Introduction

The effects of epistemology, which is accepted as an important branch of philosophy, on different disciplines have always been a matter of curiosity. Especially in the discipline of history, where abstract concepts are more involved and the process of knowledge construction is monopolised by experts (historians), epistemology-centred discussions are remarkable. Discussions on the nature, construction process and reliability of historical knowledge have been conceptualised as “epistemic cognition in history” (Maggioni et al., 2009; VanSledright & Maggioni, 2016). What kind of pedagogical effects teachers’ epistemic beliefs in history have is also a matter of curiosity. From this perspective, we aimed to adapt the Beliefs About History Questionnaire (BHQ) scale developed by Maggioni (2010) to the Turkish language and culture to determine teachers’ epistemic cognition in history. The starting point of this study was the lack of any study on epistemic cognition in history in the Turkish context. Considering that studies on epistemic cognition in history have emerged in the last 20 years (Stoel et al., 2022), we think that determining the historical epistemic cognition of individuals and especially educators in different cultures will contribute to this field. In addition, this study aims to reveal the ways in which the theoretical model proposed by Maggioni (2010) finds meaning in Turkish culture. In the following sections, information about the theoretical framework for epistemic cognition in history is given and the process of adapting the scale to Turkish culture is mentioned.

Theoretical Framework

Epistemology deals with the source, possibility, structure, limit and value of knowledge (Hofer, 2000), while epistemic beliefs are individuals’ cognitions and beliefs about what knowledge is, how it is acquired, its certainty and limits (Schommer, 1990). Educators show great interest in this field because epistemic beliefs are influential in many variables such as teaching-learning, decision-making and so on (Hofer & Bendixen, 2012; Schraw, 2013). In contemporary history teaching practice, history teachers are expected to have factual and conceptual subject knowledge as well as expertise of the epistemology of history as a discipline (Mathis & Parkes, 2020). History educators, especially in the last 20 years, have generally sought answers to the following questions to reveal how and how learners justify the accuracy of historical knowledge: (i) What is the nature of historical knowledge? (ii) What is the role of historians in producing such knowledge? (iii) Are historians merely couriers who bring information from the past to the present? or (iv) Is the past just narratives produced by historians? (v) Does historical evidence act as a tool to transfer the past into the present? (vi) Can historical evidence be considered as it is independent of the historian? (Lee & Shemilt, 2003; Maggioni et al., 2009; VanSledright, 2002). To bring logical and coherent explanations to these questions, the basic arguments of general epistemic belief models (Hofer & Pintrich, 1997; Kuhn & Weinstock, 2002) are integrated into the discipline of history, and various theoretical frameworks and understandings including historical contextualisation, historical evidence and interpretation (Lee & Shemilt, 2003; Wineburg, 2001). As a result of a series of empirical studies (Maggioni et al., 2004; Maggioni et al., 2009), which are all based on Lee and Shelmit’s (2003) six-stage development model, Maggioni (2010) developed a measurement tool consisting of 22 items related to this model. Her instrument involves three different stances (criterialist, borrower, copier) related to the nature, source and role of the knower. She then named this instrument the BHQ.

Individuals who take the copier (objectivist) stance cannot comprehend the difference between fact and history and they evaluate historical information as objective and fixed. The purpose of history is to obtain a copy of the past and catch what happened in the past as it was. Historians are the intermediaries that carry the objective and accurate information they have obtained from historical sources to the present day. Making history is also like printing out a copy of the arguably only true piece of information (Havekes et al., 2012; Stoel et al., 2017). Students in this stance are unaware that historical information can change over time due to the different perspectives and the historians adopting different ways of asking questions. The focus or tendencies of the societies that make history may change over time (Seixas, 2004).

According to individuals who take the borrower (subjectivist) stance, historical knowledge and historical sources are not fixed but have a subjective structure. History is ultimately a matter of opinion and is influenced by the historian’s personal preferences and views. What counts as history all remained in the past and it is impossible to reach the whole truth about them. For this reason, individuals having this perspective only accept the historical accounts that are suitable for their logic and only borrow from historical sources that are suitable for their historical narratives (Maggioni et al., 2009; Mierwald & Junius, 2022). From this point of view, it can be said that they are deprived of applying certain procedures such as thinking skills specific to the discipline of history and transforming the source into evidence.

According to individuals who take the criterialist stance, historical sources are not entirely objective or subjective. Historical sources are interpreted according to the questions asked and the context in which they are transformed into historical evidence. To arrive at a valid and accurate point of view reflecting multiple perspectives on the past, it is necessary to make evaluations according to the criteria (evidence, argument) based on the discipline of history. In the process of creating historical accounts and evaluating their validity, discipline-specific criteria and strategies are employed (Maggioni et al., 2009; Stoel et al., 2017).

It has been observed that the number of studies examining individuals’ and groups’ epistemic beliefs about history using BHQ has been increasing gradually (Mierwald & Junius, 2022). On the other hand, it has been reported that the three-factor structure of the scale is not supported by data and the reliability coefficients are low, both in the preliminary studies in the development process of the scale (Maggioni, 2010; Maggioni et al., 2004; Maggioni et al., 2009) and in the adaptation studies conducted in different cultures (Miguel-Revilla et al., 2017; Stoel et al., 2015). The psychometric structure of BHQ in Turkish culture is a matter of curiosity. In addition, no study has been found in the Turkish literature dealing with epistemic beliefs in the field of history. Based on this standpoint, this study aims to adapt BHQ into Turkish culture and assess its validity and reliability.

Method

The Participants

The study was carried out with 264 student social studies teachers studying at three different universities in Türkiye. 53.7% of the participants were women. It can be said that there is a balanced distribution in terms of the participants’ grade levels (First grade: 27.3%, Second grade: 26.5%, Third grade: 24.6%, Fourth grade: 21.6%).

The Instrument

To examine the participants’ epistemic cognition in history, the 22-item BHQ in the 6-point Likert type developed by Maggioni (2010) was adapted into Turkish culture.

The Process

In the first stage, the scale was translated into Turkish by the authors. Then, an expert translation evaluation form containing both Turkish and English versions of the scale was prepared. This form was presented to two experts specialising in translation/interpretation studies between English and Turkish languages. After the Turkish form of the scale was finalised in line with the suggestions from the language experts, the scale was translated from Turkish into English by another language expert, this translation was compared with the original version of the scale, and the scale was given its final form. To reveal the compatibility of the scale with the Turkish context, the scale was examined by three history education experts working in Türkiye. Their opinions were received, and necessary alterations/corrections were made on the scale. Before the main data collection phase, the opinions of four student social studies teachers who were not included in the study sample were taken with the think-aloud technique and then the final version of the scale was obtained.

In the second stage, the 6-point Likert scale was converted to a 5-point Likert type. Since most of the studies conducted in the context of Türkiye use the 5-point Likert type, this form was preferred in this study as well. Accordingly, the answers given to the items were transformed as 1: “strongly disagree”, 2: “disagree”, 3: “undecided”, 4: “agree”, 5: “completely agree”.

In the third stage, SPSS and AMOS programs were used to analyse the data. Normality tests and missing value analysis were performed to make the data ready for analysis. Seven scale forms filled incorrectly or incompletely were excluded from the analysis process. By performing multivariate normality and extreme value analysis, 13 extreme values with significant Mahalanobis D2 distances (p = 0.01) were extracted from the data set (Tabachnick & Fidell, 2019). The remaining 244 returned forms were taken into consideration. Explanatory factor analysis (EFA) and confirmatory factor analysis (CFA) were performed, respectively, to determine the construct validity and cultural adaptation level of the scale. Cronbach Alpha coefficient was reported to reveal the reliability of the obtained structure. For compliance of CFA, the values of the division of chi-square by degree of freedom (χ2/df), standardised root mean square residual (SRMR), root mean square error of approximation (RMSEA), adjusted goodness of fit index (AGFI), incremental fit index (IFI), comparative fit index (CFI) and goodness of fit index (GFI) were taken as criteria (Schumacker & Lomax, 2016).

Findings

Before the EFA, the Kaiser Meyer Olkin (KMO) sample adequacy coefficient and the Barlett Sphericity test significance values were calculated to determine the suitability of the data set for factor analysis. The KMO sample adequacy coefficient was calculated as 0.712 and the result of the Barlett Sphericity test was also significant (χ2 = 811.154, df = 136; p < 0.01). According to these results, it can be said that the data set is suitable for EFA (Field, 2013) Factors with an eigenvalue of 1 and above in principal component analysis and varimax were preferred in rotation.

At the first stage, it was found out that item 14 “It is impossible to know anything for sure about the past since no one of us was there”, which was in the subjectivist sub-dimension in the original scale, is related to both the objectivist and the subjectivist sub-dimensions. It was seen that factor loads in these two dimensions (0.452 and 0.487 respectively) were close to one another and the difference between these factor loads was not more than 0.10 (overlapping). As a result, this item was excluded from the scale and the analysis was repeated. In the second stage, Item number 22, “There is no evidence in history”, which was in the subjectivist sub-dimension of the original scale and located in the objectivist sub-dimension after this stage, was removed from the scale because it had a low factor loading (0.32). Then the analysis was repeated. In the third stage, six different sub-dimensions emerged after the rotation. When the resulting structure was examined, it was seen that the sub-dimensions were in three separate groups each one containing two sub-dimensions. For this reason, the number of sub-dimensions was reduced to three, and the analysis was repeated once again. In the fourth stage, the item “Students need to be aware that history is essentially a matter of interpretation” was excluded from the scale because it was included in the criterialist sub-dimension. Similarly, while they were supposed to be in the criterialist sub-dimension as in the original scale, item 3 “A historical account is the product of a disciplined method of inquiry” and item 11 “History is a critical inquiry about the past” were removed from the scale.

It was observed that all the remaining items were theoretically compatible with the sub-dimensions they belonged to. The factors obtained as a result of the analysis are shown in Fig. 15.1.

Fig. 15.1
A scree plot of eigenvalues versus component number has a line from (1, 3) to (17, 0.3). The first part of the line is more steep while the second part has a gradual decline. Data estimated.

Scree plot chart showing the number of factors in BHQ

Full size image

The examination of Fig. 15.1 reveals that although there are five different sub-dimensions with eigenvalues higher than 1, it does not seem possible to obtain a realistic interpretation from the scree plot graph in line with reducing the number of factors to 3. The eigenvalues and variances of the dimensions reached as a result of EFA are shown in Table 15.1.

Table 15.1 The factorial structure of the BHQ after the rotation
Full size table

It is seen from Table 15.1 that the scale consisting of three sub-factors explains 41.57% of the total variance. Kline (2011) asserts that the total variance explained in scales consisting of more than one dimension needs to be 41% or above. As a result of these findings, it can be said that the explained variance is at a sufficient level. The factor loadings of the scale items and their distribution according to the factors are shown in Table 15.2.

Table 15.2 Items in each factor and factor loads of the scale
Full size table

As can be seen in Table 15.2, the factor loads of the items forming the list of criteria vary between 0.70 and 0.46. The items that form the objectivist dimension vary between 0.77 and 0.49, and the items that form the subjectivist dimension vary between 0.67 and 0.48.

Using the same data set, a CFA was applied to the structure that emerged as a result of the EFA. As a result of the CFA, item 1 in the criterialist sub-dimension was excluded from the scale because the factor load was low (0.27). The CFA results with the remaining items were consistent with the results obtained from EFA, and the fit indices (excluding CFI and IFI) were at acceptable levels (χ2 = 209.267 df = 99, p < 0.01, χ2/df = 2.114, RMSEA = 0.06, AGFI = 0.86, GFI = 0.90, CFI = 0.83, IFI = 0.83, SRMR = 0.07).

The Cronbach Alpha internal consistency coefficient was calculated as 0.67 for the criterialist sub-dimension, 0.64 for the subjectivist sub-dimension, and 0.68 for the objectivist sub-dimension. Accordingly, it can be said that the reliability coefficients of the sub-dimensions are reasonably reliable even if they are not at a high level (Cortina, 1993).

Discussion

This paper examines the psychometric properties of BHQ in Turkish culture. Research results showed that the three sub-dimensions in the original scale were also obtained in the Turkish version. On the other hand, as a result of validity and reliability studies, five items were removed from the Turkish version of the original 22-item scale. The CFA results reveal that while the general fit indices were at an acceptable level, it was reported that the fit indices such as CFI and IFI were low. Similarly, it can be interpreted that the reliability coefficients for the sub-dimensions of the scale are not very high. These results are consistent with empirical studies reporting that BHQ has some problems in terms of validity and reliability (Hamer, 2016). Mierwald and Junius (2022) argue that this might have been caused by errors in translation into different languages, the cultural contexts, the educational levels of the participants and the terminology used in the writing of the items.

The detailed examination of the Turkish version of the BHQ revealed that the objectivist sub-dimension was clearly differentiated from the other dimensions. This means that the data collected in the Turkish context supports the clear distinction between an objectivist stance, which interprets the historical knowledge/information detached from its context, independent of its author(s)/makers and disconnected from learners’ past experiences and perspectives emphasising the interpretive nature of historical knowledge. So, this can be considered an empirical result supporting the theory (Maggioni et al., 2009; Stoel et al., 2017; VanSledright, 2010). On the other hand, one item that should theoretically be included the subjectivist sub-dimension is located in the criterialist sub-dimension, while two items in the criterialist sub-dimension are included in the subjectivist sub-dimension. These results indicate that the limited number of student teachers who participated in this study had already developed an awareness of the interpretive and subjective structure of historical knowledge. Nevertheless, they could not make a clear distinction between the subjectivist stance and the criterialist stance, which indicates the importance of discipline-specific criteria and historical research strategies. According to Stoel et al. (2017), this may be arising from the theoretical ground of BHQ. As a result of their empirical studies, the authors revealed that the items reflecting the subjectivist stance were located in both the objectivist (naïve) and the criterialist (nuanced) dimensions. It is accepted that the theoretical framework developed in the US context assumes that having a subjectivist perspective is of a lower level than adopting a criterialist stance. On the other hand, whether this may also be applied to history teachers and students living and working in different parts of the world it is a matter of question. Is the process of transforming historical sources into evidence (doing history) by means of reflecting a criterialist stance given enough space in history curricula around the world? Or, the purpose of history teaching is to transfer factual information and concepts in teacher-centred classroom contexts to raise a desired type of citizen, in which there is no room for discussion and/or inquiry? The answers to such questions differ in accordance with the epistemic beliefs of teachers as a reflection of the missions imposed on teacher education, especially in relation to the differing understandings (contextual factors) of history teaching adopted by countries or educational systems (Stoel et al., 2022). As a matter of fact, the study by Sakki and Pirttilä-Backman (2019) revealed that socio-cultural contexts affect teachers’ epistemic beliefs. In some countries (Netherlands, Austria, Germany), for example, critical thinking is at the forefront, while some countries (Estonia, Belarus, Serbia) give importance to patriotism and moral values. It has been emphasised in the relevant literature that variables such as school culture, centralised nationwide examinations, limited class time, the structure of the relevant curricula and social expectations prevent teachers to engage in questioning activities that may develop students’ epistemic beliefs in history. This situation may also force history teachers to head for a content-based learning environment (van Hover & Yeager, 2004; Voet & de Wever, 2016; Wansink et al., 2016).

Implications for Theory and Practice

Considering the Turkish context, it is natural that there is not a clear distinction between subjectivist and critical perspectives. This situation is closely related to the developmental process of history education in Türkiye. For many years, history teaching in Türkiye has been based on factual knowledge and rote learning (Simsek & Yazici, 2013). As a result of this situation, history courses were perceived by students as a boring course disconnected from daily life (Ozkal et al., 2004). It can be said that this phenomenon is still valid today. In the study conducted by Dinç and Üztemur (2017) with Turkish pre-service social studies teachers, the participants stated that secondary school social studies courses consisted of complex, boring and clichéd subjects and that these subjects were far above the level of students. As a matter of fact, although a small number of studies (Aktekin et al., 2009; Uztemur et al., 2019) aimed at preventing this perception with a student-centred history approach stand out, it can be said that there are still very few activities for making history in history lessons. In addition, history teachers’ self-efficacy in implementing doing history activities (working with first-hand sources, developing historical thinking skills, doing activities that emphasise that different perspectives can also be valuable in history, practices aimed at developing historical perspective, historical contextualisation and historical empathy skills, studies that emphasise that history and the past are not the same things and that the historian cannot be separated from the society he/she comes from, etc.) should be improved. A qualitative study conducted by Yilmaz (2008) with Turkish social studies teachers revealed that the participants had a naïve and realist epistemic belief in the nature of history, never mentioned the role of historians in the process of constructing historical knowledge, were unaware of the distinction between past and history and believed that historical knowledge was objective. When all these results are considered together, it would not be wrong to position a history teacher with a subjectivist perspective in the Turkish context closer to a relatively sophisticated historical epistemic belief. It does not seem possible to change the traditional history teaching in Türkiye, which is based on the memorisation of unchanging information and facts, quickly. Changing the belief and perception that history textbooks are unchanging and uncritical texts like sacred texts is a process that requires some time. Considering the fact that historical thinking skills and history making methodology are not addressed much in history teaching, it is natural that some items from subjectivist and critical perspectives are intertwined in the present study. As teachers’ historical epistemic beliefs become more sophisticated, they will move away from a teacher-centred and rote-based history teaching approach and become aware that history has its own methodology (criteria and strategies) in history courses. For example, in a mixed-design study conducted by Dinç and Üztemur (2017) with Turkish pre-service social studies teachers, it was observed that participants with more sophisticated epistemic beliefs opposed teacher-centred social studies lessons in the classroom, emphasised authentic learning and mentioned the importance of out-of-school learning.

Limitations and Directions for Future Research

The current study found that the three-dimensional structure of BHQ emerged in the context of Turkish culture. It can be said that there are some problems in terms of validity and reliability. As a result of CFA, some fit indices (IFI, CFI) were low. This may be due to the fact that CFA was not performed on a different data set. Due to the limited number of participants in the study, all analyses were performed on a single data set, which may have caused some problems related to validity and reliability. In addition, the relatively small number of participants might also have been reducing the power of representation. The removal of some items from the scale and the results indicating that contrary to the theory some items are located in different factors may be caused by contextual factors. To fully understand the factor structures of BHQ in Turkish culture, in-depth interviews and studies using the think-aloud technique might be helpful. Additionally, quantitative studies carried out on large samples may produce more accurate findings. Another limitation of the study is that the reliability coefficients of the sub-dimensions of BHQ were relatively low. Although the reliability coefficient was found to be low in studies in the literature, in which BHQ was adapted to different cultures (Hamer, 2016; Mierwald & Junius, 2022), studies should be conducted on different samples and comparisons between sub-dimensions should be done to find out whether this situation is specific to Turkish culture. Although the literature on epistemic beliefs in the field of history has expanded especially in the last 20 years, the present study is the first one addressing this issue in the Turkish context. In this respect, it can be said that the present study has an exploratory aspect. In Türkiye, further studies are required to assess the epistemic beliefs of teachers and students in the field of history and to reveal the effects of these beliefs on the teaching-learning processes.

Conclusion

In this study, the psychometric properties of the BHQ in Turkish culture were analysed. The results showed that the three-dimensional structure of the scale was supported in Turkish culture. Considering that the original dimensions of the scale were not supported in adaptation studies conducted in different cultures, it can be said that the results of the current research are significant. Another important result of the study was that the objectivist sub-dimension was clearly separated from the other sub-dimensions. This means that the data collected in the Turkish context supports the clear distinction between an objectivist stance, which interprets the historical knowledge/information detached from its context, independent of its author(s)/makers and disconnected from learners’ past experiences and perspectives emphasising the interpretive nature of historical knowledge. These results are noteworthy in that the findings from the Turkish context support the theory.