114 Sodobna pedagogika/Journal of Contemporary Educational Studies
Alena Letina, Suzana Tomaš and  
Alma Škugor
The Conception and Practice of Assessment 
from the Perspective of Primary Education 
Teachers in Croatia
Abstract: This paper presents the results of a study aimed at examining Croatian primary education 
teachers’ conceptions of assessment and the correlation between these conceptions and the implemen-
tation of assessment in their teaching practice. The analysis of teachers’ conceptions includes consid-
erations of the reliability and importance of assessment, its role in fostering students’ development, 
the perception of assessment as a tool for ensuring the quality of school and teacher performance, and 
its role in preparing students for examinations. The study also examines teachers’ assessment prac-
tices, analyzing whether they predominantly focus on formative assessment thus fostering student 
development, or on summative assessment by preparing students for test-related tasks. The research 
was conducted on a sample of 261 primary education teachers in Croatia. The findings reveal a sta-
tistically significant correlation between teachers’ perceptions of assessment and its implementation 
in teaching. Teachers who perceive assessment as important for students’ personal development are 
more likely to employ formative assessment in their teaching. Based on the results, the study dis-
cusses implications for practice and possibilities for improving the perception and implementation of 
assessment in primary education, emphasizing the potential of assessment to promote holistic student 
development.
Keywords: formative assessment; teachers’ self-assessment; fostering student development; summa-
tive assessment; purpose of assessment
UDC: 37.091.3
https://doi.org/10.63384/spB52z781a
Scientific article
  
Alena Letina, Associate Professor , University of Zagreb, Faculty of Teacher Education, Savska cesta 77, 
Zagreb, Croatia; e-mail address: alena.letina@ufzg,hr, 
Suzana Tomaš, Associate Professor, University of Split, Faculty of Humanities and Social Sciences, 
Poljička cesta 35, Split, Croatia; e-mail address: suzana@ffst.hr, 
Alma Škugor, Associate Professor, Josip Juraj Strossmayer University of Osijek, Faculty of Education, 
Cara Hadrijana 10, Osijek, Croatia; e-mail address: askugor@foozos.hr, 
Let./Vol. 76 (142)
Številka 2/2025
pp. 114-131

115 
Introduction
Assessment is an essential component of the educational process. At the same 
time, the teacher’s role within this process is both responsible and demanding, 
particularly in how it relates to the implementation and interpretation of assess-
ment. The way teachers perceive the nature and purpose of assessment can signifi-
cantly influence students’ understanding of it and their educational achievements 
(Brown 2008). Previous research has shown that teachers who view assessment 
as a means of supporting the learning process by providing constructive feedback 
and improving that process encourage students to adopt more effective learning 
strategies, increase their motivation to learn and enable them to achieve higher 
levels of educational outcomes (Berry 2008; Mekonnen and Melesse 2022; Toth 
and Csapo 2022). In contrast, approaches that reduce assessment solely to a sum-
mative evaluation of student achievements can foster anxiety among students and 
lead them towards superficial and reproductive learning aimed at meeting exter-
nal criteria, without deeper understanding or the development of complex cogni-
tive skills such as critical thinking and reasoning (Pekrun et al. 2017). Effective 
assessment should entail a balance between formative and summative approach-
es, ensure fairness and transparency, and promote the development of students’ 
metacognitive skills (Kožuh 2019; Stobart 2008).
In the context of this study , the term conception refers to the general, implicit 
knowledge that an individual possesses about the nature of a particular phenome-
non (Brown 2008), including its key characteristics, structure and purpose (Ajzen 
2005; Fives and Buehl 2012). Accordingly, a conception of assessment pertains to 
the understanding of the nature and purpose of methods used to monitor student 
progress, evaluate learning outcomes and assess student achievements. 
Previous research indicates that teachers’ conception, beliefs and attitudes 
can significantly influence the quality of the educational process, the achievement 
of learning outcomes, teaching methods and assessment strategies (Fives and 
Buehl 2012; Torrance and Pryor 2021). Teachers with a higher level of assessment 
literacy tend to have a clearer understanding of the purpose of assessment (Loon-
ey et al. 2017; Xu and Brown 2016). Perceiving assessment as a tool for improving 
Letina, Tomaš, Škugor

116 Sodobna pedagogika/Journal of Contemporary Educational Studies
learning correlates with teachers’ focus on fostering deeper understanding of in-
structional content during the learning process (Christoforidou et al. 2014; Lin et 
al. 2014). Meanwhile, factors such as national assessment programs often create a 
sense of pressure, which can limit the use of formative and diagnostic approaches 
to assessment (Asamoah et al. 2024; Toth and Csapo 2022). A study conducted 
by Chen and Brown (2013) among prospective teachers in China revealed that 
viewing assessment primarily as an instrument for exam success correlates with 
their negative attitudes towards assessment. Teachers with such perceptions are 
more likely to regard assessment as non-diagnostic, non-formative, irrelevant to 
the development of life skills and generally unimportant. Furthermore, Brown 
and Gao (2019) emphasize that while teachers’ perceptions of assessment often 
align with universal goals, such as supporting learning and ensuring educational 
quality, they also reflect local cultural and educational specificities. 
In 2008, Brown developed a model that categorizes the purposes of assess-
ment into three main purposes and one counter-purpose. The first perspective 
considers assessment as a tool for improving teaching and learning, emphasizing 
its formative role in identifying students’ needs and adapting instruction accord-
ingly. The second perspective highlights assessment as a means of accountability 
for schools and teachers, defining it as a tool for measuring the effectiveness of 
the educational system and teachers’ performance. The third perspective views 
assessment as a tool for fostering student accountability, emphasizing its role in 
motivating students and encouraging them to take responsibility for their own 
learning. The fourth counter-perspective regards assessment as an irrelevant as-
pect of the educational process with no meaningful impact. By combining school 
and student accountability in particular, two main purposes emerge: accountabili-
ty and improvement (Brown 2008). In the context of assessment as accountability , 
the primary goal is the evaluation and judgment of the quality of achievements. 
Conversely , assessment focused on improvement aims to identify areas for growth 
and to encourage change.
In recent decades, most educational policies worldwide have emphasized the 
creation of learning environments and practices that promote learning improve-
ment, specifically through formative assessment (Black and Wiliam 1998; Brown 
and Hattie 2012). In this context, assessment moves away from making value 
judgments about students, schools or teachers, focusing instead on reflecting on 
the quality of teaching and learning processes.
Assessment occupies a central role in educational practice, but its inherent 
duality in understanding and application creates a complex dynamic. It guides 
teachers and students towards a shared goal: identifying and overcoming learning 
difficulties. Additionally, assessment assumes a role of accountability. Assessment 
results are often used to evaluate the performance of teachers and schools, which 
in practice can create tensions and counterproductive effects. While assessment 
aimed at improvement encourages openness to recognizing and addressing weak-
nesses, the evaluative aspect, particularly when tied to negative consequences, 
can discourage this process. In environments where accountability is heavily em-
phasized, there is a risk that teachers, driven by the need to maximize perfor-
Letina, Tomaš, Škugor

117 
mance and meet the expectations of their superiors, may avoid openly confronting 
challenges associated with formative assessment. Given the ongoing tension be-
tween the functions of improvement and accountability , teachers’ perceptions and 
practices of assessment are likely to reflect the dominant approach present within 
a particular educational system, as well as the emphasis that the system places on 
each of these functions (Brown and Gao 2019).
The Croatian education system, though rooted in a strong didactic tradition, 
has increasingly incorporated Anglo-Saxon terminology and concepts, particularly 
following its 2017–2019 curricular reform. Terms such as assessment for learning, 
assessment as learning and formative assessment have been embedded in official 
documents, aligning national discourse with global trends. The National Curricu-
lum for Primary Education (MSE 2017) and the Guidelines for the Evaluation of 
Processes and the Achievement of Educational Outcomes (MSE 2019a), both pub-
lished by the Ministry of Science and Education of the Republic of Croatia (MSE), 
highlight the formative and developmental role of assessment, emphasizing feed-
back, student reflection and learning support. Foundational pedagogical principles 
reinforce a shift towards learner-centred, process-oriented instruction (Cindrić et 
al. 2010; Matijević and Radovanović 2011; Rajić 2017). The legal framework regu-
lating student assessment in Croatian primary and secondary schools, defined by 
the Ordinance on the Methods, Procedures and Elements of Student Assessment in 
Primary and Secondary Schools (MSE 2019b), distinguishes between assessment 
for, as and of learning, and promotes continuous progress monitoring with an 
increasing focus on formative strategies. However, no systematic research has yet 
explored how Croatian teachers conceptualize and apply assessment.
Methods
Aim and research questions
The aim of the study is to examine Croatian primary education teachers’ 
conceptions of the nature and purpose of assessment and to determine the extent 
to which these conceptions are reflected in their assessment practices.
Based on this aim, the study addresses the following research questions:
 – What is the dominant conception of the nature and purpose of assessment 
among primary education teachers in Croatia?
 – To what extent are teachers’ assessment practices, according to their 
self-evaluation, oriented towards formative or summative assessment?
 – To what extent do teachers’ conceptions of assessment predict their self-eval-
uated assessment practices?
 – Is there a statistically significant correlation between teachers’ conceptions 
of assessment and their self-evaluation of their assessment practices?
 – Are there statistically significant differences in teachers’ conceptions of as-
sessment and their self-evaluated assessment practices based on their edu-
cational qualifications, years of teaching experience and professional titles?
The Conception and Practice of Assessment from the Perspective of Primary ...

118 Sodobna pedagogika/Journal of Contemporary Educational Studies
Study sample
This research was conducted in April and May 2024 with a sample of 261 
primary education teachers (from 1st to 4th grade) in Croatia. Table 1 presents 
the distribution of the sample’s demographic and professional variables. The ana-
lysed variables include gender , education level, professional advancement, years of 
teaching experience and the type of environment in which the respondents work.
Variable Level N % Variable Level N %
Gender Male 10 3.8 Years of 
experience
0–5 years 35 13.4
Female 251 96.2 6–10 years 21 8.0
Education Bachelor‘s 
degree
75 28.7 11–15 years 30 11.5
Master‘s degree 151 57.9 16–20 years 31 11.9
Master‘s in 
profession
30 11.5 More than 20 
years
144 55.2
Master‘s in 
science
4 1.5 School 
area
Urban 173 66.3
PhD 1 0.4 Suburban 31 11.9
Professional 
advancement
Teacher
Teacher 
mentor
Teacher advisor
Outstanding 
teacher advisor
187
42
24
8
71.6 Rural 57 21.8
Table 1. Sample of primary education teachers in Croatia
Procedure and research instruments
The data for this study were collected using an online survey questionnaire. 
Respondents were invited via official emails sent to primary schools across Croa-
tia, which encouraged teachers to participate in this study. To be included in the 
study, participants had to be currently employed as primary school teachers. Par-
ticipation was voluntary and anonymous, and respondents were informed about 
the research objectives. The study received ethical approval from the Ethics Com-
mittee of the Faculty of Humanities and Social Sciences at the University of Split. 
Participants were also allowed to withdraw from the study at any point, ensuring 
adherence to ethical research standards, including free consent and privacy pro-
tection. The questionnaire consisted of two sections: The first section examined 
conceptions of assessment, and the second addressed self-assessment of teaching 
practices. Respondent demographic information was collected in the introductory 
part of the questionnaire.
Letina, Tomaš, Škugor

119 
To examine conceptions of assessment, the Teachers’ Conceptions and Prac-
tices of Assessment in Chinese Contexts questionnaire (Brown et al. 2011) was 
used, partially adapted to the Croatian educational system. The instrument was 
chosen for its empirical validation and alignment with established assessment 
literacy frameworks. The questionnaire consisted of 48 statements to which re-
spondents rated their agreement on a five-point Likert scale (ranging from 1 = 
strongly disagree to 5 = strongly agree). The statements were divided into four 
subscales:
 – Reliability and Importance of Assessment (RI) explored teachers’ attitudes 
towards the objectivity, reliability and importance of assessment. It included 
statements related to the accuracy and fairness of the assessment process.
 – Fostering Student Development (FSD) addressed teachers’ attitudes towards 
the importance of assessment in the learning process, focusing on fostering 
student progress, development and holistic growth. In this context, assess-
ment is viewed as a means to enhance students’ competencies.
 – Ensuring Quality of School and Teacher Performance (EQ) examined teach-
ers’ attitudes towards assessment as a tool for monitoring and improving 
the quality of school and teacher performance, emphasizing institutional ac-
countability for maintaining high educational standards.
 – Exam-Oriented Focus (E) evaluated teachers’ views on assessment’s role in 
exam preparation and success.
The teachers’ self-evaluation of teaching practices related to assessment 
was examined using the Practices of Assessment Inventory (Brown et al. 2009), 
which was adapted to this study . The instrument offers a validated framework for 
analysing teachers’ assessment practices in line with international research. The 
questionnaire consisted of 34 statements that fell into four subscales:
 – Assessment for Learning (AfL) examined the frequency of using assessment 
methods that support the learning process and provide feedback to students 
to improve the learning process.
 – Assessment as Learning (AaL) explored the frequency of applying assess-
ment strategies that encourage students to engage in self-assessment and 
reflection on their own learning.
 – Assessment of Learning (AoL) examined the use of assessment to evaluate 
outcomes and performance.
 – Preparation for Exam Tasks (PET) examined the frequency of using strate-
gies aimed at preparing students for standardized exams.
Teachers self-assessed the frequency of employing these assessment methods 
on a five-point scale, rating how often they used each strategy (1 = very rarely, 5 
= very often).
The Conception and Practice of Assessment from the Perspective of Primary ...

120 Sodobna pedagogika/Journal of Contemporary Educational Studies
Data analysis methods
The normality of distribution for all variables was tested using the Kolmogo-
rov-Smirnov test. Parametric statistical procedures were applied for further data 
analysis. Descriptive statistics were used to determine teachers’ conceptions of the 
nature and purpose of assessment, as well as the extent to which their teaching 
practices were oriented towards formative or summative assessment. The Pearson 
correlation coefficient was calculated to examine the relationship between teach-
ers’ conceptions of assessment and their assessment practices, specifically by ana-
lysing the responses across individual scales. Regression analysis was conducted 
to explore the predictive relationship between teachers’ perceptions of assessment 
and their practices. One-way analysis of variance (ANOVA) was employed to iden-
tify differences in responses based on participants’ demographic characteristics. 
Additionally, the reliability of all scales used in the questionnaire was assessed by 
calculating the Cronbach’s alpha reliability coefficient. Paired t-tests compared 
teachers’ ratings across subscales and assessment types.
Results
The descriptive statistical analysis presented in Table 2 indicates a statis-
tically significant deviation from normality in the distributions of all variables. 
However, the observed deviations in skewness and kurtosis measures are minor, 
and the distributions are relatively symmetrical, exhibiting slight positive or neg-
ative skewness. Based on these data characteristics, parametric procedures were 
used in subsequent analyses. The results of the Cronbach’s alpha reliability co-
efficient suggest acceptable reliability for the AoL scale, very good reliability for 
the Preparation for Exam Tasks scale (PET) and Summative Assessment (SUM) 
scale, and excellent reliability for all other scales (Table 2). 
Letina, Tomaš, Škugor

121 
M C D SD Skew Kurt Min Max KS α
RI total 45.8 47.0 49.0 9.06 -0.26 -0.01 22 67 0.07** .88
FSD total 52.6 53.0 52.0 11.19 -0.58 0.65 15 75 0.09** .95
EQ total 30.1 30.0 29.0 7.35 -0.14 0.05 10 50 0.08** .87
E total 17.5 18.0 20.0 3.83 -0.70 0.97 5 25 0.10** .83
AfL total 45.0 45.0 48.0 6.85 -0.36 0.31 23 60 0.07** .81
AaL total 23.1 24.0 24.0 4.25 -0.64 0.30 10 30 0.11** .83
PET total 32.1 33.0 35.0 4.90 -0.88 0.64 16 40 0.12** .77
AoL total 28.9 29.0 29.0 4.57 0.08 -0.17 16 40 0.06** .67
FOR total 68.1 68.0 66.0 10.42 -0.42 0.23 33 90 0.06* .89
SUM total 61.1 61.0 61.0 7.90 -0.47 0.61 32 80 0.07** .78
RI mean 3.1 3.1 3.3 0.60 -0.26 -0.01 1 4 0.07** .88
FSD mean 3.5 3.5 3.45 0.75 -0.58 0.65 1 5 0.09** .95
EQ mean 3.0 3.0 2.9 0.74 -0.14 0.05 1 5 0.08** .87
E mean 3.5 3.6 4.0 0.77 -0.70 0.97 1 5 0.10** .83
AfL mean 3.7 3.8 4.0 0.57 -0.36 0.31 2 5 0.07** .81
AaL mean 3.9 4.0 4.0 0.71 -0.64 0.30 2 5 0.11** .83
AoL mean 3.6 3.6 3.6 0.57 0.08 -0.17 2 5 0.06** .77
PET mean 4.0 4.1 4.4 0.61 -0.88 0.64 2 5 0.12** .67
FOR mean 3.8 3.8 3.7 0.58 -0.42 0.23 2 5 0.06* .89
SUM mean 3.8 3.8 3.8 0.49 -0.47 0.61 2 5 0.07** .78
Legend: RI - Reliability and Importance of Assessment, FSD - Fostering Student Development, EQ 
- Ensuring the Quality of Schools and Teachers, E - Exam-Oriented Focus, AfL – Assessment for 
Learning, AaL – Assessment as Learning, PET - Preparation for Exam Tasks, AoL – Assessment of 
Leraning, FOR – Formative Assessment, SUM – Summative Assessment, M – Mean, C – Median, D – 
Mode, SD – Standard Deviation, Skew – Skewness, Kurt – Kurtosis, Min – Minimum, Max – Maxi-
mum, KS – Kolmogorov-Smirnov test for normality, α – Cronbach’s alpha reliability coefficient, * p < 
0.05; ** p < 0.01.
Table 2. Descriptive statistics  of the assessed subscales and variables
Teachers’ conceptions of the nature and purpose of assessment and self-assessment 
in assessment practices
To address the first and second research questions, a descriptive statisti-
cal analysis was conducted. Arithmetic means (M) and standard deviations (SD) 
were calculated for respondents’ answers to the Conceptions of Assessment and 
Assessment Practices scales. The results are presented in Table 2. The findings 
indicate a neutral attitude towards the reliability and importance of assessment 
in the educational process (M = 3.1, SD = 0.60) and its role in improving the 
quality of schools and teachers (M = 3.0, SD = 0.74). The teachers expressed 
support for the conception of emphasizing the importance of assessment to foster 
The Conception and Practice of Assessment from the Perspective of Primary ...

122 Sodobna pedagogika/Journal of Contemporary Educational Studies
learning and student development (M = 3.5, SD = 0.75). However, the results 
also suggest the presence of a summative perception of assessment as a means 
of achieving external goals and measurable outcomes. A slightly positive attitude 
was observed towards assessment for exam preparation and success (M = 3.5, 
SD = 0.77). Based on these results, the answer to the first research question is 
that primary education teachers predominantly perceive assessment as a tool for 
fostering student development, while also preparing students for standardized 
exams. The teachers’ self-assessment of their teaching practice confirms this dis-
sonance. While teachers frequently employ formative assessment strategies (M 
= 3.8, SD = 0.58), such as providing feedback and monitoring student progress, 
exam preparation strategies remain more prevalent (M = 4.0, SD = 0.61). This 
finding suggests that, despite the teachers’ recognition of the importance of form-
ative assessment, preparing students for exams, especially to measure outcomes 
at the end of instruction, continues to play a dominant role in the participants’ 
teaching practices. Although teachers recognize the importance of fostering stu-
dents’ self-regulation and reflection on their learning, as evidenced by frequent 
use of assessment strategies to promote self- and peer-assessment (M = 3.9, SD 
= 0.71), as well as assessment for learning strategies (M = 3.7, SD = 0.57), the 
results emphasize the need to further encourage teachers to prioritize formative 
assessment over summative approaches.
A repeated measures ANOVA examining differences in the Conceptions of 
Assessment scale revealed a statistically significant difference among the average 
scores across the scale’s four subscales (F(3, 260) = 204.01, p < .01; Table 3).
M RI mean FSD mean EQ mean E mean
RI mean 3.1 – MD = -.45, p < .01 MD = -.05, p > .05 MD = -.45, p < .01
FSD mean 3.5 – – MD = .49, p < .01 MD = .00, p > .01
EQ mean 3.0 – – – MD = -.49, p < .01
E mean 3.5 – – – –
Table 3. Results of pairwise differences among all four subscales on the Conceptions of Assessment 
scale using the Bonferroni test after repeated measures ANOVA 
The respondents scored equally high (M = 3.5) on the FSD and E subscales 
(MD = .00, p > .01). Statistically significant differences were observed between 
these two subscales and the RI subscale (MD = .45, p < .01), as well as the EQ 
subscale (MD = .49, p < .01), with mean scores for the FSD and E subscales sig-
nificantly higher. The difference in mean scores between the RI and EQ subscales 
was not statistically significant (MD = -0.05, p > .05). 
A repeated measures analysis of variance conducted for the Assessment Prac-
tices scale revealed statistically significant differences among the mean scores 
across all four subscales (F(3, 260) = 40.29, p < .01; Table 4). 
Letina, Tomaš, Škugor

123 
M AfL mean AaL mean AoL mean PfT mean
AfL mean 3.7 – MD = -.10, p < .01 MD = -.13, p < .01 MD = -.27, p < .01
AaL mean 3.9 – – MD = .24, p < .01 MD = -.17, p < .01
AoL mean 3.6 – – – MD = -.40, p < .01
PET mean 4.0 – – – –
Table 4. Results of pairwise differences among all subscales on the assessment practices scale using 
the Bonferroni test after repeated measures ANOVA
In this analysis, all pairwise differences were statistically significant. The 
teachers achieved the highest mean score (M = 4.0) on the PET subscale, which 
was significantly higher compared to the subscales AoL (MD = -0.40, p < .01), AfL 
(MD = -0.27, p < .01) and AaL (MD = -0.17, p < .01). Within formative assess-
ment, the mean score on the AaL subscale (M = 3.9) was statistically significantly 
higher than the mean score on the AfL subscale (M = 3.7). Within summative 
assessment, the PET subscale (M = 4.0) had a statistically significantly higher 
mean score compared to the AoL subscale (M = 3.6).
Although differences were observed within formative and summative assess-
ment, the overall mean scores for these two categories were comparable. Paired 
samples t-test results (t = 0.97, p = .33) indicate that there is no statistically 
significant difference between these two variables. This suggests that teachers 
perceive themselves as applying both forms of assessment with similar frequency , 
thereby providing the answer to the second research question. 
Predictive value of individual conceptions of assessment on specific assessment 
practices
To examine the predictive value of individual conceptions of assessment (RI, 
FSD, EQ, E) on specific assessment practices (AaL, AfL, AoL, PET), four regres-
sion analyses were conducted. Conceptions of assessment were used as predic-
tors, while assessment practices were used as criteria. Additionally , two regression 
analyses were conducted with summative (SUM) and formative (FOR) assessment 
as the criteria. Before performing the regression analyses, the necessary assump-
tions for regression analysis were tested. This included analysing correlations 
among variables, residual autocorrelation, multicollinearity and the presence of 
outliers. Table 5 presents the correlations among the predictors and between pre-
dictors and criteria.
The Conception and Practice of Assessment from the Perspective of Primary ...

124 Sodobna pedagogika/Journal of Contemporary Educational Studies
RI FSD EQ E AfL AaL PET AoL FOR SUM
RI total 1 .88** .80** .80** .38** .34** .10 .23** .39** .20**
FSD total 1 .80** .89** .38** .38** .14* .22** .41** .22**
EQ total 1 .77** .36** .32** .15* .32** .37** .27**
E total 1 .35** .34** .16* .24** .37** .24**
FOR AfL total 1 .75** .42** .59** .60**
AaL total 1 .45** .48** .56**
SUM PET total 1 .39** .46**
AoL total 1 .58**
FOR total 1 0.62**
SUM total 1
Table 5. Intercorrelations among conceptions of assessment (predictors) and assessment practices 
(criteria) 
From Table 5, it can be observed that all predictors show small to moderate 
correlations with the criteria (r = 0.14–0.41), except for RI and PET (r = 0.10, p 
> .05). Despite this lack of correlation, RI was included in the regression analysis 
as a predictor of PET, as prior research suggests its potential relevance for this 
aspect of assessment practice.
The predictors are also intercorrelated, but these correlations do not ex-
ceed 0.90. The fact that correlations among predictors are not excessively high 
indicates no significant multicollinearity. This implies that each conception of as-
sessment provides sufficiently unique information, justifying their inclusion as 
separate predictors in the regression analysis. Additionally, the intercorrelations 
between the overall results of formative and summative assessment with the var-
iables from which they are derived were excluded from Table 5, as these repre-
sent inherent relationships that do not provide new insights into the relationships 
among variables.
Residual autocorrelations and multicollinearity are presented in Table 6. 
Durbin-Watson statistic Tolerance Variance inflation factor 
(VIF)
RI total
1.936 - 2.200
.20 5.029
FSD total .13 7.765
EQ total .30 3.286
E total .20 4.922
Table 6. Measures of residual autocorrelation (Durbin-Watson statistic) and multicollinearity (Vari-
ance inflation factor and tolerance) for regression models predicting conceptions of assessment
The range of Durbin-Watson values in Table 6 refers to the six separate re-
gression models with different criterion variables. The values range from 1.936 to 
2.200 (AfL = 1.936, AaL = 2.174, AoL = 2.191, PET = 2.200, FOR = 2.022 and 
Letina, Tomaš, Škugor

125 
SUM = 2.189), indicating no substantial autocorrelation of residuals. Tolerance 
and variance inflation factor (VIF) values, also shown in Table 6 and referring to 
the predictor variables (RI, FSD, EQ and E), were used to assess multicollinearity .  
Multicollinearity was acceptable in terms of tolerance measure, as the values are 
greater than 0.1, but weaker regarding the variance inflation factor (VIF), with 
some factors exceeding the threshold of 4. This is attributed to high correlations 
among predictors. As for outliers, they were removed prior to all analyses, ensur-
ing no influence on the results.
The regression analysis revealed that all regression coefficients were statisti-
cally significant except for the prediction of PET. While the regression coefficients 
for most criterion variables were statistically significant, the percentage of ex-
plained variance was relatively low. The highest explained variance was observed 
for AfL (16%) and AaL (15%) among individual variables. Formative assessment 
(17%) demonstrated greater predictive power compared to summative assessment 
(8%).
Beta weights analysis indicated that only two conceptions of assessment had 
a significant predictive contribution. The conception of assessment as a tool for 
FSD was a significantly positive predictor of AaL (β = .33, t = 2.07, p < .05). 
These findings suggest that teachers who perceive assessment as a means of fos-
tering student development are more likely to implement assessment as learning.
Furthermore, the conception of assessment as an EQ tool was a significant 
positive predictor of AoL (β = .37, t = 3.43, p < .01) and SUM (β = .29, t = 2.64, p 
< .01). The positive beta coefficients indicate that teachers who view assessment 
as a mechanism for improving the quality of school and teacher performance more 
frequently use strategies aimed at objectively determining student achievement, 
specifically strategies associated with assessment of learning and summative as-
sessment. 
Differences in assessment perception and self-assessment of assessment practices 
based on respondents’ demographic characteristics
To examine differences in assessment perceptions and practices based on 
demographics (educational qualifications, years of experience and professional 
rank), ANOVA were conducted. To ensure an adequate sample size for ANOVA, 
5 respondents with doctoral and master’s degrees in science were excluded from 
the analysis. The final sample (256 respondents) included 29% of teachers with 
bachelor’s degrees, 59% with a master’s degree and 12% with a professional mas-
ter’s degree. Before conducting the ANOVA, Levene’s test for homogeneity of var-
iances was applied to assess the assumption of equal variances between these 
three groups for each variable related to conceptions and practices of assessment. 
The resulting Levene’s statistic indicated no statistically significant differences in 
variances among the three groups for any variable.
The results of the one-way ANOVA revealed a statistically significant dif-
ference in the self-assessed frequency of applying assessment of learning based 
The Conception and Practice of Assessment from the Perspective of Primary ...

126 Sodobna pedagogika/Journal of Contemporary Educational Studies
on the respondents’ educational qualifications (F = 3.69, p < .05). To determine 
which groups differed significantly, Scheffe’s post-hoc tests were conducted. The 
results showed that respondents with a professional master’s degree in their field 
scored significantly higher on the AoL scale (M = 3.9, SD = 0.54) compared to 
those with a bachelor’s (M = 3.6, SD = 0.51) or master’s degree (M = 3.6, SD = 
0.60). No statistically significant difference was observed between respondents 
with a master’s or bachelor’s degree.
A one-way ANOVA was also conducted to examine differences in the percep-
tion of assessment and self-assessment of assessment practices based on years 
of teaching experience. Levene’s test showed no significant variance differences, 
confirming ANOVA assumptions. The results did not indicate a statistically signif-
icant influence of years of experience on any of the measured variables.
To analyse differences in assessment perception and practice based on pro-
fessional rank, outstanding teacher advisors (n = 8) were excluded from the sam-
ple due to insufficient representation in this category. The analysis was continued 
with the remaining sample of 253 respondents. Again, Levene’s test did not reveal 
significant results, justifying the application of a one-way ANOVA.
The results in Table 7 indicate a statistically significant difference in scores 
on the variables AfL (F(2, 255) = 5.60, p < .01), AaL (F(2, 255) = 4.16, p < .05) 
and FOR (F(2, 255) = 5.72, p < .01) based on the professional rank of primary 
education teachers.
F p
RI mean 3.02 > .05
FSD mean 2.48 > .05
EQ mean 1.19 > .05
E mean 1.21 > .05
AfL mean 5.60 < .01
AaL mean 4.16 < .05
AoL mean 0.12 > .05
PET p mean 0.71 > .05
FOR mean 5.72 < .01
SUM mean 0.52 > .05
Table 7. Results of one-way ANOVA testing based on respondents’ professional rank 
In all three cases, differences were observed only between mentor teachers 
and teachers without a professional rank. Teachers without a professional rank 
scored lower on the AfL (M = 3.7, SD = 0.55), AaL (M = 3.8, SD = 0.70) and FOR 
(M = 4.2, SD = 0.63) subscales compared to teacher mentors (M = 4.0, SD = 0.53; 
M = 4.1, SD = 0.58; and M = 4.5, SD = 0.58). These findings suggest that teacher 
mentors demonstrate a higher level of engagement in implementing formative 
assessment strategies compared to teachers who have not advanced in rank.
Letina, Tomaš, Škugor

127 
Discussion
This study examined Croatian primary school teachers’ perceptions and use 
of assessment methods, focusing on the relationship between their perceptions of 
assessment and their actual application of various assessment methods in prac-
tice. The findings reveal a low to moderate correlation between conceptions of as-
sessment and formative/summative practices. This suggests that the overall per-
ception of the nature and purpose of assessment is significantly associated with 
the active implementation of diverse forms of assessment in practice, which is 
consistent with Brown and Harris (2009).
The results also revealed no statistically significant differences in the fre-
quency of applying formative and summative assessment among primary school 
teachers. This can be interpreted in several ways. On the one hand, it may indi-
cate that primary school teachers are equally aware of the importance of both 
approaches to assessment. On the other hand, this finding may suggest that form-
ative assessment methods, which should ideally be implemented daily (Heitink et 
al. 2016), are not yet predominant in teaching practices compared to summative 
assessment, which is intended for occasional use. Additionally, it is possible that 
the instruments or methods used in this study were not sensitive enough to de-
tect existing differences in the frequency of formative and summative assessment 
application. Further research is needed to more precisely determine the balance 
between these two approaches in practice.
The regression analysis showed that teachers’ conceptions of assessment 
can predict their assessment practices, though this influence is limited and com-
plex. Although statistically significant, the relationship between conceptions and 
self-assessed practices is weak. This suggests that other factors may also influence 
assessment practices. Only the dimension of assessment as fostering student de-
velopment demonstrated a consistent and statistically significant positive impact 
on the application of formative assessment strategies. Furthermore, the study 
revealed that the frequency of applying assessment of learning and summative 
assessment is significantly associated with the perception of assessment as a tool 
for ensuring the quality of school and teacher performance. This indicates that 
teachers who view assessment in this function more frequently use strategies fo-
cused on objectively determining students’ achievements, which is consistent with 
previous findings (Brown 2008; Chen and Brown 2013). Other dimensions showed 
no consistent impact, suggesting the need to explore additional factors. Despite 
the significance of the overall model, none of the dimensions of the assessment 
conception significantly contributed to explaining the variance in the frequency 
of using AfL and PET strategies. This unexpected result suggests that the ap-
plication of these strategies is influenced by factors not addressed in this study. 
Possible factors include external pressures such as curricular demands, a lack of 
time for implementing AfL strategies or insufficient experience with their practi-
cal application. This finding in turn implies that, while conceptual understanding 
of assessment plays a role in shaping teachers’ assessment practices, other factors 
The Conception and Practice of Assessment from the Perspective of Primary ...

128 Sodobna pedagogika/Journal of Contemporary Educational Studies
influence the selection and application of assessment strategies in the classroom. 
The low explained variance may be attributed to contextual factors such as school 
policies, curricular frameworks, pressures of standardized testing and time con-
straints for implementing formative assessment, which often requires more time 
and resources compared to traditional assessment methods.
Future research should focus on identifying and analysing these factors to 
ensure alignment between theoretical conceptions of assessment and actual as-
sessment practices in education. This would support the development of strat-
egies that address external and contextual influences on assessment practices, 
thereby enhancing their effectiveness and integration into teaching.
The findings also suggest that professional advancement influences concep-
tions of assessment and formative assessment use more strongly than teaching 
qualifications or experience. Teacher mentors, who are at a higher level of profes-
sional development, likely have stronger pedagogical and methodological compe-
tencies. This is reflected in their approach to assessment and the improvement of 
educational processes, especially in their frequent use of formative assessment. 
This may result from attending a greater number of professional development 
programs, mentoring and assessment experience. These results highlight the im-
portance of ongoing teacher professional development. 
Finally, the findings of this study can serve as a guideline for future educa-
tional policies by emphasizing the need to enhance mentoring systems and pro-
fessional development for teachers. Additionally , they highlight the importance of 
encouraging all teachers to adopt a more active use of formative assessment and 
to focus on assessment strategies aimed at improving the learning process.
Conclusion
Since the implementation of the curricular reform and the adoption of new 
subject curricula in the Croatian education system, no significant research on as-
sessment in this context has been conducted, despite reform recommendations 
to adopt new approaches and assessment guidelines (MSE 2019a). This study of-
fers preliminary findings on assessment concepts and practices among Croatian 
primary teachers, showing a link between assessment purpose and its teaching 
application. Teachers who perceive assessment as a tool to improve learning and 
foster student development are more inclined to use formative assessment meth-
ods. The study also highlights that teacher advancement to higher professional 
ranks is a significant variable that influences their approach to assessment. This 
finding opens avenues for future research focused on a deeper analysis of how 
professional advancement impacts teachers’ assessment concepts and practices. 
Assessment should be viewed as a complex and multidimensional process. 
Given that assessment is an essential component of the curricular cycle, it is neces-
sary to continuously review and research it to achieve the optimal implementation 
of its key aspects – formative and summative assessment – in teaching practice. 
Developing a culture of assessment ensures quality education and holistic student 
Letina, Tomaš, Škugor

129 
development. This process entails a transformation of the traditional paradigm, 
which perceives assessment solely as a tool for evaluating student achievement, 
towards a paradigm that recognizes assessment as a process to provide feedback 
to all stakeholders in education. Furthermore, it serves as a foundation for in-
formed decision-making aimed at improving teaching practices and the overall 
educational system. This also involves continuous professional development for 
teachers, as well as collaborative communication among all stakeholders in the 
educational system, including school principals, teachers, students and parents. 
Only through such an approach can assessment primarily serve as a factor that 
supports learning and the holistic development of every student, thereby fulfill-
ing its most important purpose in the education system. To further advance the 
implementation of these principles and support the transformation of assessment 
practices, future research can determine the long-term effects of formative as-
sessment, students’ and parents’ conceptions of assessment, as well as qualitative 
analyses of the challenges that teachers face during its implementation.
References
Ajzen, I. (2005). Attitudes, Personality and Behaviour (2nd ed.). Milton-Keynes: Open Uni-
versity Press/McGraw-Hill.
Asamoah, D., Shahrill, M. and Abdul Latif, S.N. (2024). Teachers’ perceptions of school 
assessment climate and realities of assessment practices in two educational contexts. 
Frontiers in Education, 9, pp. 1-23. 
Berry, R. (2008). Assessment for learning. Hong Kong: University Press.
Black, P . and Wiliam, D. (1998). Inside the black box: Raising standards through classroom 
assessment. Phi Delta Kappan, 80, issue 2, pp. 139–144. 
Brown, G. T . L. (2008). Conceptions of Assessment: Understanding What Assessment 
Means to Teachers and Students. New York: Nova Science Publishers.
Brown, G. T . L. and Gao, L. (2019). Teachers‘ conceptions of assessment: A global phenom-
enon or a global localism. Frontiers in Education, 4, Article 16. 
Brown, G. T. L. and Harris, L. R. (2009). Unintended consequences of using tests to im-
prove learning: How improvementoriented resources heighten conceptions of assess-
ment as school accountability . Journal of MultiDisciplinary Evaluation, 6, pp. 68–91. 
Brown, G. T . L., and Hattie, J. (2012). The benefits of regular standardized assessment in 
childhood education: Guiding improved instruction and learning. In: S. Suggate and 
E. Reese (eds.). Contemporary debates in childhood education and development. Lon-
don and New York: Routledge, pp. 287–292. 
Brown, G.T .L, Hui, S.K.F ., Yu, F .W .M. and Kennedy, K.J. (2011). Teachers’ conceptions of 
assessment in Chinese contexts: A tripartite model of accountability, improvement, 
and irrelevance. International Journal of Educational Research, 50, issue 5–6, pp. 
307-320. 
Brown, G. T . L., Kennedy, K. J., Fok, P . K., Chan, J. K. S. and Yu, W . M. (2009). Assessment 
for student improvement: Understanding Hong Kong teachers‘ conceptions and prac-
tices of assessment. Assessment in Education: Principles, Policy and Practice, 16, 
issue 3, pp. 347–363.
The Conception and Practice of Assessment from the Perspective of Primary ...

130 Sodobna pedagogika/Journal of Contemporary Educational Studies
Chen, J. and Brown, G. T. L. (2013). High-stakes examination preparation that controls 
teaching: Chinese prospective teachers’ conceptions of excellent teaching and assess-
ment. Journal of Education for Teaching, 39, issue 5, pp. 541–556. 
Christoforidou, M., Kyriakides, L., Antoniou, P . and Creemers, B. P . M. (2014). Searching 
for stages of teacher’s skills in assessment. Studies in Educational Evaluation, 40, 
pp. 1–11. 
Cindrić, M., Miljković, D. and Strugar, V . (2010). Didaktika i kurikulum. Zagreb: IEP d.o.o.
Fives, H. and Buehl, M. M. (2012). Spring cleaning for the “messy” construct of teachers’ 
beliefs: What are they? Which have been examined? What can they tell us? In: K. 
R. Harris, S. Graham, T. Urdan, S. Graham, J. M. Royer and M. Zeidner (eds.). APA 
educational psychology handbook, Vol 2: Individual differences and cultural and con-
textual factors. American Psychological Association, pp. 471–499. 
Heitink, M.,C., Van der Kleij, F .,M., Veldkamp, B.,P ., Schildkamp, K. and Kippers, W .,B. 
(2016). A systematic review of prerequisites for implementing assessment for learn-
ing in classroom practice, Educational Research Review, 17, pp. 50-62. 
Kožuh, A. (2019). Perspectives on assessment at school. Sodobna pedagogika, 70, issue 2, 
pp. 160–173.
Lin, Y .C., Lee, L. and Tsai, C.C. (2014). Students’ and teachers’ conceptions of learning and 
assessment in science education: A case study in Taiwan. International Journal of 
Science Education, 36, issue 14, pp. 2331–2351. 
Looney, A., Cumming, J., van Der Kleij, F. and Harris, K. (2017). Reconceptualising the 
role of teachers as assessors: teacher assessment identity. Assessment in Education: 
Principles, Policy and Practice, 25, issue 5, pp. 442–467. 
Matijević, M. and Radovanović D. (2011). Nastava usmjerena na učenika. Zagreb: Školske 
novine.
Mekonnen, M. A. and Melesse, S. G. (2022). Primary school teachers’ conceptions and prac-
tices of assessment and feedback in mathematics classrooms. Cogent Education, 9, 
issue 1, Article 2090185. 
Ministry of Science and Education (MSE) (2017). Nacionalni kurikulum za osnovnoškol-
ski odgoj i obrazovanje. Retrieved from: https://mzom.gov.hr/istaknute-teme/
odgoj-i-obrazovanje/nacionalni-kurikulum/nacionalni-kurikulumi/531 (accessed on 
10 January 2025).
Ministry of Science and Education (MSE) (2019a). Smjernice za vrednovanje procesa i ost-
varenosti odgojno-obrazovnih ishoda u osnovnoškolskome i srednjoškolskome odgoju 
i obrazovanju. Retrieved from: https://mzom.gov.hr/vijesti/smjernice-za-vrednovan-
je-procesa-i-ostvarenosti-odgojno-obrazovnih-ishoda-u-osnovnoskolskome-i-sredn-
joskolskome-odgoju-i-obrazovanju/3250 (accessed on 10 January 2025).
Ministry of Science and Education (2019b). Pravilnik o izmjenama i dopuni Pravilnika o 
načinima, postupcima i elementima vrednovanja učenika u osnovnim i srednjim škola-
ma. Retrieved from: https://narodne-novine.nn.hr/clanci/sluzbeni/2019_09_82_1709.
html (accessed on 10 January 2025).
Pekrun, R., Lichtenfeld, S., Marsh, H. W ., Murayama, K. and Goetz, T . (2017). Achievement 
emotions and academic performance: Longitudinal models of reciprocal effects. Child 
Development, 88, issue 5, pp. 1653–1670. 
Rajić, V . (2017). Pristupi vrednovanju u obrazovanju. In: M. Matijević (ed.) School for the 
Net generation. Zagreb: Učiteljski fakultet Sveučilišta u Zagrebu, pp. 256-257.
Stobart, G. (2008). Testing times: The uses and abuses of assessment. Routledge. 
Torrance, H. and Pryor, J. (2021). Assessment conceptions and practices: Perspectives of 
primary school teachers and students. Frontiers in Education, 6, Article 631185. 
Letina, Tomaš, Škugor

131 
Toth, E. and Csapo, B. (2022). Teachers’ beliefs about assessment and accountability . Edu-
cational Assessment Evaluation and Accountability, 34, pp. 459–481 
Xu, Y. and Brown, G. T . L. (2016). Teacher assessment literacy in practice: A reconceptual-
ization. Teaching and Teacher Education, 58, pp. 149–162. 
Alena LETINA (Univerza v Zagrebu, Pedagoška fakulteta, Hrvaška)
Suzana TOMAŠ (Univerza v Splitu, Faculty of Humanities and Social Sciences, Hrvaška)
Alma ŠKUGOR (Univerza v Osjeku, Pedagoška fakulteta, Hrvaška)
ZASNOVA IN PRAKTIČNO IZVAJANJE PREVERJANJA IN OCENJEVANJA ZNANJA Z 
VIDIKA OSNOVNOŠOLSKIH UČITELJEV NA HRVAŠKEM 
 
Povzetek: V članku predstavljamo rezultate raziskave, katere namen je bil proučiti, kako osnovnošol-
ski učitelji na Hrvaškem pojmujejo preverjanje in ocenjevanje znanja in kakšna je korelacija med nji-
hovimi pojmovanji ter praktičnim izvajanjem preverjanja in ocenjevanja znanja v pedagoški praksi. 
Analiza pojmovanj učiteljev vključuje tudi njihova stališča o zanesljivosti in pomembnosti preverjanja 
in ocenjevanja znanja, o vlogi le-tega pri spodbujanju razvoja učencev, dojemanju preverjanja in ocen-
jevanja znanja kot orodja za zagotavljanje kakovosti šole in uspešnosti učiteljev ter vlogi, ki jo imajo 
ti procesi pri pripravi učencev na izpite. Raziskava ugotavlja tudi, kakšne so prakse preverjanja in 
ocenjavanja znanja učiteljev, pri čemer analizira, ali se učitelji osredotočajo predvsem na formativno 
spremljanje, ki je namenjeno  spodbujanju razvoja oz. napredka učencev, ali na sumativno ocenjevanje 
znanja, ki je namenjeno zlasti pripravi učencev na uspešno reševanje testnih nalog. Raziskava je bila 
izvedena na vzorcu 261 učiteljev v hrvaških osnovnih šolah. Ugotovitve kažejo statistično pomemb-
no korelacijo med pojmovanji učiteljev o preverjanju in ocenjevanju znanja ter izvajanjem procesov 
preverjanja in ocenjevanja znanja v njihovi neposredni praksi. Učitelji, ki preverjanje in ocenjevanje 
znanja razumejo kot pomembno za osebni razvoj učencev, so tudi bolj naklonjeni uporabi formativnega 
spremljanja v svojem poučevanju. V članku na podlagi rezultatov raziskave opravimo tudi razmislek 
o tem, kakšne so njihove implikacije za prakso in možnosti za izboljšanje percepcije in izvajanja pre-
verjanja in ocenjevanja znanja v osnovnošolskem izobraževanju, pri čemer poudarimo zlasti potencial 
preverjanja in ocenjevanja znanja za spodbujanje celostnega razvoja učencev. 
 
Ključne besede: formativno spremljanje, samoocenjevanje učiteljev, spodbujanje razvoja učencev, 
sumativno ocenjevanje znanja, cilji ocenjevanja znanja
Elektronski naslov: alena.letina@ufzg.hr
The Conception and Practice of Assessment from the Perspective of Primary ...