Profiles of teaching practices and reading skills at the first and third grade in Finland and Estonia

Abstract

The Early Childhood Classroom Observation Measure was used to observe 91 first-grade and 70 third-grade teachers in Estonia and Finland. Using a person-oriented approach, four profiles of teaching practices were identified in grade 1: the child-centred style, teacher-directed style, child-dominated style and a mixture of the child-centred and teacher-directed styles. An additional profile, the extreme-child-centred style, was found in grade 3. Children taught by child-centred teachers showed the highest reading skills, whereas children taught by child-dominated teachers showed the lowest skills. More Estonian than Finnish teachers applied the child-dominated style in grade 1 and the extreme-child-centred style in grade 3.

Full text

Teaching Practices and Reading Skills in Primary School 1
Accepted by Teaching and Teacher Education
Profiles of Teaching Practices and Reading Skills at the First and Third Grade in Finland and
Estonia
Xin Tang, Eve Kikas, Eija Pakarinen, Marja-Kristiina Lerkkanen, Joona Muotka, & Jari-Erik
Nurmi
University of Jyväskylä and Tallinn University
To cite:
Tang, X., Kikas, E., Pakarinen, E., Lerkkanen, M.-K., Muotka, J., & Nurmi, J.-E. (2017). Profiles of
teaching practices and reading skills at the first and third grade in Finland and Estonia. Teaching and
Teacher Education, 64, 150–161. https://doi.org/10.1016/j.tate.2017.01.020
Author’s Note
Xin Tang, Department of Psychology, University of Jyväskylä, Finland; P.O. Box 35
FI-40014 University of Jyväskylä, Finland; tangxin09@gmail.com
Eve Kikas, School of Natural Sciences and Health, Tallinn University, Estonia; Narva mnt 25, 10120,
Tallinn; eve.kikas@tlu.ee
Eija Pakarinen, Department of Teacher Education, University of Jyväskylä, Finland; P.O. Box 35 FI-
40014 University of Jyväskylä, Finland; eija.k.pakarinen@jyu.fi
Marja-Kristiina Lerkkanen, Department of Teacher Education, University of Jyväskylä, Finland; P.O.
Box 35 FI-40014 University of Jyväskylä, Finland; marja-kristiina.lerkkanen@jyu.fi
Joona Muotka, Department of Psychology, University of Jyväskylä, Finland; P.O. Box 35
FI-40014 University of Jyväskylä, Finland; joona.muotka@jyu.fi
Jari-Erik Nurmi, Department of Psychology, University of Jyväskylä, Finland; P.O. Box 35
FI-40014 University of Jyväskylä, Finland; jari-erik.j.nurmi@jyu.fi

Teaching Practices and Reading Skills in Primary School 2
This study was supported by grants from the Academy of Finland (No. 268 586 for 2013–2017 and
No. 277 299 for 2015–2017) and institutional research funding IUT (3-3) of the Estonian Ministry of
Education and Research. The first author thanks the support from China Scholarship Council (No.
201306090004).
Correspondence concerning this article should be addressed to Xin Tang, Department of Psychology,
P.O. Box 35 FI-40014, University of Jyväskylä, Finland. Contact: tangxin09@gmail.com

Teaching Practices and Reading Skills in Primary School 3
Abstract
The Early Childhood Classroom Observation Measure was used to observe 91 first-grade and
70 third-grade teachers in Estonia and Finland. Using a person-oriented approach, four
profiles of teaching practices were identified in grade 1: the child-centred style, teacher-
directed style, child-dominated style and a mixture of the child-centred and teacher-directed
styles. An additional profile, the extreme-child-centred style, was found in grade 3. Children
taught by child-centred teachers showed the highest reading skills, whereas children taught
by child-dominated teachers showed the lowest skills. More Estonian than Finnish teachers
applied the child-dominated style in grade 1 and the extreme-child-centred style in grade 3.
Keywords: child-centred practices, teacher-directed practices, child-dominated practices,
reading skills, primary school

Teaching Practices and Reading Skills in Primary School 4
Highlights
Using person-oriented approach to identify different profiles of teaching practices.
Child-centred style was most beneficial while child-dominated style was least
beneficial for reading skills.
Mixture of child-centred and teacher-directed style was beneficial in grade 3.
Extreme-child-centred style did not guarantee reading fluency in grade 3.
 Less beneficial styles showed more in Estonian teachers than in Finnish teachers.

Teaching Practices and Reading Skills in Primary School 5
Profiles of Teaching Practices and Reading Skills at the First and Third Grade in Finland and
Estonia
There is substantial evidence to show that teaching practices play an important role in
early learners’ academic performance (Early et al., 2007; Hamre & Pianta, 2005; Lerkkanen
et al., 2016). Teachers’ practices are typically based on their own beliefs and philosophy of
teaching, as well as their education and experience (Stipek, Daniels, Galluzzo, & Milburn,
1992; Stipek, Givvin, Salmon, & MacGyvers, 2001). Although the associations between
teaching practices and child outcomes have been studied in authentic classroom settings, only
a few studies have applied a person-oriented approach to identify subgroups of teachers who
show different teaching practices. By going beyond a variable-oriented approach and by
using person-oriented methods, the present study aimed at identifying subgroups of teachers
in Finnish and Estonian primary school classrooms on the basis of their teaching practices as
measured by the Early Childhood Classroom Observation Measure (ECCOM; Stipek &
Byler, 2005). The study further examined the extent to which these subgroups differed in
terms of children’s reading skills in the first and third grades.
Teaching Practices and Academic Outcomes
Teachers vary in the practices they use when interacting with pupils and instructing
them in the classroom (e.g., Connor, Son, Hindman, & Morrison, 2005). The previous
literature has typically focussed on child-centred and teacher-directed practices when
analysing the role of instructional approaches in children’s skill development (Lerkkanen et
al., 2016; Pressley et al., 2003). Recently, child-dominated practices also have been under
investigation (Kikas, Peets, & Hodges, 2014; Kikas, Silinskas, Jõgi, & Soodla, 2016). Child-
centred practices, which are based on the constructivist theories of learning and teaching
(Piaget, 1985; Vygotsky, 1978; for an overview, see Bransford, Brown, & Rodney, 2000), are
based on the assumption that children are active learners who construct knowledge based on

Teaching Practices and Reading Skills in Primary School 6
their prior understanding and experiences. Children’s initiatives and interests are emphasised
and children are given an appropriate level of autonomy and an active role in decision
making in the classroom. Teachers also actively use guidance and scaffolding to assist
children in developing their own knowledge and understanding and provide possibilities for
children to explore and manipulate objects (Stipek & Byler, 2004). In turn, teacher-directed
(i.e., didactic) practices, with an emphasis on concrete and rote learning (Stipek & Byler,
2004), stress that teacher make most of the decisions, control the instructional activities, and
emphasise the importance of facts and training basic skills. In child-centred practice and in
teacher-directed practices the teachers’ role is active in guiding and instructing children. By
contrast, in child-dominated practices, teachers provide children with little direction, control
or feedback (Kikas et al., 2014, 2016; Stipek & Byler, 2005). The classroom rules are often
unclear and there are no systematically designed learning tasks present. Teachers, however,
may interrupt and control activities when the children’s behaviour is out of control (Stipek &
Byler, 2005). In the present study, we examined teaching practices in light of the above-
mentioned three definitions, using an observational instrument developed by Stipek and Byler
(2004; 2005), the Early Childhood Classroom Observation Measure (ECCOM).
Because teachers who use different teaching practices have been shown to vary in
their instructional emphases (Stipek and Byler, 2004), it can be assumed that each teaching
practice plays a different role in different reading skills, that is, in basic skills, such as
decoding, and in more advanced reading skills, such as comprehension. Previous studies have
shown that the beneficial effects of different teaching practices on academic outcomes vary
depending on the skill domain, skill level, and on the age of the children. For example, child-
centred practices, in general, have a positive impact on kindergarteners’ reading performance
(Marcon, 1999), first graders’ reading fluency (Lerkkanen et al., 2016; Perry, Donohue, &
Weinstein, 2007), and the development of reading comprehension (Block, Parris, Reed,

Teaching Practices and Reading Skills in Primary School 7
Whiteley, & Cleveland, 2009). Teacher-directed practices, in turn, have been found to be
beneficial for kindergarteners’ and first graders’ basic reading skills, such as letter knowledge
and word recognition skills (Stipek et al., 1998; Stipek, Feiler, Daniels, & Milburn, 1995).
However, little research has been conducted on the effect of teacher-directed practices on
more advanced reading skills, such as reading comprehension. A recent study by Kikas et al.
(2014) showed that the effect of child-dominated practices was moderated by children’s skill
level in the classroom. In classrooms with high initial math skills, child-dominated practices
were positively associated with spelling skills and task-persistent learning behaviour, while in
classrooms with low initial math skills the impact was negative. Given the importance of
reading development at this age (e.g., Adams, 1990; Landerl & Wimmer, 2008),the present
study focusses on the development of reading skills during the early school years.
Teaching Styles
In authentic classrooms, however, teachers’ use of teaching practices is more
complex. Instead of employing predominantly one practice, they may use a combination of
different practices (Pressley et al., 2003; Rasku-Puttonen et al., 2011). The predominant use
of a specific teaching practice or combinations of different teaching practices can be
described as teaching styles (Kikas et al., 2016). Teaching styles refer to patterns or profiles
of teaching practices. However, most of the previous research has been variable-oriented and
has examined teaching practices in terms of specific dimensions. Studies aimed at identifying
different teaching styles and their combinations are rare. As far as we know, only two
previous studies have sought to identify subgroups of teachers with different profiles of
teaching practices as measured by the ECCOM (Kikas et al., 2016; Rasku-Puttonen et al.,
2011). In both studies, one in kindergarten and another in first grade, four teaching styles
were identified: the child-centred style, teacher-directed style, child-dominated style and a
mixture of the child-centred and teacher-directed style (mixture teaching style). Both studies

Teaching Practices and Reading Skills in Primary School 8
also found that most teachers showed a child-centred style. The second largest group was
those with mixture teaching style. Furthermore, Rasku-Puttonen et al. (2011) found that more
kindergarten teachers used the teacher-directed style and mixture teaching style in Finland
than in Estonia, and that more kindergarten teachers used the child-centred style and child-
dominated style in Estonia than in Finland. So far, teaching styles have not been examined in
later primary school grades using the ECCOM procedure.
Educational System and Reading Acquisition in Finland and Estonia
The Finnish and Estonian school systems are rather similar. In both countries,
compulsory formal education consists of nine years of comprehensive school, beginning in
the year the child turns seven and continuing with the same class teacher for the main
subjects. In addition, the academic demands in the early years of primary school are similar
in both national core curricula (Finnish National Board of Education, 2014; Vabariigi
Valitsus, 2011/2014). For example, in both countries, initial reading instruction in these
transparent languages is based on phonics. Both countries emphasise practice in reading
fluency and comprehension in grade 1 and their curricula include 6-7 hours of literacy lessons
per week during the first two school years.
Teacher training is also similar in both countries. Constructivist learning theories and
related teaching methods are introduced, and the individualisation of instruction is valued.
However, Estonia and Finland have experienced very different histories, which are reflected
in their educational systems. Until 1991, when Estonia was part of the Soviet Union,
authoritarian management practices and teacher-directed teaching methods were commonly
applied in schools. Even in primary schools, teacher-directed methods (e.g., whole-class
teaching, teacher-talk and assessing factual knowledge) were the predominant modes of
instruction (Ruus et al., 2008). Although reforms in schools and teacher education
institutions have taken place over the last two decades, changes in values, beliefs and

Teaching Practices and Reading Skills in Primary School 9
practices take time. Empirical studies have shown that Estonian teachers hold a variety of
child-rearing values (Tulviste & Kikas, 2010) and that teachers’ preference for teacher-
directed teaching methods depends on their age and experience: teacher-directed approaches
are favoured more by older and more experienced teachers than by younger teachers (Palu &
Kikas, 2007). In Finland, child-centred practices, adaptation of instruction according to
students’ skills and individualised support for learning are highly valued and also evident in
primary school teaching practices (Lerkkanen, Kikas, et al., 2012; Nurmi et al., 2013).
In terms of learning to read, Finnish and Estonian languages have shallow
orthographies and high transparency which affects reading skills development: the acquisition
of decoding skill and reading accuracy progresses faster in shallow orthographies than deep
orthographies (Seymour, Aro, & Erskine, 2003). Both Finnish and Estonian children have
shown high-level reading results in the Program for International Student Assessment (PISA;
OECD, 2012). However, the two countries vary in the time when children’s reading
instruction begins. Whereas in Finland formal reading instruction begins in primary school
(at age 7), in Estonia it begins in kindergarten (at age 6) (Soodla et al., 2015). In Estonia,
children are taught to decode and spell simple words already in kindergarten (Vabariigi
Valitsus, 2008/2011), while in Finland no systematic reading instruction takes place in
kindergarten. Thus, it can be assumed that more emphasis is placed on the learning of basic
reading skills in first-grade reading instruction in Finland than in Estonia. Moreover, as the
reading skills of Finnish first graders are more heterogeneous (see also Soodla et al., 2015),
more individualization in reading instruction, in accordance with the child’s reading skills, is
needed.
The Present Study: Research Questions and Hypotheses
Previous studies on the importance of different teaching practices in the development
of students’ reading skills have some limitations. First, most of the previous studies have

Teaching Practices and Reading Skills in Primary School 10
examined only one or two types of teaching practices or the composite score of two practices
(Lerkkanen, Kiuru, et al., 2012; Perry et al., 2007). Although researchers have claimed that it
is important to examine mixtures of teaching practices, such as the combination of didactic
and constructivist practices (see Pressley et al., 2003), empirical research on such
combinations is rare (for exceptions, see Kikas et al., 2016; Rasku-Puttonen et al., 2011).
Second, child-dominated practices have rarely been examined in the field, limiting the
formation of a broader view of teaching practices (for an exception, see Kikas et al., 2014;
2016). Third, most of the previous studies have been conducted in kindergarten and preschool
classrooms, and only a few at the primary school level (for exceptions, see Lerkkanen et al.,
2016; Kikas et al., 2016). Thus, the present study sought to identify subgroups of teachers
who use different combinations of teaching practices, i.e., teaching styles (Kikas et al., 2016),
in a sample of Finnish and Estonian primary school teachers, and to examine how these
subgroups differ in terms of children’s reading skills, i.e., fluency and comprehension.
Specifically, the present study examined the following research questions:
1. What kinds of profiles of teaching practices (i.e., styles) can be identified in authentic classroom
settings in grade 1 and grade 3? Based on previous findings (Kikas et al., 2016; Rasku-Puttonen et al.,
2011), we expected to find four profiles: the child-centred style, teacher-directed style, child-
dominated style, and a mixture of the child-centred and teacher-directed styles (mixture teaching
style) (Hypothesis 1).
2. To what extent do Finnish and Estonian teachers differ in their use of teaching
styles in grade 1 and grade 3? Based on the cultural and historical differences between the
educational systems in Finland and Estonia (Kikas & Lerkkanen, 2011), we expected that a
higher proportion of Finnish than Estonian teachers would use a child-centred style
(Hypothesis 2).

Teaching Practices and Reading Skills in Primary School 11
3. Do children whose teachers use different teaching styles differ in their performance in reading
fluency and reading comprehension in grade 1 and grade 3? Because both child-centred practices
(e.g., Lerkkanen et al., 2016; Marcon, 1999; Perry et al., 2007) and teacher-directed practices (e.g., G.
Adams & Carnine, 2003; Stipek et al., 1995) have been found to be positively related to basic reading
skills, we expected that, in reading fluency, children in classrooms where teachers use a child-centred
style, teacher-directed style or mixture teaching style would outperform children in classrooms where
teachers use a child-dominated style (Hypothesis 3a). Moreover, because teacher-directed teachers
have been shown to place less emphasis on high-order skills (Stipek & Byler, 2004; Tang et al., 2016),
we expected that children in classrooms where teachers use a child-centred style or mixture teaching
style would outperform children in classrooms where teachers use a teacher-directed style or child-
dominated style in reading comprehension (Hypothesis 3b). In addition, in grade 1, we expected that
children whose teachers use a child-centred style or teacher-directed style would show better reading
fluency and comprehension than children in classrooms where teachers use a child-dominated style
(Hypothesis 3c). For grade 3, no specific hypothesis was formulated, owing to the lack of empirical
evidence.
Methods
Participants and Procedures
The total sample of the present study comprised 91 first-grade teachers (32 in Finland and 59 in
Estonia) and 70 third-grade teachers (33 in Finland and 37 in Estonia), and the children in their
classrooms. Both datasets were collected as part of other ongoing longitudinal studies; the FS study in
Finland (Lerkkanen, Niemi, et al., 2006), and the KISS study and the READ study in Estonia
(AUTHORS).
The Finnish sample. This study is part of an extensive age-cohort study from kindergarten to grade 4
conducted during the years 2006–2011. The sample comprised 1,132 children from 93 classrooms
with their teachers. Thirty-two first-grade teachers (28 female, 4 male), and 33 third-grade teachers
(24 female, 9 male) were observed on a voluntary basis in the spring semester, in 2008 and 2010,
respectively. In the sample of observed classrooms, seven teachers and most of the children were the

Teaching Practices and Reading Skills in Primary School 12
same at the two measurement points. Most teachers (86% of the first-grade and 97% of the third-grade
teachers) had a master’s degree or above. Teachers’ work experience was measured by asking them to
select from one of five options (1 = less than a year, 2 = 1–5 years, 3 = 6–10 years, 4 = 11–15 years, 5
= more than 15 years). The majority of the first- and third-grade teachers had more than 15 years’
teaching experience (Median = 5, Mode = 5, for first grade; Median = 4, Mode = 5, for third grade).
The schools were in two medium-sized towns and one less urban municipality in Finland. Mean class
size was 19.22 (SD = 4.52) in grade 1 and 19.94 (SD = 5.88) in grade 3, which is the typical class size
in Finnish primary schools. The average age of the children was 7.15 years (SD = .3) when they
entered grade 1.
The Estonian sample. The Estonian grade 1 samples comprised teachers from two separate studies:
38 teachers (all female) and 869 children from the KISS study, and 21 teachers (all female) and 465
children from the READ study. All the teachers were classroom teachers and all of them had a
master’s level education. Thirty-eight first-grade teachers were observed in 2008 (from KISS) and 21
in 2012 (from READ). Thirty-seven teachers from the KISS study were observed two years later in
grade 3; almost all of them were the same (only six teachers had changed). The majority of the first-
and third-grade teachers had more than 15 years’ teaching experience (Median = 5, Mode = 5, for both
grades). The average class size was 19.72 (SD = 4.90) in grade 1, and 16.67 (SD = 4.59) in grade 3.
The average age of the children when they entered grade 1 was 7.46 years (SD = .52) and 7.34 years
(SD = .32) for the KISS and READ study samples, respectively.
Procedure. In both countries, the researchers contacted school principals and teachers first to inform
them about the project and invited them to participate. Second, parents were asked to sign an
informed consent for their children’s participation. The children’s and their parents’ background
information was measured at the start of each project.
During the spring term of the first and third grades (February–March), observations
were conducted in classrooms by experienced observers (with a master’s or doctoral degree
in education or psychology). Before starting the observations, the observers were carefully
trained until the intra-class correlation (ICC) reliability between two observers reached .81 or

Teaching Practices and Reading Skills in Primary School 13
above for each subscale. The classroom observations were conducted following the
procedures described in the ECCOM manual, and thus two observers, producing independent
ratings, were always present in a classroom (Stipek & Byler, 2005; see also Lerkkanen et al.,
2012). Each observation session lasted three lessons (i.e., at least half a day) and began at the
start of the school day. All observations included at least one literacy lesson.
The Finnish FS children’s reading skills were assessed at the beginning of the fall term of grade 1
(September) and at the end of grades 1 and 3 (April). The Estonian KISS children were assessed on
their reading skills at the beginning of grade 1 (September–October) and at the end of grade 3 (April–
May). In both countries, the grade 3 measurements of reading skills were performed with the same
instruments (see Appendix). However, different measures of reading skills were used with the grade 1
Finnish FS sample and Estonian KISS sample. The Estonian READ children were assessed on their
reading skills at the beginning of grade 1 (September–October) and at the end of grade 1 (April–May).
The same reading skills measures were used as with the Finnish FS sample (see Appendix).

Teaching Practices and Reading Skills in Primary School 14
Measures
Classroom observations. The ECCOM (Stipek & Byler, 2004; 2005) was used to
measure teaching practices on the extent (proportion of time) to which they were child-
centred, teacher-directed and child-dominated. Each practice is rated on 14 items and over the
same three subscales: management (four items), climate (four items) and instruction (six
items), as shown in Table 1. The rating scale is based on the percentage of the time that each
type of practice is demonstrated during the observation: 1 = the practice is rarely seen (0%–
20% of the time) to 5 = the practice predominates (80%–100% of the time). The use of each
of the three teaching practices was independently rated by two observers. For example, at the
end of the observation day, for a specific item (e.g., choice of activities), the use of a child-
centred practices might be rated as 4, a teacher-directed practices as 3, and a child-dominated
practices as 2 (Stipek & Byler, 2005). The mean scores of both observers were used in this
study. The inter-rater reliabilities varied between .67 - .80, which can be regarded as good or
excellent (Hallgren, 2012).
__ _ ________
Insert Table 1 about here
__ _ ________
Reading fluency. A group-administered subtest of the standardised reading test
battery (ALLU—Reading Test for Primary School; Lindeman, 1998) was used to assess
word-level reading accuracy and fluency in the grade 1 FS sample and READ sample, and
grade 3 FS sample and KISS sample,. In this speed test, a maximum of 80 items can be
attempted within a 2-minute time limit. For each item, a child was asked to read four
(phonologically similar) words and draw a line connecting a picture and the word that
semantically matched it. The score used in the analyses was the sum of correct answers
(maximum 80). In this speed test, the score reflects both the child’s fluency in reading the
stimulus words and his or her accuracy in making the correct choice from among the

Teaching Practices and Reading Skills in Primary School 15
alternatives. In a highly transparent language, such as Finnish and Estonian, only a fluency
measure can differentiate between children’s decoding skills across their primary school
years. According to the test manual (Lindeman, 1998), the Kuder–Richardson reliability
coefficient was .97 in both grade 1 and grade 3. No floor or ceiling effects were detected.
In grade 1, different measures of reading skills (i.e., phoneme awareness or phoneme-
grapheme correspondence, and the reading accuracy or fluency test) were used in the FS and
in KISS samples. In the KISS sample, the tests included phoneme-grapheme correspondence
and reading accuracy, whereas reading skills in the FS sample were measured as phoneme
awareness (identification of the initial sound from the word) and reading fluency (word level
reading accuracy and speed; ALLU test).
In the KISS sample, children's phoneme-grapheme correspondence was assessed with
nine items. In the first five items, the children were shown pictures of familiar objects.
Underneath each picture was a number of blank spaces corresponding to the number of letters
in the word for the object represented in the picture; for example, 8 blank spaces appeared
under the picture of the target word mesilane [bee]. The experimenter said the target word
aloud and the child was shown one letter in the word (e.g., the grapheme E). The child’s task
was to mark where the letter should be placed in the series of blank spaces. In the remaining
four items, the children were presented with three pictures each showing a different object.
The child’s task was to mark the object that corresponded to the word the experimenter said
aloud. Although the objects were familiar to the children, they need to listen carefully to the
names of the three objects, as they differed in the duration of the sounds (e.g., saba-saabas-
sabas; keeb-kepp-keep). For each child, the sum of correct answers was calculated (maximum
score = 9). Actual scores ranged from 1 to 9 (M = 7.63, SD = 2.08). Cronbach's alpha was .
85.

Teaching Practices and Reading Skills in Primary School 16
In the test of reading accuracy, the KISS children were given a list of seven words that
described objects needed in school (e.g., vihik [copybook]), objects that were not needed in
school (e.g., suvi [summer]), and a non-word (raamatop, which looks very similar to the word
“raamat” [book]). The children read each word and marked whether the object it named was
needed at school. Each correct answer scored 1 point. Actual scores ranged from 1 to 7.
Cronbach's alpha was .83.
In the FS sample, the initial phoneme identification test (ARMI test battery;
Lerkkanen, Poikkeus, & Ketonen, 2006) was also used. The children were shown four
pictures of objects that were named aloud by the experimenter. The children were then asked
to select the correct picture based on oral presentation of the initial phoneme of the target
word. All the words comprised one to three syllables, with two vowels and eight consonants.
The sum score was based on the number of correct items (maximum score = 10). Cronbach's
alpha was .74.
Reading comprehension. A group-administered subtest of the standardised reading
test battery (ALLU—Reading Test for Primary School; Lindeman, 1998) assessed the child’s
reading comprehension skills in gleaning factual knowledge, concepts and inferences from
text. The children were asked to answer 12 multiple-choice questions based on a silently read
text. The children received 1 point for each correct answer (maximum score 12). The
maximum time allotted was 45 mins. To ensure that task difficulty was optimal for each age,
the texts and multiple choice questions of this normed test were different for grade 1 and
grade 3. The topics of the texts were ‘Judo’ (grade 1) and ‘Operating a Camera’ (grade 3).
The Kuder–Richardson reliability coefficients, drawn from the test manual, for the reading
comprehension task in grades 1 and 3 were .85 and .75, respectively.
Mothers’ level of education. The children’s mothers’ level of education was
measured on a scale of 1 to 3 (1 = basic education, 2 = high school education, 3 = college

Teaching Practices and Reading Skills in Primary School 17
education and above). For the Finnish mothers, the distribution was 6.6% at level 1, 31.9% at
level 2, and 61.5% at level 3. For the Estonian KISS sample mothers, the corresponding
percentages were 9.5%, 58.6% and 31.9%, and for the Estonian READ sample mothers 6.2%,
45.2% and 48.6%.
Analysis Strategy
Our first aim was to identify subgroups of teachers who use different teaching
practices in classroom settings in grades 1 and 3. For this purpose, we utilised the mixture
modelling (Mplus version 7.0; Muthén & Muthén, 2012). This method allowed us to identify
teaching profiles (i.e., latent classes) from the observed data that differ from other profiles but
that are homogenous within each group. Mixture modelling also provides statistical tests to
evaluate the appropriate number of profiles. To do this, we used three criteria: (a) the model
fits, (b) mean probabilities and numbers of teachers in the latent profiles, and (c) the
interpretability of the identified profiles. The model fits were evaluated upon three criteria:
the Bayesian information criterion (BIC), adjusted Bayesian information criterion (ABIC),
and Akaike’s information criterion (AIC). For the statistical testing of the number of latent
profiles, we used the following tests: the Vuong-Lo-Mendel-Rubin test (VLMR), Lo-Mendell-
Rubin adjusted LRT test (LMR) and entropy value. Lower AIC, BIC and ABIC values
indicate a better fit, and significant test (p < .05) results indicate a higher number of profiles.
The highest log-likelihood value (log L) also indicates the best fit of the model. Classification
quality was determined by examining the posterior probabilities and entropy values (as
suggested by Celeux & Soromenho, 1996; entropy values range from 0 to 1, with 0
corresponding to randomness and 1 to a perfect classification).
Our second aim was to examine whether the Finnish teachers differ from the Estonian
teachers in their teaching styles (i.e., profiles). To do this, we conducted a cross-tabulation
analysis by using IBM SPSS statistical package. The chi-square test (p < .05) in the analysis

Teaching Practices and Reading Skills in Primary School 18
provided the overall differences between the Finnish and Estonian teachers in the use of
teaching styles. The adjusted residuals (t-values showing .05 deviation -1.96 > x > +1.96)
allowed us to compare differences in the use of each teaching style between the two
countries.
Our last aim was to examine whether children whose teachers used different teaching
styles differed from each other in reading fluency and comprehension. For this purpose, we
conducted several ANCOVAs in which children’s reading performance variables were
compared in respect to their teachers’ teaching style. In these analyses we controlled for the
children’s characteristics (i.e., age, gender), their mother’s education level and previous
reading skills to predict group differences in reading fluency and comprehension. The
samples and control variables used in these analyses differed between grade 1 and grade 3
due to practical reasons. In the grade 1 ANCOVAs, we used the Estonian READ sample (21
teachers and 397 students) and the Finnish FS sample (32 teachers and 359 students). These
samples had been administered the same measures of reading fluency and reading
comprehension in grade 1 spring, and of reading fluency in grade 1 fall. In the grade 3
ANCOVAs, the Estonian KISS sample (37 teachers and 456 students) and the Finnish FS
sample (33 teachers and 502 students) were used. These samples had used the same measures
of reading fluency and reading comprehension in grade 3 spring. However, the FS and KISS
grade 3 samples had not been administered the same measures for reading skills in grade 1.
To deal with this limitation, we standardized each of the reading measures in grade 1 and
computed a composite score for the children’s previous reading skills. In the KISS sample (Z
stands for standardized score), grade 1 reading skills were computed as Zphone-graph
+Zreading accuracy; in the FS sample, grade 1 reading skills were computed as Zphoneme
awareness+Zreading fluency.
Results

Teaching Practices and Reading Skills in Primary School 19
The descriptive statistics for the teaching practices subscales and items, separately for
the Finnish and Estonian samples, have been reported earlier (Tang et al., 2016). Both the
Finnish and Estonian teachers’ mean scores were slightly higher for child-centred practices
than teacher-directed practices. The lowest means were found for child-dominated practices
in both countries in both grades. Moreover, the Finnish teachers scored lower than the
Estonian teachers in child-dominated practices, in both the first (t = −3.95, p < .001) and third
(t = −2.37, p < .05) grades.
Latent Profiles of Teaching Practices
First grade teachers. In the mixture modelling procedure, we fitted models with
different numbers of latent profiles (Table 2). The results of the model fits showed that the
BIC, ABIC, AIC and log-likelihood values decreased as the number of classes increased.
However, the VLMR and LMR tests suggested that the four-class solution was significantly
better than the three-class solution, and that the five-class solution was not better than the
four-class solution. The entropy value of the four-class solution indicated a very good
classification (>.90). Each of the four groups had a high average value (>.92) for the
probability of group membership, and none of the groups overlapped with one another, as
indicated by other probabilities lower than .05. Finally, the interpretability of the four-class
solution was best on theoretical grounds. Consequently, the four-class solution was selected.
___________
Insert Table 2 about here
__ _ ________
Table 3 presents the descriptive statistics for the four-class solution. In grade 1, the
first and largest profile, i.e., teachers characterised by the use of the child-centred style,
comprised 43% (39) of teachers with highest means for child-centred practices. The second
profile, i.e., the users of the teacher-directed style consisted 11% (10) of the teachers with
high means for teacher-directed practices and low means for the other teaching practices. The

Teaching Practices and Reading Skills in Primary School 20
third profile, i.e., the users of the child-dominated style, comprised 11% (10) of teachers with
highest means for child-dominated practices. The fourth profile, i.e., the users of the mixture
of child-centred and teacher-directed styles (mixture teaching style), consisted 35% (32) of
teachers with nearly equal means for both child-centred and teacher-directed practices.
Next, we ran ANOVAs to examine whether the identified profiles differed in class
size, teacher’s age and teaching experience. However, none of the variables showed
significant differences between the profiles.
___________
Insert Table 3 about here
__ _ ________
Finnish versus Estonian first grade teachers. Next, we examined the extent to
which the Finnish and Estonian teachers differed with respect to the profile to which they
belonged. The results showed that most teachers in both countries were in the latent profile
labelled as child-centred style; in that group, there were slightly more Finnish teachers (47 %
of the total number of Finnish teachers) than Estonian teachers (41 % of the total number of
Estonian teachers). In the second largest latent profile, users of the mixture teaching style,
there were 44 % of the total number of Finnish teachers and 31 % of the total number of
Estonian teachers. In the third latent profile, users of the teacher-directed style, there were 9
% of the total number of Finnish teachers and 13 % of the total number of Estonian teachers.
Finally, in the latent profile of users of the child-dominated style, there were 17 % of the total
number of Estonian teachers and no Finnish teachers. We subsequently ran a cross-tabulation
analysis between teaching styles and country. The analysis of the adjusted residuals showed
that the Estonian sample contained significantly more teachers who used a child-dominated
style (adjusted standardised residual = 2.5) than the Finnish sample (adjusted standardised
residual = -2.5).

Teaching Practices and Reading Skills in Primary School 21
Third grade teachers. A similar mixture modelling procedure was conducted for the
third-grade teachers. Our results for model fit (see Table 4) again showed that the BIC, ABIC,
AIC and log-likelihood values decreased as the number of classes increased. However,
VLMR and LMR tests revealed that the five-class solution was significantly better than the
four-class solution, and that the six-class solution was not better than the five-class solution.
In addition, the entropy value of the five-class solution indicated a very good classification (>
0.91). Each of the five profile groups had a high average value (> .92) for the probability of
group membership, and none of the groups overlapped with one another, as indicated by the
values for the other probabilities, which were lower than .07. Consequently, we chose the
five-class solution as our final solution.
___________
Insert Table 4 about here
__ _ ________
The first latent profile, i.e., users of the extreme child-centred style, comprised the
teachers with extremely high mean levels for child-centred practices, and accounted for 13%
(9) of all teachers. The second profile, i.e., users of the child-centred style, included the
teachers with medium-high means on child-centred practices, and accounted for 30% (20) of
all teachers. The third profile, i.e., users of the teacher-directed style, accounted for 15% (11)
of the all teachers. The fourth profile, i.e., users of the child-dominated style accounted for
10% (7) of all teachers. The fifth profile, i.e., the users of the mixture teaching style,
accounted for 32% (23) of all teachers.
Again, no significant differences were found between the profiles in relation to class
size, teacher’s age or teaching experience (see Table 5).
___________
Insert Table 5 about here
__ _ ________

Teaching Practices and Reading Skills in Primary School 22
Finnish versus Estonian third grade teachers. The results showed that most Finnish
teachers fell into the latent profiles of users of the child-centred style (42%) and the mixture
teaching style (42%), compared to 16% and 24% of the Estonian teachers. Fewer Finnish
teachers were in the latent groups of users of the extreme child-centred style (3%), teacher-
directed style (9%) and child-dominated style (3%) compared to their Estonian counterparts:
corresponding proportions 22%, 22% and 16%. The analysis of the adjusted residuals in
cross-tabulation analysis revealed that the Finnish sample contained more teachers who
deployed the child-centred style (adjusted standardised residual = 2.4) than the Estonian
sample (adjusted standardised residual = -2.4). Moreover, the Estonian sample contained
more teachers who deployed the extreme child-centred style (adjusted standardised residual =
2.3) than the Finnish sample (adjusted standardized residual = -2.3).
Teaching Styles and Children’s Reading Skills
First grade. To examine whether reading fluency and reading comprehension would
differ across the four latent profiles of teaching practices (i.e., teaching styles), we conducted
ANCOVAs in which teaching style was an independent variable and the children’s age,
gender, mother’s level of education, and children’s reading fluency measured in first grade
fall were included as covariates. The results showed that the four latent profiles differed from
each other with respect to the children’s reading fluency, F(3) = 10.75, p < .001, and reading
comprehension, F(3) = 5.49, p = .001. The pairwise comparisons showed first that, in reading
fluency, children with teachers in the child-centred style profile scored higher than children
with teachers in the profiles of the mixture teaching style (p < .05) and the child-dominated
style (p < .001). Children with teachers in the mixture teaching style profile performed better
in reading fluency than children with teachers in the latent profile of the child-dominated
style (p < .01). Second, children whose teachers were characterised by the child-centred style
performed better in reading comprehension than children with teachers in the child-

Teaching Practices and Reading Skills in Primary School 23
dominated style profile (p < .001). Since no differences were found between the latent
profiles for class size, teacher’s age or teaching experience, these variables were not included
in the ANCOVAs.
___________
Insert Figure 1 about here
__ _ ________
Third grade. Children’s reading performance was compared for five latent profiles
by conducting ANCOVAs in which teaching style was an independent variable and the
children’s age, gender and mother’s level of education, and their previous reading skills in
grade 1 were covariates. The previous reading skills variable was a composite score of the
standardized scores of the first-grade reading measures. The results showed that the five
latent profiles differed in the children’s scores for reading fluency, F(4) = 10.99, p < .001, and
reading comprehension, F(4)=4.73, p < .01. First, the pairwise comparisons showed that
children with teachers in the child-centred style profile had higher scores in reading fluency
than children with teachers in the teacher-directed style (p < .001), extreme child-centred
style (p < .01) and the child-dominated style (p < .05) profiles. Children whose teachers
belonged to the mixture teaching style profile had higher scores in reading fluency than
children with teachers in the teacher-directed style profile (p < .001). Second, the results
showed further that children with teachers in the child-centred style, extreme child-centred
style and the mixture teaching style profiles had higher scores in reading comprehension than
children with teachers in the child-dominated style profile (p < .01; p < .01; p < .05;
respectively). Since no differences were found between the latent profiles for class size,
teacher’s age or teaching experience, these variables were included in the ANCOVAs.
Discussion

Teaching Practices and Reading Skills in Primary School 24
This study is among the few that have investigated the profiles of teaching practices
among primary school teachers. Four profiles were identified among the samples of Finnish
and Estonian grade 1 teachers: the child-centred style, teacher-directed style, child-dominated
style, and mixture teaching style. In grade 3, an additional profile, the extreme-child-centred
style, was identified. There were also differences between the Finnish and Estonian teachers
in their teaching profiles. Namely, in grade 1, a higher proportion of Estonian than Finnish
teachers used the child-dominated style, and a higher proportion of Finnish than Estonian
teachers used the child-centred style. However, in grade 3, a lower proportion of Estonian
than Finnish teachers used the extreme-child-centred style. Children in classrooms where the
teacher deployed the child-centred style showed better performance in reading fluency and
reading comprehension than children whose teachers deployed the child-dominated style.
Profiles of Teaching Practices in Grade 1 and 3
Our first aim was to examine what kinds of profiles of teaching practices can be
identified among first and third grade teachers. In line with Kikas et al. (2016) and Rasku-
Puttonen et al. (2011), we also identified four profiles in grade 1: the child-centred style,
teacher-directed style, child-dominated style and mixture teaching style. This result was
consistent with Hypothesis 1. In grade 3, we found one additional profile, namely the
extreme-child-centred style. This profile differed from the profile of the child-centred style in
that the teachers in the profile of the extreme-child-centred style had very high scores for
child-centred practices but rather low scores for teacher-directed practices.
The results also showed that, in both grades, the child-centred style (if counted
together with the extreme-child-centred style) represented the largest group of teachers,
almost half of all participants. These findings show good fit with the recent core curricula and
teacher education practices implemented in both Finland and Estonia (Kikas & Lerkkanen,
2011; Sahlberg, 2011; Vitikka, Krokfors, & Hurmerinta, 2012). Both national core curricula

Teaching Practices and Reading Skills in Primary School 25
emphasise teacher sensitivity to students’ individual differences in competence and interests,
regard for students’ perspectives, a warm and supportive classroom climate, and the
importance of collaboration and interaction in the classroom (Finnish National Board of
Education, 2014; Vabariigi Valitsus, 2011/2014). Moreover, pre-service teacher training in
both countries reflects the constructivist theory of learning and teaching. For these reasons, it
is understandable that most of the teachers were found to deploy child-centred teaching
practices. However, the emergence of the extreme-child-centred style in grade 3 may reflect
the fact that some teachers greatly emphasise child-centred practices over other teaching
practices.
In addition, the results showed that about one-third of the teachers deployed the
mixture teaching style in grades 1 and 3. This result is in line with previous studies conducted
in preschool and kindergarten (Rasku-Puttonen et al., 2011), and in primary school (Kikas et
al., 2016). Both studies found that teachers in the mixture teaching style profile represented
about 30% of all teachers. The present results correspond with the proposition by Pressley et
al. (2003) that, in an authentic classroom, teaching practices may involve both direct
transmission and constructivist elements; this they called ‘balanced teaching’.
Differences in the Profiles of Teaching Practices between Finland and Estonia
Our second aim was to compare Finnish and Estonian teachers in their teaching styles.
The results partly supported Hypothesis 2 in that some of the Estonian teachers, but none of
the Finnish first grade teachers, deployed the child-dominated style in grade 1. In grade 3,
however, the number of Finnish and Estonian teachers who used the child-dominated style
did not differ significantly. One reason for the difference in grade 1 may be related to the fact
that Estonian children are taught decoding in kindergarten (Soodla et al., 2015) whereas
systematic reading instruction in Finnish schools begins in the first grade. It is likely that
Estonian children need less support than Finnish children in acquiring basic reading skills in

Teaching Practices and Reading Skills in Primary School 26
grade 1, as Estonian children have already mastered these skills before entering primary
school. Therefore, the use of the child-dominated style by some Estonian teachers might be
related to the children’s higher level of reading skills in the Estonian first-grade classrooms
(see also Kikas et al., 2014). Later, in grade 3, when fluent reading skill has been acquired in
both countries, the average support needed in the classroom by Finnish and Estonian students
might have become similar.
In grade 3, between-country differences emerged in the use of the child-centred and
extreme-child-centred styles: more Finnish teachers deployed the child-centred style, whereas
more Estonian teachers deployed the extreme-child-centred style. Together with the finding
that Estonian teachers used the child-dominated style in grade 1, it is possible that the
Estonian teachers may have somehow misinterpreted the meaning of child-centred practices.
Given that they had mainly experienced teacher-directed practices during their own school
days, the Estonian teachers might have interpreted child-centredness as letting children
decide for themselves how to study and thus as not setting children any limits or providing
them with sufficient guidance. The Estonian teachers may also exaggerate the benefits of
constructivist practices and the importance of granting children autonomy, and in
consequence use more often extreme child-centred practices. By contrast, the Finnish
teachers tend to be more flexible in their classroom practices.
Teaching Styles and Reading Skills
The third aim of this study was to examine the extent to which children’s reading
fluency and comprehension in grade 1 and grade 3 differed according to the teaching styles
employed by their teachers. After controlling for child age and gender, previous reading skills
and mother’s level of education, the results showed that, in grade 1, children who were in
classrooms characterised by the child-centred style showed the highest performance in
reading fluency and reading comprehension, whereas children whose teachers applied the

Teaching Practices and Reading Skills in Primary School 27
child-dominated style had the lowest reading performance. In grade 3, children whose
teachers deployed either the child-centred style or mixture teaching style performed better on
reading fluency and reading comprehension than those taught by teachers who used the
child-dominated or teacher-directed styles. Overall, these results partly supported Hypotheses
3a, 3b and 3c. They are also in line with previous studies on the role of teaching practices
(Block et al., 2009; Lerkkanen et al., 2016; Perry et al., 2007) that have found constructivist
teaching to be associated with high levels of reading fluency in the first grade (Lerkkanen et
al., 2016; Perry et al., 2007) and reading comprehension (Block et al., 2009) among lower
primary students. However, children whose teachers deployed the extreme-child-centred style
showed lower reading fluency than those in the child-centred style classroom in grade 3;
however, no differences between the two groups were found in reading comprehension.
These results suggest that, although the child-centred style teaching has, overall, stronger
associations with children’s reading performance than the other styles, particularly the child-
dominated style, the extreme end of this style is not further positively related to children’s
reading performance. It should be noted that the teachers in the profile characterised by the
extreme-child-centred style placed strong emphasis on child-centred practices while scoring
lower in the other practices, whereas the teachers in the profile characterised by the child-
centred style, had medium scores in child-centred practices along with low to medium scores
in the other practices. One explanation for the absence of an association between reading
performance and the extreme-child-centred style may be that the teachers who employed this
style also showed a very low level of teacher-directed practices. In other words, some use of
teacher-directed practices might be beneficial for children in the early phases of reading skills
development. Overall, our results suggest that a combination of practices or balanced
teaching practices has stronger associations with children’s reading skill development than an
extreme emphasis on specific teaching practices. It should also be noted that the associations

Teaching Practices and Reading Skills in Primary School 28
between teaching practices and reading skills may be also highly dependent on children’s
skill level, because teachers need to adapt their practices according to the child’s individual
needs in the classrooms. However, further studies are needed to investigate this in more
detail.
Our results also showed that children who were taught by teachers with the child-
dominated style had the lowest reading fluency and reading comprehension skills in both
grades. Previous research has demonstrated that instruction with minimal guidance often fails
to produce effective learning (for a review, see Kirschner, Sweller, & Clark, 2006). These
results are understandable in that the teacher’s role in child-dominated practices is minimal
when compared to child-centred and teacher-directed practices. The teachers using the child-
dominated style tend to give their students full autonomy; presumably, they believed that
students can construct knowledge by themselves and study by following their own interests
(Kikas et al., 2016; Stipek & Byler, 2005). However, child-dominated teaching practices do
not seem to recognise children’s individual needs and individual differences in learning and
skills development (Kirschner et al., 2006), and teachers using these practices do not adapt
their instruction or support to meet the needs of each child in the classroom.
Our results also showed some differences between grades in the associations between
the teacher-directed style and reading skills. In grade 1, the performance in reading fluency
and reading comprehension of children taught by teachers using the teacher-directed style fell
in between that of the children taught by teachers using the child-centred style and that of the
children taught with the child-dominated style, whereas in grade 3, these children showed the
poorest level of reading fluency. Previous studies have shown that teacher-directed practices
are detrimental for early learners’ motivation (Lerkkanen, Kiuru, et al., 2012; Stipek et al.,
1998, 1995). Since the development of reading fluency is strongly linked with reading
motivation—the more you read, the more fluent you become—it is understandable that

Teaching Practices and Reading Skills in Primary School 29
children who were taught by teachers using the teacher-directed style showed lower reading
fluency in grade 3. This finding implies that although teacher-directed teaching may predict
children’s basic reading skills, such as decoding in the early grades, this style of teaching
does not lead to improvements in reading fluency later on. Moreover, the results of the
present study did not support Hypotheses 3a and 3c that children taught with the teacher-
directed style would show better reading fluency in grade 1 than children taught with the
child-dominated style. One reason may be due to the orthographic transparency of both
Finnish and Estonian languages. It is plausible that in countries with transparent language
children learn to read easier and faster without too much emphasis on drills and practices of
decoding and spelling that typify teacher-directed teaching practices.
The results showed further that the children whose teachers employed the mixture
teaching style had the highest reading skills scores in grade 3 despite showing an in-between
level of performance in grade 1. It may be that combining the potential benefits for
motivation and skills of constructivist practices and didactic practices can provide flexible
and effective support for children’s self-regulation and behaviour and yet still create a warm
classroom atmosphere and climate. One possible explanation for these results is that such a
mixture of teaching practices supports autonomy while also providing children with structure.
For example, Jang et al. (2010) found that teacher provision of structure and support for
autonomy was the most beneficial combination for promoting students’ engagement in
learning activities. As little teaching observation research has examined the use of mixtures or
combinations of teaching practices, more studies on the topic are needed.
Limitations
The present study has some limitations that need to be considered before generalising
the research findings. First, the sample of teachers was relatively small (91 in grade 1 and 70
in grade 3). This might have limited the power of the statistical tests. Second, for practical

Teaching Practices and Reading Skills in Primary School 30
reasons, we controlled for reading fluency when we investigated reading comprehension.
Although reading fluency typically predicts reading comprehension, the ideal situation would
nevertheless be to use identical measures as a covariate. Moreover, because identical
measures had not been used to assess grade 1 reading skills in the Finnish and Estonian grade
3 samples, the grade 1 reading scores of these two samples were standardized to compute a
composite score, which was then used as a covariate. Third, the datasets used in the grade
comparisons were cross-sectional. Fourth, teaching practices were observed over three
lessons, including at least one literacy lesson but also other subjects. Therefore, we cannot
draw any detailed conclusions on specific literacy instruction practices or the effectiveness of
methods. However, Stipek and Byler (2004) have suggested that the ECCOM focuses on the
general characteristics of teaching practices in the classroom during the school day rather
than on subject matter contents.
Conclusions and Implications for Teacher Training
By using a person-oriented approach, the present study identified different teaching
styles, i.e., patterns of teaching practices, among Finnish and Estonian primary school
teachers. The results also suggest that both a child-centred teaching style and a teaching style
including both child-centred and teacher-directed elements were typical among teachers of
students whose reading performance developed better, whereas the child-dominated style and
an extreme form of child-centred style were associated with a less positive development of
children’s reading performance. The results indicate that teaching styles which require
teachers to take a more active role benefit children’s reading skills during the early years of
primary school. The associations between the teacher-directed style and reading skills were
weaker in the third grade than in the first grade, whereas the mixture teaching style showed
stronger associations with reading skills in the third grade.

Teaching Practices and Reading Skills in Primary School 31
Our findings have also some implications for teacher training. First, teacher training
needs to be more precise and give concrete examples of the differences between teaching
practices and their specific benefits. This seems to be especially important in the case of
child-centred practices vs. extreme-child-centred practices. The extreme form of child-
centred practices does not seem to provide any additional benefits for children’s reading
fluency. Second, more effort is needed to introduce the possible negative consequences of the
child-dominated style. Although the idea of giving the child full autonomy seems appealing to
some teachers, our results showed that this teaching style is detrimental for reading skills,
especially at the lower primary school level. Third, the promising finding from our analysis
on the use of the mixture teaching style in grade 3 should be further studied and understood to
encourage teachers to use a flexible repertoire of teaching practices suited to their learners’
needs.
References
Adams, G., & Carnine, D. (2003). Direct instruction. In H. L. Swanson, K. R. Harris, & S.
Graham (Eds.), Handbook of learning disabilities (pp. 403–416). New York: Guilford
Press.
Adams, M. J. (1990). Beginning to read : thinking and learning about print. MIT Press.
Block, C. C., Parris, S. R., Reed, K. L., Whiteley, C. S., & Cleveland, M. D. (2009).
Instructional approaches that significantly increase reading comprehension. Journal of
Educational Psychology, 101(2), 262–281. https://doi.org/10.1037/a0014319
Celeux, G., & Soromenho, G. (1996). An entropy criterion for assessing the number of
clusters in a mixture model. Journal of Classification, 13(2), 195–212.
https://doi.org/10.1007/BF01246098
Connor, C. M., Son, S. H., Hindman, A. H., & Morrison, F. J. (2005). Teacher qualifications,
classroom practices, family characteristics, and preschool experience: Complex effects

Teaching Practices and Reading Skills in Primary School 32
on first graders’ vocabulary and early reading outcomes. Journal of School Psychology,
43, 343–375. https://doi.org/10.1016/j.jsp.2005.06.001
Early, D. M., Maxwell, K. L., Burchinal, M., Alva, S., Bender, R. H., Bryant, D., … Zill, N.
(2007). Teachers’ education, classroom quality, and young children’s academic skills:
results from seven studies of preschool programs. Child Development, 78(2), 558–80.
https://doi.org/10.1111/j.1467-8624.2007.01014.x
Finnish National Board of Education. (2004). National Core Curriculum for Basic Education.
Retrieved from
http://www.oph.fi/english/curricula_and_qualifications/basic_education/curricula_2004
Hallgren, K. A. (2012). Computing inter-rater reliability for observational data: An overview
and tutorial. Tutorials in Quantitative Methods for Psychology, 8(1), 23–34.
Hamre, B. K., & Pianta, R. C. (2005). Can instructional and emotional support in the first-
grade classroom make a difference for children at risk of school failure? Child
Development, 76(5), 949–67. https://doi.org/10.1111/j.1467-8624.2005.00889.x
Jang, H., Reeve, J., & Deci, E. L. (2010). Engaging students in learning activities: It is not
autonomy support or structure but autonomy support and structure. Journal of
Educational Psychology, 102(3), 588–600. https://doi.org/10.1037/a0019682
Kikas, E., & Lerkkanen, M.-K. (2011). Education in Estonia and Finland. In M. Veisson, E.
Hujala, P. K. Smith, M. Waniganayake, & E. Kikas (Eds.), Global perspectives in early
childhood education: Diversity, challenges and possibilities (pp. 33–46). Frankfurt am
Main: Peter Lang.
Kikas, E., Peets, K., & Hodges, E. V. E. (2014). Collective student characteristics alter the
effects of teaching practices on academic outcomes. Journal of Applied Developmental
Psychology, 35(4), 273–283. https://doi.org/10.1016/j.appdev.2014.04.004

Teaching Practices and Reading Skills in Primary School 33
Kikas, E., Silinskas, G., Jõgi, A.-L., & Soodla, P. (2016). Effects of teacher’s individualized
support on children’s reading skills and interest in classrooms with different teaching
styles. Learning and Individual Differences. https://doi.org/10.1016/j.lindif.2016.05.015
Kirschner, P. A., Sweller, J., & Clark, R. E. (2006). Why Minimal Guidance During
Instruction Does Not Work: An Analysis of the Failure of Constructivist, Discovery,
Problem-Based, Experiential, and Inquiry-Based Teaching. Educational Psychologist,
41(2), 75–86. https://doi.org/10.1207/s15326985ep4102_1
Landerl, K., & Wimmer, H. (2008). Development of word reading fluency and spelling in a
consistent orthography: An 8-year follow-up. Journal of Educational Psychology,
100(1), 150–161. https://doi.org/10.1037/0022-0663.100.1.150
Lerkkanen, M.-K., Kikas, E., Pakarinen, E., Trossmann, K., Poikkeus, A.-M., Rasku-
Puttonen, H., … Nurmi, J.-E. (2012). A validation of the early childhood classroom
observation measure in Finnish and Estonian kindergartens. Early Education &
Development, 23(3), 323–350. https://doi.org/10.1080/10409289.2010.527222
Lerkkanen, M.-K., Kiuru, N., Pakarinen, E., Poikkeus, A.-M., Rasku-Puttonen, H., Siekkinen,
M., & Nurmi, J.-E. (2016). Child-centered versus teacher-directed teaching practices:
Associations with the development of academic skills in the first grade at school. Early
Childhood Research Quarterly, 36, 145–156.
https://doi.org/10.1016/j.ecresq.2015.12.023
Lerkkanen, M.-K., Kiuru, N., Pakarinen, E., Viljaranta, J., Poikkeus, A., Rasku-Puttonen, H.,
… Nurmi, J.-E. (2012). The role of teaching practices in the development of children’s
interest in reading and mathematics in kindergarten. Contemporary Educational
Psychology, 37(4), 266–279. https://doi.org/10.1016/j.cedpsych.2011.03.004

Teaching Practices and Reading Skills in Primary School 34
Lerkkanen, M.-K., Niemi, P., Poikkeus, A.-M., Poskiparta, M., Siekkinen, M., & Nurmi, J.-E.
(2006). The First Steps Study. Unpublished data, Department of Psychology and
Department of Teacher Education, University of Jyväskylä, Jyväskylä, Finland.
Lerkkanen, M.-K., Poikkeus, A.-M., & Ketonen, R. (2006). ARMI--Luku- ja kirjoitus- taidon
arviointimateriaali 1. luokalle [ARMI--A tool for assessing reading and writing writing
skills in grade 1]. Helsinki: WSOY.
Lindeman, J. (1998). ALLU - Ala-asteen lukutesti [Reading test for primary school]. Turku,
Finland: University of Turku.
Marcon, R. A. (1999). Differential impact of preschool models on development and early
learning of inner-city children: A three-cohort study. Developmental Psychology, 35(2),
358–375. https://doi.org/10.1037/0012-1649.35.2.358
Muthén, L. K., & Muthén, B. O. (2012). Mplus users guide and Mplus version 7.0. Retrieved
January 10, 2014, from http://www.statmodel.com/index.shtml
Nurmi, J.-E., Kiuru, N., Lerkkanen, M.-K., Niemi, P., Poikkeus, A.-M., Ahonen, T., … Lyyra,
A.-L. (2013). Teachers adapt their instruction in reading according to individual
children’s literacy skills. Learning and Individual Differences, 23, 72–79.
https://doi.org/10.1016/j.lindif.2012.07.012
Palu, A., & Kikas, E. (2007). Primary School Teachers’ Beliefs about the Purposes and
Methods of Teaching Mathematics. Nordic Studies in Mathematics Education, 12, 5–21.
Perry, K. E., Donohue, K. M., & Weinstein, R. S. (2007). Teaching practices and the
promotion of achievement and adjustment in first grade. Journal of School Psychology,
45(3), 269–292. https://doi.org/10.1016/j.jsp.2007.02.005
Piaget, J. (1985). The equilibration of cognitive structures: The central problem of
intellectual development. Chicago, IL: University of Chicago Press.

Teaching Practices and Reading Skills in Primary School 35
Pressley, M., Roehrig, A. D., Raphael, L., Dolezal, S., Bohn, C., Mohan, L., … Hogan, K.
(2003). Teaching processes in elementary and secondary education. In Handbook of
Psychology. John Wiley & Sons, Inc. https://doi.org/10.1002/0471264385.wei0708
Rasku-Puttonen, H., Pakarinen, E., Trossmann, K., Lerkkanen, M.-K., Kikas, E., & Poikkeus,
A.-M. (2011). Classroom Practices in Finnish and Estonian preschools: Subgroups of
Observed Teaching Practices. In M. Veisson, E. Hujala, P. K. Smith, M. Waniganayake,
& E. Kikas (Eds.), Global perspectives in early childhood education: Diversity,
challenges and possibilities (pp. 313–331). Frankfurt am Main: Peter Lang.
Ruus, V.-R., Henno, I., Eisenschmidt, E., Loogma, K., Noorväli, H., Reiska, P., & Rekkor, S.
(2008). Reforms, developments and trends in Estonian education during recent decades.
In J. Mikk, M. Veisson, & P. Luik (Eds.), Reforms and innovations in Estonian
education (pp. 11–26). Frankfurt am Main: Peter Lang Publishers House.
Sahlberg, P. (2011). Finnish Lessons: What Can the World Learn from Educational Change
in Finland? New York: Teacher College Press.
Seymour, P. H. K., Aro, M., & Erskine, J. M. (2003). Foundation literacy acquisition in
European orthographies. British Journal of Psychology, 94(2), 143–174.
https://doi.org/10.1348/000712603321661859
Soodla, P., Lerkkanen, M.-K., Niemi, P., Kikas, E., Silinskas, G., & Nurmi, J.-E. (2015). Does
early reading instruction promote the rate of acquisition? A comparison of two
transparent orthographies. Learning and Instruction, 38, 14–23.
https://doi.org/10.1016/j.learninstruc.2015.02.002
Stipek, D., & Byler, P. (2004). The early childhood classroom observation measure. Early
Childhood Research Quarterly, 19(3), 375–397.
https://doi.org/10.1016/j.ecresq.2004.07.007

Teaching Practices and Reading Skills in Primary School 36
Stipek, D., & Byler, P. (2005). The early childhood classroom observation measure.
Unpublished coding manual, School of Education, Stanford University.
Stipek, D., Daniels, D., Galluzzo, D., & Milburn, S. (1992). Characterizing early childhood
education programs for poor and middle-class children. Early Childhood Research
Quarterly, 7(1), 1–19. https://doi.org/10.1016/0885-2006(92)90015-Q
Stipek, D., Feiler, R., Byler, P., Ryan, R., Milburn, S., & Salmon, J. M. (1998). Good
beginnings: What difference does the program make in preparing young children for
school? Journal of Applied Developmental Psychology, 19(1), 41–66.
https://doi.org/10.1016/S0193-3973(99)80027-6
Stipek, D., Feiler, R., Daniels, D., & Milburn, S. (1995). Effects of different instructional
approaches on young children’s achievement and motivation. Child Development, 66(1),
209–223. https://doi.org/10.2307/1131201
Stipek, D., Givvin, K., Salmon, J., & MacGyvers, V. (2001). Teachers’ beliefs and practices
related to mathematics instruction. Teaching and Teacher Education, 17(2), 213–226.
https://doi.org/10.1016/S0742-051X(00)00052-4
Tang, X., Pakarinen, E., Lerkkanen, M.-K., Kikas, E., Muotka, J., & Nurmi, J.-E. (2016).
Validating the early childhood classroom observation measure in first and third grade
classrooms. Scandinavian Journal of Educational Research, 1–20.
https://doi.org/10.1080/00313831.2015.1120237
Tulviste, T., & Kikas, E. (2010). Qualities to be developed in children at home and at school:
A comparison of the views of Estonian mothers, fathers and teachers. Journal of Applied
Developmental Psychology, 31, 315–321.
Vabariigi Valitsus [Government of the Estonian Republic]. (2008). Koolieelse lasteasutuse
riiklik õppekava [National Curriculum for Preschool Education]. Retrieved from
https://www.riigiteataja.ee/akt/13351772

Teaching Practices and Reading Skills in Primary School 37
Vabariigi Valitsus [Government of the Estonian Republic]. (2011). Põhikooli riiklik
õppekava[National curriculum for basic schools]. Retrieved from
https://www.riigiteataja.ee/akt/128082013007
Vitikka, E., Krokfors, L., & Hurmerinta, E. (2012). The Finnish National Core Curriculum:
Structure and Development. In H. Niemi, A. Toom, & A. Kallioniemi (Eds.), Miracle of
Education (pp. 83–96). Rotterdam: Sense. Retrieved from
http://curriculumredesign.org/wp-content/uploads/The-Finnish-National-Core-
Curriculum_Vitikka-et-al.-2011.pdf
Vygotsky, L. S. (1978). Mind in society: The development of higher psychological processes.
Cambridge, MA: Harvard University Press.

Teaching Practices and Reading Skills in Primary School 38
Table1
Description of the teaching practices, subscales, and items used in the ECCOM
Subscale and item
Teaching Practices
Child-Centred Teacher-Directed Child-Dominated
Management
1. Child Responsibility Children are allowed to take
responsibility to the degree that they
are able.
Children are not given opportunities
to take responsibility (teacher
control).
No one seems to take responsibility
for maintaining an orderly
environment.
2. Management Teacher has clear but somewhat
flexible classroom rules and
routines.
Teacher has clearly communicated
expectations and classroom rules
that are rigidly adhered to.
There are no clearly defined
expectations or rules. The classroom
is chaotic.
3. Choice of Activities There is a mixture of teacher and
child choice.
Teacher makes most of the choices. Children make most of the choices.
4. Discipline Strategies Conflict resolution is smooth;
consequences are appropriate and
apply equally.
Discipline is imposed without
explanation or discussion;
consequences are inconsistent.
Teacher rarely disciplines.
Climate
5. Support for Communication
Skills
Teacher encourages children to
engage in conversation and elaborate
on their thoughts.
Teacher does not encourage children
to engage in conversation (teacher-
controlled conversation).
Teacher does not engage children in
interactive conversation.
6. Support for Interpersonal Skills Teacher provides opportunities for
cooperative, small-group activities
that promote peer interactions.
Teacher does not provide
opportunities for children to develop
interpersonal skills.
There are opportunities but no
support for the development of
children’s interpersonal skills.
7. Student Engagement Teacher attempts to engage all
children in ways that will improve
their skills and understanding.
Teacher engages children in rote
activities (e.g., rigid expectations
about being engaged in work).
Teacher makes no systematic effort
to engage children in productive
activity.
8. Individualisation of Learning
Activities
Teacher is attentive to children’s
individual skill levels and adapts
tasks accordingly.
Tasks are not flexible or adapted to
children’s individual needs (e.g., all
do the same tasks).
Teacher does not address children’s
individual needs.
(Continued)
Table1. Continued

Teaching Practices and Reading Skills in Primary School 39
Subscale and item
Teaching Practices
Child-Centred Teacher-Directed Child-Dominated
Instruction
9. Learning Standards Teacher holds children accountable
for attaining some individualised
standard (assists and challenges
children at their respective levels).
Teacher rigidly holds children
accountable for completing work
and for attaining a universal standard
(e.g., standards are rigid and
invariable).
Teacher does not hold children
accountable for completing work
and has no apparent standards.
10. Coherence of Instructional
Activities
There are connections between and
within academic lessons
(concepts/skills are embedded into a
broader set of goals).
Academic lessons are distinct and
disconnected (concepts/skills are
presented as an isolated set of facts
or skills to be learned).
Lessons are disjointed and the focus
is unclear (connections are on a
superficial level with no unifying
concept).
11. Teaching Concepts Tasks and lessons are designed to
teach identifiable concepts and
develop understanding.
Tasks are designed to help children
learn facts or procedures. Problem
solving is constrained.
The specific concept of tasks is
unclear.
12. Instructional Conversation Teacher solicits children’s questions,
ideas, solutions or interpretations
around a clearly defined topic.
Teacher dominates instructional
conversation; children’s participation
is limited.
Teacher does not engage in
instructional conversations with
children, or topics are unfocused or
unclear.
13. Literacy Instruction Teacher provides a broad array of
literacy experiences and
instructional practices.
Teacher’s literacy instruction places
a heavy emphasis on phonics and
paper-and-pencil tasks.
Teacher provides no instruction on
phonics or reading comprehension
strategies.
14. Math Instruction Math instruction emphasises
developing understanding.
Math instruction emphasises rote
memorisation and drill and practice.
There is little evidence of math
instruction or conversation about
math concepts.
Notes: Based on Stipek and Byler (2005). Observers rate classrooms on each of the14 scale items, giving one code for Child-Centred, one code for Teacher-
Directed and one code for Child-Dominated. All items are rated on a scale of 1 to 5 (1=these practices are rarely seen, less than 20% of the time; 5=these
practices predominate, 80%–100% of the time).

Teaching Practices and Reading Skills in Primary School 40
Table 2
Indices for Mixture Models with Different Numbers of Latent Classes, Grade 1
Clas
s
log L BIC ABIC AIC VLMR LMR Entropy N
1-335.72 698.50 679.56 683.43 1.00 91
2-290.04 625.18 593.62 600.07 .00 .01 .98 11/80
3-247.42 558.00 513.81 522.85 .04 .05 .89 11/29/51
4 -228.00 537.20 480.39 492.00 .02 .02 .91 39/10/10/32
5-217.03 533.31 463.87 478.87 .44 .46 .92 10/30/38/4/9
6-207.84 532.96 450.90 467.68 .35 .37 .89 4/30/24/10/14/9
Notes: log L = Log-likelihood; BIC = Bayesian Information Criterion; ABIC = Adjusted
Bayesian Information Criterion; AIC = Akaike Information Criterion; VLMR = Vuong–Lo–
Mendell–Rubin test p-value; LMR = Lo–Mendell–Rubin test p-value;

Teaching Practices and Reading Skills in Primary School 41
Table 3
Descriptive Statistics of the Four Profiles of Teaching Styles in Grade 1
Child-
centred
Style
M(SD)
Teacher-
directed
Style
M(SD)
Child-
dominated
Style
M(SD)
Mixture
Teaching Style
M(SD)
F value
N39 10 10 32
Percentage 43% 11% 11% 35%
Score of Child-
centred Practices
3.71a (.36) 1.59c
(.28)
1.98c (.34) 2.66b (.31) 161.97***
Score of Teacher-
directed Practices
1.87c (.49) 4.13a
(.50)
2.50b (.57) 2.91b (.43) 67.80***
Score of Child-
dominated
Practices
1.18c (.28) 1.43bc
(.43)
3.54a (.55) 1.45b (.35) 116.89***
Classroom size 18.92
(5.24)
19.55
(5.36)
20.89
(5.09)
19.91 (4.74) ns
Teacher’s age 43.33
(9.36)
46.8
(11.82)
37.2
(10.16)
39.94 (10.93) ns
Teaching
experience
Med = 5 Med = 5 Med = 4 Med = 4 ns
Less than a
year
5.3% 10% 20% 9.7%
1–5 years 5.3% 10% 20% 12.9%
6–10 years 13.2% - 10% 16.1%
11–15 years 13.2% 10% - 16.1%
More than 15
years
63.2% 70% 50% 45.2%
Notes. Pairs with the same subscript letters do not differ significantly (p>.05) based on
ANOVA post-hoc comparisons. Tamhane's T2 was used when variances were not equal; in
other cases, Bonferroni post hoc comparisons were used. Med = Median.

Teaching Practices and Reading Skills in Primary School 42
Table 4
Indices for Mixture Models with Different Numbers of Latent Classes, Grade 3
Class log L BIC ABIC AIC VLMR LMR Entropy N
1 -252.60 530.68 511.68 517.78 70
2 -210.46 463.41 431.91 440.93 .05 .05 .99 9/61
3 -182.06 423.61 379.51 392.13 .07 .08 .89 33/7/30
4 -167.16 410.79 354.09 370.32 .55 .57 .86 19/28/16/7
5 -152.65 398.77 329.47 349.30 .03 .04 .93 23/11/9/20/7
6 -142.91 396.27 314.38 337.81 .23 .25 .94 23/20/9/7/2/9
7 -133.22 393.90 299.40 326.44 .09 .10 .94 6/9/23/20/9/1/2

Teaching Practices and Reading Skills in Primary School 43
Table 5
Descriptive Statistics of the Five Profiles of Teaching Styles in Grade 3
Extreme
Child-
centred
Style
M(SD)
Child-
centred
Style
M(SD)
Teacher-
directed
Style
M(SD)
Child-
dominate
d Style
M(SD)
Mixture
Teaching
Style
M(SD)
F values
N9 20 11 7 23
Percentage 13% 30% 15% 10% 32%
Score of Child-
centred
Practices
4.65a (.31) 3.53b (.31) 1.68e
(.27)
2.13d (.31) 2.68c (.29) 159.28***
Score of
Teacher-
directed
1.31d (.28) 1.98c (.29) 4.02a
(.46)
2.09bcd
(.82)
2.94b (.42) 64.49***
Score of Child-
dominated
1.04b (.13) 1.12b (.15) 1.37b
(.39)
2.95a (.37) 1.19b (.22) 82.68***
Classroom size 15.66
(5.78)
18.8 (5.54) 17.18
(5.15)
16.29
(2.69)
19.78 (5.76) ns
Teacher’s age 44.11
(13.10)
43.10 (9.30) 47.18
(9.69)
42.00
(9.51)
43.48
(10.07)
ns
Teaching
experience
Med =5 Med =4.5 Med =5 Med=4.5 Med=4.5 ns
Less than a
year
- - - - -
1–5 years 11.1% 25% 9.1% - 9.1%
6–10 years 22.2% 15% 9.1% 33.3% 27.3%
11–15 years - 10% - 16.7% 13.6%
More than
15 years
66.7% 50% 81.8% 50% 50.0%
Notes. The pairs with the same subscript letters do not differ significantly (p>.05) based on
ANOVA post-hoc comparisons. Tamhane's T2 was used when variances were not equal; in
other cases, Bonferroni post hoc comparisons were used. Med = Median.

Teaching Practices and Reading Skills in Primary School 44
Figure 1. Differences between the teaching styles on reading fluency and comprehension in
Grade 1 and Grade 3.
Notes. Pairs with the same subscript letters do not differ significantly (p>.05) based on post-
hoc comparisons; extr-cc = extreme child-centred style; cc = child-centred style; cd = child-
dominated style; td = teacher-directed style; mix = mixture of child-centred and teacher-
directed styles.

Teaching Practices and Reading Skills in Primary School 45
Appendix. Description of identical measures in reading skills
Identical measures of reading skills in ANCOVAs
Finn FS EST-READ EST-KISS
1st grade
fall
reading fluency reading fluency
1st grade
spring
reading fluency
reading comprehension
reading fluency
reading comprehension
3rd grade
spring
reading fluency
reading comprehension
reading fluency
reading comprehension