Perspectives on teaching: Conceptions of teaching and epistemological beliefs of university academics and students in different domains

Abstract

This study examines how academics and students from different disciplines perceive teaching and knowledge acquisition. University academics and students from both hard and soft disciplines completed the Domain-specific Epistemological Beliefs Questionnaire and the Approaches to Teaching Inventory. Results showed that academics held more student-/learning-centred and less teacher-/content-centred conceptions about teaching than students. Furthermore, prior findings on different beliefs about knowledge and teaching in different academic domains were confirmed for the entire sample. Finally, less sophisticated epistemological beliefs were associated positively with an information-transmission/teacher-centred and negatively with a student-/learning-centred view of good teaching. The findings are discussed against the background of effective teaching and the professionalization of university academics.

Keywords

approaches to teaching conceptions of teaching epistemological beliefs higher education

Conceptions of teaching

Many professional development initiatives for university academics, that is, faculty, focus on making academics aware of their conceptions of teaching, that is, their beliefs about teaching and learning (Amundsen and Wilson, 2012). Kember’s (1997) review of university academics’ conceptions of teaching revealed that these can be subsumed within two orientations: a teacher-/content-centred and a student-/learning-centred orientation. These two orientations have also been confirmed with large multidisciplinary samples (Postareff and Lindblom-Ylänne, 2008; Stes and Van Petegem, 2014). They differ mostly in terms of student and teacher activity: the teacher-/content-centred orientation is characterized by high teacher and low student activity (Trigwell et al., 2005). Here, good teachers possess sound academic knowledge, and their main function is to impart (well-structured) information to students who are viewed as passive recipients (Kember, 1997). In the student-/learning-centred orientation, the teachers’ focus shifts away from themselves and the content towards the students and their learning processes. The teachers’ main function is then to create an appropriate learning environment in which students’ conceptions about the topics can develop and their experiences are utilized. The students are viewed as active learners (Kember, 1997).

The Approaches to Teaching Inventory (ATI) was developed to measure ‘descriptions of qualitatively different conceptions’ (Trigwell et al., 2005: 351), the variation in teaching strategies and the underlying intentions within a particular teaching context. It is also used in professional development to help academics become aware of their conceptions of teaching (Trigwell et al., 2005). Its two scales, Information Transmission/Teacher Focus (ITTF) and Conceptual Change/Student Focus (CCSF), represent the two key orientations proposed by, among others, Kember (1997). Because conceptions are viewed as being composed of intentions, beliefs and actions (Pratt, 1992), the items in the ATI are appropriate indicators for conceptions of teaching. Therefore, although both constructs – conceptions of teaching and approaches to teaching – can be distinguished theoretically, they also have overlapping and strongly related contents.

Approaches to teaching are one topic in teachers’ beliefs research. These beliefs are considered to frame problems and tasks, and thereby impact actions (Fives and Buehl, 2012). Even though this impact is often hard to observe (see Devlin, 2006, for a critical overview of the attitude–behaviour phenomenon related to conceptions of teaching), development in teaching is considered to occur only when academics change their conceptions of teaching (Amundsen and Wilson, 2012). In the university context, Trigwell et al. (1999) showed that students of teachers taking an information-transmission approach used more learning strategies focusing on the reproduction of content (surface approach to learning) and fewer of the deep strategies accompanied by more cognitive involvement that generally led to better learning outcomes.

Students’ conceptions of teaching are also relevant to the teaching and learning process, because ‘students come to learning situations with their perceptions, prior experiences and personal motivation, and quality in learning can only be partly regulated by the teachers’ activities’ (Virtanen and Lindblom-Ylänne, 2010: 355). Destructive friction between university teachers’ and students’ expectations regarding good teaching hinders effective learning (Vermunt and Verloop, 1999). Nonetheless, students’ conceptions of teaching have been found to be more knowledge centred and teacher centred, and to include fewer dimensions than those of academics (Päuler and Jucks, 2013; Virtanen and Lindblom-Ylänne, 2010). Much research on conceptions of teaching has focused on the variation between academic disciplines (e.g. Lindblom-Ylänne et al., 2006; Lübeck, 2009). These are grouped most often according to their epistemological characteristics into a non-paradigmatically soft domain – containing social sciences and the arts – and a paradigmatically hard domain – containing the natural sciences and mathematics (Muis et al., 2006; Neumann et al., 2002). The majority of studies in this field have found that university teachers in the soft domain score higher on the Student Focus scale in the ATI than their colleagues in the hard domain, whereas this relation reverses for the Teacher Focus scale (Lindblom-Ylänne et al., 2006). The disparities between academic domains are often attributed to epistemological differences between the hard and soft domains (Kember and Leung, 2011).

Students’ teaching preferences differ somewhat similarly between disciplines: students of behavioural sciences (soft domain) emphasize interaction, group work and critical thinking whereas students of veterinary medicine (hard domain) prefer traditional lectures, opportunities to ask questions and teaching for understanding (Kember and Leung, 2011; Parpala et al., 2011). However, as yet, there has been no research on students’ conceptions of teaching, on how these differ from those of university academics and on whether the differences between academic domains found for university teachers generalize to students as well.

Epistemological beliefs

Epistemological beliefs are intimately related to differences between academic domains. They refer to the nature of knowledge and the process of knowing (Hofer, 2000) and impact on thinking and learning processes, that is, on metacognitive processes and self-regulation processes. They influence teaching and therefore play an important role in a variety of academic experiences (Hofer, 2000). An inspection of the literature reveals a multidimensional structure of epistemological beliefs (see Hofer and Bendixen, 2012, for an overview). For example, Hofer (2000) uses the dimensions of certainty and simplicity (the latter in the sense of how discrete and concrete knowledge is and not that it is less difficult) of knowledge to distinguish beliefs about the nature of knowledge from beliefs about the nature or process of knowing that involve the dimensions source of knowledge and justification of knowledge. Each of these dimensions is regarded as a continuum ranging from objectivist beliefs with a dualistic conception (e.g. knowledge is true or false) to relativistic beliefs that view knowledge as being rather complex (e.g. knowledge is context dependent; Hofer and Pintrich, 1997). It is widely assumed that epistemological beliefs develop from objectivist to relativistic beliefs through learning experiences over the life span, and are influenced particularly by education at school and university (Hofer and Pintrich, 1997; Trautwein and Lüdtke, 2007). This developmental perspective on epistemological beliefs indicates the need to examine whether academics hold more developed, that is, more relativistic beliefs than students.

In addition to general epistemological beliefs, a substantial amount of research has dealt with domain-specific epistemological beliefs (see Muis et al., 2006; Richardson, 2013, for overviews). On the one hand, domain-specific epistemological beliefs have been shown to differ among students of different disciplines (interindividual approach; Muis et al. 2006): students of soft disciplines show more relativistic epistemological beliefs than students of hard disciplines. This is attributed not only to self-selection and to socialization processes within the discipline (Trautwein and Lüdtke, 2007) but also to the disciplines’ curricula that might vary in how far they support the development of relativistic epistemological beliefs (Lindblom-Ylänne and Lonka, 1999). On the other hand, people also have different beliefs about knowledge in different domains (intra-individual approach). For example, Hofer (2000) found that students evaluated knowledge in hard sciences to be less relativistic than knowledge in psychology. Research on teachers’ epistemological beliefs is still developing (Hofer and Bendixen, 2012). There is a need to examine whether the differences in epistemological beliefs over academic domains found for students also generalize to university teachers.

In addition, teachers’ epistemological beliefs impact those of their students (Bendixen and Feucht, 2010). For example, Van Rossum and Hamer (2010) describe how epistemological mismatches between teachers and students can result in students’ dissatisfaction. With regard to domain-specific epistemological beliefs, research has yet to examine how academics’ expertise, that is, the existence of extensive experience in one academic domain/discipline compared to no experiences in others, impacts these beliefs.

Association between conceptions of teaching and epistemological beliefs

Because both previously described constructs are assumed to influence teaching and learning, it is interesting to see whether and how the two relate. From a theoretical perspective, an association is likely, because ‘epistemological differences have been used as an explanation for disparities in ratings and approaches to teaching between science and the humanities’ (Kember and Leung, 2011: 295). Teachers’ more objectivist epistemologies have been found to be associated with less of a student orientation (Roth and Weinstock, 2013). Other research – however, mostly with an Asian cultural background – has revealed rather inconsistent findings, and reported both negative and positive associations of the learning-centred orientation towards some aspects of global objectivist epistemological beliefs (Otting et al., 2010). Stes and Van Petegem (2014) clustered lecturers on the basis of their belief structures and concluded that many lecturers could not be described as having congruent epistemological beliefs and conceptions of teaching.

Therefore, there is a need to elicit associations between the two constructs in both academics and students, and to analyse these associations against the conceptual background of domain-specific epistemological beliefs as well as to examine the epistemological beliefs and conceptions of teaching within different academic domains. The following hypotheses are formulated:

Hypothesis 1: Members of the hard domain will show a higher teacher focus and a lower student focus in their conceptions of teaching, and they will report fewer relativistic epistemological beliefs than members of the soft domain.

Hypothesis 2: Academics will show more relativistic domain-specific epistemological beliefs regarding their own domain compared to the particular other domain (expertise hypothesis).

Hypothesis 3: Academics and students will differ with regard to their conceptions of teaching and epistemological beliefs with academics having more relativistic epistemological beliefs than students.

Hypothesis 4: More relativistic epistemological beliefs will relate positively to a student-focused and negatively to a teacher-focused conception of teaching.

Methods and materials

Design and procedure

Academics and students from different disciplines, clustered into two domains (hard and soft), completed a paper-and-pencil questionnaire package. The survey was sent to academics via internal mail at a large German university; students received it at the end of particular lectures. To provide a warm-up to the questionnaire, participants were first asked to complete the sentence: ‘I believe that for teaching to be good, it is important that …’. On the first page of the questionnaire package, academics were also asked to rate the following three statements on 5-point scales ranging from 1 (not agree at all) to 5 (agree totally): ‘Research is important to me’, ‘Teaching is important to me’ and ‘Administration is important to me’. The questionnaire package also contained the ATI, the Domain-specific Epistemological Beliefs Questionnaire (DEBQ) and questions gathering demographic information (discipline, age, gender, teaching experience, vocational training, etc.). Note that in order to gather data on intra-individual differences in epistemological beliefs in different domains, academics received the DEBQ in two versions: one for their own academic domain and the other for the particular other domain (within-subjects factor: evaluated domain).

Participants

Academic sample

A total of 81 academics – randomly selected from the population of academics currently enrolled in teaching at the university – completed the questionnaire package (response rate: 44%). Nine datasets could not be used for further analyses due to missing values. Of the remaining 72 lecturers, 37 belonged to a mathematical or natural science faculty (chemistry, physics, mathematics) and 35 were members of a social science or arts faculty (German studies, history, communication science, sociology, politics). As a reward for their participation, academics were entered into a lottery in which they could win one of several packs of presentation materials for teachers with a monetary value of 77€ (105US$). A total of 47 of the participants were research and teaching associates on either pre- or post-doctoral levels (65.3%), 23 were full professors (31.9%) and one participant was a lecturer with teaching but no research obligations (1.4%). The academics’ mean age was 40.24 years (SD = 11.79), and 25% were female. Their teaching experience ranged from 0.5 to 40.0 years (Md = 8.5, M = 11.67, SD = 10.94).

Student sample

A total of 318 undergraduates from the same faculties in the same university as the academics filled in questionnaires at the end of lectures in these faculties. Due to missing values, the responses of nine participants were excluded. The primary study subject of 198 of the remaining 309 students belonged to the mathematics or natural science domain, whereas 111 students were studying a social science or arts subject. The students’ mean age was 22.44 years (SD = 6.88), they were studying on average in their fifth semester (M = 4.96, SD = 1.97) and 42% were female.

Dependent Measures

ATI

Participants’ conceptions of teaching were assessed with Trigwell et al.’s (2005) Approaches to Teaching Inventory–Revised (ATI-R; German translation by Lübeck, 2009). This contains 11 items assessing an ITTF orientation (e.g. ‘In this subject, students should focus on studying what the lecturer provides them with’) and 11 items assessing a CCSF orientation towards teaching (e.g. ‘I see teaching as helping students develop new ways of thinking in this subject’). Participants rated the items with reference to a particular course on 5-point scales ranging from 1 (not agree at all) to 5 (agree totally). The original time-based range (Trigwell et al., 2005; almost never to almost always) was changed to this agreement-based response range in order to compare the approaches to teaching in the sense of general beliefs. The instructions and items in the ATI-R were likewise adapted to assess conceptions of teaching within the seminar type of teaching course (e.g. ‘In courses, I find it important to build up a dialogue with the students about the topics we are studying’). For students, these items were further adapted to refer to courses from the students’ point of view (e.g. ‘In courses, I find it important that the teacher tries to build up a dialogue with us students about the topics we are studying’). Internal consistencies were satisfactory with α = 0.76 for academics and α = 0.79 for students on the CCSF scale, and α = 0.63 for academics and α = 0.71 for students on the ITTF scale. These values correspond to those reported by Lübeck (2009).

DEBQ

Richter’s (2004) German version of Hofer’s (2000) DEBQ was used. This instrument uses 18 items to assess beliefs about knowledge and knowing on four scales: certainty/simplicity of knowledge (eight items), source: external authorities (four items), justification: personal (four items) and attainability of truth (two items). Participants were instructed to apply the items to their own domain (natural sciences or social sciences and arts). Additionally, academics received a second version of the DEBQ in which they were asked to assess the particular other domain. Therefore, items were rephrased to be answered for the own domain (e.g. ‘Truth is unchanging in my subject’) or for the other domain (e.g. ‘Truth is unchanging in these subjects’). Students received only one version for their own study subject. Participants rated each item on 5-point scales ranging from 1 (not agree at all) to 5 (agree totally). Higher scores on all scales indicated more absolutistic and objectivist beliefs. Because only the Certainty/Simplicity scale had satisfactory internal consistencies for rating the own domain with α = 0.87 for academics and α = 0.84 for students, this scale alone was used in further analyses (Cronbach’s α for the other subscales ranged from 0.32–0.70). As the certainty dimension is the central component of epistemological reasoning (Trautwein and Lüdtke, 2007), valuable and relevant insights can be gained using this dimension. Internal consistency for the Certainty/Simplicity scale for academics rating the other domain was α = 0.79. Knowledge at one end of the certainty/simplicity dimension ‘would be viewed as discrete elements that are known with certainty; at the other end, knowledge would be more complex and interwoven, subject to greater interpretation’ (Hofer, 2000: 398). More developed views on this dimension would be represented through lower agreement with, for example, the statement: ‘All experts in this field understand the field in the same way’. To use only this dimension is theoretically justifiable, because the certainty dimension can be viewed as a core component and the most reliable descriptor of epistemological reasoning (Trautwein and Lüdtke, 2007).

Results

Preliminary analyses revealed that most data were non-normal and heteroscedastic. Therefore, all analyses of covariance (ANCOVAs) were conducted with Winsorized values of the original data as more robust estimators for the dependent variables (Erceg-Hurn and Mirosevich, 2008). In Winsorization, the lowest and highest, for example, 10% of scores ordered by size are replaced by a particular cut-off value, thereby eliminating potential outliers. The Winsorized variance is then calculated on the basis of these trimmed scores using the same formula as that used to calculate the (ordinary-least-squares) variance (Erceg-Hurn and Mirosevich, 2008).

Preliminary analyses of the assessed covariates revealed significantly fewer women in the group from the mathematics or natural science faculties than in the group from the social science and arts faculties in both the academic and student samples (academics: 8% vs 43%, χ²(1, N = 71) = 11.18, p = 0.001; students: 30% vs 62%, χ²(1, N = 308) = 29.32, p < 0.001). Because of the relevance of gender-related patterns in both personal epistemology (Richardson, 2013) and approaches to teaching (e.g. Stes and Van Petegem, 2014) and significant correlations with most of the dependent variables, r_{g, ITTF} = −0.03, p > 0.05, r_{g, CCSF} = −0.11, p = 0.033, r_{g, certainty} = 0.15, p = 0.004, gender was entered as a control variable in these ANCOVAs.

Students from the social sciences and arts were in a higher study semester (ss) than students from the natural sciences, U = 9018.50, z = −3.17, p = 0.002 (M_soft = 5.35, SD_soft = 2.12; M_hard = 4.74, SD_hard = 1.84), and study semester correlated significantly with all dependent variables (r_{ss, ITTF} = −0.24, p < 0.001; r_{ss, CCSF} = 0.18, p = 0.002; r_{ss, certainty} = −0.18, p = 0.001). Moderation analyses were conducted to examine the interactions between study semester and domain because academic experience is said to influence (particularly domain-specific) epistemological beliefs (Muis et al., 2006; Van Rossum and Hamer, 2010): the longer students study, the more relativistic their epistemological beliefs will become.

No other covariates were included in the analyses, because there were either no significant group differences (academics’ age, teaching experience and research experience: smallest U value = 527.00; largest z value in absolute terms = −1.00, p > 0.05) or the particular covariate did not correlate significantly with any of the dependent variables (students’ age and academics’ vocational training, largest absolute correlation value r = 0.21, p > 0.05).

Missing values in the covariates (gender, semester of study) were replaced by the mean of the particular group (students or academics). However, values missing in the dependent variables were not replaced, and scale means were computed on the basis of the number of valid values. As a measure of effect size, we calculated partial eta-square, and we interpreted the results in line with Cohen (1988) with $η_{p}^{2}$ = 0.01 as a small, $η_{p}^{2}$ = 0.06 as a medium and $η_{p}^{2}$ = .14 as a large effect. For a better overview, we split the presentation of results between our two main research foci (group and domain) and the third focus on the association between the two constructs. Before addressing the results, the results of the introductory questions for the academics about the importance of research, teaching and administration are reported.

Importance of research, teaching and administration

Analyses revealed that academics from the soft disciplines attached greater importance to administration than academics from the hard domain, F(1, 69) = 20.64, p < 0.001, $η_{p}^{2}$ = 0.23, whereas there were no significant differences regarding the importance of research or teaching – the largest F(1, 69) was 1.48, p > 0.05. The covariate gender showed one main effect: women found administration less important than men (M_wom = 2.22, SD_wom = 1.06; M_men = 2.66, SD_men = 1.11), F(1, 69) = 11.34, p = 0.001, $η_{p}^{2}$ = 0.14.

Domain focus

There were significant main effects of domain on all dependent variables. The analyses for the ATI-R subscales revealed that members of the hard domain had a higher teacher focus, F(1, 376) = 21.63, p < 0.001, $η_{p}^{2}$ = 0.05 (medium effect), and a lower student focus, F(1, 376) = 41.50, p < 0.001, $η_{p}^{2}$ = 0.10 (medium to large effect), than members of the soft domain. Neither the covariate gender nor the interaction between the two factors attained significance – the largest F(1, 376) was 2.43, p > 0.05.

The analysis with the DEBQ subscale certainty/simplicity showed that members of the hard domain believed knowledge to be more certain and simple than members of the soft domain, F(1, 376) = 157.64, p < 0.001, $η_{p}^{2}$ = 0.30 (large effect). Gender did not have a significant effect in this ANCOVA, F(1, 376) = 0.27, p > 0.05. Table 1 depicts the Winsorized means of the ATI-R subscales and the DEBQ subscale for domain and group. Results showed that members of the hard domain revealed a higher teacher focus and a lower student focus in their conceptions of teaching, and they reported fewer relativistic epistemological beliefs than members of the soft domain.

Table 1.

Winsorized means and standard deviations by domain and group for the ATI-R subscales ITTF and CCSF plus the DEBQ subscale certainty/simplicity.

Dependent variable	Domain	Group
		AcademicsM (SD)	StudentsM (SD)
Teacher focus (ATI-R: ITTF)	Hard disciplines	3.18 (0.35)	3.58 (0.36)
Teacher focus (ATI-R: ITTF)	Soft disciplines	2.97 (0.30)	3.34 (0.46)
Student focus (ATI-R: CCSF)	Hard disciplines	3.70 (0.44)	3.35 (0.47)
Student focus (ATI-R: CCSF)	Soft disciplines	4.10 (0.24)	3.73 (0.41)
Certainty/simplicity (DEBQ)	Hard disciplines	2.98 (0.76)	2.84 (0.72)
Certainty/simplicity (DEBQ)	Soft disciplines	1.58 (0.29)	1.98 (0.59)

ITTF: Information Transmission/Teacher Focus; CCSF: Conceptual Change/Student Focus; ATI-R: Approaches to Teaching Inventory–Revised; DEBQ: Domain-specific Epistemological Beliefs Questionnaire.

With regard to whether or not academics will show more relativistic domain-specific epistemological beliefs regarding their own domain compared to the particular other domain, the ANCOVA with repeated measures showed that academics, independently of their own domain, evaluated knowledge in the hard domain as being more certain and simple than knowledge in the soft domain (MD_hard–soft = 0.99, p < 0.001), F(1, 69) = 7.62, p = 0.007, $η_{p}^{2}$ = 0.10 (medium to large effect). Alongside this general domain difference in the evaluation of knowledge, there was a significant interaction effect between own domain and evaluated other domain, F(1, 69) = 7.99, p = 0.007, $η_{p}^{2}$ = 0.10 (medium to large effect). Academics from the soft domain evaluated knowledge in their own domain as being less certain than knowledge in the other domain. However, academics from the hard domain did not show such an expertise effect but evaluated knowledge in their own domain as more certain and simple than knowledge in the other domain. In summary, academics from the hard domain did not show an expertise effect. Descriptive results of this analysis are presented in Table 2. The interaction effect of evaluated domain and gender was not significant, F(1, 69) = 0.01, p > 0.05.

Table 2.

Mean values and standard deviations on the DEBQ subscale certainty/simplicity by group of academics (between-subjects factor) and evaluated knowledge domain (within-subjects factor).

Group of academics	Evaluated knowledge domain
	Hard domainM (SD)	Soft domainM (SD)
Academics from hard domain	3.00 (0.78)	1.77 (0.37)
Academics from soft domain	2.23 (0.57)	1.49 (0.36)
Total	2.62 (0.88)	1.64 (0.43)

Note: DEBQ: Domain-specific Epistemological Beliefs Questionnaire.

Interestingly, this intra-individual comparison showed that expertise in one of the two domains did not automatically lead to more relativistic beliefs about knowledge in that particular domain, but that academics socialized in the hard sciences evaluated knowledge in this domain as quite certain and simple.

Group focus

The ANCOVAs with the ATI-R subscales, ITTF and CCSF, showed that academics attained a higher student focus, F(1, 376) = 39.80, p < 0.001, $η_{p}^{2}$ = 0.10 (medium to large effect), and a lower teacher focus, F(1, 376) = 50.89, p < 0.001, $η_{p}^{2}$ = 0.12 (large effect), than students.

For the epistemological beliefs, the ANCOVA for the DEBQ subscale certainty/simplicity revealed no significant main effect of group, F(1, 376) = 1.72, p > 0.05. However, post hoc comparisons following a significant interaction effect of domain and group, F(1, 376) = 9.30, p = 0.002, $η_{p}^{2}$ = 0.02 (small effect), showed that academics in the soft domain believed knowledge to be less certain and simple than their students, F(1, 143) = 12.76, p < 0.001, $η_{p}^{2}$ = 0.08 (medium effect), whereas academics and students in the hard domain did not differ significantly, F(1, 232) = 1.16, p > 0.05.

Thus, academics and students differed with regard to their conceptions of teaching with academics having more relativistic epistemological beliefs than students. For epistemological beliefs, however, differences were found only in the soft domain. Academics from the hard disciplines did not show more relativistic beliefs about the certainty and simplicity of knowledge in their domain than their students.

For the student sample, a moderating effect of study semester on the certainty/simplicity dimension showed that compared to students from the hard disciplines, students from the soft disciplines developed increasingly relativistic beliefs the longer they studied, F(1, 305) = 5.25, p = 0.023, $η_{p}^{2}$ = 0.02 (small effect). No moderating effect of study semester for any of the approaches to teaching was found, largest F(1, 205) = 3.26, p > 0.05. To clarify the results of the moderation analysis, Figure 1 presents the mean values of the Winsorized DEBQ subscale and ATI-R subscales for the two extreme groups of students with high and low study semester (defined by the mean of study semester ± 1 SD) for the two domains calculated from the regression term. An additional interesting result of these analyses was that the higher the specified semester of study, the higher the students scored on the Student Focus scale, F(1, 305) = 4.46, p = 0.036, $η_{p}^{2}$ = 0.001 (small effect), and the lower they scored on the Teacher Focus scale, F(1, 305) = 10.66, p = 0.001, $η_{p}^{2}$ = 0.01 (small effect).

Figure 1.

Winsorized means of the ATI-R subscales and the DEBQ subscale for high and low semester of study and hard and soft disciplines.

Association

The fourth research question addressed the relationship between attitudes towards teaching and epistemological beliefs. Hence, we calculated partial correlations between the two ATI-R subscales and the DEBQ subscale certainty/simplicity while controlling for age and gender. Age was considered as an additional covariate here, because results were not broken down into groups for this analysis.

The analysis revealed significant medium-sized positive correlations between certainty/simplicity scores and the scores for the information-transmission/teacher-focused approach to teaching, r = 0.33, p < 0.001, and negative correlations between certainty/simplicity values and the learning-centred/student-focused approach, r = −0.33, p < 0.001. Therefore, we found a significant association between the conceptions of teaching and epistemological beliefs; that is, more relativistic epistemological beliefs about certainty/simplicity of knowledge related positively to a student-focused and negatively to a teacher-/content-focused orientation towards teaching.

Discussion and conclusion

In order to examine important perspectives on teaching and knowing, the study described in this article was conducted with university academics and students from hard and soft disciplines. Using two widely recognized instruments to measure both domain-specific epistemological beliefs and approaches to teaching, significant differences were found for both group and discipline.

Group focus: conceptions of teaching

Academics considered their own activity and information transmission to be less important for good teaching, and they focused more on student activity and conceptual development than the students did. This has not been shown before. Results from data gathered with an open-ended question at the beginning of the questionnaire, however, support this result (Päuler and Jucks, 2013). One possible explanation for this finding could be that professional development programmes and literature about teaching in higher education sensitise academics to conceptions of teaching and provide them with facilitative, activating methods (Amundsen and Wilson, 2012). In the sample, values on the ATI scales showed no statistically significant dependence on the academics’ vocational training. This might be due to insufficient statistical power because of too small a sample. Less than one-half of the academics in the sample had attended any vocational training course, and only 13 academics had attended more than one. Power was reduced even further because the factor domain also needed to be considered in this analysis.

Students, on the other side, might come to university expecting lecture-like courses and having ill-informed preconceptions about what it means to be an independent learner and to be personally responsible for one’s learning (Sander et al., 2000). As a result, it is only then that they begin to develop the ability to learn independently (Kember et al., 2012), and their beliefs about good teaching develop simultaneously. This assumption is supported by the moderation effect of study semester: beliefs about teaching of students in higher semesters are more learning centred than the beliefs of students in lower semesters. Note that in contrast to Virtanen and Lindblom-Ylänne’s (2010) study, which also reported substantial differences between academics’ and first-year students’ views of good teaching, this sample consisted mostly of students at the end of their second year (Md = 4.00 semesters) and not their first. This undermines the assumption that differences between academics and students might (only) be due to misconceptions during the transition from school to university (cf. Kember et al., 2012) but indicates a general developmental perspective on teaching and learning. Future work will need to examine the development of beliefs over the academic career not only within the student group but also among the academics.

However, another possible explanation for the findings on the differences between groups could be that academics adopt teacher- and content-centred methods in (some) of their courses despite their student-centred beliefs about teaching (Lindblom-Ylänne et al., 2006). Students experience those teaching styles ‘in action’ that shape what students consider adequate teaching. This study does not provide any data on the academics’ actual teaching practice; however, this is a widely discussed topic in the literature on conceptions of teaching (see, for example, Devlin, 2006, for an overview).

Group focus: epistemological beliefs

Regarding the differences in epistemological beliefs between groups, the main effect of group in the data showed that academics viewed knowledge as being significantly (one-tailed p) less certain and simple than students did. These cross-sectional results support the development hypothesis that university students develop more relativistic epistemological beliefs the longer they remain in an academic environment (Hofer, 2000; Muis et al., 2006). However, multivariate results on the DEBQ subscale certainty/simplicity suggest a slightly different picture: whereas the difference between academics and students in the soft domain was quite striking, the differences within the hard domain were not significant, indicating that the significant main effect was due mostly to the difference within the soft sciences. Additionally, on a descriptive level, academics from the hard domain viewed knowledge in their domain as being even more certain and simple than students did. This might indicate how the different epistemological character of the hard compared to the soft domain shapes members’ beliefs over their academic careers. These findings are in line with longitudinal findings reported by Trautwein and Lüdtke (2007): the development of epistemological beliefs over the (academic) life span does not seem to apply to the hard disciplines; socialization processes might shape domain-specific epistemological beliefs within these disciplines; and these processes in the hard disciplines might not include many experiences of events demonstrating the uncertainty and complexity of knowledge (Van Rossum and Hamer, 2010).

Domain focus

With regard to the domain focus, the study confirms research showing that members of the hard disciplines view knowledge in their domain as being more certain and simple. In other words, once answers have been found, they are rather stable and experts have the same understanding. In addition, they consider good teaching to be more focused on knowledge transmission and teacher activity and less oriented towards students’ constructivist learning processes than members of the soft disciplines do. However, these findings should not be interpreted as indicating less intellectual sophistication among academics from ‘hard’ disciplines, but rather as representing widely shared views of the discipline (Hofer, 2000) resulting in more tightly structured technical terms, mostly empirical methodology and a resultantly broader consensus on what constitutes knowledge (cf. Muis et al., 2006). This also becomes clear in the main effect of the intra-individual domain-specific epistemological beliefs: independent of their own disciplinary background, knowledge in the soft sciences was evaluated as being more complex and less certain than knowledge in the hard sciences. This once again reflects the fact that disciplines have different values, ways of behaving and practices in relation to their knowledge and ideas (e.g. Lindblom-Ylänne et al., 2006). Note that the science of education is included in the soft domain here, which makes the shift of thinking as educators reflecting about teaching more demanding for ‘hard’ than for ‘soft’ faculty. Because most research on disciplinary differences and epistemological beliefs has been conducted with students or student teachers (e.g. Hofer, 2000; Trautwein and Lüdtke, 2007), the findings expand this field of research by including the university teachers’ perspective. Similarly, most research on conceptions of teaching and their different characteristics in academic disciplines has been based on teachers (e.g. Lindblom-Ylänne et al., 2006; Lübeck, 2009), whereas this study delivers new insights into the students’ perspective on teaching.

In summary, results show that beliefs about knowledge and teaching differ not only between domains but also between groups of academics and students. These differences have the following main implications: first, if academics with more student-centred beliefs about teaching than their students want to use more student-centred teaching methods, they need not only to shape their own role accordingly but also to prepare their students for the change in their role (Otting et al., 2010). Academics should be aware of and address their students’ more teacher-centred and less student-oriented beliefs, just as they need to be aware of and able to influence their students’ approaches to learning (Trigwell et al., 1999). It might seem more convenient for students to put the academics in charge of teaching and make them responsible for the learning processes instead of actively arranging teaching–learning situations within the course frame – even though they will gain a deeper learning by, for example, expressing criticism (e.g. Roth and Weinstock, 2013). However, as soon as academics realize this, they will be able to communicate what they expect from the students more clearly. It is important for academics to ‘provid[e] students with rationales for teachers’ expectations, rather than expecting that everyone should hold the same perspective as the teacher’ (Roth and Weinstock, 2013: 405).

Second, academic domains have their own particular epistemological characteristics. For example, most mathematical facts are certain, unchanging and atomistic; and good teaching of these facts involves rather content- or teacher-centred teaching methods such as a teacher demonstrating a mathematical argument on the chalkboard at the front of a class. Academics’ interdisciplinary professional development initiatives hold the potential to point out which content-centred teaching situations can be reorganized to support student learning better, and thereby broaden academics’ conceptions about teaching against the background of their epistemological beliefs. Teaching methods that are suitable for ‘hard’ content such as peer learning and problem-based learning need to be fostered more strongly (Hellmann et al., 2014). For ‘hard’ faculty themselves, it could be helpful to take their students’ perspective more often and to bear in mind that active learning, rather than passive listening, promises deeper learning. This could make the individual’s repertoire of teaching methods broader and more flexible, and maximize both the learning and the content orientation (cf. Devlin, 2006). As a means to this end, and fortified by our results on a significant association between conceptions of teaching and certainty beliefs about knowledge, it is necessary to address not only academics’ conceptions of teaching but also their epistemological beliefs.

The participants in this study were university academics and students from the same university who came from the same disciplines. However, no information was available as to which lecturer might have taught which students, thus forming one of the study’s limitations. Future studies should focus on academics’ and their particular students’ conceptions of teaching and epistemological beliefs, because this might provide further insights into dynamics and reciprocal effects. Furthermore, the development of conceptions of teaching over the span of tertiary education needs to be examined in more depth and with longitudinal designs. First indicators that they change from teacher oriented and instructional to student oriented and constructivist have been obtained in this study. For academics, an interesting question would be whether differences could be found between status groups (i.e. professors, post-doctoral students and pre-doctoral research assistants). Furthermore, future research could focus on academics’ conceptions of teaching and epistemological beliefs as a function of their attendance of vocational training.

A second limitation of the study is that the academics participated on a voluntary basis. This may have led to a selection bias and attracted those who were interested in teaching and who might have reflected on their teaching more than others. Therefore, the academic sample might not represent the whole range of university academics and the different amounts of time and effort they invest in teaching. Academics’ and students’ epistemological beliefs and conceptions of teaching were investigated with two widely used inventories. However, the insufficient internal consistencies for three of the four subscales in the DEBQ reflect a well-known problem with epistemological beliefs questionnaires (e.g. Muis et al., 2006). Concerning the dimensionality of teaching approaches, Postareff and Lindblom-Ylänne (2008) state that when contrasting the two approaches, more facets need to be explored in order to capture the phenomenon adequately. These include the teaching process (planning, practices of teaching and assessment) and the learning environment (teacher’s role, students’ role, interaction and atmosphere). Hence, the ATI might focus too narrowly on teaching practices that are intended to represent the two orientations in these concepts of teaching.

This article sheds light on previously insufficiently examined perspectives on teaching and knowledge within the university context. The comparison of academics’ and students’ conceptions of teaching has revealed substantial differences in how the two interacting groups define their roles in the teaching–learning process. This implies that university teachers should make their expectations and understanding of effective teaching and learning more explicit to their students. The comparison of domain-specific beliefs on the certainty and simplicity of knowledge between academics and students, between hard and soft domains, and between their own and the particular other domain extends the field of research on epistemological beliefs and their development. It seems that socialization effects within the discipline impact the development of epistemological beliefs strongly, resulting in less relativistic epistemological beliefs within the hard domain at more advanced stages of academic life. Finally, the association between the two constructs – epistemological beliefs and conceptions of teaching – contributes to a better understanding of how both constructs interact with each other in university teaching.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Author biographies

Lena Päuler-Kuppinger is a PhD student and a researcher at the University of Münster, Germany. Her research focuses on attitudes towards teaching and their impact on communication between academic professionals and students. Address: Centre for Teaching in Higher Education (ZHL), Westfälische Wilhelms-Universität Münster, Fliednerstr. 21, 48149 Münster, Germany. [email: Lena.Paeuler@uni-muenster.de]

Regina Jucks is a full Professor in the Institute of Psychology for Education at the University of Münster, Germany. She received her PhD in psychology in 2001 and her post-doctoral habilitation in 2005. Major topics in her ongoing research are computer-mediated communication and interaction processes between lecturers and students at university. Since 2011, she has also been the director of the local Centre of Teaching in Higher Education. Address: Institute of Psychology for Education, Westfälische Wilhelms-Universität Münster, Fliednerstr. 21, 48149 Münster, Germany. [email: jucks@uni-muenster.de]

References

Amundsen

Wilson

(2012) Are we asking the right questions? A conceptual review of the educational development literature in higher education. Review of Educational Research 82(1): 90–126.

Bendixen

Feucht

(2010) Personal epistemology in the classroom: What does research and theory tell us and where do we need to go next? In: Bendixen

Feucht

(eds) Personal Epistemology in the Classroom: Theory, Research, and Implications for Practice. New York: Cambridge University Press, pp. 555–586.

Cohen

(1988) Statistical Power Analysis for the Behavioral Sciences, 2nd edn. Hillsdale, NJ: Lawrence Erlbaum.

Devlin

(2006) Challenging accepted wisdom about the place of conceptions of teaching in university teaching improvement. International Journal of Teaching and Learning in Higher Education 18(2): 112–119.

Erceg-Hurn

Mirosevich

(2008) Modern robust statistical methods – An easy way to maximize accuracy and power of your research. American Psychologist 63(7): 591–601.

Fives

Buehl

(2012) Spring cleaning for the ‘messy’ construct of teachers’ beliefs: What are they? Which have been examined? What can they tell us? In: Harris

Graham

Urdan

(eds) APA Educational Psychology Handbook, Vol. 2: Individual Differences and Cultural and Contextual Factors. Washington, DC: APA Psychological Association, pp. 471–499.

Hellmann

Paus

Jucks

(2014) How can innovative teaching be taught? Insights from higher education. Psychology Learning & Teaching 13(1): 43–50.

Hofer

(2000) Dimensionality and disciplinary differences in personal epistemology. Contemporary Educational Psychology 25(4): 378–405.

Hofer

Bendixen

(2012) Personal epistemology: Theory, research, and future directions. In: Harris

Graham

Urdan

(eds) APA Educational Psychology Handbook, Vol. 1: Theories, Constructs, and Critical Issues. Washington, DC: American Psychological Association, pp. 227–256.

10.

Hofer

Pintrich

(1997) The development of epistemological theories: Beliefs about knowledge and knowing and their relation to learning. Review of Educational Research 67(1): 88–140.

11.

Kember

(1997) A reconceptualisation of the research into university academics’ conceptions of teaching. Learning and Instruction 7(3): 255–275.

12.

Kember

Hong

(2012) From model answers to multiple perspectives: Adapting study approaches to suit university study. Active Learning in Higher Education 14(1): 23–35.

13.

Kember

Leung

DYP

(2011) Disciplinary differences in student ratings of teaching quality. Research in Higher Education 52(3): 278–299.

14.

Lindblom-Ylänne

Lonka

(1999) Individual ways of interacting with the learning environment – Are they related to study success? Learning and Instruction 9(1): 1–18.

15.

Lindblom-Ylänne

Trigwell

Nevgi

. (2006) How approaches to teaching are affected by discipline and teaching context. Studies in Higher Education 31(3): 285–298.

16.

Lübeck

(2009) Lehransätze in der Hochschullehre [Teaching approaches in higher education]. PhD Thesis, Freie Universität Berlin, Germany. Available at: http://www.diss.fu-berlin.de/diss/content/below/index.xml (accessed 9 June 2015).

17.

Muis

Bendixen

Haerle

(2006) Domain-generality and domain-specificity in personal epistemology research: Philosophical and empirical reflections in the development of a theoretical framework. Educational Psychology Review 18(1): 3–54.

18.

Neumann

Parry

Becher

(2002) Teaching and learning in their disciplinary context: A conceptual analysis. Studies in Higher Education 27(4): 405–417.

19.

Otting

Zwaal

Tempelaar

. (2010) The structural relationship between students’ epistemological beliefs and conceptions of teaching and learning. Studies in Higher Education 35(7): 741–760.

20.

Parpala

Lindblom-Ylänne

Rytkönen

(2011) Students’ conceptions of good teaching in three different disciplines. Assessment and Evaluation in Higher Education 36(5): 549–563.

21.

Päuler

Jucks

(2013) Direkt erfragt: Die Messung von Lehrauffassungen per Fragebogen mit offenem Antwortformat [Directly asked: Measuring approaches to teaching via open question formats]. Zeitschrift für Hochschulentwicklung 8(3): 95–109.

22.

Postareff

Lindblom-Ylänne

(2008) Variation in teachers’ descriptions of teaching: Broadening the understanding of teaching in higher education. Learning and Instruction 18(2): 109–120.

23.

Pratt

(1992) Conceptions of teaching. Adult Education Quarterly 42(4): 203–220.

24.

Richardson

JTE

(2013) Epistemological development in higher education. Educational Research Review 9: 191–206.

25.

Richter

(2004) Deutsche Version des DEBQ von Hofer [German version of Hofer’s DEBQ]. Psychologisches Institut, Universität Köln, Germany.

26.

Roth

Weinstock

(2013) Teachers’ epistemological beliefs as an antecedent of autonomy-supportive teaching. Motivation and Emotion 37(3): 402–412.

27.

Sander

Stevenson

King

. (2000) University students’ expectations of teaching. Studies in Higher Education 25: 309–323.

28.

Stes

Van Petegem

(2014) Profiling approaches to teaching in higher education: A cluster analytic study. Studies in Higher Education 39(4): 644–658.

29.

Trautwein

Lüdtke

(2007) Epistemological beliefs, school achievement, and college major: A large-scale, longitudinal study on the impact of certainty beliefs. Contemporary Educational Psychology 32(3): 348–366.

30.

Trigwell

Prosser

Ginns

(2005) Phenomenographic pedagogy and a revised Approaches to Teaching Inventory. Higher Education Research and Development 24(4): 349–360.

31.

Trigwell

Prosser

Waterhouse

(1999) Relations between teachers’ approaches to teaching and students’ approaches to learning. Higher Education 37(1): 57–70.

32.

Van Rossum

Hamer

(2010) The Meaning of Learning and Knowing. Rotterdam: Sense Publishers.

33.

Vermunt

Verloop

(1999) Congruence and friction between learning and teaching. Learning and Instruction 9(3): 257–280.

34.

Virtanen

Lindblom-Ylänne

(2010) University students’ and teachers’ conceptions of teaching and learning in the biosciences. Instructional Science 38(4): 355–370.