Epistemic and Ontological Cognition of Entering and Postdidactic Occupational Therapy Students

Research Design

The study used a mixed-methods cross-sectional design to examine the research question. The study was granted exempt status by the university’s institutional review board.

Participants and Setting

Participants were two groups of entry-level MOT students enrolled in a health science center campus in the mid-South region of the United States. All of the 21 students who were in their 1st wk of the program and 33 of the 35 students who had completed the didactic coursework volunteered to participate. All students had completed 90 credit hours of prerequisite coursework; they may or may not have earned a bachelor’s degree. The program consisted of 81 credit hours of basic science and occupational therapy coursework, including three 2-wk Level 1 fieldwork experiences (one during each of the three 6-mo didactic terms) and three 3-mo Level 2 fieldwork experiences (after the 18-mo didactic coursework).

Instruments and Procedures

Two quantitative self-report measures were used: the Epistemic Beliefs Inventory (EBI; Schraw, Bendixen, & Dunkle, 2002) and Schraw and Olafson’s (2008) Four-Quadrant Scale of Ontology and Epistemology (FQS), with instructions adapted for occupational therapy students (mFQS). The EBI consists of 32 items representing Certain Knowledge, Quick Learning, Simple Knowledge, Omniscient Authority, and Fixed Ability factors. Participants rate the strength of their beliefs on a 5-point Likert scale ranging from 1 = strongly disagree to 5 = strongly agree. The Simple Knowledge and Certain Knowledge (SCK) factors were combined and used as dependent variables, along with the Omniscient Authority (OA) factor. EBI internal consistency reliabilities range from .50 to .65. Test–retest reliability ranges from .62 to .81. Studies have shown that the EBI explains around 40% of sample variance. Evidence of construct validity is available for the five factors of the EBI, and the measure has demonstrated modest but significant predictive validity for reading comprehension (Schraw et al., 2002).

On the FQS, respondents rate their beliefs along a continuum from realist to relativist for both epistemology and ontology. The realist end of each continuum represents beliefs in certain ontology or epistemology, and the relativist end represents beliefs in changing, tentative ontology or epistemology. The epistemology continuum is horizontal, intersecting at right angles with the vertical ontology continuum. The result is four quadrants: realist–realist, realist–relativist, relativist–realist, and relativist–relativist. Each axis of the scale measures 150 mm in length, allowing both ratings to be scored on a scale of 1–150 using a ruler (Schraw & Olafson, 2008). The participants provided narrative explanations for their self-ratings.

No published reliability or validity evidence is available for the FQS, but Schraw and Olafson (2008) piloted the FQS with practicing teachers enrolled in graduate programs and found a statistically significant positive relationship between their epistemologies and ontologies. Consistent with the EOCD model, a naive ontological worldview (OW) did not appear to be compatible with a sophisticated epistemic worldview (EW). Greene et al. (2008, 2010) recognized that additional EOC dimensions may exist, and the mFQS allowed participants to describe dimensions other than those specified by the EOCD model. Schraw and Olafson also suggested that using measures of separate EOC such as the EBI simultaneously with more holistic measures such as the FQS may allow for cross-validation of the measures.

The EBI, followed by the mFQS, was administered to students in a classroom setting either during the 1st wk of the occupational therapy program or within 2 wk of the end of the didactic coursework (i.e., after 18 mo in the program, before Level 2 fieldwork). No time limits were used, and students were assured both verbally and in the written instructions that there were no right or wrong answers.

Data Analysis

The data were first checked for accuracy. Internal consistency reliability of the scales, means, standard deviations, and effect sizes were calculated. An iterative process was used to analyze students’ written explanations on the mFQS; responses were read and categorized on the basis of patterns and similarities and then categories were reviewed and revised as necessary. Categories were either based on the EOCD model or emerged from the data.

Means and standard deviations for the EBI SCK and OA factors and the mFQS EW and OW dimensions were used, along with the narrative data from the mFQS, to describe the students’ EOC. A multivariate analysis of variance (MANOVA) was performed to test for differences among the entering and postdidactic students’ SCK, OA, EW, and OW scores.

Internal Consistency Reliability of the EBI Factor Scales

Initially, the internal consistency reliabilities of the SCK and OA scales of the EBI were .53 and .55, respectively. Elimination of Items 11, 18, and 22 from the SCK scale yielded a 12-item scale with a Cronbach’s α of .62. Elimination of Item 7 from the OA scale yielded a 4-item scale that also had an internal consistency reliability of .62. These amended scales were used in the subsequent analyses.

Descriptive Data

Dependent t tests revealed that for both groups of students, SCK scores were significantly lower than OA scores: entering students, t(20) = −7.969, p < .001; postdidactic students, t(32) = −7.515, p < .001. This finding suggests stronger beliefs in justification by authority than in simple and certain knowledge. Dependent t tests revealed no significant differences between EW and OW mean scores for the entering students, t(20) = −1.262, p = .221. OW scores were significantly lower than EW scores for the postdidactic students, t(32) = 2.211, p = .034, suggesting more sophisticated OW than EW (see Table 1).

Differences Between Entering and Postdidactic Students

Preliminary analyses of all dependent variables (SCK, OA, OW, and EW) using skewness and kurtosis statistics indicated that none departed significantly from normal distribution curves. One postdidactic participant was an outlier at the multivariate level. Profiling revealed nothing remarkable except that this participant placed himself or herself at the extreme relativist end of the EW scale and at the extreme realist end of the OW scale. Removal of this participant resulted in no substantive effect on the means, standard deviations, or MANOVA results. Therefore, the results for all 54 participants are reported. Both the multivariate test of equality of the variance–covariance matrixes, Box’s M = 6.878, F(10, 8462) = 0.626, p = .793, and the univariate tests of homogeneity of variance were nonsignificant at p < .01: SCK, F(1, 52) = 0.002, p = .964; OA, F(1, 52) = 2.315, p = .134; EW, F(1, 52) = 2.785, p = .101; and OW, F(1, 52) = 5.864, p = .019.

The MANOVA was statistically significant, Wilks’ lambda = 0.698; F(4, 49) = 5.303, p = .001, η_p ² = .302; D ² = 1.75, indicating group differences in scores on the set of dependent variables. Bonferroni-adjusted (α = .0125) univariate analyses of variance were used to determine which of the four dependent variables contributed to group differences. As shown in Table 2, the univariate tests indicated that only the OW dimension of the mFQS contributed to the multivariate significance. The postdidactic students’ OW scores were lower than the entering students’ scores, indicating more sophisticated OW; η_p ² = .250.

Figure 1 presents the percentage of students in each group whose self-ratings placed them in each of the four mFQS quadrants. In each group, 1 student’s ratings were at the intersection of the axes, resulting in no quadrant. Thus, the percentages total less than 100.

OPEN IN VIEWER

Figure 1.

Percentage of each group placing themselves in each quadrant of the modified Four-Quadrant Scale.

Substantial differences existed in the percentages of entering and postdidactic students who made comments related to three themes when explaining their mFQS ratings: client as authority, client–practitioner collaboration, and individual realities. None of the entering students made statements related to the client as the authority in intervention or to individual realities, whereas 48.5% and 66.7% of the postdidactic students, respectively, made such comments. Of the 16 postdidactic students who described the client as the authority in making intervention decisions, 15 were in the relativist–relativist quadrant, and the other was in the realist OW–relativist EW quadrant.

The postdidactic students also described how the client’s individual reality and unique beliefs and values would influence the approach used in treatment. One wrote, “I lean more towards both epistemological and ontological relativism because I believe treatment techniques should be different for every client because they all have different values and beliefs.” Of the 22 postdidactic students who mentioned this theme in their comments, 17 were in the relativist–relativist quadrant, 3 were in the realist OW–relativist EW quadrant, 1 was in the relativist OW–realist EW quadrant, and 1 was in no quadrant.

One theme emerged from the entering students’ explanations but not from the postdidactic students’ explanations: Close to half of the entering students made statements related to client–practitioner collaboration, whereas none of the postdidactic students made these types of comments. Of the 11 students who made comments related to client–practitioner collaboration, 6 were in the relativist–relativist quadrant, 3 were in the realist OW–relativist EW quadrant, and 1 was in no quadrant.

Ontological Cognition

According to Greene et al. (2008, 2010), only the realist holds strong views about the simplicity and certainty of knowledge (the term strong views has not been operationally defined). The mean ontological cognition scores of both groups in the current study were lower than the median of 3 on the 5-point EBI scale, and the two groups did not differ significantly on this scale. The scores suggest that both groups had moved beyond the realist position, consistent with Greene et al.’s (2008, 2010) hypothesis that the realist position is characteristic of children ages 4–12 yr. In a study of 740 middle school through graduate school students, Greene et al. (2010) found that those classified as realists had lower educational levels than had been attained by these entry-level MOT students. In this study, the students’ narrative comments suggested movement beyond the realist position; they made statements that suggested a degree of belief in simple and certain knowledge but not an absolute adherence to this point of view. One entering student commented,

In general, the students’ written responses demonstrated evidence of beliefs in both certain and uncertain knowledge consistent with Baxter Magolda’s (2002) description of transitional knowing, in which knowledge is certain in some areas and uncertain in others, and compatible with Schommer-Aikins’ (2002) view of differing EOC as a matter of frequency. According to Schommer-Aikins, more mature thinkers demonstrate beliefs that are primarily—but not exclusively—flexible and contextually relative, allowing incorporation of new ideas, adaptation of old ones, and preservation of foundational beliefs. Because occupational therapy knowledge incorporates information from both well-structured domains such as anatomy and ill-structured domains such as psychology, it is not surprising that students who had completed prerequisite and occupational therapy coursework described belief in both certain and uncertain knowledge.

Despite the similarity in EBI scores, significant differences were found in the occupational therapy–specific mFQS ontological cognition scores and the narrative comments made by the two groups, with evidence of more sophisticated ontological cognition in the postdidactic students. The postdidactic students appeared to have a growing appreciation of multiple perspectives and recognition that factual information needs to be adapted and applied differently according to the context. They also acknowledged that theory and treatment approaches change over time and that multiple assessments can be necessary to achieve a comprehensive view of the client and make intervention decisions. According to one postdidactic student,

Hofer (2006a) described the development of expertise in a discipline, stating that a critical step in this development involves integration of the discipline’s epistemic assumptions. These data may indicate that the postdidactic students were beginning to grasp the EOC and assumptions that are characteristic of occupational therapy (American Occupational Therapy Association, 2014) and to more accurately assign characteristics to the ontological categories of occupational therapy knowledge (Greene et al., 2008, 2010).

Epistemic Cognition

Both groups’ mean scores on the OA factor of the EBI were slightly above the median score of 3 on the 5-point scale, with no statistically significant differences between groups on either measure of epistemic cognition. The EBI scores suggest that participants had strong views about justification of knowledge by authority, which could indicate a dogmatist position. However, the EBI does not include a personal justification factor that might allow discrimination among the dogmatist; the skeptic, who has strong beliefs in personal justification; and the rationalist, who considers multiple sources of evidence (including authority and personal experience), the quality of the evidence, and the context (Greene et al., 2008, 2010). Nevertheless, the theme of justification by authority, specifically research and other practitioners, was apparent in the narrative data for both groups of students. The narrative statements related to justification by authority differed by group, however. For the entering students, the theme of collaboration between the client and practitioner during intervention was apparent; however, an extension of the idea of omniscient authority to the client–practitioner relationship also emerged. That is, one-third of the entering students made statements that pointed to the practitioner as the authority for the client. For example, one student remarked,

Such beliefs conflict with the client-centered approach advocated by the occupational therapy profession (American Occupational Therapy Association, 2014). At entry into the program, these students had not yet been exposed to the philosophy of occupational therapy and the process of building rapport and developing a working alliance with the client.

In contrast, the postdidactic students made statements suggesting consideration of the client as more than a source of knowledge. These students discussed the active role of the client in collaborating with the practitioner and making intervention decisions. In fact, the theme of the client as authority became apparent in the postdidactic students’ comments. For example, one student commented,

For the postdidactic students, other prominent themes related to epistemic cognition included the use of multiple sources to justify knowledge, particularly previous experience, client input, and research. The understanding of different client perspectives and their influence on the intervention process was also a prominent theme for the postdidactic students.

Two of the primary sources of knowledge identified by the entering students were personal experiences and logic and reasoning, with emphasis on logic and reasoning. The entering students considered contextual aspects such as client variables; however, the evidence was limited that they weighed multiple sources of evidence and considered the quality of the evidence, as the rationalist does. These patterns are consistent with previous research suggesting that few people reach the rationalist position (Greene et al., 2010), although Greene et al. (2010) hypothesized that by mid- to late college, students could demonstrate a rationalist EOC in an ill-structured domain such as occupational therapy.

Overall, this study found evidence of both dogmatist and skeptic EOC in the entering students’ statements. Some of the ideas that emerged from their narrative responses, such as the view of the practitioner as the authority for the client, the notion of the practitioner “allowing” the client to make intervention decisions, and the use of research as a source of knowledge, provide evidence of dogmatism. Conversely, like the skeptic, all of the entering students made statements related to personal justification, specifically justification through personal logic and reasoning. Although a skeptic stance may have been more prominent, the dogmatist view was also apparent.

The postdidactic students described less dogmatist beliefs, deemphasizing the role of the omniscient authority in the justification of occupational therapy–specific knowledge. However, the prominence of the personal justification theme suggests that many of the postdidactic students continued to hold the skeptic position. In addition, almost half of these students described beliefs in multiple, equally valid approaches to intervention and answers to problems, similar to the multiplist position described by Kuhn et al. (2000). Despite this evidence, other evidence suggested that the postdidactic students may have moved toward a more rationalist position in terms of occupational therapy–specific knowledge. These students appeared to place less emphasis on the authority of the experienced clinician, looking instead to the client as the authority for making intervention decisions. In fact, only one postdidactic student expressed the idea of the practitioner as an authority for the client, a theme that was fairly prominent for the entering students. Moreover, the postdidactic students’ statements reflected more moderate beliefs in the authority of research; they described the limits of research evidence. Overall, the postdidactic students appeared to have a clearer understanding of the EOC used by occupational therapy practitioners, more commonly recognizing the need to consider contextual variables and vary treatment approaches accordingly, to be client centered, and to use multiple knowledge sources to make intervention decisions.

In summary, the evidence suggests that the postdidactic students’ EOC was more sophisticated than the entering students’. Although beliefs that are characteristic of the skeptic were evident in both groups, the entering students demonstrated some evidence of dogmatist views and little evidence of rationalist beliefs, whereas statements suggesting a rationalist position were more evident in the postdidactic group, and their occupational therapy–specific ontological cognition was more sophisticated than that of the entering students.