)  (Black)
Factor                 (n=20)    (n= 19)   (n=27)   (n= 17)

Communication of
appropriate
information(b)
   X                     8.00      7.26      7.52     5.76
   SD                    2.22      1.72      1.67     1.48
   Range                 3-10      3-10      3-10     2-8

Thoroughness of
report
   X                     7.60      7.00       7.44    5.88
   SD                    2.26      1.89       1.72    1.69
   Range                 2-10      3-10       3-10    2-8

Organization of
information(c)
   X                     7.10      6.79       6.9     4.94
   SD                    2.51      2.09       1.87    2.38
   Range                 1-10      2-10       1-9     1-10

Clarity of
presentation(d)
   X                     7.90      6.53       6.63    3.18
   SD                    1.89      1.80       1.74    1.74
   Range                 4-10      3-10       3-9     1-6

Maintaining interest(e)
   X                     7.00      6.84       6.2     64.47
   SD                    2.71      2.01       1.85     2.21
   Range                 1-10      3-10       2-9     2-9

Overall rating(f)
   X                     7.60      7.00       7.07    4.76
   SD                    1.85      1.91       1.49    1.78
   Range                 3-10      3-10       3-9     1-8

(a) A one-way analysis of variance revealed statistically significant differences (P [is less than] .02) between groups for all rated factors and the overall ratings.

(b) Post hoc analysis via the Scheffe test for differences between groups revealed statistically significant differences (P [is less than] .05) between groups 1 and 4 and groups 3 and 4.

(c) Post hoc analysis via the Scheffe test for differences between groups revealed statistically significant differences (P [is less than] .05) between groups 1 and 4.

(d) Post hoc analysis via the Scheffe test for differences between groups revealed statistically significant differences (P [is less than] .05) between groups 1 and 4, groups 1 and 4, and groups 3 and 4.

(e) Post hoc analysis via the Scheffe test for differences between groups revealed statistically significant differences (P [is less than] .05) between groups 1 and 4 and groups 2 and 4.

(f) Post hoc analysis via the Scheffe test for differences between groups revealed statistically significant differences (P [is less than] .05) between groups 1 and 4, groups 2 and 4, and groups 3 and 4.

The comments written by the participants were reviewed and categorized as to being negative, positive, or related to the testing situation: Those comments relating to the testing situation included comments regarding the noisy background of the area adjacent to the exhibit hall and the small amount of time given to assess the student. Included in this category also were comments made by subjects about having missed the practitioner's question at the beginning of the videotape and about being hard of hearing.

Negative comments were grouped further into those that related to the students' personal style and those that related to the content of the presentation. Comments relating to the students' personal style included comments about an accent that was difficult to understand, poor speech patterns (such as not enunciating clearly or speaking in a monotone), speaking too fast or rushing (included not giving the referring practitioner time to respond to a question), poor eye contact, inappropriate body movement (eg, fidgeting, waving hands, moving body while talking), and appearing to lack confidence or knowledge (Tab. 2).

Table 2. Negative Comments Relating to Personal Style and Content by Students' Race/Ethnicity

                   Frequency of Response
Categorized        Group 1  Group 2     Group 3  Group 4
Comments           (White)  (Hispanic)  (Asian)  (Black)

Accent hard to
understand                                          7

Poor speech
pattern                                     1       6

Spoke too fast,
rushed                           3          2       6

Poor eye contact      2                             1

Inappropriate
body movement                    1          1       1

Lacked confidence                1                  3

Poorly organized      1                             1

Inappropriate         5          4          4       2

Total                 8          9          8      27

Negative comments relating to the content of the presentation included stating that a student was poorly organized or that the content of the presentation was inappropriate (incomplete, too detailed, or not appropriate for what the referring practitioner wants or needs to hear) (Tab. 2).

The black student received the highest number of negative comments (n=24) about her personal style, and the lowest number of negative comments (n=3) about the content of her presentation. The white student received the highest number of negative comments (n=6) about the content of her presentation and the lowest number of negative comments (n=2) about her personal style. Eighty-nine percent of the negative comments received by the black student related to her personal style, whereas only 25% of the negative comments received by the white student related to her personal style. Seventy-five percent of the negative comments received by the white student were about the content of her presentation, whereas only 11% of the negative comments received by the black student were about the content of her presentation. The Hispanic and Asian students both received about the same number of negative comments regarding their personal style and the content of their presentations.

Table 3 illustrates the distribution of positive comments. Positive comments about the students' personal style included comments about appearing confident, caring, or knowledgeable and speaking clearly. Only 4 of the 83 subjects made positive comments in this area. Positive comments about the content of the presentation included comments about the student being well organized or thorough and giving a good or excellent report (including having good questions for the referring practitioner).

Table 3. Positive Comments Regarding Personal Style and Content by Students' Race/Ethnicity

                    Frequency of Response

Categorized         Group 1   Group 2    Group 3  Group 4
Comments            (White)  (Hispanic)  (Asian)  (Black)

Confident, caring,              1          1        1
knowledgeable

Spoke clearly         1

Well-organized,
thorough              2                    1

Good or excellent
report                4         1          1        1

Total                 7         2          3        2

Table 4 displays the distribution of negative and positive comments. Of the 52 negative comments made, 27 comments (52%) were directed at the black student. Of the 14 positive comments made, 7 comments (50%) were directed at the white student. The black student received the highest number of comments (n=29), of which 27 comments (93%) were negative and 2 comments (7%) were positive. The white student received the second highest number of comments (n=15), of which 8 comments (53%) were negative and 7 comments (47%) were positive. The ratio of negative to positive comments was higher for the Hispanic and Asian students than for the white student, but lower than the ratio of negative to positive comments for the black student. Chi-square analysis revealed the distribution of positive and negative comments to be significant (P [isless than] .05).

Table 4. Comparison of Negative and Positive Comments by Students' Race/ Ethnicity

                    Frequency of Response (%)(a)

Categorized        Group 1  Group 2     Group 3  Group 4
Comments           (White)  (Hispanic)  (Asian)  (Black)

Total negative
comments
Count                8          9          8       27
Row percentage      15.4       17.3       15.4     51.9

Column percentage   53.3       81.8       72.7     93.1

Total positive
comments
Count                7          2          3        2
Row percentage      50.0       14.3       21.4     14.3

Column percentage   46.7       18.2       27.3      6.9

Total comments      15         11         11       29

(a) P= .022.

Discussion

There were marked differences in the students' ratings. The black student consistently received the lower ratings, despite reciting the identical script recited by the white, Asian, and Hispanic students, indicating that the black student may have been treated differentially based on her race or Haitian ethnicity or her accent. Previous research on ratings has been performed on other types of students. Dorsey and Colliver[25] found no widespread grading bias in testing for white and African American medical students, whereas Powell and Collier[26] found bias in the oral performance scoring of Latino, Asian, and African American college students as compared with white students. Powell and Collier concluded that the white students mainstream style of speaking caused them to be perceived as more competent and that the nonwhite students were negatively criticized because of their accents or cultural communication styles. The black student's lower ratings in our study may have been due more to her accent, and the negative stereotypes associated with accents, than to her race. Our data do not allow us to isolate the variable of accent from that of race.

A few of the participants made comments regarding the testing situation. Interestingly, although the test setting was noisy, it seemed to be most bothersome to those participants viewing the videotape of the black student. Coupled with the black student's accent, the noise in the testing area may have created an insurmountable barrier for some of those viewing her videotape. Yet, rather than not evaluating the student, viewers of videotape 4 gave her lower ratings across the board. This finding may have implications for clinical instructors supervising students with accents. Clinical instructors need to be aware of their reactions to the students' accents and, where the students' accents do not interfere with student-patient communication, to strive to keep their assessment focused on the students' behaviors and outcomes rather than on their accents.

Many of the participants who viewed the videotape of the white student wrote comments critical of the content of her presentation. Typical comments related to her presentation referred to her being poorly organized, too detailed, or not detailed enough. Only one person commented on her speaking too fast, and two participants commented on her having poor eye contact. Yet, despite receiving the highest number of negative comments on the content of the report, the white student received a higher overall rating than the black student. This finding is consistent with the observation that white students' speaking style leads to their being perceived as competent.[26]

The black student received the highest number of comments, most of which were negative. Most of the negative comments addressed the student's presentation style, and very few comments addressed the content or appropriateness of her presentation. Typical comments related to the black student's presentation were that she had poor eye contact, that she spoke too quickly or too softly, that she did not enunciate, and that her accent was difficult to understand. For the Hispanic and Asian students, negative comments addressed both content and presentation. Although there was no difference in the ratings of the white, Hispanic, and Asian students, both the Hispanic and the Asian students received proportionately more negative comments than positive comments when compared with the white student and more negative comments relating to the style of their presentations than the white student received. These results are consistent with those of other studies that showed that white teachers are more likely to treat minorities differentially and are more likely to criticize minority students[11] and that teachers are more likely to criticize students with accents.[26]

Of all of the positive comments offered by the evaluators, half were directed at the white student, and there were few positive comments for the Hispanic, Asian, and black students. This finding is also consistent with studies that indicate that teachers reward and praise white students more often than they praise minority students.[11]

Limitations

A limitation of this study is that we were unable to determine whether the low ratings received by the black student were due to her skin color or to her very apparent accent, or both. Nonetheless, the coinicidence of these factors (race, ethnicity, and accent) is not uncommon in students, and the results indicate that bias exists against students who possess this combination of characteristics. According to Garrott,[27] Haitian students face difficulties because of attitudes about Haitian culture and racial bias manifested in negative classroom interactions. Pailliotet[28] asserts that bilingual and ethnic minority students experience many racial tensions as they attempt to reconcile conflicts in language and communication. Further study is needed to explore the impact of a regional accent, color alone, accent alone, and color with accent on the evaluation of physical therapist students.

Another limitation of this study could be the rating scale used to assess the verbal progress reports. Because no definitions or examples were given for the rating scale measures (eg, "communication of appropriate information"), it is possible that the participants interpreted the rating scale measures differently. There was no reliability testing of the instrument. None of the written comments, however, indicated that any of the participants had questions about the rating scale.

A further limitation of this study could be that the results may not be generalizable to other areas of the country or to male students. Demetrulias[29] studied student teachers' perceptions of written reports that varied only by the ethnically identifiable surnames of hypothetical student writers. He found that perceptions were altered by the difference in surnames, but that this alteration was different in different parts of the country, and recommended that further research be conducted to investigate the impact of regional differences. The state of Florida certainly represents a unique mixture of racial, ethnic, and cultural groups, and therefore the results we obtained may have been unique to our area. Several studies[30-33] have confirmed gender bias at the college level. Future studies need to be conducted in different geographic locations throughout the United States and using male as well as female students.

Conclusion and Implications

The results of our study seem to indicate that physical therapy practitioners in the United States may not be prepared to work with students of different racial or ethnic backgrounds. We believe that our study shows that there is a need for physical therapy practitioners who supervise students to become aware of and acknowledge their own biases and to actively seek to educate themselves about cultural diversity.

Our study may also have implications for physical therapy clinicians and their relations with other practitioners and clients from diverse backgrounds. Professional, postprofessional, and continuing education programs, in our view, need to educate physical therapists and physical therapist assistants to achieve a multicultural, global perspective.

The lack of training for members of majority cultural and racial groups regarding minority cultural and racial groups has been blamed in part for the failure of our educational system to provide equality and opportunity. [34] There are monocultural teachers who teach from an ethnocentric point of view and who hold ethnocentric expectations.34 Our educational system is "culturally encumbered,"35 reflecting only the values of the dominant group. Pahnos and Butt state, "In order for schools to provide equal education for all students, educators must confront their own biases and teach from a global perspective."[34](p120)

Acknowledgments

We gratefully acknowledge the assistance of two generations of physical therapist students (now physical therapists) who assisted in updating the review of the literature: Cary DeVarona, Jacqueline Enright, Rachel A Jean, Rosa Maria Lee, and Holly Meyerhofer. We also appreciate the statistical consulting help of Dr Paulette Johnson and the suggestions for research design offered by Dr Samuel S Shapiro. Most of all, we are grateful to Karen Fisher for contributing to the formulation of this study, and for the support of the FPTA Council on Cultural Diversity, particularly James Ball, immediate past President of the FPTA, and Craig Crosby, Executive Director.

(*) SPSS Inc, 444 N Michigan Ave, Chicago, IL 60611.

This study was approved by the University Research Council of Florida International University.

This research was cosponsored by the Council on Cultural Diversity of the Florida Physical Therapy Association and presented in poster format at the Spring Conference of the Florida Physical Therapy Association; Orlando, Fla; April 8, 1995.

This article was submitted June 12, 1996, and was accepted October 7, 1996.

References

[1] Committee to Study the Role of Allied Health Personnel, Institute of Medicine. Allied Health Services. Avoiding Crises. Washington, DC: National Academy Press; 1989.

[2] Maguire J. Reversing the recent decline in minority participation in higher education. In: Minorities in Higher Education: At a Turning Point. Washington, DC: American Association of State Colleges and Universities; 1988:21-41.

[3] Polakow-Suransky S, Ulaby N. Students take action to combat racism. Phi Delta Kappan. 1990;71:601-606.

[4] Sedlacek WE, Brooks GC. Racism in American Education: A Model for Change. Chicago, Ill: Nelson-Hall; 1976.

[5] Rist RC. Cited by Garrott CL: Understanding the Haitian Student: Resources for the French Instructor. Murfreesboro, NC: Chowan College; 1983. (ERIC Document Reproduction Series No. ED 362 015)

[6] Darr RF. Teacher characteristics. Presented at the Annual Meeting of the Eastern Educational Research Association; February 14-17, 1990; Clearwater, Fla.

[7] Jones DJ, Watson BC. High-Risk Students and Higher Education: Future Trends. ASHE-ERIC Higher Education Report No. 3. Washington, DC: George Washington University, School of Education and Human Development; 1990.

[8] Cheng L-RL. Recognizing diversity: a need for a paradigm shift. American Behavioral Scientist. 1990;34:263-278.

[9] Wong S, Wong J. Problems in teaching ethnic minority students. J Adv Nurs. 1982;7:255-259.

[10] Jenkins CA, Bainer DL. Common instructional problems in multi-cultural classrooms. Presented at the Lilly Conference on College Teaching; March 1990; Lake Arrowhead, Calif.

[11] Allen BP, Niss JF. A chill in the college classroom? Phi Delta Kappan. 1990;71:607-609.

[12] White-Hood M. African-American males in middle school: one school's efforts at empowerment. The Clearing House. 1991;65(1):4-8.

[13] Mehan H. Understanding inequality in schools: the contribution of interpretive studies. Sociology of Education. 1992;65:1-20.

[14] May WW, Morgan BJ, Lemke JC, et al. Model for ability-based assessment in physical therapy education. Journal of Physical Therapy Education. 1995;9(1):3-6.

[15] Felice LG. Black student dropout behavior: disengagement from school, rejection and racial discrimination. The Journal of Negro Education. 1981;50:415-424.

[16] McJamerson EC, Larke PJ. Data-driven retention research: using institutional research to inform institutional practice. Presented at the Annual Meeting of the American Educational Research Association; March 27-31, 1989; San Francisco, Calif.

[17] Upton JN, Pruitt AS. Financial Assistance to Black Students in Two Adams States: A Policy Analysis. Atlanta, Ga: Southern Education; 1986.

[18] Rendon LI. Facilitating retention and transfer for first generation students in community colleges. Presented at the Rural Community College Initiative; March 1, 1995; Espanola, NM.

[19] Swager S. An analysis of student motivations for withdrawal in a community college. Presented at the Annual Forum of the Association for Institutional Research; May 28-31, 1995; Boston, Mass.

[20] Carter DJ, Wilson R. Minorities in Higher Education: 1992 Eleventh Annual Status Report. Washington, DC: American Council on Education; 1993.

[21] Gupta GC. Student attrition a challenge for allied health education programs. JAMA. 1991;266:963-967.

[22] Minority Membership Statistics. Alexandria, Va: American Physical Therapy Association, Department of Minority/International Affairs; 1994.

[23] Williams J, Rogers S. The multicultural workplace: preparing preceptors. Journal of Continuing Education in Nursing 1993;24:101-104.

[24] Kurtz ME. Students clinical assessments: Are they affected by stereotyping? Journal of Social Work Education. 1989;25(1):3-12.

[25] Dorsey JK, Colliver JA. Effect of anonymous test grading on passing rates as related to gender and race. Acad Med. 1995;70:321-323.

[26] Powell RP, Collier MJ. Public speaking instruction and cultural bias: the future of the basic course. American Behavioral Scientist. 1990;34: 240-250.

[27] Garrott CL. Understanding the Haitian Student: Resources for the French Instructor. Murfreesboro, NC: Chowan College; 1983. (ERIC Document Reproduction Series No. ED 362 015)

[28] Pailliotet AW. Tensions of a language and ethnic minority college student during teacher preparation: a case study. Presented at the Annual Meeting of the American Educational Research Association; April 18-22, 1995; San Francisco, Calif.

[29] Demetrulias DM. Ethnic surnames. Educational Research Quarterly. 1990;14(3):2-6.

[30] Sadker M, Sadker D. Sexism in the classroom: from grade school to graduate school. Phi Delta Kappan. 1986;67:512-515.

[31] Sternglantz S, Lyberger-Ficek S. Sex differences in student-teacher interactions in the college classroom. Sex Roles. 1977;3:345-351.

[32] Hall RM, Sandler BR. The Classroom Climate: A Chilly One for Women? Washington, DC: Project on the Status and Education of Women, American Association of Colleges; 1982.

[33] Treichler P, Kramarae C. Women's talk in the ivory tower. Communication Quarterly. 1983;31:118-132.

[34] Pahnos ML, Butt KL. Ethnocentrism--a universal pride in one's ethnic background: its impact on teaching and learning. Education. 1993;113:118-120.

[35] Howe KR. Liberal democracy, equal educational opportunity and the challenge of multiculturalism. American Educational Research Journal 1992;29:455-470.

Appendix.

Script

Physician

(off camera): How is Mrs Henry doing?

Student:

Oh, she's doing all right, but I'm a little concerned about two things: Her knee flexion is abaut 60 degrees, and she's complaining abaut a lot of pain. She's moving in bed with minimal assistance, and she needs moderate assistance with transfers. She's walking about 7.6 m [25 ft] with a walker, with weight bearing as tolerated. We'll be starting stair climbing in a day or so. What do you think about discharge planning?. . . I'd like to see her go home and get some home health physical therapy. I'd like to do something for her pain.... Could you take a look at her and tell me what you think?

Invited Commentaries

Following are three invited commentaries on "Covert Bias in Evaluation of Physical Therapist Students' Clinical Performance."

This article addresses an ongoing issue that has been quietly discussed in many minority circles but rarely brought out of the closet. The results of the study were not surprising; however, the individuals who need to become aware of and sensitive to their inappropriate behaviors may not consider covert bias a relevant issue and may choose not to read the article. Covert bias is just that--hidden--even to those involved. In addition, covert bias is not owned by white Americans. Each of us is able to attack the surface demons we all carry, but the values, perceptions, biases, and myths that were very carefully taught by our families and loved ones often can be confused with facts. These have been and are still passed on from generation to generation by the people we love and trust. Special training programs, slogans, vision statements, and declarations of corporate values and workplace cultures cannot erase these "truths," which we hold self-evident.

The American Physical Therapy Association has unequivocally espoused a commitment to equal opportunity, the beauty of cultural diversity, and all of the other modern-day concerns with political correctness. The 1992 Combined Sections' workshops in San Francisco, however, did not play to a standing-room-only crowd. We were preaching to the choir. Our members are products of the society at large, so even our foreign-born and foreign-educated members come to the profession replete with individual experiences and teachings that are biased at the very least and outright hostile if an "ugly American" experience is buried in the subconscious. Each one of us carries his or her own bag of cultural myths. Many of us can mouth the right thing when it is expedient or sometimes when objective measures are used, but we apply covert biases when our tasks are subjective. Subtle acts of disrespect can be seen in clinics, among professionals in the work site, and at our professional meeting places. Some of our actions are based in ignorance, but many of our actions are based on our tendency to relate to the individual based on some preconceived notion about a group. These notions may be positive or negative, depending on our own experiences and teachings.

The researchers attempted to control the variables in many ways, but it was unclear how similar the African-American respondent's ratings were. In addition, we must question whether some of the negative responses for the minority students were indeed based on a failure to meet minimum standards for this level of professional, no matter the ethnic group. There are, after all, some skills that must be mastered by all members, and it is wrong to attribute the inability to achieve these requirements to poor race relations. The only consistent element in the study was the written script. Each participant received negative comments on this element. However, I have known individuals who could read a telephone book with great sophistication. Even a poorly prepared script can be finessed by a skilled communicator. If all of the students had the same "acting" abilities, dressed alike, and came from the same socioeconomic backgrounds, the importance of race as the determining factor, rather than class, may have been more apparent.

William A Gamson, in Power and Discontent, stated that "not all social organizations can control who is let in, but many exercise considerable selectivity.... Entry is not an all-or-nothing state. Once in, members have differential access to resources and communication opportunities."[1](p84) Thus, those who lack the requisite skills and training to adequately communicate with physicians (either orally or in written forms), to display confidence in defending treatment choices, and to demonstrate acceptable social skills will be identified as having an absence of "proper" professional qualifications; therefore, a justification on highly legitimate and widely accepted criteria to exclude the disadvantaged group. Clinical coordinators have little control over who gains entry into this social organization we call "physical therapy," but they can ensure "selective exit." Gamson[1] stated that organizations maintain systems of sanctions to reward the "responsible" and to punish the "irresponsible or deviant"--the desirables vis-a-vis the undesirables. We tend to define desirables as those most like ourselves. This tendency often has nothing to do with ethnicity. Class drives acceptance in many instances, but in some cases, the seeds of past teachings still bear bitter fruits, and reaching the class status does not protect the individual from the covert bias based on the group identity.

So, what does one do with the findings of this study? I agree that it might be worthwhile to investigate whether there is a significant difference in findings in other parts of the country, and it would be of interest to some if an African American without an accent is rated the same way, but I doubt that more workshops will change biased clinicians significantly. If anything, they may cause the same backlash that the term "affirmative action" has wrought. Critics hear the term and leap to the conclusion that one is suggesting quotas and diminution of standards.[2] Cultural diversity workshops have become exercises that must be endured, counted, and reported to the CEO. One has only to remember the Texaco fiasco to know that words and slogans are easy. Day-to-day encounters are something else again. Until the kudzu of intolerance is eradicated, those who suffer slights and indignities must not allow themselves or others to drop out of the competitions of life. We must strive for excellence and be aware of the biases in ourselves and others, but use them as opportunities to grow and to teach others in a nonthreatening way. I also recommend some of Colin Powell's Rules[3](p613):

* Get mad, then get over it.

* You can't make someone else's choices. You shouldn't let someone else make yours.

* Don't take counsel of your fears or naysayers.

* Perpetual optimism is a force multiplier.

We must follow the lead and attain the skills of our many successful role models in the society as a whole and in this profession specifically so that our qualifications and abilities cannot be questioned and our self-confidence will preclude having to (as in Marvin Gaye's tune) make somebody our own piece of clay in order to calm the fears of our inadequacies.

References

[1] Gamson WA. Power and Discontent. Reprinted in: Gamson WA. SIMSOC: Simulated Society. 2nd ed. New York, NY: The Free Press; 1969:84-85.

[2] Waterman RH. What America Does Right. New York, NY: Penguin Books; 1995.

[3] Powell CL; with Persico JE. My American Journey. New York, NY: Random House Inc; 1995.

Senora D Simpson, DrPH, PT Washington, DC

The authors have undertaken a sensitive and essential issue for the profession in their study of "covert bias" in the evaluative judgments of physical therapists. Their study was motivated in part through their personal observation that nontraditional and minority students are more apt than other students to experience conflict with clinical instructors. The authors were prompted to ask whether such difficulties were less a matter of personality differences and more the result of covert racial or ethnic bias. Hence, the current research. There is ample evidence that racial and ethnic bias exists in the United States, and that such bias finds expression in our culturally pluralistic health care delivery system.[1] The authors, therefore, are well justified in choosing to investigate the articulation of such bias in physical therapy education. Their study question, findings, and recommendations raise important considerations for the profession.

The research presented subjects with the task of rating a videotaped performance by one of four ethnically different female physical therapist students. Each student recited an identical script about a hypothetical patient. Unfortunately, as the authors acknowledge, there are a few critical methodological issues that were part of the implementation of the study. The most significant issue was the confounding variable related to the black physical therapist student, who also had a strong Haitian accent. As the authors note, this confounding variable does not allow one to differentiate clearly on the basis of race, ethnicity, or accent. A second methodological concern is the use of the evaluation tool for observing the videotape, which was a 1 to 10 scale where there were no anchors given for the ratings, so the participants were likely to be using the scale differently.

In spite of these concerns, this research provides us with several important points for dialogue and strategic planning within the professional community. Americans are increasingly aware of the heterogeneous character of our society. Accompanying this awareness is the moral responsibility to safeguard against discrimination. Physical therapists will note that there is strong antidiscrimination language in the American Physical Therapy Association's (APTA) Code of Ethics ("Principle 1: Physical therapists respect the rights and dignity of all individuals."[2]) and in APTA's nondiscrimination policy ("The APTA prohibits preferential or adverse discrimination on the basis of race, creed, color, gender, age, national or ethnic origin, sexual orientation, disability or health status in all areas including, but not limited to, its qualifications for membership, rights of membership, policies, programs, activities, and employment practices."[3]). The Association has further underscored its commitment to principles of inclusion through APTA's Plan to Foster Minority Representation and Participation in Physical Therapy, which is the basis for many of the Association's activities, including establishment of the Department of Minority/International Affairs in 1988.[4]

The authors remind all of us in professional education of the importance of preparing students with the skills and sensitivities needed for cross-cultural competence.[5] Cultural competence is a set of congruent behaviors, attitudes, and policies that enable professionals in a system, agency, or institution to work effectively in cross-cultural situations.[6] The importance of such competence for clinicians working in culturally pluralistic health care settings has been well documented by medical anthropologists and others who have observed that cultural misunderstandings between caregivers and patients have led to instances of frustration, confusion, nonadherence, and even death.

Culture is a dynamic and multifaceted social phenomenon that affects all aspects of our lives. In addition to ethnicity and country of origin, a person's culture is strongly influenced by his or her age, gender, gender roles, sexual orientation, religion, socioeconomic status, family background, environment, illness, and other life experiences. One of the first steps in multicultural education is recognition of our own cultural frameworks and how they influence all that we do and value. For instance, culture influences what we think is tasty, beautiful, healthy, helpful, articulate, intelligent, productive, and respectful. Language is the primary mode of transmitting culture, and it is one of the foremost artifacts of a culture. The words that make up our vocabularies and the structure and logic of our sentences are obvious cultural constructions.

Less conspicuous cultural features of language are the kinds of nuanced communication cues that were significant to the research on covert bias. Because each student spoke from a script, the evaluators' rankings were influenced by such factors as the students' posture, eye contact, gestures, pacing, and tone of voice. Although they may have thought they were judging the content of the presentation, the evaluators were undoubtedly influenced by each student's presentation style. Sociolinguists have observed that communication cues such as those named above can have a profound discriminatory influence on our interaction with others. The researchers on covert bias have also raised our awareness of this danger. Acting on us like an irritant or, as linguist John Gumperz[7] says, like a fingernail on a blackboard, these subtle differences in presentation style can cause offense where no offense was intended or can sway opinions based on less-than-objective reasoning. These inappropriate feelings and judgments result from ethnocentrism, that is, the tendency for us to view our own cultural orientation--in this case, communication style--as the only legitimate, or most highly valued, perspective on reality.[1]

The research on covert bias among physical therapists raises our awareness of the importance of recognizing and guarding against the harmfulness of this human tendency. Although cross-culturally we may share many beliefs, values, and attitudes that guide our behaviors, we must not assume that our views and lifestyles are universally accepted or esteemed, nor must we judge others without conscientious examination of the potentially damaging influence of our culturally constructed biases.

The research leads us to ask of ourselves reflectively, as health care professionals: How can we best support our moral obligations to create a more just and humane society and health care system?

References

[1] Krebs GL, Kunimoto EN. Effective Communication in Multicultural Health Care Settings. Thousand Oaks, Calif: Sage Publications Inc; 1994.

[2] Code of Ethics. Alexandria, Va: American Physical Therapy Association; 1996.

[3] House of Delegates Policies. Alexandria, Va: American Physical Therapy Association; effective dune 1995.

[4] Monahan B. Reaching out for a more diverse profession. PT-Magazine of Physical Therapy. 1994;2(2):34-45.

[5] Brody C, Wallace J, eds. Ethical and Social Issues in Professional Education. Albany, NY: State University of New York Press; 1994.

[6] Workshop materials from Ethnogeriatric Curriculum: What should we teach and how should we teach it? Stanford, Calif: Stanford Geriatric Education Center;January 10-11, 1992.

[7] Gumperz JJ. Discourse Strategies. New York, NY: Cambridge University Press; 1982.

Gail M Jensen, PhD, PT Associate Professor Departments of Physical Therapy and Occupational Therapy School of Pharmacy and Allied Health Creighton University Omaha, NE 68178

Kate Brown, PhD Associate Professor Department of Occupational Therapy and Center for Health Policy and Ethics Creighton University

In 1989, I wrote an article for the special minority recruitment issue of the Journal of Physical Therapy Education. I began that article with a challenge to the reader to imagine a half glass of water and determine whether it were half full or half empty. The point of using this analogy was to raise the question of our perceptions as we view, observe, and judge minority students in physical therapy. My position at that time was that invariably minority students are judged to be half empty rather than half full. This current research by Haskins and colleagues reconfirms that there is still a big problem within our profession regarding the performance evaluations of nonwhite students. The intent of this research must be applauded. I hope that this research will serve as a wake-up call and a strong beginning for a larger national study that will permit a more thorough analysis and a stronger ability to generalize the findings to the profession at large. This is one of very few studies regarding minority students in which the investigators studied a population of instructors. It is most disconcerting to discover that at a time when our client population is becoming extraordinarily diverse, our profession appears not to be advancing its numbers or its posture with respect to "valuing diversity." Do these data not beg the issue of how these therapists regard the people of color who are their patients? Is this an indication that there is a lack of respect for people of color? Is this a posture we can tolerate in a global society? I think not!

The issues surrounding race, ethnicity, and culture should be paramount within each physical therapy curriculum in our nation. Each curriculum is a reflection of the collective values and beliefs of a faculty. Therefore, until faculty, both clinical and didactic, make the commitment to thread the appropriate concepts throughout the curriculum, we will remain a profession that is viewed as not participating fully in meeting the needs of our pluralistic society. Change will not occur until we are willing to become analytical at a very deep level and make a commitment to move beyond the awareness level on the behavioral hierarchy.

These authors have been extremely careful in interpreting their data. Their findings are quite profound. I would urge each reader, however, to be less cautious and focus on the questions that are implicit in this research report. Use them as a stimulus for assessment of personal and professional development with regard to your knowledge, perceptions, and actions relevant to race, ethnicity, and culture. I am confident that the evidence is available to support that these variables are indeed salient and merit inclusion in physical therapy education, practice, and research. These data have some frightening implications.

Lynda D Woodruff PhD, PT Professor Graduate Program in Physical Therapy North Georgia College Dahlonega, GA 30597

Author Response

We are grateful for the commentaries provided by Dr Simpson, Dr Jensen and Dr Brown, and Dr Woodruff. Throughout the years that we have been conducting this research, we have often been asked, "So what?" It seems to us that the commentaries are powerful reminders of that "so what" from practice, philosophical, and academic perspectives. We are gratified that our distinguished colleagues deem this study sufficiently important to the profession and have thus taken their time to provide us with their views.

Dr Simpson raises important issues from a practice perspective. Dealing on a daily basis with the effects of covert bias, particularly in a society that equates providing equal opportunity with a lowering of standards, can be a disheartening and embittering experience. Newspaper headlines such as "Black America Radicalizing: Survey Uncovers More Pessimism" (The Miami Herald; April 15, 1994; p 7A) remind us that a large segment of our population is cynical about the equity of our legal and economic systems. Dr Simpson also addresses the issue of trying to educate our colleagues through workshops. We are all too familiar with "preaching to the choir," as Dr Simpson describes, and that was one of the motives for embarking on this area of research. Like Dr Simpson, however, we ascribe to the principle of "perpetual optimism." We would like to believe that a physical therapist who hears about workshops on cultural diversity and says, "I don't need this. I'm not prejudiced," will now stop and say, "Maybe I should attend. It's not just `them'; it's me, too," and be spurred on to further explore his or her own attitudes and awareness.

Dr Jensen and Dr Brown raise valid and important issues from a philosophical viewpoint. They articulate in an eloquent fashion the importance of culture and ethnocentrism in our perceptions of reality. We are grateful to them for providing this insight and discussion. Addressing a methodological problem, Dr Jensen and Dr Brown note the confounding variable of accent in the Haitian student, as we acknowledged in our article. Our own unpublished data from a project that used a black student with no accent indicated that covert bias was still present.

We agree with Dr Woodruff that, at a time when American society considers affirmative-action efforts no longer necessary, we "still have a big problem" with the evaluation of nonwhite students in our profession. In addition to raising an alarm that our profession seems not to have advanced in its "valuing [of] diversity," Dr Woodruff raises important issues regarding the academic curriculum. She urges academicians to "thread the appropriate [racial, ethnic, and cultural] concepts throughout the curriculum."

The physical therapy profession is about 20 years late, compared with education or medicine,[1,2] in publicly uncovering covert bias and discrimination in its ranks, but we are not too late to do something about it. That "something" starts with increasing one's awareness; it progresses by vigilantly monitoring one's behaviors and the consequences of those behaviors; it is advanced by extending one's interest to monitoring the environment and its climate on an ongoing basis; and it climaxes with joyously appreciating that "[w]e draw strength from the fact that we all sing America with different accents and dance it with different rhythms" (Pitts L. "But Daddy, I'm tan, not black." The Miami Herald; May 16, 1996; p 1F).

References

[1] Sedlacek WE, Brooks GC. Racism in American Education: A Model for Change. Chicago, Ill: Nelson-Hall; 1976.

[2] Johnson DG, Smith VC, Tarnoff SL. Recruitment and progress of minority medical school entrants, 1970-1972. J Med Educ. 1975;50:713-766.

AR Haskins, EdD, PT, is Associate Professor and Chairperson, Department of Physical Therapy, Florida International University, Miami, FL 33199 (USA) ([e-mail]). Address all correspondence to Dr Haskins.

C Rose-St Prix, PT, is Assistant Professor and Coordinator of Minority Student Recruitment, Department of Physical Therapy, Florida International University.

L Elbaum, EdD, PT, is Associate Professor, Department of Physical Therapy, Florida International University.