The Relationship between Epistemological Beliefs and Assessment Conceptions among Pre-Service Teachers

1.1 Statement of the Problem

Despite Ethiopia’s policy changes and reforms aimed at enhancing educational quality at all levels, significant challenges persist. Various initiatives have been launched, including the Higher Diploma Programs at tertiary institutions, Continuous Professional Development at primary and secondary schools, and initiatives like Strengthening Mathematics and Science Education in Ethiopia. Additionally, extensive early grade assessments such as EGRA for reading and EGMA for mathematics, alongside assessment for learning (AfL) and continuous classroom assessment at lower primary schools and teacher education colleges have been implemented. However, national-level learning assessments reveal persistent concerns regarding educational quality and student performance. For instance, the sixth National Learning Assessment (NEAEA, 2020) reported that grade 4 students’ composite score across four subjects was only 37.71%, significantly below the 50% average. The most recent data from the seventh round (NEAEA, 2023) show a slight improvement to 39.7% with a standard deviation of 13.8, underscoring ongoing issues with educational quality.

Even though assessments are critical for monitoring, supporting, and crediting student progress (Hofer, 2001; Ismail, 2017), there was a significant knowledge gap among Ethiopian teachers, from primary to tertiary levels, about the purpose and implementation of assessments (Moges, 2018; Tulu & Tolosa, 2018). Often, this results in the preference for continuous testing over meaningful assessment practices. Thus, pre-service teachers must bridge the gap between assessment theory and practice by broadening their understanding of assessment, thinking critically about how they conceptualize assessment, and putting their assessment views into practice (Prastikawati, Mujiyanto, Saleh, & WuliFitriati, 2022), as their assessment conceptualization has far-reaching implications for the teaching and learning process (Lutovac & Flores, 2022). Since there are serious consequences for educational quality if teachers’ epistemological beliefs and assessment roles are not in line with these principles, these gaps in knowledge, beliefs, and practices underscore the need for more research, especially on pre-service teachers’ epistemological beliefs and comprehension of assessment conceptions.

Although there are many studies on the relationship between pre-service teachers’ epistemological beliefs and teaching and learning conceptions globally, there seem to be relatively few on pre-service teachers’ epistemological beliefs and how they might find their pattern in and inform assessment conceptions. This issue is meager in the Ethiopian context. These underexplored allegations sparked our research and inspired the researchers to conduct the current investigation. Subsequently, using the Schommer (1990) model of epistemological beliefs as its theoretical base, this study investigated pre-service teachers’ assessment conceptions, epistemological beliefs, and potential links between these two features targeting to enhance the preparation and effectiveness of teachers in education colleges, which could ultimately improve educational practices and quality across schools.

This investigation aims to bridge the gaps in understanding and practice, thereby contributing positively to the future landscape of educational quality in Ethiopia.

1.2 Theoretical Framework

The concept of epistemological beliefs concerns how individuals perceive knowledge and the processes of acquiring it. This field gained traction following Piaget’s introduction of genetic epistemology in the 1950s, and further developed through Perry’s studies in the 1960s. Piaget’s theory, reflecting the convergence of philosophy and psychology, outlines the stages of intellectual development and the acquisition of knowledge (Hofer & Pintrich, 1997). Perry extended this understanding by exploring college students’ epistemic growth through his own theoretical framework, identifying four developmental stages: dualism, multiplicity, relativism, and commitment (Perry, 1970).

These stages progress from viewing knowledge as absolute truths to recognizing its contextual and evaluative nature, culminating in a commitment to synthesized ideas and values (Hofer, 2001). Challenging this linear perspective, Schommer (1990) argued against the unidimensional and developmental approach of epistemological beliefs as proposed by Perry. Instead, Schommer introduced a multidimensional framework that consists of five dimensions: certainty of knowledge, source of knowledge, structure of knowledge, control of knowledge acquisition, and speed of knowledge acquisition. These five dimensions constitute two broad categories at the heart of individual’s epistemological theories: beliefs about the nature of knowledge and beliefs about the nature of knowing. While the source, control, and speed of knowledge acquisition indicate the nature of knowing, the certainty, and structure of knowledge reflect the nature of knowing.

Schommer used the terms Simple Knowledge, Certain Authority, Omniscient Authority, Quick Learning, and Innate Ability to characterize these five aspects of believing. Simple knowledge is the belief in the structure of knowledge, which ranges from discrete pieces to integrated concepts, while certain authority is the belief in the dependability of knowledge, which ranges from certain to changing. Omniscient authority refers to beliefs in the validity of the source of knowledge, which can range from being inherited from superiors to being based on evidence and reason. Innate ability is concerned with an individual’s belief that their learning capacity is fixed at birth or subject to change over time, whereas quick learning is the notion that the speed of learning ranges from quick or not at all to gradual.

These belief dimensions range from naïve to sophisticated beliefs and emphasize the evolving and complex nature of knowledge, suggesting that learners can exhibit different levels of sophistication across these dimensions independently (Schommer, 2008). This indicates that learners with sophisticated beliefs consider knowledge as a complex, interconnected, and comprehensive structure, while naive learners see it as simple and discrete elements (Mardiha & Alibakhshi, 2020). Naive learners want to assume transmission of knowledge through authority, that learning is permanent and hereditary, and consider experts as the source of knowledge (Soleimani, 2020). Sophisticated learners, on the other hand, believe that knowledge acquisition is not only through specialists but also via lifelong experiences and observations (Schommer, 2008). Individuals who hold advanced epistemological beliefs feel that improving learning involves substantial effort (Önal & Kırmızıgül, 2021).

Research has demonstrated the connection between the epistemological perspectives of learners and teachers, teaching and learning conceptions, conceptual learning, assessment conceptions, and motivating concepts such as self-efficacy and intrinsic worth (Ismail, 2017; Jiang, Wang, Bonner, & Yau, 2021; Sengul, 2024; Soleimani, 2020). Understanding learners’ epistemological beliefs can help us better understand how they engage in learning activities, and how best they can be prepared for the work in schools.

In this study, Schommer's framework that views epistemological beliefs through five dimensions: Simple Knowledge, Certain Authority, Omniscient Authority, Quick Learning, and Innate Ability were employed. To assess the epistemological beliefs of pre-service teachers, the researchers utilized the Epistemological Beliefs Questionnaire developed by Schraw, Bendixen, and Dunkle (2002). Moreover, the researchers utilized a scale adapted from Kul (2013) to categorize the epistemological beliefs of pre-service teachers into naïve, moderate, and sophisticated. This scale, which employs a five-point metric with four intervals, delineates the range of beliefs as follows: a mean score from 1 to 2.6 indicates naïve beliefs, from 2.61 to 3.4 indicates moderate beliefs, and from 3.41 to 5 indicates sophisticated beliefs. This scale provides a structured method to assess the complexity of pre-service teachers’ epistemological beliefs, facilitating a deeper understanding of their implications for educational practices (Figure 1).

Figure 1

A scale for Determining naïve, moderate, and sophisticated epistemological beliefs of pre-service teachers adapted from Kul (2013).

Assessment as an ongoing process that involves identifying, collecting, recording, scoring, and interpreting information to reflect on the overall learners’ attainment of goals and purposes, it can sustain the intended performance and growth of each student according to his or her learning needs, style, and speed (Black & Wiliam, 1998; Gebremedhin & Syoum, 2015). Through assessment, teachers can make educational decisions about their learners to provide learners with ongoing feedback on their progress, their weaknesses, and strengths, judge the effectiveness of the instruction and the adequacy of the curriculum, and inform policy (Ghaicha, 2016). It establishes how well learners are supported during the learning process and the outcome of learning. This attribute has a lot to do with instructors’ understanding of nature, essence, and various approaches to conceiving the subjects they teach.

As mentioned above, whether at the national level, local level, or classroom level, assessments should have a variety of functions. Above all, teachers should view and use assessments to guide their teaching practices (Black & Wiliam, 2009). However, research and theory on teachers’ perspectives have revealed that teachers’ conceptions about professional activities such as instruction and assessment have their own contribution to how they practice their work (Barnes, Fives, & Dacey, 2015; Fives & Buehl, 2012) specifically the way teachers conceptualize assessment influences how they engage their students in assessment activities.

The word “conceptions” refers to “mental structure encompassing beliefs, meanings, concepts propositions, rules, mental images, preferences and the like” (Thompson, 1992, p. 130). Brown (2004, 2006) pioneered a line of research in assessment that used this construction to investigate and quantify teachers’ assessment beliefs. He created and used the Conception of Assessment (COA-III) tool to test teachers’ conceptions about the nature of assessment. This assessment tool tried to describe the purpose of assessment based on the existing literature. In light of this, he gave four perspectives on assessment purposes: improvement, student accountability, school accountability, and assessment as irrelevant. The first conception of assessments informs instruction and encourages learners’ learning development. Assessment for student accountability informs either instruction or determines grades. The third is school accountability assessment, which is used to evaluate and make decisions regarding the quality and performance of schools and teachers. These conceptions utilize the results of an assessment to illustrate publicly that teachers or schools are performing a good job. Assessment for improvement deals with promoting teaching and learning, and this informs instruction and students’ learning progress. In this study, assessment conceptions are used based on the assessment purpose modeled by Brown’s (2008) as a framework to analyze assessment conceptions of pre-service teachers.

Teachers’ and students’ beliefs on the nature of knowledge and how to acquire it have been shown to have an impact on their pedagogical conceptions and instructional strategies. It is possible to say that the epistemological beliefs of learners and teachers are related to their conceptions of assessment, yet there are limited studies reflecting the association between epistemological beliefs of pre-service teacher and their assessment conceptions in relation to assessment purpose. Based on these theoretical considerations, we transformed the study’s two primary goals: finding pre-service teachers’ epistemological beliefs and assessment conceptions and determining the relationship between these beliefs and their assessment conceptions to address the following research questions about pre-service teachers’ espoused epistemological beliefs and assessment conceptions.

1. What are the pre-service teachers’ conceptions of assessment?

2. What are the pre-service teachers’ epistemological beliefs about assessment?

3. What is the relationship between pre-service teachers’ conceptions of assessment and their epistemological beliefs?

4. To what extent do epistemological beliefs and their dimensions explain variations in pre-service teachers’ conceptions of assessment?

2.1 Context of the Study and Participants

The research was carried out at Fitche College of Teacher Education in Oromia, Ethiopia. The participants of the study came from all parts of the Oromia region. Afan Oromo was the medium of instruction from grades 1 to 8, while English was used from grades 9 to 12. In the teacher education college, the medium of instruction is Afan Oromo. The college offers a three-year and a four-year diploma in regular and extension programs in different specializations like mathematics, integrated & environmental science, Language (Afan Oromo and English) and social science. Upon completion of the diploma program, pre-service teachers are required to teach in a primary school: grade 1 through grade 8. The age of the participants ranges from 20 to 23 years. All pre-service teachers studied a variety of common courses throughout their time in the college, including pedagogical classes, subject courses, subject area methodology courses, and practicums in their respective departments. This study involved 227 pre-service elementary teachers from Fitche College of Teacher Education in the Oromia region of Ethiopia, who were enrolled in pedagogical, subject content, subject area methodology courses, and practicum. Of these, 197 responded adequately to the questionnaires, and the results are analyzed based on these respondents. From 197 participants, 119 of them are males. The number of participants in this study from each department/stream/is given in Table 1.

2.2 Data Collection Tools and Procedure

Data were gathered using two standardized questionnaires. The Epistemological Beliefs Questionnaire by Schraw et al. (2002) and the Assessment Conceptions Questionnaire by Brown (2002). Schraw et al. (2002) identified five dimensions of epistemological beliefs: Simple Knowledge, Certainty Knowledge, Omniscient Authority, Quick Learning, and Innate Ability. Brown categorized assessment conceptions into four dimensions: Improvement in Teaching and Learning, School Accountability, Student Accountability, and the Irrelevance of Assessment. The Epistemological Beliefs Questionnaire consisted of 26 items rated on a 5-point Likert scale (strongly disagree to strongly agree), and the Assessment Conceptions Questionnaire included 22 items on a 6-point scale (1–6, where, 1 = strongly disagree, 2 = mostly disagree, 3 = slightly agree, 4 = moderately agree, 5 = mostly agree and 6 = strongly agree).

2.3 Reliability and Validity

Both questionnaires were translated into Afan Oromo to reduce bias and ensure accurate responses. The translations were reviewed by language experts, measurement and evaluation specialists, and mathematics educators, leading to revisions based on their feedback. The tools were originally validated, but to check contextual suitability, we had the plan to pilot them in a teacher education college. However, there were no available students around and the tools were pilot-tested with 112 teachers from Fitche primary schools by choosing those who studied in the College. Reliability analyses yielded a Cronbach alpha coefficient of 0.84 for the Epistemological Beliefs Questionnaire and 0.87 for the Assessment Conceptions Questionnaire. To ensure the reliability of the final collected data, we also calculated the Cronbach alpha coefficients after collecting the actual data used for the study and found a Cronbach alpha coefficient of 0.78 for the Epistemological Beliefs Questionnaire and 0.84 for the Assessment Conceptions Questionnaire both of which are within the acceptable range.

2.4 Data Analysis

Data were analyzed using descriptive and inferential statistics, including means, standard deviations, multiple regression, and correlation. Prior to using correlation and multiple regression, the researchers examined several assumptions, such as normality, homogeneity of variance, and linearity. The data were evaluated for normality using skewness and kurtosis, and the findings were within the desired range, indicating an approximately normal distribution. The researchers utilized a scatter plot to check the linearity assumption of the epistemological beliefs and assessment conceptions data, and the distribution of data was approximately close to a straight line, indicating that linearity is assumed and this is indicated in Appendix 1.

The researchers utilized Levene’s test to check for the homogeneity of variances; the ANOVA p-value for all components of assessment conceptions and epistemological beliefs was greater than 0.05, meaning that the assumption of homogeneity of variances was not violated. In terms of multicollinearity, the relationship between pre-service teachers’ epistemological beliefs and sub-components of their conception of assessment was commuted. The outcome indicated that all variables had pairwise correlation coefficients are less than 0.8, proving that the multicollinearity assumption was upheld and that multiple regression computation feasible.

2.5 Ethical Issues

In this study, meticulous attention was given to ethical considerations to uphold the integrity and ethical standards of the research. Participants were first thoroughly informed about the study’s objectives, the procedures involved, and any potential risks. Their informed consent was obtained before participation, ensuring that they were fully aware and agreeable to the terms. To protect participants’ privacy, strict confidentiality measures were put in place. This ensured that their identities and responses remained secure, and accessible only to the research team. Participation in the study was entirely voluntary, allowing participants the freedom to withdraw at any point without facing any negative repercussions. Furthermore, the study had received prior approval from the appropriate institutional review boards, affirming its adherence to ethical research standards. Additionally, cultural sensitivity was carefully considered. Questionnaires were adapted and translated to reflect the cultural contexts and languages of the participants, which helped respect cultural differences and improved comprehension. These ethical practices were critical in conducting the research responsibly and respectfully, consistent with both international standards and local ethical guidelines.

3.1 Epistemological Beliefs of Pre-Service Teachers

The pre-service teachers’ epistemological beliefs were assessed using mean scores and standard deviations. Table 2 presents the descriptive statistics, highlighting that the pre-service teachers generally possess sophisticated epistemological beliefs with mean scores ranging from 2.56 to 3.56 on a five-point scale.

From the data, the dimension of “Omniscient Authority” exhibits the highest mean score (M = 3.56), suggesting that pre-service teachers value the role of individual critical inquiry over external authoritative sources in their knowledge acquisition. This is closely followed by “Simple Knowledge” (M = 3.45), indicating a preference for understanding knowledge as interconnected concepts rather than isolated facts.

“Certainty of Knowledge” records a mean of 3.17, reflecting a sophisticated view among the teachers that knowledge is tentative and subject to change. In contrast, “Quick Learning” has the lowest mean score (M = 2.56), which indicates a belief that acquiring knowledge requires significant time and effort. The “Innate Ability” dimension, with a mean of 3.02, implies a prevailing belief that learning abilities are acquired and dynamic, rather than fixed from birth.

Finally, the composite mean score for overall epistemological beliefs (M = 3.15) is significantly above the midpoint of the scale, reinforcing the indication that the participants hold relatively sophisticated epistemological beliefs.

3.2 Pre-Service Teachers’ Conceptions of Assessment

The descriptive statistics used to analyze the pre-service teachers’ conceptions of assessment included mean scores and standard deviations across four dimensions. These statistics help identify the level of the pre-service teachers’ conceptions regarding different aspects of assessment.

Table 3 indicates that the concept of “Improvement” receives the highest level of agreement among the participants (M = 4.55, SD = 1.1), signifying a strong endorsement of assessment as a tool for enhancing learning. This is followed by “Student Accountability” (M = 4.45, SD = 1.16), which underscores the importance of making students accountable for their learning.

The concept of “School Accountability” also receives substantial agreement (M = 4.41, SD = 1.19), suggesting that pre-service teachers recognize the role of assessments in holding schools accountable, though to a slightly lesser extent than individual improvement and student accountability. Meanwhile, “Irrelevance” of assessments has got a mean score (M = 3.56, SD = 0.92), indicating that students couldn’t definitely decide on the irrelevance of assessment, but it also highlights that some students tend to rate assessment as irrelevant, demanding action to revert such conceptions.

Despite analyzing epistemological beliefs and assessment conceptions as distinct areas, the relationships between these domains, including how dimensions of one may influence or correlate with dimensions of the other, were further investigated to understand the variations and interconnections more deeply.

3.3 Relationship between Epistemological Beliefs and Assessment Conceptions

Table 4 reveals that “Student Accountability” does not exhibit statistically significant correlations with any dimensions of epistemological beliefs, whereas other assessment conceptions do. Notably, “Improvement” and “School Accountability” show strong correlations with “Omniscient Authority” and “Simple Knowledge,” indicating a significant relationship between these epistemological beliefs and assessment conceptions.

A composite correlation coefficient of 0.435 (statistically significant) was determined, indicating a moderate overall association between the dimensions of epistemological beliefs and assessment conceptions. Regression analysis further revealed a model where epistemological beliefs predict assessment conceptions with a formula of Conception = 2.617 + 0.516 × Epistemological Belief and an R-squared value of 0.189. This suggests that epistemological beliefs explain approximately 18.9% of the variance in assessment conceptions as outlined in Table 5.

3.4 School Accountability as a Function of Components of Epistemological Beliefs

Significant predictors of school accountability include “Omniscient Authority” and “Simple Knowledge,” while other dimensions showed no significant impact. This model explains 15% of the variance in school accountability conceptions (Table 6).

3.5 Improvement as a Function of Components of Epistemological Beliefs

For “Improvement,” significant predictors include “Omniscient Authority,” “Simple Knowledge,” and “Innate Ability,” explaining 23% of the variance in improvement conceptions (Table 7).

3.6 Irrelevance as a Function of Components of Epistemological Beliefs

The “Irrelevance” model indicates that the dimension “Simple Knowledge” negatively contributes to the variation in irrelevance perceptions, with the model explaining 12% of the variance in these conceptions (Table 8).

3.7 Student Accountability a Function of Components of Epistemological Beliefs

The results from Table 9 demonstrate that the epistemological beliefs of pre-service teachers collectively contribute to the prediction of student accountability assessment conceptions, though they explain only 9% of the variance. This suggests that while these beliefs play a role, other factors significantly contribute to the variation of these conceptions, accounting for approximately 91% of the variance.

Among the predictors, “Omniscient Authority” emerges as a significant contributor to the variation of assessment conceptions (β = 0.258, t = 3.748, and p < 0.001), indicating that beliefs related to the importance of individual authority in knowledge generation strongly predict conceptions of student accountability in assessments. Other dimensions, including “Simple Knowledge,” “Certainty Knowledge,” “Quick Learning,” and “Innate Ability,” do not show a statistically significant contribution to the variation in student accountability conceptions. The regression equation for predicting student accountability assessment conceptions is: SAC = 2.727 + 0.005(SK) + 0.071(CK) + 0.008(QL) + 0.364(OA) + 0.032(IA)SAC = 2.727 + 0.005(SK) + 0.071(CK) + 0.008(QL) + 0.364(OA) + 0.032(IA)

This model allows for a deeper understanding of how specific epistemological beliefs contribute to the variation in assessment conceptions related to student accountability.

5.1 Recommendations

This study has established that the epistemological beliefs of pre-service teachers significantly contribute to the variation in their conceptions of assessment. Accordingly, teacher education programs should prioritize these beliefs. The research indicates that pre-service teachers possess moderately sophisticated epistemological beliefs, suggesting a need for teacher educators to implement targeted strategies to refine these beliefs. Furthermore, addressing and mitigating irrelevant conceptions is also essential.

The current study concentrated on general epistemological beliefs. However, epistemological beliefs and conceptions of assessment could vary depending on the subject nature. Hence, future research should explore subject-specific epistemological beliefs, such as those related to mathematics, science, language, etc., to provide a more nuanced understanding. Additionally, investigating effective methods to alter pre-service teachers’ epistemological beliefs could yield practical benefits.

Beyond the scope of this study, it would be valuable to examine the interplay between epistemological beliefs and other critical educational outcomes, such as higher-order critical thinking, communication skills, imagination, and creativity. This broader approach could further illuminate the contribution of foundational beliefs to various aspects of teaching and learning. Furthermore, since the data were only collected through a questionnaire, this might have its own limitations on the study results, and therefore, future studies should include interviews and class observation as a means of data collection, even for triangulation.

5.2 Limitations of the Study

This study has some limitations. Some limitations include the following: data were collected through self-responded questionnaires. Such data could have been strengthened through classroom observation and interview/group discussion that may have improved the study’s outcomes. The use of a self-responded questionnaire might have restricted the generalizability of the findings.