South African Journal of Education, Volume 43, Number 3, August 2023 1

Art. #2100, 11 pages, https://doi.org/10.15700/saje.v43n3a2100

Effects and challenges to implement differentiated mathematics teaching among fourth

graders in Montenegro

Veselin Mićanović and Dijana Vučković

Teacher Training Department, Faculty of Philosophy, University of Montenegro, Podgorica, Montenegro
veselinm@ucg.ac.me
Biljana Maslovarić
Preschool Teacher Training Department, Faculty of Philosophy, University of Montenegro, Podgorica, Montenegro
Nada Šakotić and Tatjana Novović
Department of Pedagogy, Faculty of Philosophy, University of Montenegro, Podgorica, Montenegro

In this quasi-experimental pretest-posttest study we examined the effects of differentiated instruction (DI) in within-class
ability groupings of 246 Montenegrin fourth-graders and their ability to solve algebraic equations. We assessed 2 parallel
student groups at equal achievement levels to compare DI, in which teaching and work modes were adapted to students’
grouping according to previous achievement and pretest scores, and traditional whole-class instruction. Pretest-posttest
evaluations were administered to both groups, and observation indicators were evaluated to assess the level and type of
student activities, engagement, and individualisation. Students in the homogeneous DI experimental groups with tailored
instructions were significantly more successful at solving algebraic tasks than their peers in the traditional whole-class
instruction control group. DI improved students’ results, but teachers required specific training and significantly more
preparation time.

Keywords: algebra; differentiated instruction; mathematical skills; quasi-experimental pretest-posttest; within-class ability
grouping

Introduction
In 2017, the Montenegrin elementary school curricula were redesigned to envelop specific learning outcomes
and knowledge standards as a basis to adjust instruction to student capabilities (Pavićević, Vučeljić, Lalić,
Pavićević & Kostić, 2017). The curricula reform emphasised improving mathematical literacy, a global trend to
equip children with life-long skills in the 21st century (Mejer, Turchetti & Gere, 2011). Montenegrin students
continually scored poorly in mathematical literacy testing in the Program for International Student Assessment
(PISA) (Organisation for Economic Co-operation and Development [OECD], 2018). Educators posit that
mathematical instructions were insufficiently adapted to meet student needs in early childhood (Prica, Čolić &
Baronijan, 2014). Montenegrin teachers were also insufficiently adapted to the new curriculum and often
inflexible in implementing diverse teaching methods (Prast, Van de Weijer-Bergsma, Kroesbergen & Van Luit,
2018; Prica et al., 2014), which is important since teaching methods organise instructions and help to implement
a curriculum (Abah, 2020).
Experts proposed differentiated instruction (DI) as an appropriate remedial classroom instructional
strategy. DI facilitates teachers’ adaption of learning content, process, and products (curricular elements) to fit
students’ interests, readiness, and learning profiles, i.e., their characteristics (Prast et al., 2018; Prica et al., 2014;
Tomlinson, 2014). It is a student-centred approach that positions the teacher as a guide who facilitates students’
participation in their learning (Hackenberg, Creager & Eker, 2021). Conversely, non-differentiated or
one-size-fits-all teaching is more general (Hertberg-Davis, 2009). This traditional prescription-oriented
approach to teaching (George, 2005) is difficult for students and consequently they progress slowly through the
curriculum (Gamble, 2011). Tradition is not intransigence but rather adherence to long-standing practices that
are familiar and comfortable (Abah, 2020). The teacher and the curriculum are the focus in traditional
teacher-directed teaching (Harris & Johnson, n.d.). In traditional mathematics instruction, the instructor teaches
the same content to all students who are simultaneously engaged in the same activity (Kesteloot, 2011). This
method is often criticised for showing little regard for students’ differences (Sammons, 2010).
DI implementation in mathematics involves adapting the curriculum, instruction, and assessment to the
needs and abilities of diverse student groups (Livers, Paxton, O’Grady & Tontillo, 2018; McKeen, 2019).
However, motivating teachers to implement DI strategies in elementary mathematics may prove difficult
because of their poor understanding of the underlying concepts or disbelief in its effectiveness (Livers et al.,
2018; Van Geel, Keuning, Frèrejean, Dolmans, Van Merriënboer & Visscher, 2019). For instance, Gaitas and
Martins’ (2017) evaluation of 273 Portuguese teachers found that they experienced difficulty implementing DI
in regular classes because of student characteristics. Instead, it was easier for teachers to not to apply
differentiation. Some authors admitted moderate success using DI in mathematics instructions. Maxey’s (2013)
causal-comparative design of second-graders at a United States of America (USA) military base found that high
groups improved compared to average and low groups. Correspondingly, Prast et al.’s (2018) survey found
2 Mićanović, Vučković, Maslovarić, Šakotić, Novović

positive effects in student achievement growth with DI in elementary Mathematics and ability grouping
DI mathematics in a large-scale teacher Ability grouping of elementary students in the
professional development program. Similarly, learning of mathematics is a polemic issue among
Livers et al.’s (2018) phenomenologically themed educators (Anthony & Hunter, 2017). Nevertheless,
collaboration illustrated how teacher candidates most would concur that students are unique and
gained confidence using DI to teach elementary learn at different rates. Flexible grouping in DI of
mathematics through curriculum compacting. The content areas such as algebra allows teachers to
latter allows teachers to bypass already mastered group students based on their strengths and abilities
content, provide advanced students with more (McKeen, 2019). Some educators frown upon the
challenging content, and focus on students with less practice as they contend that it promotes lower self-
mastery. esteem among students of average ability and
In this paper, through the results of a reinforces social inequality (Boaler, Wiliam &
pedagogical experiment involving two groups of Brown, 2000). However, Boaler (2013) observed
Montenegrin fourth-graders, we assess the relative that students scored higher in international tests in
effectiveness of differentiated and non- countries that were more flexible about grouping,
differentiated mathematics instruction to enhance unlike the USA and European countries where
students’ ability in algebra. By the fourth grade, students structured into rigidly fixed groups scored
mathematics becomes more abstract and complex significantly lower in similar exams. Consequently,
so students experience difficulties when algebra is struggling or low achievers often developed a
introduced (Bender, 2013). We also identify negative mindset towards learning and their
teaching challenges in implementing DI and capabilities (Boaler, 2013; Boaler et al., 2000), thus
propose recommendations. widening the achievement gap between low and
high performers (Papanastasiou, 2002).
Literature Review Ability grouping coupled with curricula
Elementary Mathematics and DI revision or differentiation may result in substantial
Students’ success in learning mathematics depends achievement gains for high achievers (Livers et al.,
on the way in which mathematics is taught (Boaler, 2018; Tieso, 2003). In Finland, differentiation is
2002; Papanastasiou, 2002). Teaching mathematics paired with early identification and flexible support
at elementary school is important as basic arrangements. Using a multimethod approach that
knowledge is acquired and habits and learning surveyed special education and mathematic
styles are developed (Doubet & Hockett, 2017; teachers in 55 schools, Ekstam, Linnanmäki and
Prica et al., 2014). DI gained importance when its Aunio (2015:75) assessed the benefits of the “pull
benefits and successes for teachers and students out of the classroom” method for low-performing
within the classroom were highlighted students to a three-tier incremental support model.
(Papanastasiou, 2002; Prast et al., 2018; Sousa & This involved general support through a whole
Tomlinson, 2011). The approach presents a class approach at Tier I; intensified support for a
solution for teaching mathematics to students at limited time with periodic evaluation at Tier 2; and
different levels of mastery (Livers et al., 2018), as special support and an individual educational plan
teaching and learning can be tailored to meet the that required higher-level intervention involving
students’ needs, abilities, and learning profiles teacher, parent, student, and principal at Tier 3.
(Prast et al., 2018). However, Tomlinson and Schoolwide cluster groups were also perceived as a
Imbeau (2010:15) observed that it was first positive influence on student performance
necessary to differentiate the four main curriculum (Matthews, Ritchotte & McBee, 2013).
elements, namely “content, process, product, and
affect.” DI in elementary Mathematics teaching and learning
Mathematics teaching should be organised DI adapts teaching to students’ abilities,
based on differentiation principles if teachers want knowledge, interests, and needs. The overall aim is
students to achieve better results (Light & Pierson, to ensure that teachers focus on processes and
2014; Maxey, 2013; Prast et al., 2018). Algebra and procedures that ensure effective learning
fourth-graders are a significant coupling as (Kesteloot, 2011). Other factors can affect
problem-solving becomes more complex and elementary students’ mathematical learning skills,
students mature (Bender, 2013). Therefore, such as the presence of cognitive ability like
applying the same approach to all students at this memory, reasoning, processing speed, and oral
stage may be difficult due to the unevenness of expression (Fuchs, Fuchs, Compton, Powell,
their progression (Tomlinson, 2014). Thus, this Seethaler, Capizzi, Schatschneider & Fletcher,
awareness of student needs and subject knowledge 2006). In a qualitative study of Malaysian
are key factors to influence successful DI elementary mathematics students it was found that
(Papanastasiou, 2002). these factors impeded their problem-solving skills
 South African Journal of Education, Volume 43, Number 3, August 2023 3

and consequently students showed little interest in (Hackenberg et al., 2021). The implementation of
solving particularly challenging assignments DI in the teaching of algebra can positively impact
(Tambychika & Meerah, 2010). As students student learning (Maxey, 2013).
mature, the mathematics curriculum becomes more
complex (Bender, 2013), adding to the disinterest. Methodology
DI fosters a successful understanding of In a quasi-experimental pretest-posttest design with
concepts such as algebra, as it considers different six groups we focused specifically on algebraic
levels of students’ knowledge (Cowan & Powell, content and equation solving (general form
2014). Classroom management is key, as equations and equations with textual tasks), which
Hackenberg et al. (2021) found that teaching in are part of the mainstream syllabus. DI and
whole-class discussions, attending to small groups, traditional teaching approaches were tested using
and responding during group work impeded DI. non-differentiated teaching methods as the
Teachers’ struggles to meet all students’ needs independent variable and student achievements in
results in teaching to the middle, thus solving the mathematical assignments as the
disenfranchising both low and high achievers dependent variable. Demographic variables were
(Abah, 2020). However, modelling creative not included in this research design, although it is
mathematical in-class assignments and accounting quite possible that some of them (e.g., gender) may
for task complexity with students’ individualised have an influence on the students’ achievement.
abilities develop their mathematical competencies However, the teaching approach (DI vs. traditional
(Ashley, 2016). Modelling enables mathematically teaching) was the focus of this study, and since it is
proficient students to apply their knowledge to a very complex phenomenon, it was decided not to
daily problem-solving, such as writing a include demographic variables but to have full
multiplication equation or comparing candy prices. focus on the teaching approach. The within-class
Differentiation in mathematics teaching leads to DI groups comprised students whose previous
higher student motivation, greater academic achievements were similar (experimental group),
achievements (Bal, 2016; Prast et al., 2018), and whereas students in the control group were
greater cooperation between students of similar heterogeneous within-class groups (with different
abilities (Hertberg-Davis, 2009). More importantly, previous achievements) who all received the same
DI mathematics instruction improves learning instruction. Teachers’ instruction in the DI group
outcomes (Tieso, 2003). was adapted to students’ different learning
DI requires that teaching methods, approaches and needs. Both groups completed the
instructions, and materials should be adapted for same assignments.
small group learning (Lou, Abrami, Spence,
Poulsen, Chambers & D’Apollonia, 1996). Compliance with Ethical Standards
Mathematics teachers must be able to differentiate Participant volunteers were informed about the
between content and learning methods to respond research purpose. Parents gave consent during
to distinct abilities, interests, and learning styles of parent/teacher meetings and the Montenegrin
students in the same class (Ashley, 2016; Bureau of Education Services approved the study
Chamberlin & Powers, 2010). Process and grade (Confirmation number [no.] 01-646).
levels are, therefore, crucial. Normally, DI research
on numerical operations in elementary mathematics Sample
largely focuses on older grades (Bal, 2016), Two hundred and forty-six Montenegrin fourth-
therefore, this research is significant as we sought graders were recruited from four urban elementary
to address this gap by investigating fourth-graders schools with a combined population of 3,750
and their learning of algebra. students. Using random sampling, we created two
control and two experimental groups (three control
Research Hypothesis groups [CGs] and three experimental groups [EGs]:
The use of DI to teach algebraic content contributes n × 4 = 31; 1 CG and 1 EG: n × 2 = 30). Eight
to better overall task achievement than whole-class teachers participated.
instruction. Control and experimental group
homogenisation were determined based on student
Theoretical Framework achievements in a preliminary test. Montenegro
DI is rooted in the constructivist theory. The schools require all students to undergo a
framework engages mathematics learners in psychological test on enrolment, after which they
activities that match their strengths and preferences are homogenised into classes. Before the study,
and ensures that educators focus on effective participants completed a pretest comprising five
teaching that will benefit them (Tomlinson & objective tasks. Results were compared and
McTighe, 2006). DI positions the teacher as a students were assigned to two parallel classes in
facilitator of students’ participation in their learning each school. Students’ achievements in the initial
by redesigning and implementing the content testing were similar in each parallel class. To
4 Mićanović, Vučković, Maslovarić, Šakotić, Novović

calculate the power of the two sample tests we used method during initial teacher education. The
the function pwr.t.test() from the PWR package in teaching groups planned their classes as usual,
the R software. For more details we refer readers to focusing on the average group.
Champely (2020). One of the input values is the so- Differentiation was based on students’ current
called Cohen’s d. According to Cohen (1988) it is achievement levels as determined from the pretest
the most common way to measure effect size. scores and previous mathematics grades (Roy,
Therefore, for the n = 246, d = 0.55 and Guay & Valois, 2013), which provided a basis for
significance level of 5% it was found that the “cognitive or readiness-based differentiation”
achieved power was 0.9. (Prast et al., 2018:22). These two variables were
used for agglomerative hierarchical clustering,
Procedure where observations were initially grouped into one
Two researchers trained the four EG teachers for 4 cluster, which was then successively partitioned.
days in the planning, preparation, and teaching of Ward’s clustering with squared Euclidian distance
algebraic content for fourth-graders based on the was then conducted to create compact, even
DI model. To maintain consistency in all teachers’ clusters (Murtagh & Legendre, 2014). The elbow
approaches, the EG teachers and researchers step was located at the 120th observation; hence,
planned the task differentiation process together. resulting in three final clusters. After completing
Teachers in the CG received no additional or the analysis to determine cluster membership, we
special training because we chose to compare the applied the k-means technique (k = 3) to partition
effects of traditional and DI instruction, and CG each observation into the cluster with the nearest
teachers were asked to work in the way that they mean. This technique was applied since the
normally did. All were familiar with the application of the Shapiro Wilk’s test indicates that
experimental programme and possessed basic samples are normally distributed. Table 1 shows
knowledge of differentiation, which is a the descriptive statistics after analysis including p
compulsory didactic principle and special teaching values obtained via Shapiro-Wilk’s test.

Table 1 Number of units per experimental group cluster

Correlation between
p value Mean value pretest and previous
(Shapiro Mean value (previous math grades 95% CI for the mean
Cluster N Wilk’s test) (test) grades) (p < .001) value
1 43 0.7124 88.95 4.86 0.98 86.08–90.05
4.29–4.95
2 43 0.5785 77.05 3.87 0.96 75.21–78.99
3.28–3.97
3 37 0.4725 55.42 3.05 0.92 51.11–59.54
2.98–3.18

Differentiation focused on three achievement Evaluation and Data Analysis

levels: high performers (n = 43), moderately high We used an assessment scale to measure the impact
performers (n = 43), and average performers (n = of the differentiated approach to solve the
37). EGs and CGs were organised into four classes mathematical tasks, using knowledge testing (a
taught by students’ regular mathematics teachers. series of objective tasks) to compare the effects.
Once parallel classes were selected, DI was Ultimately, we did a posttest to assess students’
provided to the EG, and non-DI was given to the ability to solve the equations. All students solved
CG during 12 sessions. In the DI classes, the lesson identical tasks without extra aid. The number of
was not adapted regarding volume, depth, or successfully solved tasks formed the evaluation
planned concepts/content. The teaching method criterion for both pretest and posttest. Students’
was, however, adjusted at the level of instruction achievements were assessed using five objective
and formation of small workgroups. Workgroups tasks. Both teachers and researchers maintained
were homogeneous in the EGs and heterogeneous that this aptly measured students’ abilities to solve
in the CGs. Students in the EGs were given special general form equations and equations with textual
instructions to solve the assignments, whereas the tasks. Tests were scored on a range of 0
CGs received identical explanations. The (unsuccessfully solved tasks) to 5 (successfully
mathematical assignments with complex forms solved tasks).
included several different operations. All students We included two additional experimental
performed the same tasks from the common techniques: observation and evaluation of direct
curriculum, albeit using different work methods. teaching practice from which the effects of the
 South African Journal of Education, Volume 43, Number 3, August 2023 5

experimental factors were determined. Indicators 1 (exceptionally high) to 5 (exceptionally low).

were developed to assess the domains of Data analysis measured objective statistical
individualisation, student activity, and interest indicators.
during learning. Two observers independently
monitored student activities in the parallel classes Results
and completed observation protocols based on Table 2 is a summary of the pretest results. The
specific indicators. The scores for each indicator EGs and CGs were balanced. The average
were averaged, scored protocols were compared, achievement in the parallel groups was similar, as
and we agreed on a position regarding the well as the means for both groups. However, within
applicability of the learning strategy for each class. both groups, there were at least three distinct
The protocols provided five assessment options for student levels: excellent grades, very good grades,
each indicator; the first two categories ranged from and sufficient grades.

Table 2 Pretest results

No. of No. of students with successfully solved tasks
Group students 5 4 3 2 1 M SD
Experimental 123 N 43 43 25 12 0 3.95 0.94
% 34.95 34.95 20.32 9.75 0.00
Control 123 N 44 42 26 11 0 3.96 0.95
% 35.77 34.14 21.13 8.94 0.00
Total 246 N 87 85 51 23 0 3.95 0.95
% 35.36 34.55 20.73 9.34 0.00

In order to be sure that parametric statistics is got p > .05 meaning that all samples are within
an appropriate choice for this analysis we normal distribution.
employed Shapiro-Wilk’s test. For all groups we

Table 3 Comparing the results of the initial test between control and experimental groups
t test
Group Control Experimental Levene test p value (p value)
i. 61.24 62.03 0.432 -0.342
(N = 30) (N = 30) (p = 0.221)
ii. 58.96 57.07 0.336 0.954
(N = 31) (N = 31) (p = 0.339)
iii. 60.08 60.78 0.479 -0.188
(N = 31) (N = 31) (p = 0.785)
iv. 62.11 61.82 0.663 0.449
(N = 31) (N = 31) (p = 0.663)

Parametric statistics can now follow. In Table In Table 4, the EG demonstrates a higher
3, Levene’s test indicates that there was no proportion of successfully solved tasks than for the
statistical difference in variability between any CG. In all four experimental subgroups, only
pairs of CG and EG (p > .05). The t-tests indicated 2.43% of those students completed more than three
that there was no statistical difference in pretest tasks compared to 24.37% of the CG. A line chart
scores either within or between the CGs and EGs (p of the results from Table 4 shows the EG and CG
> .05). differences, with the ascent line for the EG being
greater than that of the CG (cf. Figure 1).

Table 4 Posttest results for the experimental and control groups

No. of students who successfully solved tasks
No. of
Group students 5 4 3 2 1 M SD
Experimental 123 N 79 35 6 3 0 4.54 0.95
% 64.22 28.45 4.87 2.43 0.00
Control 123 N 46 47 15 9 6 3.75 0.93
% 37.39 38.21 12.19 7.31 4.87
 6 Mićanović, Vučković, Maslovarić, Šakotić, Novović

Figure 1 Student trends in the parallel groups

Welch’s adjusted test was conducted because presented in Table 5 confirm significantly higher
Levene’s test indicated heterogeneity in-group scores for the EG across all four schools (p < .05).
variance. The degrees of freedom were calculated We also calculated size effect, which was large in
via the Welch–Satterthwaite equation. The results all four schools.

Table 5 Comparison of the final test results between control and experimental groups
Control Experimental Levene’s test t-test for equality of M
School N M SD N M SD F p df t p Cohen’s d Power
i. 30 66.98 4.79 30 71.98 5.22 6.21 0.028 48.06 −3.214 .025 0.99 0.964
ii. 31 61.42 4.94 31 68.78 5.18 5.98 0.039 42.29 −4.251 .038 1.45 0.999
iii. 31 63.25 5.04 31 69.99 5.47 4.98 0.045 56.12 −5.324 .042 1.28 0.998
iv. 31 66.18 4.87 31 70.15 5.15 6.31 0.026 58.69 −3.247 .034 0.80 0.872

The pretest and posttest results are compered students in the EG decreased, whereas the CG
in Table 6. Mean scores among the CG were increased with medium size effect.
slightly decreased. Dispersion of results among the

Table 6 Comparative results for the initial and final test results for the parallel groups
Testing Levene’s test t-test
Group *I-F n M SD F **Sig df t p Cohen’s d Power
Experimental I 123 3.95 0.94 5.924 0.027 228.5 −5.52 0.01 0.70 0.999
F 123 4.54 0.72
Control I 123 3.96 0.97 1.742 0.479 244 0.0756 0.94
F 23 0.95 0.1
Note. *I = initial; F = final. **Sig = p value of Levene’s test.

With the comparative effectiveness of the significantly more intense student activity and
differentiated approach for the teaching of interest in the EG classes than in the CGs, and none
mathematical content, the control, and of the students demonstrated low or extremely low
experimental class observations we also sought to activity. Overall, the percentage of students in the
evaluate the activity and engagement levels in the EG exhibiting average or lower activity was
teaching process regarding knowledge processing. approximately 8%, compared to 33% of the CG;
As summarised in Table 7, we documented 18.7% demonstrated low or extremely low activity.
 South African Journal of Education, Volume 43, Number 3, August 2023 7

Table 7 Students’ in-class activity and engagement levels

Extremely
Group high High Average Low Extremely low Total *p
Experimental N 95 27 1 0 0 123 p < 0.001
% 77.23 21.95 0.81 0.00 0.00 100
Control N 37 45 18 17 6 123
% 30.08 36.58 14.63 13.82 4.87 100
Total N 132 72 19 17 6 246
% 53.65 29.26 7.72 6.91 2.43 100
Note. *Based on the chi-square test results.

A chi-square test with Yates’ correction EG, the use of teaching materials, and the
resulted in a p value of < 0.001. Cramér’s V value individualisation development processes. We
was 0.72, indicating a strong and significant assessed each class activity on the scale from
relationship between groups and student activity. extremely low to extremely high presence of three
Using R function pwr.chisq.test() we were able to selected dimensions (activities, teaching materials
calculate the power of the chi square test. It is not used, and individualisation). The final assessment
difficult to calculate the sample size as w = 0.31. (Table 8) was the result of our discussion and
Therefore, for a sample size = 246 and for a agreement on each dimension and its observed
significance level of 5%, the achieved power was presence. The development of individualisation in
0.89. the EG was significantly greater than in the CG; all
Additionally, the observations of the activities of the students in the former group showed high or
of the teachers and students in the CG and EG, the extremely high individualisation development,
use of teaching materials, and the individualisation whereas 57.7% of the CG exhibited average or
development processes were documented (Table 8). lower development in this area, including 23% who
The results of the qualitative assessment were demonstrated low or extremely low
obtained by two researchers who applied systemic individualisation. Chi-square testing resulted in a p
observation of the classes. Each of us value near 0. Cramér’s V value was 0.68. There
independently completed the observation protocol appeared to be a significant relationship between
that consisted of several items describing the individualisation development and the study group.
activities of the teachers and students in the CG and

Table 8 Development of individualisation in the teaching process

Extremely
Group high High Average Low Extremely low Total p
Experimental N 99 24 0 0 0 123 < 0.001
% 80.48 19.51 0.00 0.00 0.00 100
Control N 25 27 39 19 13 123
% 20.32 21.95 31.70 15.44 10.56 100
Total N 124 51 39 19 13 246
% 50.40 20.73 15.85 7.72 5.28 100

Table 9 presents a summary of the levels at CG. The spectrum between low and high was
which students adapted six strategies based on coded in five intervals. Application of learning
teachers’ and researchers’ evaluations. The strategies measured students’ use of six strategies
assessments of learning strategies were the result of (Table 8) and scored in a reverse manner, i.e., on a
systemic observations by two researchers and the range between 1 (not used) and 5 (used to a great
teachers’ self-evaluation. The final assessment extent). Observations of established learning
(Table 9) was conducted after discussions and strategies in the CG demonstrated that whole-class
agreement among the observers and teachers. The instruction and solving tasks with the same
EG demonstrated high use of learning adapted to requirements were used without any differentiated
student ability, assignments with adjusted support. These strategies were minimally used in
requirements, tasks with work assistance, and the EG.
individual work, none of which were used in the
8 Mićanović, Vučković, Maslovarić, Šakotić, Novović

Table 9 Application of learning strategies in parallel groups

Learning strategy Experimental group Control group
Teaching adapted to student abilities 4 1
Tasks with adapted assignments 4 1
Tasks with work assistance 5 1
Whole class instruction 2 5
Solving tasks with homogenous requirements without work assistance 1 5
Individual work 4 1
Note. 1, 0%–20% use; 2, 21%–40% use; 3, 41%–60% use; 4, 61%–80% use; 5, over 80% use.

Discussion instruction levels for each of the 12 sessions (N =

The pretest and posttest results show significant 36 or more tasks). Conversely, teachers in the CG
differences between the EG and CG. The overall only had to prepare one common instruction for
ratio of successfully to unsuccessfully solved tasks each task. Since tailored explanations had to be
demonstrated that the EG achieved a significantly provided to average and struggling learner groups,
higher mean score than the CG, and the EG made EG teachers received additional training. We found
significant progress from the initial to the final their competencies to deliver DI and their
testing. By contrast, the CG appeared to stagnate. knowledge of the learning content were weak (e.g.,
Therefore, the research hypothesis was accepted. the gradualness of introducing and understanding
Additionally, based on the observed levels of equations). In addition, students had to be inspired
individualisation, interest, and activities when to participate (Van Geel et al., 2019).
solving problems in the EG and CG, we agreed that Although teachers in the EG were greatly
tasks given in the EG stimulated students’ interest interested in DI and encouraged by good results, it
much more than in the CG. is uncertain whether they will continue the practice.
DI in mathematics teaching increased student Several studies found that differentiation was
activity, interest and achievements, indicating that difficult to achieve (Gaitas & Martins, 2017) and
differentiation successfully catered to students’ teachers were often unsuccessful in its
learning needs and opportunities to progress (Bal, implementation (Van Tassel-Baska & Stambaugh,
2016; Gamble, 2011; Tomlinson, 2014). This 2005), possibly due to the extra work involved (De
supports findings that DI can improve mathematics Graaf, Westbroek & Janssen, 2019). This is why
teaching and learning in younger students (Bal, most teachers preferred to plan for the average
2016; Prast et al., 2018; Wilson, 2014). The results student (Hertberg-Davis, 2009). Although De Graaf
also support findings that a differentiated approach et al.’s (2019) investigation of the practical
to develop students’ ability to successfully solve application of DI confirmed our attitudinal
algebraic tasks is methodologically justified for findings, DI was positively assessed by both
fostering achievement (Bal, 2016). Also, using a teachers and students.
differentiated approach to teach mathematics to The degree to which teachers understand their
fourth-graders led to enhanced learning outcomes students is important in DI (Tomlinson, 2008).
compared to non-differentiated teaching. Accordingly, the findings had implications for the
Observations of class activities in the parallel ability grouping debate (Boaler & Wiliam, 2001).
groups identified six recognisable learning Although scholars continue to advocate for its
strategies for differentiated versus traditional abandonment (Boaler, 2013; Francis, Archer,
instruction. The EG had a high degree of Hodgen, Pepper, Taylor & Travers, 2017; Wiliam
applicability for assignments with work support, & Bartholomew, 2004), DI advocates found that
learning adapted to student ability, assignments teachers were satisfied with the effectiveness of
with adjusted requirements, and individual work, student support and success when they
whereas the CG primarily engaged in a whole-class differentiated mathematics teaching by ability
approach to teaching and solving tasks with (Ekstam et al., 2015). Steenbergen-Hu, Makel and
homogenous requirements and no working support. Olszewski-Kubilius (2016) found that although
These findings further support DI benefits for the between-class homogenous groupings were
successful resolution and enhancement of students’ generally disadvantageous for students, the effects
interest in mathematics tasks (Ashley, 2016; of within-class groupings were more positive. The
Wilson, 2014). findings indicate that besides within-class ability
Although we found significant effects of DI groupings, the teaching approach may significantly
on student achievement, we also observed that to influence students’ learning. If students are
achieve this success teachers had to have enhanced grouped by ability with no further differentiation,
training and spend more preparation time on this may harm their self-esteem, particularly based
algebraic content. Designing differentiation for the on the reduced achievement levels of lower ability
experimental class involved a significant increase students (Boaler, 2013; Wiliam & Bartholomew,
in time for teaching preparation and planning, as 2004). Within-class groupings and implementation
the EG teachers had to plan for three different of DI among the EGs suggest potential benefits of
 South African Journal of Education, Volume 43, Number 3, August 2023 9

more flexible and mixed approaches to ability- through curriculum design. It is especially crucial
grouping (Steenbergen-Hu et al., 2016; Tieso, for systems challenged by poor results in PISA and
2003). One final strength of this study was its similar tests.
ability to observe the teaching process as well as
Acknowledgements
the quantitative indicators of student performance.
We express our appreciation to all research
This study was significant because the use of
participants. We extend special gratitude to the
DI to teach algebraic content to lower elementary
students who actively participated and to our
school graders has not previously been explored.
teacher colleagues, whose ideas significantly
There are limitations, such as the thematic scope
improved the research.
and study duration. Therefore, future DI research
should focus on other mathematics topics over Authors’ Contributions
longer periods. Additionally, we did not analyse This article was written by the first (VM) and the
factors such as students’ gender or socioeconomic second author (DV) with support from the third
status, which may impact academic achievement, (BM), the fourth (NŠ) and the fifth author (TN)
particularly in ability grouping (Wiliam & who also helped in data collection (designing the
Bartholomew, 2004). Finally, future studies should study, and implementing the experimental
ensure that all teachers are trained to be equally programme). The experimental programme was
prepared across both groups so that the findings are developed with the participation of each author,
not adversely impacted. Although no significant based on the ideas of the first author (VM). BM
differences were observed in teachers’ content and NŠ were in charge of quantitative data
knowledge during teaching implementation, at least processing, while the fifth author (TN) was in
some of the gaps between the EG and CGs’ charge of qualitative data processing. Data
achievement and engagement could be attributed to discussion was created by the second (DV) and the
variations in knowledge. first (VM) author. The percentage of contributions
is indicated by the order of authorship provided.
Conclusion
The authors read and approved the final
In this study we evaluated the efficacy of DI among
manuscript.
learning groups in Montenegro fourth-graders and
demonstrated that the teaching and learning Notes
approach optimised students’ algebraic capabilities. i. Published under a Creative Commons Attribution
This resulted in greater success at task resolution, Licence.
greater engagement and activity, and increased ii. DATES: Received: 6 November 2020; Revised: 15
March 2023; Accepted: 24 June 2023; Published: 31
persistence. Teachers have a central role in
August 2023.
planning and implementing DI as they monitor
students’ work and progress. These findings References
support previous research that encourages more Abah JA 2020. An appeal in the case involving
teachers to apply the DI model when teaching conventional teaching: Emphasizing the
mathematics. transformation to enhanced conventional teaching
Our research also confirmed the need for in mathematics education. VillageMath
Educational Review, 1(1):1–10.
additional teacher training in implementing https://doi.org/10.5281/zenodo.3860320
differentiation strategies. Teachers must possess Anthony G & Hunter R 2017. Grouping practices in New
sound knowledge of pedagogical content, Zealand mathematics classrooms: Where are we at
understand the relationships between tasks, the and where should we be? New Zealand Journal of
concepts of gradual learning, and students’ Educational Studies, 52(1):73–92.
cognitive characteristics in order to adequately https://doi.org/10.1007/s40841-016-0054-z
respond to the learning needs. Ashley LM 2016. Implementation of a math workshop
Accordingly, we recommend the following: model in the elementary classroom: Understanding
• A differentiated model should be designed with how teachers differentiate instruction. PhD thesis.
tasks providing separate instructions tailored for the Boston, MA: Northeastern University. Available at
needs of advanced, in-between, and struggling https://repository.library.northeastern.edu/files/neu:
learners. Teachers must have adequate time to get to cj82m2381/fulltext.pdf. Accessed 31 August 2023.
know their students and other teaching materials. Bal AP 2016. The effect of the differentiated teaching
approach in the algebraic learning field on
• Montenegrin teachers should receive additional DI
students’ academic achievements. Eurasian
training. Teachers must become thoroughly familiar
Journal of Educational Research, 63:185–204.
with its conceptualisation and application. This
Available at
teaching model is important because diversity
https://dergipark.org.tr/en/download/article-
within classrooms should be regarded as an
file/225130. Accessed 31 August 2023.
opportunity to assist all student groups.
Bender WN 2013. Differentiating math instruction, K-8:
• Educational decision-makers should carefully plan
Common core mathematics in the 21st century
the introduction and assessment of DI models.
classroom. Thousand Oaks, CA: Corwin Press.
Students’ mathematical literacy can be successfully
enhanced using DI and this should be recognised
10 Mićanović, Vučković, Maslovarić, Šakotić, Novović

Boaler J 2002. Experiencing school mathematics: Gaitas S & Martins MA 2017. Teacher perceived
Traditional and reform approaches to teaching and difficulty in implementing differentiated
their impact on student learning (Rev. ed). London, instructional strategies in primary school.
England: Routledge. International Journal of Inclusive Education,
Boaler J 2013. Ability and Mathematics: The mindset 21(5):544–556.
revolution that is reshaping education. Forum, https://doi.org/10.1080/13603116.2016.1223180
55(1):143–152. Available at Gamble V 2011. The impact of differentiated versus
http://www.youcubed.org/wp- traditional instruction on math achievement and
content/uploads/14_Boaler_FORUM_55_1_web.p student attitudes. PhD dissertation. Minneapolis,
df. Accessed 31 August 2023. MN: Walden University. Available at
Boaler J & Wiliam D 2001. ‘We’ve still got to learn!’ https://scholarworks.waldenu.edu/cgi/viewcontent.
Students’ perspectives on ability grouping and cgi?article=1922&context=dissertations. Accessed
mathematics achievement. In P Gates (ed). Issues 31 August 2023.
in mathematics teaching. London, England: George PS 2005. A rationale for differentiating
Routledge. instruction in the regular classroom. Theory Into
Boaler J, Wiliam D & Brown M 2000. Students’ Practice, 44(3):185–193.
experiences of ability grouping—disaffection, https://doi.org/10.1207/s15430421tip4403_2
polarisation and the construction of failure. British Hackenberg AJ, Creager M & Eker A 2021. Teaching
Educational Research Journal, 26(5):631–648. practices for differentiating mathematics
https://doi.org/10.1080/713651583 instruction for middle school students.
Chamberlin M & Powers A 2010. The promise of Mathematical Thinking and Learning, 23(2):95–
differentiated instruction for enhancing the 124.
mathematical understandings of college students. https://doi.org/10.1080/10986065.2020.1731656
Teaching Mathematics and its Applications: An Harris P & Johnson R n.d. Non-traditional teaching and
International Journal of the IMA, 29(3):113–139. learning strategies. Available at
https://doi.org/10.1093/teamat/hrq006 http://www.montana.edu/facultyexcellence/Papers/
Champely S 2020. pwr: Basic functions for power activelearn2.html. Accessed 16 May 2019.
analysis. R package version 1.3-0. Available at Hertberg-Davis H 2009. Myth 7: Differentiation in the
https://CRAN.R-project.org/package=pwr. regular classroom is equivalent to gifted programs
Accessed 31 August 2023. and is sufficient: Classroom teachers have the time,
Cohen J 1988. Statistical power analysis for the the skill, and the will to differentiate adequately.
behavioral sciences (2nd ed). Hillsdale, NJ: Gifted Child Quarterly, 53(4):251–253.
Lawrence Erlbaum. https://doi.org/10.1177/0016986209346927
Cowan R & Powell D 2014. The contributions of Kesteloot BA 2011. Effects of differentiated mathematics
domain-general and numerical factors to third- instruction in a fourth grade classroom. MSc
grade arithmetic skills and mathematical learning dissertation. Marshall MN: Southwest Minnesota
disability. Journal of Educational Psychology, State University. Available at
106(1):214–229. https://doi.org/10.1037/a0034097 https://www.proquest.com/docview/894122407?pq
De Graaf A, Westbroek H & Janssen F 2019. A practical -origsite=gscholar&fromopenview=true. Accessed
approach to differentiated instruction: How biology 31 August 2023.
teachers redesigned their genetics and ecology Light D & Pierson E 2014. Increasing student
lessons. Journal of Science Teacher Education, engagement in math: The use of Khan Academy in
30(1):6–23. Chilean classrooms. International Journal of
https://doi.org/10.1080/1046560X.2018.1523646 Education and Development using Information and
Doubet KJ & Hockett JA 2017. Differentiation in the Communication Technology, 10(2):103–119.
elementary grades: Strategies to engage and equip Available at
all learners. Alexandria, VA: ASCD. https://www.learntechlib.org/p/147457/. Accessed
Ekstam U, Linnanmäki K & Aunio P 2015. Educational 1 June 2019.
support for low-performing students in Livers SD, Paxton M, O’Grady N & Tontillo M 2018.
mathematics: The three-tier support model in Embracing curriculum compacting: Teacher
Finnish lower secondary schools. European candidates supporting differentiated instruction in
Journal of Special Needs Education, 30(1):75–92. elementary mathematics. School-University
https://doi.org/10.1080/08856257.2014.964578 Partnerships, 11(1):19–25. Available at
Francis B, Archer L, Hodgen J, Pepper D, Taylor B & https://files.eric.ed.gov/fulltext/EJ1179969.pdf.
Travers MC 2017. Exploring the relative lack of Accessed 31 August 2023.
impact of research on ‘ability grouping’ in Lou Y, Abrami PC, Spence JC, Poulsen C, Chambers B
England: A discourse analytic account. Cambridge & D’Apollonia S 1996. Within-class grouping: A
Journal of Education, 47(1):1–17. meta-analysis. Review of Educational Research,
https://doi.org/10.1080/0305764X.2015.1093095 66(4):423–458. https://doi.org/10.2307/1170650
Fuchs LS, Fuchs D, Compton DL, Powell SR, Seethaler Matthews MS, Ritchotte JA & McBee MT 2013. Effects
PM, Capizzi AM, Schatschneider C & Fletcher JM of schoolwide cluster grouping and within-class
2006. The cognitive correlates of third-grade skill ability grouping on elementary school students’
in arithmetic, algorithmic computation, and academic achievement growth. High Ability
arithmetic word problems. Journal of Educational Studies, 24(2):81–97.
Psychology, 98(1):29–43. https://doi.org/10.1080/13598139.2013.846251
https://doi.org/10.1037/0022-0663.98.1.29 Maxey KS 2013. Differentiated instruction: Effects on
 South African Journal of Education, Volume 43, Number 3, August 2023 11

primary students’ mathematics achievement. PhD mixed methods in social and behavioral research
dissertation. Scottsdale, AZ: Northcentral (2nd ed). Thousand Oaks, CA: Sage.
University. Available at Sousa DA & Tomlinson CA 2011. Differentiation and
https://www.proquest.com/docview/1449660560?p the brain: How neuroscience supports the learner-
q-origsite=gscholar&fromopenview=true. friendly classroom. Bloomington, IN: Solution
Accessed 31 August 2023. Tree Press.
McKeen H 2019. The impact of grade level flexible Steenbergen-Hu S, Makel MC & Olszewski-Kubilius P
grouping on Math Achievement Scores. Georgia 2016. What one hundred years of research says
Educational Researcher, 16(1):48–62. about the effects of ability grouping and
https://doi.org/10.20429/ger.2019.160105 acceleration on K–12 Students’ academic
Mejer L, Turchetti P & Gere E 2011. Trends in European achievement: Findings of two second-order meta-
education during the last decade. Available at analyses. Review of Educational Research,
https://ec.europa.eu/eurostat/documents/3433488/5 86(4):849–899.
579668/KS-SF-11-054-EN.PDF.pdf/a98c93d3- https://doi.org/10.3102/0034654316675417
dff5-479e-8ac1-aa15975ecb37?t=1414693417000. Tambychika T & Meerah TSM 2010. Students’
Accessed 31 August 2023. difficulties in mathematics problem-solving: What
Murtagh F & Legendre P 2014. Ward’s hierarchical do they say? Procedia - Social and Behavioral
agglomerative clustering method: Which Sciences, 8:142–151.
algorithms implement Ward’s criterion? Journal of https://doi.org/10.1016/j.sbspro.2010.12.020
Classification, 31:274–295. Tieso CL 2003. Ability grouping is not just tracking
https://doi.org/10.1007/s00357-014-9161-z anymore. Roeper Review, 26(1):29–36.
Organisation for Economic Co-operation and https://doi.org/10.1080/02783190309554236
Development 2018. PISA 2015: Results in focus. Tomlinson CA 2008. The goals of differentiation:
Paris, France: Author. Available at Differentiated instruction helps students not only
https://www.oecd.org/pisa/pisa-2015-results-in- master content, but also form their own identities
focus.pdf. Accessed 31 August 2023. as learners. Educational Leadership, 66(3):1–6.
Papanastasiou E 2002. Factors that differentiate Tomlinson CA 2014. The differentiated classroom:
mathematics students in Cyprus, Hong Kong, and Responding to the needs of all learners (2nd ed).
the USA. Educational Research and Evaluation, Alexandria, VA: ASCD.
8(1):129–146. Tomlinson CA & Imbeau MB 2010. Leading and
https://doi.org/10.1076/edre.8.1.129.6919 managing a differentiated classroom. Alexandria,
Pavićević Ž, Vučeljić M, Lalić M, Pavićević I & Kostić VA: ASCD.
A 2017. Teaching subject mathematics, curriculum Tomlinson CA & McTighe J 2006. Integrating
for grades I, II, III, IV, V, VI, VII, VIII and IX of differentiated instruction and understanding by
elementary school. Podgorica, Montenegro: design: Connecting content and kids. Alexandria,
Ministry of Education and Science. VA: Association for Supervision and Curriculum
Prast EJ, Van de Weijer-Bergsma E, Kroesbergen EH & Development.
Van Luit JEH 2018. Differentiated instruction in Van Geel M, Keuning T, Frèrejean J, Dolmans D, Van
primary mathematics: Effects of teacher Merriënboer J & Visscher AJ 2019. Capturing the
professional development on student achievement. complexity of differentiated instruction. School
Learning and Instruction, 54:22–34. Effectiveness and School Improvement, 30(1):51–
https://doi.org/10.1016/j.learninstruc.2018.01.009 67.
Prica I, Čolić L & Baronijan H 2014. Studija o ulaganju https://doi.org/10.1080/09243453.2018.1539013
u rano obrazovanje djece u Crnoj Gori [A study on Van Tassel-Baska J & Stambaugh T 2005. Challenges
investment in early childhood education in and possibilities for serving gifted learners in the
Montenegro]. Podgorica, Montenegro: UNICEF. regular classroom. Theory Into Practice,
Available at 44(3):211–217.
https://www.unicef.org/montenegro/media/2781/fil https://doi.org/10.1207/s15430421tip4403_5
e/MNE-media-MNEpublication34.pdf. Accessed Wiliam D & Bartholomew H 2004. It’s not which school
31 August 2023. but which set you’re in that matters: The influence
Roy A, Guay F & Valois P 2013. Teaching to address of ability grouping practices on student progress in
diverse learning needs: Development and mathematics. British Educational Research
validation of a Differentiated Instruction Scale. Journal, 30(2):279–293.
International Journal of Inclusive Education, https://doi.org/10.1080/0141192042000195245
17(11):1186–1204. Wilson G 2014. Activating assessment for all students:
https://doi.org/10.1080/13603116.2012.743604 Differentiated instruction and informative methods
Sammons P 2010. The contribution of mixed methods to in math and science. Mathematics Teacher,
recent research on educational effectiveness. In A 108(1):76–77.
Tashakkori & C Teddlie (eds). Sage handbook of