Early Childhood Research Quarterly 43 (2018) 62–72

Contents lists available at ScienceDirect

Early Childhood Research Quarterly

Individual differences in basic arithmetic skills in children with and

without developmental language disorder: Role of home numeracy
experiences
Tijs Kleemans ∗ , Eliane Segers, Ludo Verhoeven
Behavioural Science Institute, Radboud University, Montessorilaan 3, P.O. Box 9104, 6500 HE Nijmegen, The Netherlands

a r t i c l e i n f o a b s t r a c t

Article history: The present study investigated the role of kindergarten home numeracy experiences in predicting indi-
Received 12 May 2017 vidual differences in second-grade basic arithmetic (i.e., small problem sizes and large problem sizes)
Received in revised form among children with developmental language disorders (DLD) and their peers with normal language
23 November 2017
achievement (NLA), after controlling for kindergarten cognitive and linguistic capacities and first-grade
Accepted 17 January 2018
Available online 9 February 2018
basic arithmetic. Forty children with DLD and 103 children with NLA were tested on cognitive, linguistic,
and basic arithmetic skills, and their parents filled in a questionnaire on home numeracy activities and
numeracy expectations. The results showed that children with DLD scored below their peers with NLA on
Keywords:
Basic arithmetic skills basic arithmetic skills, with larger group differences on arithmetic with large problem sizes than small
Home numeracy experiences problem sizes. Furthermore, for both groups of children, home numeracy experiences were positively
Parent–child activities related to arithmetic with large, but not small problem sizes, suggesting that the role of home numeracy
Parents’ expectations in basic arithmetic might be restricted to those arithmetic skills that children have not fully mastered
Developmental language disorder yet.
© 2018 Elsevier Inc. All rights reserved.

1. Introduction
Experiences at home impact children’s academic abilities. This
goes for both home literacy experiences (Sénéchal & LeFevre, 2014)
as well as home numeracy experiences (Skwarchuk, Sowinski, &
LeFevre, 2014). Children with learning problems may have a dual
risk; besides having a learning problem, their home environment
(in terms of informal learning experiences) is often below that
of children without learning problems (Martin, Volkmar, & Lewis,
2007). With regard to the development of numeracy, children with
developmental language disorder (DLD) form an interesting group.
These children show significant delays within the domains of recep-
tive and/or expressive language, as well as phonological working
memory skills (Bishop, Snowling, Thompson, & Greenhalgh, 2016),
which hinders them to form adequate representations of count-
ing words and number facts (Fazio, 1994, 1996). As a consequence,
they also lag behind in early numeracy and basic arithmetic skills
(e.g., Durkin, Mok, & Conti-Ramsden, 2013). Furthermore, the home
numeracy experiences of children with DLD have been found to
be poorer than that of normal language achieving (NLA) children,
∗ Corresponding author.
E-mail address: m.kleemans@bsi.ru.nl (T. Kleemans).
and this has been found to impact their basic arithmetic skills in
first grade (Kleemans, Segers, & Verhoeven, 2013). In second grade,
arithmetic starts to rely more on phonological working memory
as well as language-based fact retrieval, and the impact of lin-
guistic skills on arithmetic in this phase is well established (e.g.,
Durkin et al., 2013; LeFevre, Fast et al., 2010). It remains unclear,
however, to what extent home numeracy experiences continue to
play a role in predicting individual differences in basic arithmetic
skills in second grade. In the present study, we investigated this in
a longitudinal design, in which children with DLD and NLA were
compared.
1.1. Child factors of basic arithmetic skills
Basic arithmetic skills comprise the addition and subtraction
of single digits (i.e., 1–9) in first and second grades, and are pre-
ceded by the acquisition of early numeracy skills in kindergarten
(Desoete & Grégoire, 2006; Jordan, Kaplan, Locuniak, & Ramineni,
2007). Evidence from neurocognitive research suggests that a dis-
tinction can be made between arithmetic with small problem sizes
(i.e., arithmetic with sums and minuends below 10) and arithmetic
with large problem sizes (i.e., arithmetic with sums and minuends
above 10 and less than 20, including carryover operations) (e.g.,
Stanescu-Cosson et al., 2000). The solution process of arithmetic

https://doi.org/10.1016/j.ecresq.2018.01.005
0885-2006/© 2018 Elsevier Inc. All rights reserved.
 T. Kleemans et al. / Early Childhood Research Quarterly 43 (2018) 62–72
with small problem sizes relies on having access to arithmetic facts
that are stored in long-term memory (De Smedt, Taylor, Archibald,
& Ansari, 2010). For arithmetic with large problem sizes, phono-
logical working memory (i.e., phonological loop) is required as
well, as such problems exceed the 10, and a solution procedure
needs to be carried out before the correct answer can be given (e.g.,
7 + 5 = 7 + 3 = 10 + 2 = 12). This solution procedure relies on verbal
strategies that may vary from using verbal counting (e.g., seven
plus three is “seven,”,’ “eight,” “nine,” “ten”) to temporarily main-
taining arithmetic facts during the solution process (e.g., seven plus
three is ten), which are both assumed to rely on phonological codes
(Dehaene, Piazza, Pinel, & Cohen, 2003) and are supported by the
phonological loop (Lee & Kang, 2002).
From the literature on individual differences in basic arithmetic
skills in children with DLD and NLA, a theoretical framework can
be derived that includes cognitive as well as linguistic child factors
that influence the development of these skills (see e.g., Kleemans,
Segers, & Verhoeven, 2016; LeFevre, Fast et al., 2010). To begin
with, both general intelligence and working memory as cogni-
tive factors have been found to predict the development of basic
arithmetic skills. The role of general intelligence in basic arith-
metic skills has been identified in both children with DLD (Cowan
et al., 2005) and children with NLA (De Smedt et al., 2009; Stock,
Desoete, & Roeyers, 2009). For working memory, three related
systems can be identified (Baddeley, 1996): one is a primary sys-
tem (i.e., central executive functioning) that is responsible for the
control and regulation of cognitive processes, and two slave sys-
tems that are involved in processing verbal (i.e., phonological loop)
and visual–spatial information (i.e., visual–spatial sketchpad). In
a meta-analysis by Friso-van den Bos, Kroesbergen, Van der Ven,
and Van Luit (2013), it was found that the central executive (as
measured by a backward digit span) was a strong predictor of math-
ematical achievement throughout primary school, whereas the role
of the slave systems was related to children’s strategy use. To be
more precise, the visual–spatial sketchpad was found to be asso-
ciated with finger counting (Rasmussen & Bisanz, 2005), whereas
the phonological loop was found to be related to more mature
strategies that relied on verbal counting or fact retrieval (De Smedt,
Holloway, & Ansari, 2011; Hecht, Torgesen, Wagner, & Rashotte,
2001; Siegler, 1996).
Next to cognitive factors, linguistic factors such as phonologi-
cal awareness and grammatical ability also predict the acquisition
of basic arithmetic. Phonological awareness was uniquely related
to small additions and small subtractions in both children with
NLA (Hecht et al., 2001; Simmons & Singleton, 2008) and children
with DLD (Kleemans, Segers, & Verhoeven, 2012), as the quality of
verbal codes (e.g., counting words) relies on phonological repre-
sentations that are stored in long-term memory (De Smedt et al.,
2010). Furthermore, grammatical ability and basic arithmetic skills
are based on the same underlying rules and develop on the princi-
ple of recursion (e.g., Hauser, Chomsky, & Fitch, 2002). For example,
the order of words in a sentence determines its meaning (“The boy
chased the dog” vs. “The dog chased the boy”), which is also the
case in the order of the numbers and operations in an arithmetical
problem (“15 − 8 = ” vs. “8 − 15 = ”). Grammatical ability has been
found to predict single-digit addition and subtraction problems
(i.e., small and large problem sizes) in children with DLD (Cowan
et al., 2005; Durkin et al., 2013) and children with NLA (Kleemans
et al., 2014). Finally, naming speed as an additional child factor
should be taken into consideration as well, because the solution
process of arithmetic in second grade requires fast retrieval of lin-
guistically encoded arithmetic facts (Koponen, Mononen, Rasanen,
& Ahonen, 2006). Interestingly, naming speed differentially pre-
dicted the numerical skills of children with DLD in kindergarten as
well as first grade (Kleemans, Segers, & Verhoeven, 2011; Kleemans
63
et al., 2012); a better naming speed might reduce the problems in
basic arithmetic skills.
1.2. Home factors of basic arithmetic skills
Not only child factors predict learning outcomes in the early
years of primary education, but individual differences in the home
environment are also related to academic success (Melhuish et al.,
2008). It is well established that home literacy experiences are asso-
ciated with children’s literacy skills (see e.g., Sénéchal & LeFevre,
2014; Sénéchal & Young, 2008). In a similar way, home numer-
acy experiences relate to children’s numeracy skills (Blevins-Knabe,
2016; Huntsinger, Jose, & Luo, 2016; Skwarchuk et al., 2014). Home
numeracy experiences can be seen as a multi-componential, but
interrelated construct that facilitate numeracy learning. The two
main components of home numeracy experiences are parent–child
numeracy activities and parents’ numeracy expectations.
With respect to parent–child numeracy activities, LeFevre,
Clarke, and Stringer (2002) found the frequency of parent–child
numeracy activities to be positively related with the counting
skills of French- and English-speaking Canadian children. Compa-
rable results have been found in other cultures (e.g., Turkish- and
Chinese-speaking children) as well (Cankaya & LeFevre, 2016): the
higher the frequency of parent–child numeracy activities, the better
the numeracy skills of the children. Furthermore, Skwarchuk (2009)
showed that parents’ involvement in activities with basic and com-
plex numeracy goals was positively related to children’s early
numeracy skills. Longitudinal relationships between parent–child
numeracy activities and later basic arithmetic have also been
reported. To begin with, in a sample of 609 German kinder-
gartners, Niklas and Schneider (2013) showed that the reported
frequency of parental engagement in numeracy activities at home
was positively associated with basic arithmetic skills one year
later. Furthermore, Huntsinger et al. (2000) reported the frequency
of parent–child numeracy activities to be positively related to
later basic arithmetic. LeFevre et al. (2009) found not only simi-
lar results, but also controlled for cognitive (i.e., spatial memory)
and linguistic (i.e., vocabulary) child factors. Although most studies
found positive relationships between the frequency of parent–child
numeracy activities and basic arithmetic skills, negative relation-
ships have been reported in both cross-sectional (Blevins-Knabe
& Musun-Miller, 1996) and longitudinal studies (Ciping, Silisinkas,
Wei, & Georgiou, 2015) as well. These negative relationships were
explained by the fact that when children are starting to lag behind at
school, parents may increase the frequency of numeracy activities
in the home (Ciping et al., 2015).
With respect to parents’ numeracy expectations, it was found
that higher expectations on the part of the parent resulted in more
numeracy-related practices, which was related to better numeracy
achievement on the part of the child (LeFevre, Polyzoi, Skwarchuk,
Fast, & Sowinksi, 2010). A more recent line of research combined
both components of home numeracy experiences to determine
the unique role of each factor in predicting numerical abilities.
Cross-sectional evidence has been provided by Kleemans, Peeters
et al. (2012), who found both aspects to be positively related with
the early numeracy skills of 89 typically developing kindergart-
ners, after controlling for cognitive and linguistic child factors.
In a longitudinal study by Skwarchuk et al. (2014), formal home
numeracy practices (e.g., the parent practicing simple sums) as well
as parents’ expectations were positively associated with symbolic
number knowledge in first grade. In addition, in children with DLD,
the role of home numeracy experiences in first-grade arithmetic
was evidenced: both parent–child numeracy activities and parents’
numeracy expectations were positively related to arithmetic with
small problem sizes in first grade, after controlling for cognitive and
linguistic child factors. Those relations were stronger for parents of
64 T. Kleemans et al. / Early Childhood Research Quarterly 43 (2018) 62–72
children with DLD, as compared to parents of children with NLA
(Kleemans et al., 2013).
1.3. The present study
The aim of the present study was to examine the role of
home numeracy experiences in predicting individual differences in
second-grade basic arithmetic in children with DLD and their peers
with NLA. The present study extended previous research in two
ways. First, it explored more in depth how children with DLD differ
with their peers with NLA in basic arithmetic skills in second grade.
Previous research on individual differences in basic arithmetic skills
children did not take into account the distinction between arith-
metic with small problem sizes and arithmetic with large problem
sizes. Given that second-grade arithmetic with large problem sizes
not only depends on language-based fact retrieval, but also on
phonological working memory (De Smedt et al., 2009), it can be
expected that, when compared to children with NLA, children with
DLD have even more problems in solving arithmetic with large
problem sizes. Second, the present study was a longitudinal study
on the role of home numeracy experiences, as measured in kinder-
garten, in second grade basic arithmetic, after taking into account
cognitive and linguistic child factors. From the research conducted
so far, it has become clear that both aspects of home numeracy
experiences are related to early numeracy skills in kindergarten
as well as basic arithmetic skills in first grade. However, the rel-
ative contribution of each of the child and home factors beyond
the start of formal arithmetic education in first grade is still unex-
plored. Moreover, when contrasted to typical populations, less is
known about the home numeracy experiences of atypical popula-
tions, which is important with an eye on early identification and
prevention of math problems in these groups (Kaufmann, 2008).
The question rises whether home numeracy experiences are also
related to second-grade basic arithmetic, and we investigated this
in the present study as a follow-up to Kleemans et al. (2013). To
determine the unique contribution of each of these factors, we also
took into account the autoregressive effect of prior arithmetic skills
in first grade.
The present study was carried out in the Netherlands. In Dutch
schools, children do not receive formal reading and arithmetic edu-
cation until first grade. Halfway through kindergarten, it is expected
that children gain phonological awareness (e.g., rhyming, blending)
and have knowledge of some letters of the alphabet. Moreover,
they should be able to arrange, count, and organize objects (e.g.,
apples), and are expected to solve elementary arithmetic problems
(e.g., one apple plus one apple makes two apples). Halfway through
first grade, children are expected to solve arithmetic problems with
small problem sizes, and halfway through second grade, children
should not only instantly retrieve the solution of small problem
sizes (i.e., arithmetic problems with sums and minuends below 10),
but are also assumed to successfully solve large problem sizes (i.e.,
arithmetic problems with sums and minuends above 10 and less
than 20 that contain carryover operations) (Van der Stap, 2012).
In the Netherlands, children who are diagnosed with DLD are
referred to schools for special education in which they receive an
adapted curriculum in language as well as math in relatively small
classrooms. The diagnosis of DLD is given by a multidisciplinary
team consisting of clinical linguists, educational psychologists,
speech therapists, and a physician. To receive this diagnosis, chil-
dren should have severe problems in receptive or productive
language domains (>1.5 SD on a minimum of two subtests out of
a standardized series of tests). Furthermore, in the present study,
children with hearing impairment (>30 dB), intellectual disabilities,
physical disabilities, or autism spectrum disorders were excluded.
The following research questions were addressed:
1. How do children with DLD and NLA differ in their basic arith-
metic skills?
2. To what extent can the basic arithmetic skills of children with
DLD and NLA in second grade be predicted from their home
numeracy experiences, as measured in kindergarten, when cog-
nitive and linguistic child factors are being controlled for?
With respect to the first research question, we expected chil-
dren with DLD to have lower basic arithmetic skills than children
with NLA, with larger group differences on arithmetic with large
problem sizes. With respect to the second research question, we
expected that home numeracy experiences were positively related
to the basic arithmetic skills (i.e., small problem sizes and large
problem sizes) of both children with DLD and NLA, after controlling
for their cognitive and linguistic skills.
2. Method
2.1. Participants
Participants of the present study were part of a three-year lon-
gitudinal study on the role of cognitive and linguistic precursors
in numerical development (see also Kleemans et al., 2013). The
relevant characteristics of both groups are displayed in Table 1.
Between Time 1 (i.e., second year of kindergarten) and Time 3,
two years later (i.e., second grade), twenty-nine children dropped
out of the study due to having moved or staying an extra year in
kindergarten or first grade. In addition, no differences were found
on the predictor and outcome measures between the group of
children who dropped out after kindergarten or first grade and
the group of children who participated in second grade. Further-
more, all children spoke Dutch as their native language. In addition,
both groups had a nonverbal intelligence within normal range
(M = 5.00, SD = 2.00). Finally, both groups did not differ in nonver-
bal intelligence (kindergarten: p = .274; second grade: p = .139), age
(kindergarten: p = .135; second grade: p = .089), and socioeconomic
status (kindergarten: p = .507; second grade: p = .105).
2.2. Materials
2.2.1. Child factors (Time 1: kindergarten)
2.2.1.1. General intelligence. The Raven’s Coloured Progressive
Matrices (Raven, 1956) was used to assess the capacity of general
intelligence. In this task, thirty-six visual patterns were presented,
with increasing difficulty. In each puzzle, a piece of information
was missing, and the child had to choose the correct piece out of six
alternatives. All correct answers were counted and converted into
standard scores, based on Dutch norms (Van Bon, 1986). This task
had good reliability (Cronbach’s alpha = .90; Van Bon, 1986). Fur-
thermore, this test had sufficient criterion and construct validity
according to the Dutch Committee on Diagnostic Assessment Mat-
ters [Commissie Testaangelegenheden Nederland; Cotan] (Cotan,
2017).
2.2.1.2. Working memory: central executive functioning. Central
executive functioning was assessed by means of the Number Recall
subtest from the Kaufman Assessment Battery for Children, sec-
ond edition (KABC-II) (Kaufman & Kaufman, 2004). In this task, the
experimenter pronounces sequences of digits that the child is asked
to repeat in backward order. The series of digits increased from two
to nine digits. Testing was terminated after three consecutive mis-
takes. The number of correctly recalled sequences was counted.
Cronbach’s alpha was .85, reflecting good reliability (Kaufman,
Lichtenberger, Fletcher-Janzen, & Kaufman, 2005). In addition, the
 T. Kleemans et al. / Early Childhood Research Quarterly 43 (2018) 62–72
Table 1
Descriptive characteristics of the children with DLD and children with NLA.
Time 1 (kindergarten)
DLD (n = 61)
Variable M (SD)
Age (months) 73.95 (5.578)
Socioeconomic statusa 2.43 (.562)
Nonverbal intelligence 4.42 (2.391)
Note. DLD = developmental language disorder; NLA = normal language achievement.
a
Parents were asked for highest completed education, varying from 1 (only postsecondary education, i.e., high school), 2 (intermediate education, i.e., education oriented
towards vocational training), 3 (higher education, i.e., university of applied sciences), to 4 (research university).
criterion and construct validity were evaluated as sufficient (Cotan,
2017).
2.2.1.3. Working memory: phonological loop. The capacity of the
phonological loop was measured using the subtest Woorden en
Zinnen Nazeggen [Repeating Words and Sentences] from the ESM
toets [Test for Children With SLI] (Verhoeven, 2005). The test was
administered in two parts. In the first part, 12 sequences of words
were presented to the child, with the number of words increasing
as testing progresses. An interval of one second was left between
each word. The child had to reproduce the sequence of word in
the correct order. In the second part, the child had to recall 12
sentences, with increasing difficulty. In both parts, testing was ter-
minated after four consecutive failures. The number of correctly
recalled sequences or sentences was counted. Cronbach’s alpha was
.88, indicating good reliability (Verhoeven, 2005). Furthermore, this
test was found to have good criterion validity and sufficient con-
struct validity (Cotan, 2017).
2.2.1.4. Working memory: visual-spatial sketchpad. The subtest
Geheugenspan [Memory Span] from the Revisie Amsterdamse Kinder
Intelligentie Test [Revision Amsterdam Child Intelligence Test]
(RAKIT) (Bleichrodt, Drenth, Zaal, & Resing, 1987) was used to
assess the visual–spatial sketchpad. A sequence of abstract fig-
ures was shown to the child, with the number of abstract figures
increasing as testing progresses. The task of the child was to cor-
rectly reproduce this sequence, by using little blocks on which these
abstract figures were depicted. When the child failed on three con-
secutive sequences, testing was terminated. Correctly reproduced
sequences were then counted. Reliability of this task was good
(Cronbach’s alpha in the present sample = .81). In addition, criterion
and construct validity were found to be good (Cotan, 2017).
2.2.1.5. Phonological awareness. Phonological awareness was mea-
sured using the Screenings instrument Beginnende Geletterdheid
[Diagnostic Instrument for Emergent Literacy] (Vloedgraven,
Keuning, & Verhoeven, 2009). This test consisted of four tasks.
During the first three tasks (i.e., rhyming, blending, and synthe-
sis), children were shown three response alternatives (both visually
and auditory), followed by the auditory presentation of the target
word. The instructions for the subsequent part of each task differed.
For rhyming, the child was asked which picture (e.g., “bee,” “ball,”
and “monkey”) rhymed with the target word (e.g., “What rhymes
with wall?”). For blending, the child had to select the picture start-
ing with the same initial sound (e.g., “c for cat”). For synthesis, the
child had to indicate which of the three pictures matched the orally
presented letter-sound sequence (e.g., “c-a-t”). Finally, for recep-
tive letter knowledge, four lowercase graphemes were presented on
a computer screen. The child heard a phoneme (e.g., “l for lips”)
and had to click on the corresponding grapheme. On each task, the
number of correct answers was counted. Cronbach’s alpha in the
current sample was >.90, indicating excellent reliability. Further-
65
Time 3 (second grade)
NLA (n = 111) DLD (n = 40) NLA (n = 103)
M (SD) M (SD) M (SD)
72.70 (4.420) 98.23 (5.385) 96.71 (4.421)
2.37 (.485) 2.53 (.544) 2.39 (.480)
4.88 (2.773) 4.26 (2.421) 4.99 (2.715)
more, this test was found to have sufficient criterion validity and
good construct validity (Cotan, 2017).
2.2.1.6. Grammatical ability. Three grammar tasks from the Taal-
toets Alle Kinderen [Language Test for All Children] (Verhoeven &
Vermeer, 1993) were used to assess grammatical ability. Each item
started with the presentation of three pictures. Then, the child
heard a sentence that matched one of the three pictures and had to
select the picture that corresponded with the auditory-presented
sentence. On each task, a different skill of grammatical ability was
assessed. First, the child had to complete 42 items that all contained
different kinds of function words (e.g., “Which glass is full?”). Sec-
ond, 32 sentences structures (e.g., “She pushes him”) were presented,
and finally, the child heard 32 sentences that all reflect different
kinds of nuances (e.g., “The glass is not full yet”). The number of cor-
rect answers was the score on each task. Cronbach’s alpha was at
least .82, indicating good reliability (Verhoeven & Vermeer, 2006).
In addition, both criterion validity and construct validity were eval-
uated as good (Cotan, 2017).
2.2.1.7. Naming speed. Naming speed was measured using the sub-
test Rapid naming: Plaatjes [Rapid Naming: Pictures] from the
ESM toets [Test for Children With Specific Language Impairment]
(Verhoeven, 2005). The task consisted of five columns comprising
120 high-frequent randomly presented items: schoen [shoe], eend
[duck], bril [glasses], huis [house], and kam [comb]. After a prac-
tice trial of ten items, the child was to name as many pictures as
possible within a minute. The score on this task was the number
of correctly named pictures. Cronbach’s alpha was .95, indicating
excellent reliability (Verhoeven, 2005). Furthermore, this test was
found to have good criterion validity and sufficient construct valid-
ity (Cotan, 2017).
2.2.2. Home factors (Time 1: kindergarten)
Parents of the participants filled in a questionnaire on home
numeracy practices. This questionnaire was based on previous
questionnaires on home numeracy practices (cf. Kleemans et al.,
2013; LeFevre et al., 2009). To verify that the questionnaire
indeed covered the hypothesized home numeracy constructs (i.e.,
parent–child numeracy activities and parents’ numeracy expecta-
tions), a principal component analysis (PCA) with varimax rotation
was used. The number of factors was based on theoretical inter-
pretability as well as acceptable eigenvalues (>1; Allen & Yen, 1979).
An item was assigned to a particular factor, when its factor load-
ing was over .35. A two-factor structure emerged from the results,
each factor accounting for at least 26.72% of the variability. The first
factor covered parent–child numeracy activities; the second factor
consisted of parents’ numeracy expectations. Comparable results
were found for principal axis factoring (PAF) and maximum likeli-
hood (ML) extraction methods. The results are shown in Table 2.
2.2.2.1. Parent–child numeracy activities. Parents had to rate six
items on the extent to which they participate in numeracy activities
66 T. Kleemans et al. / Early Childhood Research Quarterly 43 (2018) 62–72
Table 2
Results of the principal axis factoring analysis with promax rotation on aspects of home numeracy experiences (n = 143).
Variable
Parent–child numeracy activities
Making arithmetic puzzles
Talking about money when shopping
Doing counting activities
Playing counting games, using child computer or arithmetic software
Practicing numerical conceptual knowledge
Rehearsing counting rhymes
Parents’ numeracy expectations
Expectation of child’s ability to count till 20 (forward and backward)
Expectation of child’s ability to count without hands
Expectations of child’s ability to count in groups of 2, 5 or 10
Expectation of child’s performance on addition till 10
Expectation of child’s performance on addition till 20
Expectation of child’s performance on subtraction till 10
Expectation of child’s performance on subtraction till 20
with their child. Parents were asked: “How often did you and your
child engage in the following activities?” For each item, a 5-point
scale (1–5) was used. The higher the score, the more frequent the
parent participated in numeracy-related activities. The complete
list can be found in Appendix A. Cronbach’s alpha in the current
sample was .93, indicating excellent reliability.
2.2.2.2. Parents’ numeracy expectations. In this part, parents had to
fill out seven items on their expectations regarding numeracy with
answering options on a four-point scale. Parents were asked: “To
what extent do you expect your child will master the following
numeracy skills at the end of second grade?” Higher scores impli-
cated higher expectations on numeracy achievement. The list of
items is included in Appendix A. Reliability analysis showed that
these items had good reliability (Cronbach’s alpha in the current
sample was .90).
2.2.3. Basic arithmetic skills: small and large problem sizes
The tasks to measure basic arithmetic skills were based on pre-
vious research on arithmetic with small and large problem sizes
(Cowan et al., 2005). In first grade (Time 2), a battery of 40 addi-
tion and 40 subtraction problems with sums and minuends below
10 was presented to the child (i.e., small problem sizes). In sec-
ond grade (Time 3), the same battery of addition and subtraction
items below 10 was administered, followed by another battery, of
large problem sizes, consisting of 30 addition and 30 subtractions
problems with sums and minuends above 10 and less than 20, all
comprising carryover operations. Each child started with the task
that contained the addition problems followed by the task with the
subtraction problems. On each task, the child had to correctly solve
as many as arithmetical problems within a time limit of two min-
utes. All correct answers were counted. The complete list of items
can be found in Appendix B.
E-prime software (Schneider, Eschmann, & Zuccolotto, 2002)
was used to administer the various tasks. On each task, the same
procedure was used to present the items. First, an asterisk (i.e.,
‘*’) was shown to draw the child’s attention to the middle of the
screen, followed by the item (i.e., the arithmetical problem in Arabic
notation) in horizontal format. The task of the child was to pro-
vide the correct answer by using a numeric Universal Serial Bus
(USB) keyboard, with the Arabic digits (i.e., 0–9) depicted on it.
The items were presented in quasi-random order (i.e., an Arabic
digit presented in either the problem or solution did not occur in
a previous or following item) to prevent priming effect (Jackson &
Coney, 2007). Furthermore, the subtraction items were derived as
an inverse of the addition items (e.g., 2 + 4 = 6 and 6 − 4 = 2). Each
task had a good reliability (Cronbach’s alpha >.83). This also goes
for both sets of problem sizes (Cronbach’s alpha for small problem
Eigenvalue Percent explained variance ˛ Factor loading
5.355 41.194 .93
.847
.797
.878
.859
.861
.850
3.474 26.724 .90
.597
.768
.679
.834
.888
.836
.875
sizes = .84; Cronbach’s alpha for large problem sizes = .83). Finally,
preliminary testing of each of these tasks in a pilot study with a
similar age group showed sufficient construct validity with a stan-
dardized arithmetic achievement test (r > .80).
2.3. Procedure
The home numeracy questionnaire was sent to the parents after
the schools agreed to participate. The questionnaire was adminis-
tered halfway through the second year of kindergarten. The parent,
who was most involved in the numeracy-related activities with the
child, was asked to fill in the questionnaires. Completed question-
naires were returned by post.
Eight graduate student testers in educational psychology went
to the schools after receiving training by an educational psycholo-
gist. Each of the testers already had experience for at least two years
in testing children between the ages of 5–9 years. The training con-
sisted of a three-hour session in which the educational psychologist
(i.e., the first author) explained the standardized testing protocol
that each of the testers was expected to follow during testing. Then,
each of the testers practiced the standardized testing protocol with
a kindergartner (Time 1), first grader (Time 2), or second grader
(Time 3). Before the testers went to the schools, another meeting
with the educational psychologist was held to answer the remain-
ing questions.
Each child was assigned to one tester, and all children were
tested in a quiet room in their own school. The test battery con-
sisted of four randomly presented blocks, half an hour each. Within
a block, the same order of tests was used. The testing period com-
prised a period of two months, halfway through the second year
of kindergarten (Time 1), halfway through first grade (Time 2), and
halfway through second grade (Time 3).
3. Results
To answer the first research question, on differences between
the group of children with DLD and the group of children with NLA
in second grade, we first computed t-tests for independent sam-
ples with Holm–Bonferroni correction (Holm, 1979) on all child
and home factors. The results indicated that children with DLD
scored below the group of children with NLA on all variables, with
medium to high effect sizes (d), varying between .377 and 2.688. As
expected, children with DLD did not differ from children with NLA
on general intelligence. The results are presented in Table 3.
Next, we conducted a multivariate analysis of variance with
Problem type (small problem sizes, large problem sizes) and
Operation (addition, subtraction) as within-subjects factors and
Group (DLD, NLA) as between-subjects factor. Next to main
 T. Kleemans et al. / Early Childhood Research Quarterly 43 (2018) 62–72 67

Table 3
Descriptive statistics for the NLA group (n = 103) and DLD group (n = 40); results of t-tests for independent samples; and Cohen’s d.

Variable NLA DLD t d

M SD M SD

General intelligence 4.99 2.715 4.26 2.421 1.486 .249

Central executive functioning 3.96 1.468 1.78 1.527 7.903*** 1.331
Phonological loop 13.70 5.339 6.75 4.349 7.336*** 1.234
Visual–spatial sketchpad 5.62 1.890 3.13 1.305 8.984*** 1.775

Phonological awareness
Rhyming 9.22 1.137 5.98 2.082 9.343*** 2.688
Blending 7.94 2.367 4.75 2.085 7.472*** 1.259
Synthesis 8.44 1.949 5.23 2.247 8.469*** 1.426
Letter knowledge 21.86 5.557 11.85 5.432 9.732*** 1.639

Grammatical ability
Function words 35.83 2.753 29.33 3.951 9.545*** 2.589
Sentences structures 27.99 2.599 21.23 4.560 8.842*** 2.522
Nuances 26.92 2.976 21.63 3.656 8.945*** 1.501
Naming speed 39.92 12.596 34.25 11.591 2.466* .415

Parent–child numeracy activities

Making arithmetic puzzles 2.52 1.228 1.98 .660 3.439** .611
Talking about money when shopping 2.78 .885 2.38 .705 2.839** .603
Doing counting activities 2.50 1.065 1.85 .802 3.977*** .821
Playing counting games, using child computer or arithmetic software 2.53 1.101 1.85 .864 3.922*** .827
Practicing numerical conceptual knowledge 2.59 1.124 2.20 .939 2.117* .461
Rehearsing counting rhymes 2.26 1.111 1.70 1.043 2.762*** .465

Parents’ numeracy expectations

Expectation of child’s ability to count till 20 (forward and backward) 3.85 .364 3.25 .670 5.318*** 1.532
Expectation of child’s ability to count without hands 3.14 .852 2.78 .800 2.311* .389
Expectations of child’s ability to count in groups of 2,5 or 10 3.51 .752 3.20 .758 2.240* .377
Expectation of child’s performance on addition till 10 3.78 .418 3.23 .947 3.522*** 1.059
Expectation of child’s performance on addition till 20 3.38 .756 2.93 1.023 2.906** .489
Expectation of child’s performance on subtraction till 10 3.43 .824 2.93 .859 3.234*** .545
Expectation of child’s performance on subtraction till 20 3.12 .878 2.48 .933 3.854*** .649

Small problem sizes (first grade)

Addition till 10 26.36 10.037 15.50 8.470 3.884*** .654
Subtraction till 10 17.36 7.969 11.68 7.549 6.053*** 1.020

Small problem sizes (second grade)

Addition till 10 31.50 6.191 26.40 6.222 4.420*** .744
Subtraction till 10 25.99 8.022 20.23 4.588 5.373*** .976

Large problem sizes (second grade)

Addition till 20 22.72 7.689 14.10 7.182 6.125*** 1.032
Subtraction till 20 18.54 7.674 9.18 6.823 6.751*** 1.137

Note. Compared to our previous study (Kleemans et al., 2013), some children dropped out due to having moved or staying an extra year in first grade. Therefore, in the present
study, t-tests were computed again.
*
p < .05
**
p < .01.
***
p < .001.

effects of Operation (F(1, 141) = 225.858, p < .001, 2p = .616), tions between child and home factors and later basic arithmetic
Group (F(1, 141) = 40.734, p < .001, 2p = .224) and Problem type skills that were not affected by prior arithmetic skills. Further-
(F(1, 141) = 404.138, p < .001, 2p = .741), the results showed a more, to explore the relative contribution of the various cognitive
and linguistic child factors, these predictors were added in Step 2,
Group × Problem type interaction (F(1, 141) = 13.065, p < .001, 2p =
including a dummy group variable (NLA = 0, DLD = 1). In Step 3, both
.085), indicating that the differences between children with DLD
parent–child numeracy activities and parents’ numeracy expecta-
and their peers with NLA were larger for large problem sizes
tions were added to determine the unique contribution of each of
(B = 17.987, t = 6.946, p <. 001, 2p = .255) than for small problem
the home factors. Finally, to investigate whether there were addi-
sizes (B = 10.870, t = 4.654, p < . 001, 2p = .133).
tional differences between the groups in predicting basic arithmetic
To answer the second research question, we first computed
skills, all corresponding interaction terms between the child and
correlations between the child factors, home factors, and basic
home factors and group were added in Step 4. Adjusted R2 statistics
arithmetic skills. As can be seen in Table 4, correlations with the
were used to evaluate relative contribution (in terms of explained
dependent variables (i.e., small problem sizes and large problem
variance) of each step to the previous one as well as to correct for
sizes) were all significant.
the growing number of predictors in the model (Voeten & Van den
We then performed hierarchical regression analyses on both
Bercken, 2003).
outcome measures (i.e., small problem sizes and large problem
After controlling for prior arithmetic skills (i.e., arithmetic with
sizes). Prior to the regression analyses, we centered both child
small problem sizes in first grade) in Step 1, the results in Step 2
and home factors to prevent multicollinearity (Jewell, 2004). In
showed that general intelligence, phonological loop, phonological
Step 1, we first added the autoregressive effect of prior arithmetic
awareness, and grammatical ability as measured in kindergarten
skills, that is, arithmetic with small problem sizes in first grade.
positively predicted both basic arithmetic skills in second grade:
By controlling for this effect, we were able to investigate the rela-
68 T. Kleemans et al. / Early Childhood Research Quarterly 43 (2018) 62–72

Table 4
Correlations among the child factors, home factors, and basic arithmetic skills (n = 143).

Variable 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12.

1. General intelligence –
2. Central executive functioning .217** –
3: Phonological loop .218** .527*** –
4. Visual–spatial sketchpad .100 .418*** .387*** –
5. Phonological awareness .133 .536*** .595*** .413*** –
6. Grammatical ability .245** .569*** .589*** .451*** .702*** –
7. Naming speed .076 .179* .179* .210* .215* .245** –
8. Parent–child numeracy activities .147 .324*** .173 .243** .211* .241** .164 –
9. Parents’ numeracy expectations .187* .333*** .337*** .201* .320*** .355*** .238** .208* –
10. Small problem sizes (first grade) .109 .227** .248** .188* .322*** .402*** .002 .229** .195* –
11. Small problem sizes (second grade) .333*** .437*** .526*** .307*** .516*** .528*** .206* .225** .305*** .366***
12. Large problem sizes (second grade) .317*** .497*** .580*** .426*** .587*** .607*** .168* .415*** .506*** .373*** .735*** –
*
p < .05
**
p < .01.
***
p < .001.

Table 5
Child and home factors of basic arithmetic skills in children with DLD (n = 40) and children with NLA (n = 103); and Cohen’s f2.

Predictor Small problem sizes (second grade) Large problem sizes (second grade)

R 2
f 2
ˇ R2 f2 ␤
*** ***
Step 1 .134 .577 .139 .595
Small problem sizes (first grade) .366*** .373***
*** ***
Step 2 .343 1.045 .393 1.322
General intelligence .174** .153*
Central executive functioning .102 .088
Phonological loop .219* .217**
Visual–spatial sketchpad .079 .153*
Phonological awareness .281** .244*
Grammatical ability .243* .220*
Naming speed .073 −.003
Group .368** .171

Step 3 .003 .006 .098*** .311

Parent–child numeracy activities .030 .216***
Parents’ numeracy expectations .049 .258***

Step 4 .049 .128 .048 .304

Group × Small problem sizes (first grade) −.087 −.026
Group × General intelligence −.057 −.068
Group × Central executive functioning −.078 .060
Group × Phonological loop .054 −.101
Group × Visual–spatial sketchpad −.048 −.173
Group × Phonological awareness −.126 −.058
Group × Grammatical ability −.135 −.164
Group × Naming speed .194* .193*
Group × Parent–child numeracy activities −.021 .195*
Group × Parents’ numeracy expectations .092 −.074

Total Radj
2
.446*** .624***

Note. Only significant interaction effects are displayed.

*
p < .05.
**
p < .01.
***
p < .001.

higher scores on general intelligence, working memory, phonolog- nificant. The results thus suggest identical relationships between
ical awareness, and grammatical ability resulted in higher scores children with DLD and children with NLA in predicting basic arith-
on arithmetic with small problem sizes as well as large problem metic skills. All significant steps in the hierarchical regression
sizes. In addition, unique effects of the visual–spatial sketchpad in analysis (in terms of Cohen’s f2 ) had sufficient power (>.80) at the
kindergarten were found for arithmetic with large problems in sec- .05 level (two-tailed). The results are presented in Table 5.
ond grade: higher scores on the visual–spatial sketchpad resulted in
higher scores on arithmetic with large, but not small problem sizes.
4. Discussion
Furthermore, it was found in Step 3 that both parent–child numer-
acy activities and parents’ numeracy expectations in kindergarten
The present study investigated the longitudinal effects of the
positively predicted arithmetic with large problem sizes: higher
aspects of home numeracy experiences (i.e., parent–child numer-
scores on parent–child numeracy activities and parents’ numeracy
acy activities and parents’ numeracy expectations) on the basic
expectations were associated with higher scores on arithmetic with
arithmetic skills of children with DLD and their peers with NLA
large problem sizes, but not small problem sizes. Finally, in Step 4,
in second grade. First, differences between children with DLD and
the addition of explained variance in predicting arithmetic with
children with NLA on basic arithmetic skills were investigated. We
small problem sizes as well as large problem sizes was not sig-
found second-grade children with DLD to score below their peers
 T. Kleemans et al. / Early Childhood Research Quarterly 43 (2018) 62–72
on arithmetic with both small and large problem sizes, with a larger
difference for the large problem sizes. Given that arithmetic with
small problems relies on having access to arithmetic facts that are
stored in long-term memory (De Smedt et al., 2010), the impaired
linguistic skills of children with DLD may hinder them in acquir-
ing those accurate arithmetic representations. The ability to solve
arithmetic problems with large problem sizes not only depends on
applying the right solution procedure, but also on having instant
access to these arithmetic facts.
With respect to the longitudinal effects of aspects of home
numeracy experiences, it was first found that general intelligence,
the phonological loop, phonological awareness, and grammatical
ability predicted the basic arithmetic skills (i.e., small and large
problem sizes) of children with DLD and their peers with NLA. These
results extended earlier research (cf. Durkin et al., 2013; Kleemans
et al., 2014) in showing that both cognitive and linguistic factors
account for the acquisition of not only first-grade basic arithmetic
skills, but also the more advanced second-grade basic arithmetic
skills.
Furthermore, the results indicated that for both groups of chil-
dren, parent–child numeracy activities and parents’ numeracy
expectations were positively related to second-grade arithmetic
achievement with large problem sizes. As an extension of previ-
ous research (e.g., Kleemans et al., 2013; Skwarchuk et al., 2014),
the present study was the first to provide an in-depth view on
the unique role of each of the child and home factors on basic
arithmetic skills beyond the start of formal arithmetic educa-
tion in first grade. It is interesting to note that home numeracy
experiences did not relate to second-grade arithmetic with small
problem sizes. This might be due to the fact that home numer-
acy experiences are only related to those skills that are within the
zone of proximal development (Vygotsky, 1978) of the child and
that are a subject of the educational curriculum at that time (cf.
Skwarchuk et al., 2014). According to the final terms of the Dutch
national curriculum (Van der Stap, 2012), the second graders in
the present study were expected to instantly retrieve the solu-
tion of arithmetic problems with small problem sizes. Furthermore,
they should successfully solve arithmetic with large problem sizes,
which is primarily dependent on the ability to shift from a finger
counting strategy (i.e., visual–spatial sketchpad) to a more ver-
bal strategy (i.e., phonological loop) (Siegler, 1996). These verbal
strategies (i.e., verbal counting and temporarily maintaining arith-
metic facts during the solution process), are considered to be of
crucial importance in order to develop strong problem–answer
associations in long-term memory (De Smedt et al., 2009). As can
be seen in Table 5, the visual–spatial sketchpad as a cognitive fac-
tor was still significant in predicting arithmetic with large problem
sizes, supporting the idea that the children were still in the pro-
cess of automatizing arithmetic with large problem sizes. The role
of home numeracy experiences in basic arithmetic might thus be
restricted to those arithmetic skills that children have not fully
mastered yet. It may, therefore, be expected that the effects of
the aspects of home numeracy experiences may further reduce in
strength when children have acquired basic arithmetic skills, and
other mathematical skills (e.g., geometry and algebra) are becom-
ing increasingly important (Kleemans et al., 2016).
In contrast to what was found in the present sample one year
earlier (Kleemans et al., 2013), the results of the present study
did not indicate that parent–child numeracy activities were more
strongly related to the basic arithmetic skills of children with DLD
than their peers with NLA. Moreover, it was also found in our pre-
vious research that naming speed differentially predicted the basic
arithmetic skills of children with DLD. The absence of both effects
in the current study might be due to the relative small sample
size of children with DLD and the corresponding loss of statisti-
cal power. Indeed, although the results indicated that Step 4 as a
69
whole did not add a significant percentage of explained variance,
it might be noted, that the individual interaction term between
parent–child numeracy activities and Group in predicting arith-
metic with large problem sizes did reach significance (ˇ = .195;
p = .021). If allowed to further analyze this effect, it would have
led to the same conclusions as described in previous studies: chil-
dren with DLD and lower parent–child numeracy activities (<−1 SD
below the mean score of the NLA group) performed worse on arith-
metic with large problem sizes, when compared to children with
DLD and higher parent–child numeracy activities (>−1 SD below
the mean score of the NLA group). This was also the case for the
Naming speed × Group interaction (small problem sizes: ˇ = .194;
p = .035 and large problem sizes: ˇ = .193; p = .012).
The results of the present study have some limitations. First,
the present study did not have a representative distribution of
SES. Parents of both children with DLD and NLA had relatively
high educational levels (i.e., over 70 percent were middle or highly
educated), which makes it difficult to generalize these findings to
families of children with a lower SES. Previous research showed
that children that come from lower SES families face numerous
stressors, such as financial constraints and lower education. This
may limit parents’ ability to support their children’s academic
development and influence the type of number-related activities
parents report engaging in with their children at home (Ramani
& Siegler, 2014). Indeed, it has been found that differences in the
home environment were more strongly related to the numeracy
skills of children with a lower SES (Melhuish et al., 2008; Siegler &
Ramani, 2009), and that especially these children may benefit a lot
from playing informal numeracy activities to reduce their delays
in numerical estimation skills, one of the most important prereq-
uisites for mathematics (see Siegler & Ramani, 2008). The stronger
relationships with the home numeracy environment may thus be
more localized on the lower ability range. We, therefore, cannot
rule out the possibility that home numeracy experiences are also
related to arithmetic with small problems in second grade when
the full range of SES was included. Future research could focus on
this.
Second, another limitation was that we did not assess home lit-
eracy experiences. Manolitsis, Georgiou, and Tziraki (2013) found
that parent–child numeracy activities were associated with early
math acquisition, but when they replaced home numeracy expe-
riences with home literacy experiences in a path analysis, parent
literacy teaching was also positively related to early math acqui-
sition. However, Segers, Kleemans, and Verhoeven (2015) only
found unique positive effects of home numeracy experiences after
controlling for home literacy in a sample of 60 kindergartners. Pre-
vious research indicated that parents of children with DLD report
engaging in fewer literacy activities and also have lower literacy
expectations than parents of children with NLA (Skibbe, Justice,
Zucker, & McGinty, 2008). We, therefore, cannot rule out the pos-
sibility that children with DLD are less interested in numeracy
related activities because of their impaired linguistic skills and, as a
consequence, parents are less motivated to engage in linguistically-
oriented home practices. This could be a topic for future research.
Finally, because we used self-reports to assess home numeracy
experiences in children with DLD, we cannot rule out the possi-
bility that the group differences on home numeracy experiences
may be caused because of language problems on the part of the
parent. Indeed, DLD has been found to be highly heritable (Bishop
& Snowling, 2004). It would, therefore, be recommended in future
research to also control for the language level of the parent when
testing group differences on home numeracy experiences. A related
issue that needs to be considered is the fact that the use of ques-
tionnaires in general may be biased because of social desirability
(Richman, Kiesler, Weisband, & Drasgow, 1999). Other ways of
assessing home numeracy skills may overcome such limitations.
70 T. Kleemans et al. / Early Childhood Research Quarterly 43 (2018) 62–72

For example, Bjorklund, Hubertz, and Reubens (2004) found the 3. Expectations of child’s ability to count in groups of 2, 5 or 10
quality of parent–child interactions (by means of structured obser- 4. Expectation of child’s ability on addition till 10
vations) to be related to children’s numeracy skills as well. It is 5. Expectation of child’s ability on addition till 20
still unclear, however, to what extent such a measure as opposed 6. Expectation of child’s ability on subtraction till 10
to the use of questionnaires may be a better, more valid way to 7. Expectation of child’s ability on subtraction till 20
assess home numeracy experiences. Therefore, validation studies
to determine the most optimal way to assess home numeracy are Note. A scale ranging from 1 to 4 was included for each item.
called for (Skwarchuk et al., 2014).
To conclude, the results point out that home numeracy factors
relate to second-grade arithmetic with large, but not small prob- Appendix B.
lem sizes. This finding has implications for both the home and
school environment. First, parents should be made aware that they See Tables B1 and B2.
can make a difference when it comes to stimulating their child’s
arithmetic skills. To be more specific, parents could provide chil- Table B1
dren with those numeracy-related activities that are just above Small problem sizes.
their current competence level (Wells, 1999). Second, it can be rec- Addition Subtraction
ommended to introduce numeracy activities at home before the
3+6=9 9−6=3
start of formal arithmetic education. Consequently, professional 1+1=2 2−1=1
communication between teachers and parents is crucial. This impli- 5+4=9 9−4=5
cates that teachers should inform parents about their crucial role 2+1=3 3−1=2
in stimulating their child’s math skills along with providing them a 6+2=8 8−2=6
4+3=7 7−3=4
realistic image of their child’s current math skills. Finally, when it
1+5=6 6−5=1
comes to intervention on reducing delays in basic arithmetic skills, 3 + 7 = 10 10 − 7 = 3
it is important to take into account both child and home factors that 4+5=9 9−5=4
children bring into the classroom. 1+2=3 3−2=1
4+4=8 8−4=4
7 + 3 = 10 10 − 3 = 7
Appendix A. 4+2=6 6−2=4
5+3=8 8−3=5
Parent–Child Numeracy Activities 2+7=9 9−7=2
3+1=4 4−1=3
5+2=7 7−2=5
Question 4 + 6 = 10 10 − 6 = 4
How often did you and your child engage in the following activ- 7+2=9 9−2=7
ities? Circle 1 if the activity did not occur, circle 2 if it occurred on 3+5=8 8−5=3
1 + 9 = 10 10 − 9 = 1
a monthly base, circle 3 if it occurred on a weekly base, circle 4 if
2+4=6 6−4=2
it occurred on a daily base, and circle 5 if it occurred a few times a 1+8=9 9−8=1
day. 6 + 4 = 10 10 − 4 = 6
3+2=5 5−2=3
2+6=8 8−6=2
1. Making arithmetic puzzles (e.g., connect-the-dot activities)
1+3=4 4−3=1
2. Doing counting activities (e.g., playing with child cash register; 2 + 8 = 10 10 − 8 = 2
playing with number wall; playing with die) 1+4=5 5−4=1
3. Talking about money when shopping 3+3=6 6−3=3
4. Playing counting games, using child computer or arithmetic 2+5=7 7−5=2
6+3=9 9−3=6
software (e.g., playing with ‘My first computer’ (to practice arith-
8 + 2 = 10 10 − 2 = 8
metic and counting skills), playing Disney preschool) 3+4=7 7−4=3
5. Practicing numerical conceptual knowledge (e.g., ordering 5 + 5 = 10 10 − 5 = 5
objects by size, shape, colour; arranging objects by size, height; 1+6=7 7−6=1
2+3=5 5−3=2
mass, number; what is more/less)
9 + 1 = 10 10 − 1 = 9
6. Rehearsing counting rhymes 2+2=4 4−2=2
1+7=8 8−7=1
Note. A scale ranging from 1 to 5 was included for each item.

Parents’ Numeracy Expectations

Table B2
Question Large problem sizes.
To what extent do you expect your child to master the following
Addition Subtraction
numeracy skills at the end of kindergarten? Circle 1 if you expect
your child not to master a particular skill at all, circle 2 when you 8 + 9 = 17 17 − 9 = 8
5 + 6 = 11 11 − 6 = 5
expect your child to master the particular skill a little, circle 3 if
8 + 4 = 12 12 − 4 = 8
you expect your child to sufficiently master the particular skill, 9 + 7 = 16 16 − 7 = 9
and circle 4 when you expect your child to completely master the 8 + 3 = 11 11 − 3 = 8
particular skill. 5 + 9 = 14 14 − 9 = 5
6 + 7 = 13 13 − 7 = 6
9 + 2 = 11 11 − 2 = 9
1. Expectation of child’s ability to count till 20 (forward and back- 8 + 7 = 15 15 − 7 = 8
ward) 4 + 9 = 13 13 − 9 = 4
2. Expectation of child’s ability to count without hands 7 + 5 = 12 12 − 5 = 7
 T. Kleemans et al. / Early Childhood Research Quarterly 43 (2018) 62–72 71

Table B2 (Continued) Developmental Science, 13, 508–520. http://dx.doi.org/10.1111/j.1467-7687.

2009.00897.x
Addition Subtraction
De Smedt, Holloway, I. D., & Ansari, D. (2011). Effects of problem size and
8 + 6 = 14 14 − 6 = 8 arithmetic operation on brain activation during calculation in children with
9 + 3 = 12 12 − 3 = 9 varying levels of arithmetical fluency. Neuroimage, 57(3), 771–781. http://dx.
6 + 5 = 11 11 − 5 = 6 doi.org/10.1016/j.neuroimage.2010.12.037
Dehaene, S., Piazza, M., Pinel, P., & Cohen, L. (2003). Three parietal circuits for
4 + 8 = 12 12 − 8 = 4
number processing. Cognitive Neuropsychology, 20(3–6), 487–506. http://dx.
7 + 6 = 13 13 − 6 = 7
doi.org/10.1080/02643290244000239
9 + 5 = 14 14 − 5 = 9 Desoete, A., & Grégoire, J. (2006). Numerical competence in young children and in
3 + 8 = 11 11 − 8 = 3 children with mathematical learning disabilities. Learning and Individual
6 + 9 = 15 15 − 9 = 6 Differences, 16, 351–367. http://dx.doi.org/10.1016/j.lindif.2006.12.006
4 + 7 = 11 11 − 7 = 4 Durkin, K., Mok, P. L., & Conti-Ramsden, G. (2013). Severity of specific language
5 + 8 = 13 13 − 8 = 5 impairment predicts delayed development in number skills. Frontiers in
2 + 9 = 11 11 − 9 = 2 Psychology, 4, 1–10. http://dx.doi.org/10.3389/fpsyg.2013.00581
7 + 8 = 15 15 − 8 = 7 Fazio, B. B. (1994). The counting abilities of children with specific language
9 + 4 = 13 13 − 4 = 9 impairment: A comparison of oral and gestural tasks. Journal of Speech,
5 + 7 = 12 12 − 7 = 5 Language, and Hearing Research, 37(2), 358–368. http://dx.doi.org/10.1044/jshr.
6 + 8 = 14 14 − 8 = 6 3702.358
3 + 9 = 12 12 − 9 = 3 Fazio, B. B. (1996). Mathematical abilities of children with specific language
impairment: A 2-year follow-up. Journal of Speech, Language, and Hearing
7 + 4 = 11 11 − 4 = 7
Research, 39(4), 839–849. http://dx.doi.org/10.1044/jshr.3904.839
8 + 5 = 13 13 − 5 = 8
Friso-van den Bos, I., Van der Ven, S. H. G., Kroesbergen, E. H., & Van Luit, J. E. H.
7 + 9 = 16 16 − 9 = 7
(2013). Working memory and mathematics in primary school children: A
meta-analysis. Educational Research Review, 10, 29–44. http://dx.doi.org/10.
1016/j.edurev.2013.05.003
Appendix C. Supplementary data Hauser, M. D., Chomsky, N., & Fitch, W. T. (2002). The faculty of language: What is
it, who has it and how did it evolve? Science, 298, 1569–1579. http://dx.doi.
org/10.1126/science.298.5598.1569
Supplementary data associated with this article can be found, Hecht, S. A., Torgesen, J. K., Wagner, R. K., & Rashotte, C. A. (2001). The relations
in the online version, at https://doi.org/10.1016/j.ecresq.2018.01. between phonological processing abilities and emerging individual differences
005. in mathematical computation skills: A longitudinal study from second to fifth
grades. Journal of Experimental Child Psychology, 79, 192–227. http://dx.doi.org/
10.1006/jecp.2000.2586
References Holm, S. (1979). A simple sequentially rejective multiple test procedure.
Scandinavian Journal of Statistics, 6, 65–70.
Allen, M. J., & Yen, W. M. (1979). Introduction to measurement theory. Monterey, CA: Huntsinger, C. S., Jose, P. E., Larson, S. L., Balsink Krieg, D., & Shaligram, C. (2000).
Brooks/Cole. Mathematics, vocabulary, and reading development in Chinese American and
Baddeley, A. D. (1996). Exploring the central executive. Quarterly Journal of European American children over the primary school years. Journal of
Experimental Psychology, 49, 5–28. http://dx.doi.org/10.1080/ Educational Psychology, 92(4), 745–760. http://dx.doi.org/10.1037//0022-0663.
027249896392784 92.4.745
Bishop, D. V. M., & Snowling, M. J. (2004). Developmental dyslexia and specific Huntsinger, C. S., Jose, P. E., & Luo, Z. (2016). Parental facilitation of early
language impairment: Same or different? Psychological Bulletin, 130(6), mathematics and reading skills and knowledge through encouragement of
858–886. http://dx.doi.org/10.1037/0033-2909.130.6.858 home-based activities. Early Childhood Research Quarterly, 37, 1–15. http://dx.
Bishop, D. V., Snowling, M. J., Thompson, P. A., & Greenhalgh, T. (2016). CATALISE: doi.org/10.1016/j.ecresq.2016.02.005
A multinational and multidisciplinary Delphi consensus study. Identifying Jackson, N., & Coney, J. (2007). Simple arithmetic processing. Surface form effects
language impairments in children. Public Library of Science, 11(7) http://dx.doi. in a priming task. Acta Psychologica, 125(1), 1–19. http://dx.doi.org/10.1016/j.
org/10.1371/journal.pone.0158753 actpsy.2006.05.003
Bjorklund, D. F., Hubertz, M. J., & Reubens, A. C. (2004). Young children’s arithmetic Jewell, P. (2004). Statistics for epidemiology. Boca Raton, FL: Chapman & Hall/CRC.
strategies in social context: How parents contribute to children’s strategy Jordan, N. C., Kaplan, D., Locuniak, M. N., & Ramineni, C. (2007). Predicting
development while playing games. International Journal of Behavioral first-grade math achievement from developmental number sense trajectories.
Development, 28(4), 347–357. Learning Disabilities Research & Practice, 22, 36–46. http://dx.doi.org/10.1111/j.
Bleichrodt, N., Drenth, P. J. D., Zaal, J. N., & Resing, W. C. M. (1987). RAKIT. 1540-5826.2007.00229.x
Handleiding bij de revisie Amsterdamse kinder intelligentie test [Manual for the Kaufman, A. S., & Kaufman, N. L. (2004). Kaufman assessment battery for children
revised Amsterdam child intelligence test]. Lisse, the Netherlands: Swets & (2nd ed.). Circle Pines, MN: American Guidance Service.
Zeitlinger. Kaufman, A. S., Lichtenberger, E. O., Fletcher-Janzen, E., & Kaufman, N. L. (2005).
Blevins-Knabe, B., & Musun-Miller, L. (1996). Number use at home by children and Essentials of KABC-II assessment. Hoboken, NJ: Wiley & Sons.
their parents and its relationship to early mathematical performance. Early Kaufmann, L. (2008). Dyscalculia: Neuroscience and education. Educational
Development and Parenting, 5, 35–45. http://dx.doi.org/10.1002/(SICI)1099- Research, 50(2), 163–175. http://dx.doi.org/10.1080/00131880802082658
0917(199603)5:1<35:AID-EDP113 Kleemans, T., Segers, E., & Verhoeven, L. (2011). Precursors to numeracy in
Blevins-Knabe, B. (2016). Early mathematical development: How the home kindergartners with specific language impairment. Research in Developmental
environment matters. In B. Blevins-Knabe, & A. Berghout Austin (Eds.), Early Disabilities, 32(6), 2901–2908. http://dx.doi.org/10.1016/j.ridd.2011.05.013
childhood mathematics skill development in the home environment (pp. 7–28). Kleemans, T., Segers, E., & Verhoeven, L. (2012). Naming speed as a clinical marker
Cham, Switzerland: Springer International Publishing. in predicting basic calculation skills in children with specific language
Cankaya, O., & LeFevre, J. A. (2016). The home numeracy environment: What do impairment. Research in Developmental Disabilities, 33(3), 882–886.
cross-cultural comparisons tell us about how to scaffold young children’s Kleemans, T., Segers, E., & Verhoeven, L. (2013). Relations between home
mathematical skills? In B. Blevins-Knabe, & A. Berghout Austin (Eds.), Early numeracy experiences and basic calculation skills in children with and without
childhood mathematics skill development in the home environment (pp. 87–104). specific language impairment. Early Childhood Research Quarterly, 28(2),
Cham, Switzerland: Springer International Publishing. 415–423. http://dx.doi.org/10.1016/j.ecresq.2012.10.004
Ciping, D., Silinskas, G., Wei, W., & Georgiou, G. K. (2015). Cross-lagged Kleemans, T., Segers, E., & Verhoeven, L. (2014). Cognitive and linguistic predictors
relationships between home learning environment and academic achievement of basic arithmetic skills: Evidence from first and second language learners.
in Chinese. Early Childhood Research Quarterly, 33, 12–20. http://dx.doi.org/10. International Journal of Development Disability and Education, 61(3), 306–316.
1016/j.ecresq.2015.05.001 http://dx.doi.org/10.1080/1034912X2014.934017
Cotan. (2017). Cotan Documentatie [Cotan Documentation]. Retrieved from Kleemans, T., Segers, E., & Verhoeven, L. (2016). Towards a theoretical framework
https://www.cotandocumentatie.nl. on individual differences in numerical abilities: Role of home numeracy
Cowan, R., Donlan, C., Newton, E. J., & Lloyd, D. (2005). Number skills and experiences. In B. Blevins-Knabe, & A. Berghout Austin (Eds.), Early childhood
knowledge in children with specific language impairment. Journal of mathematics skill development in the home environment (pp. 87–104). Cham,
Educational Psychology, 97, 732–744. http://dx.doi.org/10.1037/0022-0663.97. Switzerland: Springer International Publishing.
4.732 Kleemans, T., Peeters, M., Segers, E., & Verhoeven, L. (2012). Child and home
De Smedt, B., Janssen, R., Bouwens, K., Verschaffel, L., Boets, B., & Ghesquière, P. predictors of early numeracy skills in kindergarten. Early Childhood Research
(2009). Working memory and individual differences in mathematics Quarterly, 27, 471–477. http://dx.doi.org/10.1016/j.ecresq.2011.12.004
achievement: A longitudinal study from first grade to second grade. Journal of Kleemans, T., Segers, E., & Verhoeven, L. (2012). Naming speed as a clinical marker
Experimental Child Psychology, 103, 86–201. http://dx.doi.org/10.1016/j.jecp. in predicting basic calculation skills in children with specific language
2009.01.004 impairment. Research in Developmental Disabilities, 33(3), 882–886. http://dx.
De Smedt, B., Taylor, J., Archibald, L., & Ansari, D. (2010). How is phonological doi.org/10.1016/j.ridd.2011.12.007
processing related to individual differences in children’s arithmetic?
72 T. Kleemans et al. / Early Childhood Research Quarterly 43 (2018) 62–72
Koponen, T., Mononen, R., Räsänen, P., & Ahonen, T. (2006). Basic numeracy in
children with specific language impairment: Heterogeneity and connections to
language. Journal of Speech, Language, and Hearing Research, 49, 58–73. http://
dx.doi.org/10.1044/1092-4388(2006/005)
LeFevre, J., Clarke, T., & Stringer, A. P. (2002). Influences of language and parental
involvement on the development of counting skills: Comparisons of French-
and English-speaking Canadian children. Early Child Development and Care, 172,
283–300. http://dx.doi.org/10.1080/03004430212127
LeFevre, J., Skwarchuk, S., Smith-Chant, B. L., Fast, L., Kamawar, D., & Bisanz, J.
(2009). Home numeracy experiences and children’s math performance in the
early school years. Canadian Journal of Behavioural Science, 41, 55–66. http://dx.
doi.org/10.1037/a0014532
LeFevre, J., Fast, L., Skwarchuk, S. L., Smith-Chant, B. L., Bisanz, J., Kamawar, D., et al.
(2010). Pathways to mathematics: Longitudinal predictors of performance.
Child Development, 81(6), 1753–1767. http://dx.doi.org/10.1111/j.1467-8624.
2010.01508.x
LeFevre, J., Polyzoi, E., Skwarchuk, S., Fast, L., & Sowinski, C. (2010). Do home
numeracy and literacy practices of Greek and Canadian parents predict the
numeracy skills of kindergarten children? International Journal of Early Years
Education, 18(1), 55–70. http://dx.doi.org/10.1080/09669761003693926
Lee, K. M., & Kang, S. Y. (2002). Arithmetic operation and working memory:
Differential suppression in dual tasks. Cognition, 83(3), 63–68. http://dx.doi.
org/10.1016/s0010-0277(02)00010-0
Manolitsis, G., Georgiou, G. K., & Tziraki, N. (2013). Examining the effects of home
literacy and numeracy environment on early reading and math acquisition.
Early Childhood Research Quarterly, 28, 692–703. http://dx.doi.org/10.1016/j.
ecresq.2013.05.004
Martin, A., Volkmar, F. R., & Lewis, M. (2007). Lewis’s child and adolescent psychiatry.
A comprehensive textbook. Philadelphia, PA: Lippincott, Williams & Wilkins.
Melhuish, E. C., Sylva, K., Sammons, P., Siraj-Blatchford, I., Taggart, B., Phan, M.,
et al. (2008). Preschool influences on mathematics achievement. Science, 321,
1161–1162. http://dx.doi.org/10.1126/science.1158808
Niklas, F., & Schneider, W. (2013). Casting the die before the die is cast: The
importance of the home numeracy environment for preschool children.
European Journal of Psychology of Education, 29(3), 327–345. http://dx.doi.org/
10.1007/s10212-013-0201-6
Ramani, G. B., & Siegler, R. S. (2014). How informal learning activities can promote
children’s numerical knowledge. In R. C. Kadosh, & A. Dowker (Eds.), Oxford
handbooks online http://dx.doi.org/10.1093/oxfordhb/9780199642342.013.012
Rasmussen, C., & Bisanz, J. (2005). Representation and working memory in early
arithmetic. Journal of Experimental Child Psychology, 91, 137–157. http://dx.doi.
org/10.1016/j.jecp.2005.01.004
Raven, J. C. (1956). Guide to using the coloured progressive matrices. London, United
Kingdom: H.K. Lewis & Co.
Richman, W. L., Kiesler, S., Weisband, S., & Drasgow, F. (1999). A meta-analytic
study of social desirability distortion in computer-administered
questionnaires, traditional questionnaires, and interviews. Journal of Applied
Psychology, 84(5), 754–775. http://dx.doi.org/10.1037/0021-9010.84.5.754
Sénéchal, M., & LeFevre, J. (2014). Continuity and change in the home literacy
environment as predictors of growth in vocabulary and reading. Child
Development, 85, 1535–1551. http://dx.doi.org/10.1111/cdev.12222
Sénéchal, M., & Young, L. (2008). The effect of family literacy interventions on
children’s acquisition of reading from kindergarten to grade 3: A meta-analytic
review. Review of Educational Research, 78, 880–907. http://dx.doi.org/10.3102/
0034654308320319
Schneider, W., Eschmann, A., & Zuccolotto, A. (2002). E-Prime user’s guide.
Pittsburgh, PA: Psychology Software Tools, Inc.
Segers, E., Kleemans, T., & Verhoeven, L. (2015). Role of parent literacy and
numeracy expectations and activities in predicting early numeracy skills.
Mathematical Thinking and Learning, 17(2–3), 219–236. http://dx.doi.org/10.
1080/10986065.2015.1016819
Siegler, R. S., & Ramani, G. B. (2008). Playing linear numerical board games
promotes low-income children’s numerical development. Developmental
Science, 11(5), 655–661. http://dx.doi.org/10.1111/j.1467-7687.2008.00714.x
Siegler, R. S., & Ramani, G. B. (2009). Playing linear number board games—but not
circular ones—improves low-income preschoolers’ numerical understanding.
Journal of Educational Psychology, 101(3), 545. http://dx.doi.org/10.1037/
a0014239
Siegler, R. S. (1996). Emerging minds: The process of change in children’s thinking.
New York, NY: Oxford University Press.
Simmons, F., & Singleton, C. (2008). Do weak phonological representations impact
on arithmetic development? A review of research into arithmetic and dyslexia.
Dyslexia, 14(2), 77–94. http://dx.doi.org/10.1002/dys.341
Skibbe, L. E., Justice, L. M., Zucker, T. A., & McGinty, A. S. (2008). Relations among
maternal literacy beliefs, home literacy practices, and the early literacy skills of
preschoolers with specific language impairment. Early Education and
Development, 19, 68–88. http://dx.doi.org/10.1080/10409280701839015
Skwarchuk, S.-L., Sowinkski, C., & LeFevre, J. (2014). Formal and informal home
learning activities in relation to children’s early numeracy and literacy skills:
The development of a home numeracy model. Journal of Experimental Child
Psychology, 121, 63–84. http://dx.doi.org/10.1016/j.jecp.2013.11.006
Skwarchuk, S.-L. (2009). How do parents support preschoolers’ numeracy
experiences at home? Early Childhood Education Journal, 37(3), 189–197. http://
dx.doi.org/10.1007/s10643-009-0340-1
Stanescu-Cosson, R., Pinel, P., Van De Moortele, P. F., Le Bihan, D., Cohen, L., &
Dehaene, S. (2000). Understanding dissociations in dyscalculia: A brain
imaging study of the impact of number size on the cerebral networks for exact
and approximate calculation. Brain, 123, 2240–2255. http://dx.doi.org/10.
1093/brain/123.11.2240
Stock, P., Desoete, A., & Roeyers, H. (2009). Predicting arithmetic abilities: The role
of preparatory arithmetic markers and intelligence. Journal of
Psychoeducational Assessment, 27(3), 237–251. http://dx.doi.org/10.1177/
0734282908330587
Van Bon, W. H. J. (1986). Raven’s Coloured Progressive Matrices. Nederlandse normen
en enige andere uitkomsten van onderzoek [Raven’s Coloured Progressive
Matrices. Dutch Norms and some other results of research]. Lisse, the
Netherlands: Swets & Zeitlinger.
Van der Stap, M. (2012). Van kerndoel tot leerlijn. Concretisering van de kerndoelen
voor het speciaal onderwijs [Elaboration of the primary objectives in special
education]. Amsterdam, the Netherlands: Uitgeverij SWP.
Verhoeven, L., & Vermeer, A. (1993). Taaltoets alle kinderen. handleiding [Language
test for all children. Manual]. Arnhem, the Netherlands: Cito.
Verhoeven, L., & Vermeer, A. (2006). Taaltoets alle kinderen. Verantwoording
[Language test for all children. Psychometric analysis]. Arnhem, the Netherlands:
Cito.
Verhoeven, L. (2005). ESM-Toets [Test for children with specific language
impairment]. Arnhem, the Netherlands: Cito.
Vloedgraven, J., Keuning, J., & Verhoeven, L. (2009). Screeningsinstrument
beginnende geletterdheid [Diagnostic instrument for emergent literacy]. Arnhem,
the Netherlands: Cito.
Voeten, M. J. M., & Van den Bercken, J. H. L. (2003). Lineaire regressieanalyse [Linear
regression analysis]. Groningen, the Netherlands: Stenfert-Kroese.
Vygotsky, L. S. (1978). Mind in society: The development of higher psychological
processes. Cambridge, MA: Harvard University Press.
Wells, G. (1999). Dialogic inquiry: Towards a socio-cultural practice and theory of
education. Cambridge University Press.