Do Different Pairings of Teaching Styles and Learning Styles Make a Difference? Preceptor and Resident Perceptions

Abstract

Background: Certain teaching style (TS) and learning style (LS) combinations may enhance learning. Purpose: The objective was to examine the effects of combinations of TS and LS in preceptor–resident (PR–RE) dyads in a long-term teaching–learning environment. Methods: Forty-four pediatric PR–RE pairs responded to Grasha's TS and LS inventories, the Clinical Teacher Characteristics Instrument (CTCI) and the Preceptor–Resident Relationship Inventory (PRPRI). Combined CTCI and PRRI means were compared using the t test. Results: The facilitator or personal model TS and collaborative LS pairs supported a healthy teaching–learning environment. Independent learners rated their preceptors' teaching characteristics positively. Competitive LS residents rated the relationship with the preceptor and preceptor teaching effectiveness less favorably. Residents overall indicated a less favorable relationship and clinical teaching characteristics than preceptors. Conclusions: Certain combinations of TS with LS are perceived by preceptors and residents as more positive than others. This suggests TS and LS should be considered when pairing residents and preceptors.

Would you prefer to be paired with a preceptor whose teaching style complements your learning style or would you prefer that pairing not be based on teaching and learning styles? And if this pairing did occur, would it affect the teaching–learning relationship? The question of combining teaching style and learning style has been the center of much debate in the educational field. Although it has been suggested that learners can benefit from complementary teaching and learning styles, ^{1 , 2} it is also possible that learners can be stimulated when these pairings are deliberately noncomplementary. ^{3 , 4 , 5 , 6 , 7} Furthermore, if a fixed combination of teaching style and learning style occurs regularly, it is possible that learners may become bored with the learning environment. Grasha, ^{6 , 8} Dember, ⁹ and others in education and the social sciences have suggested that optimal learning occurs when there is some amount of tension in the learning environment so that learners feel challenged while learning. What is less clear within this debate is whether combining certain teaching styles and learning styles affects the teaching–learning relationship, and if so in what way.

Sometimes educational objectives dictate which teaching style and learning style will be the most effective. For example, one cannot learn to do procedures or perform an ophthalmoscopic examination effectively by only reading texts, although this may be the preferred style of some learners. Furthermore, some teaching and learning styles may not always be the most suitable for specific environments. For example, the emergency room or a busy office is often not the place for slow, reflective processing, although again, this may be a preferred style of some learners. Because medical precepting involves a complex set of teaching, supervision, and mentoring skills, ¹⁰ it is important that we recognize the variety of factors that can affect the interaction between teacher and learner. Three aspects of this important learning dyad are teaching styles, learning styles, and the interaction of teaching and learning styles.

TEACHING STYLES

Teachers tend to have preferred teaching styles with which they are comfortable and revert to in chaotic situations. Montauk and Grasha ¹⁰ outlined five positive preceptor styles with a similar progression and include Expert, Formal Authority, Personal Model, Facilitator, and Delegator. (These styles have been summarized in more detail in other references. ^{11 , 12} ) According to Montauk and Grasha's model, Expert preceptors are likely to possess knowledge and expertise. They tend to direct learners and emphasize factual information. Formal Authority preceptors are similar to expert preceptors, but they have status among learners because of their authority and position in addition to their knowledge. Formal Authority preceptors usually follow “traditions” and standards of medical practice while focusing on the expected rules for learners. The Personal Model preceptor is likely to lead by personal example while suggesting guidelines for appropriate behavior. This preceptor often shows learners how to do things and wants learners to observe and emulate their approach to medicine. The Facilitator preceptor emphasizes the personal nature of the teaching-learning relationship and is more likely to ask questions; explore options with learners; and emphasize learner responsibility, independence, and initiative. The Delegator preceptor encourages learner responsibility and initiative but acts more as a “resource person” with the goal of providing autonomy to the learner. The Delegator preceptor will answer learner questions and periodically review learner progress. All of these preceptor styles are associated with particular teaching roles, attitudes, behaviors, and preferred methods and tend to complement certain preferred learning styles, as described next. All teaching styles have particular advantages and disadvantages, which have been described elsewhere. ^{11 , 12} Consciously adopting a particular teaching style requires consideration of learning style of learner, capability of learner, maintenance of productive interpersonal relationship with learner, type of setting, content, available time, and comfort/conceptual base of preceptor. Adaptability to all teaching styles is an important tool that prepares preceptors for a variety of teaching conditions where they can appeal to a greater variety of learners and their learning styles.

Some research has been conducted examining teaching styles with regard to academic rank of professors, course level, gender, and academic discipline. ¹¹ More relevant to this study is the research about one-on-one teaching–learning situations such as graduate students with thesis/dissertation advisors. In these situations, learners were more likely to evaluate their advisors as employing the Delegator and Facilitator styles more often than the other teaching styles. ¹³ However, in clinical teaching, the teaching styles of Personal Model, Facilitator, and Delegator occurred almost evenly among faculty (25%, 22%, and 24%, respectively).

LEARNING STYLES

Likewise, many learners would prefer to remain within their “comfort zone” by utilizing their dominant learning style. ³ These preferred styles can change over time, especially upon exposure to different teaching styles, but change requires effort and work. Learning style models emphasize various aspects of learning, cognition, and thinking and usually include both the learner and the learning environment in the theoretical conceptualization. ¹⁴ Learning style “attempts to explain learning variation between individuals in the way they approach learning tasks” ¹⁵ (pp. 226–7).

Grasha ¹¹ described six predominant learning styles based on how learners interact with their peers and their instructors. The styles are Independent, Dependent, Collaborative, Avoidant, Participant, and Competitive, and these styles have clear ties to Grasha's teaching styles described above. Independent learners prefer to think for themselves and are confident about their learning abilities. They prefer to work alone, learning content that they think is important. Dependent learners show little intellectual curiosity and learn only what is required. They look to authority figures, teachers, and peers for specific guidelines on structure, support, and what to do. Collaborative learners enjoy working with peers and teachers, and they believe they can learn by sharing ideas and talents. Avoidant learners tend to be uninterested and/or overwhelmed by the learning situation. They are not enthusiastic and do not participate in the learning process. Participant learners are the “good citizens.” They are eager to please and will do as much as is required to meet requirements. They enjoy most learning activities and are likely to participate actively in the learning process. Competitive learners compete with their peers for grades and like to be the center of attention receiving recognition for their accomplishments. All of the learning styles have particular advantages and disadvantages, which have been described elsewhere. ^{11 , 12}

Grasha and others have examined the distribution of learning styles in university students. In one noteworthy study, Montauk and Grasha ¹⁰ examined the learning styles of 150 entering 1st-year medical students compared to a national sample of 1,000 students (nonmedical). Overall, both the entering medical students and the national norm students displayed higher scores in the Independent, Collaborative, Dependent, and Participant styles and lower scores on the Avoidant and Competitive styles. However, the entering medical students, as compared to the national sample of students, saw themselves as more Avoidant, Collaborative, and Dependent. Grasha noted that these findings diminish the stereotype of medical students as “grade-grubbers” who are overly competitive in their learning. In another study of 1,678 students and 84 faculty members at a small Midwestern university, Grasha ¹¹ found that when students and faculty took his learning styles inventory (faculty responded to the test as if they were undergraduate students themselves), the prevalence of the six learning styles among students and faculty were similar. Grasha wanted to know the extent to which the learning styles of students resembled those of their faculty member teachers. Grasha believed that the results suggested that the learning styles of faculty were important to some extent in the modes of instruction they chose to employ. Thus, in some ways, faculty may be teaching to “a projected image of themselves” meaning that professors adopt teaching styles that complement learning styles that they were comfortable with as students. In this study, faculty members, as compared to their students, had higher scores for Independent, Collaborative, and Participant learning styles. Other research less relevant for the current study has examined the relationship of learning styles to academic major, level of student (undergraduate vs. graduate), type of institution (4-year vs. 2-year), gender, and age. Finally, with regard to grades and learning styles, it is no surprise that students with an Avoidant learning style tend to get lower grades as compared to students with other learning styles. ¹¹

INTERACTION OF TEACHING AND LEARNING STYLES IN THE LEARNING ENVIRONMENT

In Grasha's typology of teaching and learning styles, learning styles tend to correspond to or complement certain teaching styles so that the resulting dyad draws on the strengths and avoids the weaknesses of both the teaching style and the learning style. Grasha termed these complementary styles “teaching and learning style clusters.” These dyads have been described in detail elsewhere, ^{11 , 12} however, a brief, summary table (Table 1) is included below.

TABLE 1 Grasha's Teaching and Learning Style Clusters

Teaching style	Learning Style
Expert/Formal Authority	Dependent, Participant, Competitive
Personal Model/Expert/Formal Authority	Participant, Dependent, Collaborative
Facilitator/Personal Model/Expert	Collaborative, Participant, Independent
Delegator/Facilitative/ Expert	Independent, Collaborative, Participant

Grasha noted that learning styles are not “fixed” and can be modified depending on the teaching styles and teaching methods the teacher uses. Teachers have several options for managing their diversity of learners with regard to learning styles: accommodate particular styles, provide creative mismatches, or provide a variety of instructional approaches so that learners are both accommodated and mismatched at times. Andrews ¹⁶ used the Grasha–Reichmann learning styles in a study with chemistry students and found that Collaborative learners reported more benefits when they participated in a peer-centered discussion section. However, Competitive learners in this same study reported benefits from an instructor-centered class. Overall, Andrews noted that students with strong interpersonal styles (e.g., Collaborative, Participant, Dependent) gained the most benefit from participating in review sessions, reading study questions, and learning from other students. In contrast, those students with more “impersonal” styles (e.g., Independent, Avoidant, Competitive) found the textbooks, handouts, and lectures to be most beneficial.

In one-on-one teaching–learning situations, there are similar patterns to traditional classroom teaching in terms of teaching styles and learning styles. Grasha ¹⁷ reported, however, that 77% of the interaction in one-on-one teaching situations can be attributed to the clusters of teaching and learning styles described above whereas in classrooms these clusters describe 92% of the interactions. In addition, the Expert-Formal Authority cluster is more popular in the classroom as compared to one-on-one teaching-learning situations (38% classroom, 6% one on one).

The interpersonal relationship between medical teacher and learner has been examined previously. ¹⁸ Residents' satisfaction with their preceptor and perceptions of their preceptors' effectiveness were negatively correlated with the amount of emotional distance they perceived in the relationship. This study did not, however, consider the influence of teaching and learning styles on the relationship between preceptor and resident.

PURPOSE

In the area of medical education, the preclinical years have traditionally been taught using the didactic approach based on lectures to large groups of medical students. During the clinical years that follow, the apprenticeship model of bedside, rounds-based teaching is introduced and persists into graduate medical education, where it is the cornerstone of resident education. In the last few years increasing emphasis on continuity of care in primary care settings has taken the apprenticeship model to the community physician's office and created a unique teaching and learning environment. In this long-term relationship between preceptor and resident, one would expect teaching and learning styles to emerge as important factors in a successful educational experience. The purpose of this study was to examine the effects of combined teaching and learning styles in preceptor–resident dyads in a long-term teaching–learning environment.

METHODS

Participants

Forty-five pediatric preceptor–resident pairs (90 participants) from a large midwestern pediatric medical center, who were engaged in a continuity training experience (resident spends one to two afternoons/week with primary preceptor for 3 years of residency training), were solicited to respond to a questionnaire containing several instruments. Ninety-eight percent (44 of 45) of the preceptor–resident pairs completed all of the surveys and the teaching and learning styles inventories. The assignments of resident to preceptor were random except that residents could choose which type of setting they preferred (30% were assigned to a hospital clinic, 18% to an off-campus clinic, and 52% to a private pediatric office). Preceptors and residents varied by gender, race/ethnic group, and year of involvement in residency program. All preceptors and residents were personally contacted to participate in the study and had the right to refuse participation. The study was reviewed and approved by the Institutional Review Board of the medical center.

Instruments

The 1:1 Teaching Styles Inventory (1:1 TSI)

This inventory is a modification of Grasha's Teaching Style Inventory, Version 3.0 (TSI¹¹) which categorizes responses into five teaching style categories: Expert, Formal Authority, Personal Model, Facilitator, and Delegator. The modified 1:1 TSI emphasizes one-on-one teacher–learner interactions in a 40-item scale. Preceptors were asked to complete the scale about themselves and their teaching preferences. Each item is scored using a 7-point Likert scale from 1 (very unimportant aspect of my approach to teaching/supervising this learner) and 7 (very important aspect of my approach to teaching/ supervising this learner). Mean score ranges for each of the sets of items related to the individual teaching styles are then calculated, and the mean scores are categorized as either low, moderate, or high (based on standards developed by Grasha) where high corresponds to a preferred teaching style. Because the Likert scale scores represent a spectrum of response to each item (the scoring is not a ranking of items), many responders score themselves high in enough items in different sets of items corresponding to a TS category that they are categorized as having more than one TS. This results in a category of TS labeled “mixed preferences” (where they are categorized as having more than one TS). It also results in more TS than preceptor–resident pairs. Coefficient alphas on the 1:1 TSI for this study's sample were .85 for the preceptors and .87 for the residents.

The Grasha–Vaughn Medical Resident Learning Styles Scale (G-VMRLSS)

This scale is a 60-item self-report measure of medical residents' preferred learning styles. The G-VMRLSS was adapted from the Grasha–Riechmann Student Learning Styles Scales (GRSLSS¹¹), which target high school and college/university students. Both the G-VMRLSS and the GRSLSS are used to identify learner preferences for interacting with peers and the instructor/supervisor in the teaching–learning situation. The internal consistency of items for the GRSLSS range from 0.65 to 0.71.¹¹ Both the G-VMRLSS and the GRSLSS contain 60 items with 10 items per each subscale. Responses are categorized into six social learning style subscales: Independent, Dependent, Competitive, Collaborative, Avoidant, and Participant. Residents were asked to complete the scale about their own preferred learning styles using a 5-point rating scale that ranges from 1 (strongly disagree) to 5 (strongly agree). Similar to the TSI mean score, ranges for each of the sets of items related to the individual learning styles are then calculated and the mean scores are categorized as low, moderate, or high (based on standards developed by Grasha) where high corresponds to a preferred learning style. Because the Likert scale scores represent a spectrum of response to each item (the scoring is not a ranking of items), many responders score themselves high in enough items in different sets of items corresponding to an LS category that they are categorized as having more than one LS. This results in a category of LS labeled “mixed preferences” (where they are categorized as having more than one LS). It also results in more LS than preceptor–resident pairs. Coefficient alphas on the G-VMRLSS for this study's sample were .56 for the preceptors and .63 for the residents.

The Clinical Teacher Characteristics Instrument (CTCI¹⁹)

This instrument is a measure of teaching effectiveness that, although somewhat dated, provides an excellent measure of clinical teaching effectiveness, which includes the teacher–learner relationship and personal attributes as subscales. It is composed of a 20-item scale that has three subscales: Relationship with Students (e.g., “The preceptor encourages residents to feel free to ask questions or to ask for help”), Professional Competence (e.g., “The preceptor relates underlying theory to medical practice”), and Personal Attributes (e.g., “The preceptor is flexible when the occasion calls for it”). Each item is scored using a 5-point Likert scale, from 1 (does not meet this characteristic at all) to 5 (very effectively meets this characteristic). All items are positive characteristics so higher Likert scale scores represent more positive teaching characteristics. Residents were asked to complete the scale about their preceptor, and preceptors were asked to complete the scale about themselves. The CTCI was validated for content by graduate nursing students and faculty ¹⁹ and had coefficient alphas of .87 (preceptors) and .93 (residents) for this study's sample.

The Preceptor–Resident Relationship Inventory (PRRI)

This inventory is a modification of Graen's Leader-member Exchange Scale (LMX²⁰), which was used in this study to measure satisfaction in the preceptor–resident relationship. Graen's original LMX scale consisted of seven items and was designed to measure the quality of exchange between supervisors and subordinates. All items are positive characteristics so higher Likert scale scores represent more positive relationships. An internal consistency coefficient for the LMX scale of .80 was reported by Graen et al., and scores have been shown to be a predictor of employee turnover. ²⁰ Five of the original seven items were used in the current study. The items were modified by (a) substituting the original word supervisor with preceptor and employee with resident and (b) inverting the end points of the 7-point Likert scale to more closely resemble the scale formats of the other instruments and avoid confusion. (Graen's original scale endpoints of 1 [more negative evaluation] and 7 [more positive evaluation] were changed to 1 [more positive evaluation] and 7 [more negative evaluation]). Both preceptors and residents were asked to respond to the measure using their continuity relationship as a frame of reference. Sample items in this scale included, “How well do you feel that your continuity preceptor understands your problems and needs?” and “How would you characterize your working relationship with your continuity preceptor?”

Coefficient alphas on the PRRI for this study's sample were .70 for the preceptors and .89 for the residents.

A Brief Demographic Questionnaire

This questionnaire, developed by the researchers, is a nine-item scale that asked relevant background information about the participants.

Data Analysis

Data were analyzed using SPSS for Windows, release 11.5.0 (September 2002), standard version. The TS was defined for the purposes of this study as the calculated average score for each of the TS which occurred in the “high” range (Expert = 4.9–7.0, Formal Authority = 5.5–7.0, Personal Model = 5.8–7.0, Facilitator = 5.4–7.0, Delegator = 4.3–7.0). Similarly, the LS was defined for this study as the calculated average scores for each of the LS, which occurred in the “high” range (Independent = 3.9–5.0, Avoidant = 3.2–5.0, Collaborative = 3.5–5.0, Dependent = 4.1–5.0, Competitive = 2.9–5.0, Participant = 4.2–5.0).

Standard univariate statistics were used to describe the results of the resident LS and preceptor TS. To examine specific combinations between preceptor TS and resident LS, all possible combinations of preceptor TS and resident LS were tabulated (30 possible combinations) for number of pairings (Table 3). For example, 8 preceptor–resident dyads had an Expert TS in the preceptor and an Independent LS in the resident. Sixteen dyads had an Expert TS in the preceptor and a Collaborative LS in the resident. No dyads had an Expert TS in the preceptor and Participant LS in the resident. Because resident assignments to preceptors were originally made randomly, many preceptor–resident dyad combinations of TS with LS are missing.

For each possible combination of TS and LS, a “Pair” designation was assigned for each particular TS–LS pairing (e.g., 8 dyads with preceptor Expert TS and resident Independent LS) and “All Others” was assigned for all other combinations (36 dyads that were not Expert TS and Independent LS). Means of the two groups (Pair vs. All Others) were compared using the t test on combined items in the CTCI and the PRRI for the preceptors and the residents. Results were tabulated by mean, standard deviation, standardized mean difference (d), and p value from the t test. A p value of .05 or less was considered significant.

RESULTS

Teaching Style and Learning Style Preferences of Participants

Table 2 shows the results of the preceptor TS and resident LS. As explained above, many of the preceptors (28) scored in the “high” category in more than one TS, indicating more than one preferred TS. The majority of these were compatible with Grasha's Teaching Clusters. For example, 13 scored in the high category in the Delegator/Facilitative/Expert cluster and 21 in the Facilitative/Delegator category (includes the previous 13). Three preceptors did not have high scores in any TS category (“No preference”).

TABLE 2 Preceptor Teaching Styles and Resident Learning Styles

TS	No. Preceptors ^a (No. with Single TS)	LS	No. Residents ^a (No. with Single LS)
Expert	17	Independent	18
Formal Authority	1	Avoidant	0
Personal Model	8	Collaborative	42 (23 ^b )
Facilitator	23	Dependent	1
Delegator	41(13 ^b )	Competitive	7
Mixed preferences	28	Participant	1
No preference	3	Mixed preferences	19
		No preference	2

Similarly, in the area of LS, many learners scored in the “high” category in more than one LS (19), indicating more than one preferred LS. Many of these were compatible with Grasha's Learning Style Clusters. For example, 14 of the 19 total were in the Independent/Collaborative/Participant Cluster. Two residents did not have high scores in any LS category.

Assessment of Compatibility of Preceptor Teaching Styles and Resident Learning Styles

In order to assess compatibility of preceptor teaching styles and resident learning styles in the context of enhancing the teaching–learning environment, we evaluated all of the possible combinations of preceptor TS with resident LS as related to the CTCI and the PRRI (as described above in the “data analysis” section). Table 3 (column 2) shows the number of combinations that were available to be evaluated. Because the assignments of resident to preceptor were random (except for type of site), all possible combinations are not present. We compared CTCI and PRRI scores with preceptor–resident TS–LS pairs only when there were at least four combinations to analyze (and only these TS–LS pairs are indicated in Table 3).

TABLE 3 Standardized Mean Differences of PRRI and CTCI Across Combinations

		PRRI Resident			PRRI Preceptor			CTCI Resident			CTCI Preceptor
TS/LS Pairs	No. Pairs	Pair M (SD)	All Others M (SD)	d	Pair M (SD)	All Others M (SD)	d	Pair M (SD)	All Others M (SD)	d	Pair M (SD)	All Others M (SD)	d
Expert/ Independent	8	28.12 (5.49)	27.89 (4.73)	0.04 ^c	27.25 (2.82)	25.03 (4.31)	0.61 ^b	91.75 (8.56)	88.27 (8.83)	0.40 ^c	81.05 (8.86)	82.11 (7.07)	0.13 ^c
Expert/ Collaborator	16	27.56 (4.79)	28.14 (4.89)	−0.12 ^c	26.25 (4.16)	24.96 (4.13)	0.31 ^c	91.34 (7.98)	87.48 (9.07)	0.45 ^c	82.62 (8.67)	81.64 (6.56)	0.13 ^c
Personal Model/ Independent	4	28.75 (5.32)	27.85 (4.82)	0.18 ^c	28.50 (3.11)	25.12 (4.13)	0.92 ^a	95.25 (3.40)	88.27 (8.93)	1.03 ^a	87.50 (5.57)	81.45 (7.29)	0.93 ^a
Personal Model/ Collaborator	8	27.75 (4.89)	27.97 (4.85)	−0.05 ^c	27.75 (4.06)	24.92 (4.03)	0.70 ^b	92.81 (7.62)	88.03 (8.89)	0.58 ^b	87.38 (5.24)	80.81 (7.22)	1.04 ^a *
Facilitator/ Independent	9	28.00 (6.36)	27.91 (4.43)	0.02 ^c	27.00 (3.39)	25.03 (4.26)	0.51 ^b	89.44 (11.08)	88.78 (8.27)	0.07 ^c	84.33 (6.00)	81.40 (7.57)	0.43 ^c
Facilitator/ Collaborator	22	27.64 (5.42)	28.23 (4.21)	−0.12 ^c	26.54 (3.30)	24.32 (4.64)	0.55 ^b	89.25 (10.07)	88.57 (7.45)	0.08 ^c	86.32 (5.66)	77.68 (6.20)	1.46 ^a **
Facilitator/ Competitive	4	23.50 (5.80)	28.38 (4.54)	−0.94 ^a	25.00 (3.56)	25.48 (4.23)	–0.12 ^c	81.00 (9.20)	89.73 (8.45)	−0.99 ^a	85.50 (3.51)	81.65 (7.53)	0.66 ^b
Delegator/ Independent	17	27.71 (5.77)	28.07 (4.20)	−0.07 ^c	24.76 (3.99)	25.85 (4.25)	–0.26 ^c	86.88 (9.87)	90.25 (7.92)	−0.38 ^c	79.71 (7.30)	83.44 (7.08)	−0.52 ^b
Delegator/ Collaborative	39	27.74 (4.97)	29.40 (3.29)	−0.39 ^c	25.67 (4.13)	23.60 (4.16)	0.50 ^b	88.83 (9.17)	89.60 (5.68)	−0.10 ^c	82.46 (7.52)	78.40 (4.34)	0.66 ^b
Delegator/ Competitive	7	25.57 (6.12)	28.38 (4.48)	0.52 ^b	24.43 (3.55)	25.62 (4.26)	–0.30 ^c	84.28 (7.91)	89.82 (8.77)	0.66 ^b	83.28 (4.07)	81.76 (7.79)	0.24
Note: d = standardized mean difference.

Table 3 also shows the standardized mean difference (effect size) in PRRI and CTCI total scores across combinations. Items from both the CTCI and the RRPI are all oriented from negative (less desirable trait = Likert scale 1) to positive (more desirable trait = Likert scale 7 [RRPI] or 5 [CTCI]). A positive mean difference (d) represents a more favorable response from the specific TS–LS pair, whereas a negative d indicates a less favorable response. Two combinations of TS and LS showed significance on the t test and had large effect sizes: Personal model TS - Collaborative LS and Facilitator TS – Collaborative LS.

Table 4 shows a summary of the data related to the PRRI and CTCI by item. Overall Competitive LS residents give a less favorable evaluation of the relationship and preceptor teacher characteristics regardless of the preceptors' TS. In contrast, Independent and Collaborative LS residents report a more favorable evaluation of the relationship with and the teacher characteristics of the preceptor across TS. When the preceptors evaluated the interpersonal relationship and their own teaching characteristics, a similar favorable pattern with Collaborative and Independent LS emerged. There was a tendency for more favorable evaluations of the various combinations as rated by the preceptor (34 positive items and 5 negative items) compared to 13 positive items and 10 negative items as rated by the residents.

TABLE 4 Summary of Significant PRRI and CTCI by Item

Resident as Evaluator		Preceptor as Evaluator
Pair	No. of Significant PRRI Items (Positive ^a or Negative ^b )	Pair	No. of Significant PRRI Items (Positive ^a or Negative ^b )
Facilitator TS – Competitive LS	1 (Negative)	Expert TS – Independent LS	1 (Positive)
Delegator TS – Competitive LS	1 (Negative)	Personal Model TS – Collaborative LS	1 (Positive)
		Facilitator TS – Collaborative LS	3 (Positive)
		Facilitator TS – Independent LS	1 (Positive)
		Delegator TS – Collaborative LS	1 (Positive)
Note: PRRI = Preceptor–Resident Relationship Inventory; CTCI = Clinical Teacher Characteristics Instrument; TS = teaching style; LS = learning style.
Pair	No. of Significant CTCI Items (Positive ^a or Negative ^b )	Pair	No. of Significant CTCI Items (Positive ^a or Negative ^b )
Expert TS – Independent LS	1 (Positive)	Personal Model TS – Independent LS	3 (Positive)
Expert TS – Collaborative LS	1 (Positive)	Personal Model TS – Collaborative LS	3 (Positive)
Personal Model TS – Independent LS	8 (Positive)	Facilitator TS – Independent LS	2 (Positive)
Personal Model TS – Collaborative LS	3 (Positive)	Facilitator TS – Competitive LS	2 (Positive)
Facilitator TS – Competitive LS	5 (Negative)	Facilitator TS – Collaborative LS	13 (Positive)
Delegator TS – Collaborative LS	2 (Negative)	Delegator TS – Collaborative LS	5 (4 Positive, 1 Negative)
Delegator TS - Competitive LS	1 (Negative)	Delegator TS – Independent LS	4 (Negative)

DISCUSSION AND CONCLUSIONS

The purpose of this study was to examine the effects of different combinations of teaching and learning styles in preceptor–resident dyads in a long-term teaching–learning environment based on Grasha's model of teaching and learning styles. Several findings of the study are noteworthy.

When residents and preceptors rated the interpersonal relationships and teacher characteristics, the pairing of Facilitator or Personal model TS and Collaborative LS supported a healthy teaching–learning environment that fits Grasha's concept of the teaching and learning style clusters. This suggests that the combinations with learners who have a Collaborative learning style were perceived more positively overall by the preceptors across two different teaching styles. This may be in part because of the general characteristics of the typical collaborative learner, which includes the enjoyment of learning with peers and teachers. This learning characteristic may also make it easier for preceptors to both get along with and to feel more positive and comfortable trying out various teaching styles with the collaborative learner. In this way, collaborative learners may encourage preceptor flexibility to the teaching–learning environment, which could result in overall positive feelings about the relationship and the preceptor's own teaching abilities.

Collaborative learners were the largest group of learning styles represented among the residents. When paired with personal model preceptors, collaborative learners rated their preceptors higher than preceptors with a Delegator TS, which is intuitive because collaborators require interactive, interpersonal relationships which are less likely to occur by preceptors with Delegator teaching styles because of the nature of the delegator style, which can be perceived as more distant and impersonal.

Independent learners, especially when paired with personal model preceptors, tended to rate their preceptors' teaching characteristics positively. Consistent with Grasha's clusters, this pairing is likely to be effective according to the learners because they are able to watch the personal model preceptor and use a “take it or leave it” policy with regard to learning. In addition, both the independent learner and the personal model preceptor seem to have similar styles in terms of doing things their own way, which may contribute to greater empathic understanding between them about their approaches. Both of these pairings were represented in Grasha's teaching and learning style clusters.

In the TS and LS combinations with a Competitive learning style resident, the resident rated the relationship with the preceptor less favorably and evaluated the teaching effectiveness of the preceptor less favorably regardless of teaching style of the preceptor. This was consistent with the teaching and learning style clusters in which Expert and Formal Authority TS were the only TS compatible with Competitive LS as described by Grasha. ¹¹ A possible explanation for this negative effect on teaching and learning is that competitive learners may place less emphasis on the interpersonal relationship with the teacher and focus more on what they can gain from the relationship to move ahead in their education. In fact, Kiedaisch ²¹ reported that when competitive learners have to collaborate, tension sometimes increases and learning is less effective.

A final noteworthy finding of the study is apparent in Table 4. When the residents were completing the evaluations (the RRPI and CTCI), 10 of 23 areas of significance indicated a negative evaluation (less favorable preceptor–resident relationship and less favorable clinical teaching characteristics for the preceptor–resident pairings). By comparison, when the preceptors completed the evaluations, the majority of the pairings (34 of 39 areas of significance) showed a favorable relationship or teaching characteristics. There are several issues that may explain this finding. The first is that residents have greater expectations of the teaching and learning process as a result of recent changes in American Council for Graduate Medical Education evaluation policies and the increasing emphasis in general on medical education methodologies and evaluation, including increasing emphasis on faculty educator development. A second possibility may relate to the location of the continuity site and the preceptor. Many if not most residency programs are now sending residents to private practices in the community where the preceptor is a practicing primary care physician. As more emphasis is being placed on evidence-based medicine in teaching programs, some residents may not place as much value on the experience of practicing physicians and the importance of the art of medicine. This might result in less favorable evaluations of teaching in the practice setting. By contrast, however, many practicing physicians do find value in having a resident in their office to stimulate teaching and learning and, in so doing, to learn from them. Thus they might find more value in the interpersonal relationship aspect of the preceptor–resident pairing and place more value in the process with more favorable evaluation of teaching. It is also likely that the age difference between preceptor and learner may influence the evaluation process.

A limitation of this study is inherent in the random assignments of resident to preceptor without regard to teaching or learning styles. The result is that there are several combinations we could not examine. Therefore, it is possible that there are some combinations that would have demonstrated significant positive correlations as measured by the PPRI and the CTCI. In addition, because the study comes from a single institution, the numbers of preceptor–resident dyads examined is small. Larger numbers might have demonstrated significance in more teaching and learning style pairs. A second limitation relates to the inventories we used to categorize the TS and LS. As mentioned in the Methods section, when using the 1:1 TSI and G-VMRLSS it is possible to have a preceptor self-score as having more than one TS and a resident as more than one LS. If we had used a tool that used a ranking of items (e.g., the Learning Preference Inventory ² ), only one TS or LS would have been identified for each of the study participants resulting, perhaps, in different results and a cleaner study. We selected Grasha's Teaching and Learning Styles inventories for two reasons. First, we recognize that, in fact, most teachers and learners have a mixture of styles, even though one may predominate. Second, Grasha provides both a TS and LS inventory which are related and can be used to examine clusters of complementary TS and LS.

Future Research

There are many interesting questions that could have been examined using the current data; however, because of a small sample size, such analyses were not conducted. Future research could include comparisons by level of training, ethnicity, type of practice setting, or an examination of the effects of mixed TS and LS preferences on the perceptions of the learners. An even more important study would be to examine whether certain TS–LS combinations improve learning outcomes and lead to improved physician behaviors. In an age of technology, it would also be interesting to consider the role of TS and LS in online educational experiences.

Should preceptors and learners be taught about the concept of TS and LS and be aware of each others teaching and learning style preferences in order to improve their interactions prior to placement in a long-term educational experience? Should preceptor–resident assignments be manipulated to increase the number of certain TS–LS pairs? Although this study does not permit a definitive answer to these questions, it would appear that residents with competitive LS may present a challenge in the learning environment regardless of the preceptors' TS. Conversely, residents with collaborative LS are able to accommodate and do well with most preceptors regardless of their TS. Pairing preceptors and learners regardless of TS and LS may encourage the preceptor to use variety in order to appeal to both stronger and weaker learning styles of the learners and at the same time promote flexibility and versatility in the learners' learning styles. Being adaptable both as a teacher and a learner can ultimately contribute to lifelong learning skills and thus enhance the teaching–learning situation for all involved. This idea of “creative mismatch” (deliberate noncomplementary pairing of TS and LS) in terms of TS and LS has been suggested previously. ^{8 , 11 , 12} Strengthening less-preferred learning styles helps students expand the scope of their learning, become more versatile learners, and adapt to the requisites of the real world. ²² By using a variety of teaching methods and styles, teachers can expose learners to both familiar and unfamiliar ways of learning without designating any individual learner as “different.” This provides both tension and comfort during the process, which encourages adaptability to multiple learning styles and encourages lifelong learning. Situational “matching” and “mismatching” of teaching and learning styles should be considered. Intentional “matching” can be helpful to teachers and learners when one may be weaker in a particular area or be working in an unfamiliar content area. Purposive “mismatching” may be useful when both teacher and learner are highly capable in a particular area and working with familiar material.

Notes

^a Because of the method by which the teaching style (TS) and learning style (LS) inventories are scored, it is possible to have more than one TS or LS per person (see text); that is, more than one mean score in the “high” range; these are referred to as mixed preferences.

^b The number in parentheses refers to the number of preceptors or residents who scored in only one category of teaching or learning style. Thirteen of 44 preceptors scored themselves as solely Delegator TS; 23 of 44 residents scored themselves as solely Collaborative learners. All other preceptors (28) and residents (19) scored themselves in the “high” category of more than one teaching or learning style.

^a Large effect size > .8.

^b Medium effect size > .5 to .79.

^c Small effect size < .5.

*p = .02 (based on t test).

**p < .001 (based on t test).

^a More favorable relationship on PRRI, more favorable preceptor teaching on CTCI.

^b Less favorable relationship on PRRI, less favorable preceptor teaching on CTCI.