Patterns of school readiness among selected
subgroups of Canadian children: Children with special
needs and children with diverse language
backgrounds
Magdalena Janus
Debra Hughes
Eric Duku
Offord Centre for Child Studies, McMaster University
February 2010

Patterns of school readiness among selected subgroups of Canadian children: Children with
special needs and children with diverse language backgrounds
2
This work was funded by a contribution from the Canadian Council on Learning
All documents produced by the Canadian Council on Learning (CCL) will be available in both
French and English. However, documents produced by external organizations for CCL will be
posted on the website only in the language of origin. When a full translation is not available,
CCL will ensure that an executive summary is available in the other official language.
The opinions expressed herein are solely those of the authors. The Canadian Council on
Learning bears no responsibility for its content.

Patterns of school readiness among selected subgroups of Canadian children: Children with
special needs and children with diverse language backgrounds
3
Executive Summary
School readiness is usually understood as a child’s ability to take advantage of the
educational opportunities offered by the school environment. It is a holistic concept including
several developmental areas: physical health and well-being, social competence, emotional
maturity, language and cognitive development, and communication skills and general
knowledge. The Early Development Instrument (EDI) is a Canadian checklist tool measuring
children’s school readiness in all these five areas. Teachers complete the checklist for each child
in their classroom when children are in the second half of their kindergarten year. The EDI also
includes information on children’s characteristics like gender, their first language, or special
education needs. Since 2000, standardized data on children’s readiness to learn at school have
been collected for over half a million children in many communities in Canada with the EDI.
This project had been designed in response to community-level requests for empirical
data on special and small populations. The current document reports school readiness patterns
among 5-year-old Canadian children belonging to two types of important, yet low-frequency
populations: children with special needs and children from diverse language backgrounds - those
for whom the language of instruction is not their first language. The majority of analyses in this
report come from the most recent three-year cycle of data (from 2004/5 to 2006-7), called the
Normative II database, which contains data for over 160,000 children from most regions in
Canada.
EDI scores for all five domains are calculated based on the Normative II database. These
are provided for all children, as well as for boys and girls separately. These values are often used
by communities to establish baselines for comparisons with their own community EDI data.
Small groups, like children with special educational needs, and children with a first language
other than French or English (diverse language backgrounds) are often too small at a local level
to achieve adequate representation for examination of school readiness. In this report, we
examined patterns of school readiness in those small groups in comparison with children who do
not have special needs, and those who have English or French as their first and only language.
As well as providing results to communities, we expect that our findings will inform policy
decisions at many levels and provide the direction for early interventions.

Patterns of school readiness among selected subgroups of Canadian children: Children with
special needs and children with diverse language backgrounds
4
For the purpose of this report, not only general patterns of school readiness against a
control group were examined, but also even smaller groups with more detailed characteristics
were identified, and their school readiness outcomes in five domains of the EDI examined.
When the results are described, each of the groups is placed in a context based on recent data on
Canadians as provided mostly by Statistics Canada, gathered through Census every five years.
Children with special needs represented 3.8% of the population in the Normative II
database. Boys outnumbered girls. As a rule, children with special needs had consistently low
EDI scores in all domains, and the differences between them and the controls were of large
magnitude. The domain of Communication Skills and General Knowledge showed the largest
magnitude of difference. There were also, however, disparities among groups in relation to the
type of needs: for example, students with visual and hearing impairments, while functioning
below their control group peers, did not fare as poorly as those children having other types of
problems.
Children with diverse language background were those whose first language was neither
English nor French. They were further classified into those who were 1) not fluent in the
language of instruction (English or French, depending on the school board’s language) – thus
having only one, non-official language, and classified as Second Language Learners (SLL), and
2) bilingual – those that had English or French as well as another language and were not
classified as English as a Second Language or French as a Second Language Learners. We also
created a language-control group with children who were monolingual in the language of
instruction. Twelve language groupings were identified as numerous enough in the Normative II
database to allow for detailed analyses. Punjabi, Spanish, and Cantonese were the three most
frequent, reflecting the overall Canadian population data. The SLL group of children had
consistently lower outcomes than the bilingual or language-control groups. For some of the
language groups, however, the strong differences were only shown in the language and
communication areas.
Some groups of bilingual children had better outcomes than the language-control group.
The extent of this finding varied in terms of the developmental areas and the magnitude of
difference. The most common pattern was that bilingual children did better than the controls in
the physical development, social and sometimes emotional development, did as well in language,
and tended to do slightly worse in the communication areas.

Patterns of school readiness among selected subgroups of Canadian children: Children with
special needs and children with diverse language backgrounds
5
The findings, on the whole, confirmed the expectations formed on smaller samples. With
the large-scale population data, we were able to show that the two groups of interest, even
though they are often considered as a whole, have a considerable diversity that can only be
untangled with such data as provided by the EDI. The striking differences in school readiness
within language groups in relation to the children’s fluency in the language of instruction are a
very clear indication that more effort has to be directed toward language support for families
with young children.
Placed within the appropriate context, these results can guide practical intervention and
prevention strategies at school and before, as well as provide broad-based background for more
detailed, in depth investigations of specific populations.

Patterns of school readiness among selected subgroups of Canadian children: Children with
special needs and children with diverse language backgrounds
6
Background
Concept of school readiness
The shape of school trajectories is often established early in a child’s school career
(Jimerson, Egeland, & Teo, 1999; Entwisle & Alexander, 1999; Ramey & Ramey, 2004). Thus,
the first years are critical for setting children on a trajectory of success – a course which pays out
future dividends in the areas of later educational attainment and adult health (Zuckerman &
Halfon, 2003). In order to develop optimally in this context, a child needs to be able to profit by
the social and academic environment which the school affords (Janus & Offord, 2000). In other
words, a child needs to be ready to learn at school. According to Forget-Dubois, et al. (2007, p.
406), school readiness is a “multidimensional construct that refers to the cognitive,
communicational, behavioral, and emotional skills, as well as basic knowledge that facilitate the
child’s learning and adjustment at school entry.” It includes both cognitive and non-cognitive
factors, hard and soft skills, general knowledge and approaches to learning, academic and socio-
emotional components (Janus & Duku, 2007). Readiness is a process which develops over the
preschool years, is influenced by the school context and is an outcome of the child’s progress to
date (Meisels, 1999). The challenge then for child development researchers investigating
diversity issues in the context of school readiness is to have a tool which is adequate to measure
the multiple domains of development and to have access to substantive numbers in low-
frequency, heterogeneous groups. The current body of literature is reflective of the difficulty of
this task. For example, in LaParo and Pianta’s (2000) meta-analytic review of more than seventy
longitudinal studies measuring several domains of school readiness on subsequent primary
school performance, the researched populations were small in number (typically no more than a
few hundred children), relied heavily on individual testing or screening measures, and consisted

Patterns of school readiness among selected subgroups of Canadian children: Children with
special needs and children with diverse language backgrounds
7
primarily of Caucasians, with limited representation of children from differing ethnic, cultural
and economic backgrounds.
School readiness should be understood as a holistic concept involving several
developmental areas such as cognitive, socio-emotional and physical (Meisels, 1999; Jimerson et
al., 1999). Competence in all these areas will ensure that children are ready to benefit from
educational activities offered in the school environment (Janus & Offord, 2000). In 1998-1999,
a new instrument was developed at McMaster University to capture this holistic concept in the
measurement of school readiness. The Early Development Instrument (EDI) is a teacher-
completed checklist, containing just over 100 core items grouped into five developmental
domains (Janus & Offord, 2007). For the purpose of the monitoring function, school readiness is
conceptualized in the EDI as a proxy for a holistic view of outcomes of the child’s early years,
within the context of family and neighbourhood. Therefore, at a population level, it is intended
to capture the status of children’s early development in the context of that child’s community.
Since its inception, use of the EDI has proliferated in regions across Canada and
internationally. Its large database provides ample opportunities for comparison studies and, in
combination with community mapping projects, for research into societal factors affecting
children’s physical, emotional and intellectual growth. Based on EDI results, community leaders
can determine patterns of skills and deficits or vulnerabilities that supersede differences in family
circumstances and reflect the larger environment affecting children.
Diversity in a classroom
Diversity characterizes the Canadian classroom. With the increasing commitment in our
classes to inclusion of students with exceptionalities, and with a shifting demographic driven by

Patterns of school readiness among selected subgroups of Canadian children: Children with
special needs and children with diverse language backgrounds
8
new immigration patterns, schools and communities are encountering both new challenges and
opportunities. Indeed, the face of our student population is changing.
It is to be expected that the state of children’s school readiness might be affected by these
changes in the make-up of the classroom. In Canada, school education is under provincial
jurisdiction. The provincial Ministries of Education throughout the country have established
policies and protocols addressing needs of individual children (e.g. Janus, Lefort, Cameron &
Kopechanski, 2007). For example, children who have difficulties in comprehending the
classroom language are offered English (or French) as a Second Language classes, while
children with special educational needs undergo an assessment process supervised by a special
committee (in Ontario, for example, the Identification, Placement, and Review Committee)
intended to determine what specific resources and adjustments might be called for in the
classroom.
While these policies often work for individuals, there is currently no overall, country-
level information on the status of school readiness among children from diverse populations that
might be useful for a broadly-based approach, which addresses these needs before children enter
school. One of the difficulties in creating such information is that these subpopulations are often
both heterogeneous and numerically small. In the face of this, often the greatest challenge is to
gather data which are representative of these subpopulations of interest. These groups might
include, for example, children who belong to low-frequency categories, such as those with
various minority languages and those with a variety of special needs. Although local data may
reflect the demographics of a particular community fairly, small numbers do not allow for
adequate analyses of vulnerability and strength patterns. Small numbers are insufficient for

Patterns of school readiness among selected subgroups of Canadian children: Children with
special needs and children with diverse language backgrounds
9
developing a solid foundation for either program implementation or policy recommendation.
Small numbers capture pixels of the overall picture.
Research questions and the databases
The research questions addressed in this report can be summarized as follows:
1)
Are patterns of school readiness different for children with special needs than for
typically-developing children? Are these differences systematic with respect to specific
problems or conditions?
2)
Are patterns of school readiness different for children with diverse language background
than for children whose first language is the language of the classroom? Are these
differences systematic with respect to specific language? Are these differences consistent
regardless of a child’s fluency in the classroom language?
The Offord Centre at McMaster University houses an extensive national database on the
developmental status of Canadian kindergarten children. This database is used to investigate the
research questions. In many communities in Canada standardized data on children’s readiness to
learn at school have been collected over the past six years with the Early Development
Instrument (EDI). The EDI provides kindergarten outcome measures in the domains of Physical
Health and Well-being, Social Competence, Emotional Maturity, Language and Cognitive
Development, and Communication and General Knowledge
1
(Janus & Offord, 2007). Four of
these could be further subdivided into subdomains. In total, there are 16 subdomains. Table 1
outlines the domains, subdomains, and gives examples of the items.
1
More details on the domains are provided in the Methods section

Patterns of school readiness among selected subgroups of Canadian children: Children with
special needs and children with diverse language backgrounds
10
Table 1
Domains, subdomains, and sample questions on the EDI
EDI Domains
Subdomains
Example items
Physical readiness for school day
arrives at school hungry
Physical independence
has well-coordinated
movements
Physical Health
and Well-being
Gross and fine motor skills
is able to manipulate objects
Overall social competence
is able to get along with other
children
Responsibility and respect
accepts responsibility for
actions
Approaches to learning
works independently
Social
Competence
Readiness to explore new things
is eager to explore new items
Prosocial and helping behaviour
helps other children in distress
Anxious and fearful behaviour
appears unhappy or sad
Aggressive behaviour
gets into physical fights
Emotional
Maturity
Hyperactivity and inattention
is restless
Basic literacy
is able to write own name
Interest in literacy/numeracy, and uses
memory
is interested in games
involving numbers
Advanced literacy
is able to read sentences
Language and
Cognitive
Development
Basic numeracy
is able to count to 20
Communication
Skills and General
Knowledge
(No subdomains)
is able to clearly communicate
one’s own needs and
understand others
shows interest in general
knowledge about the world
These domains and their variations have been widely recognized as fundamental
components of developmental health (Love & Raikes, 2004) and are significant factors for later
school success (; Doherty, 1997; Keating, 2007). Federal governments in both Canada and the
United States have commissioned reports addressing children’s school readiness and both

Patterns of school readiness among selected subgroups of Canadian children: Children with
special needs and children with diverse language backgrounds
11
identify these functional capacities. In Canada the components of school readiness have been
outlined in Zero to Six: The Basis for School Readiness (Doherty, 1997), and in the United
States in Goal One of the National Education Goals (Kagan, Moore & Bredekamp, 1995;
Shepard, Kagan & Wurtz, 1998).
The Early Development Instrument is a psychometrically sound tool which is adequate to
measure the requisite multiple domains of development. And, due to the scope of its
implementation in Canada over the past decade, the EDI database is impressive in both number
and geographic scale. A representative national sample of over 116,000 kindergarten children
during the years from 2000 to 2004 was collected. A second sample followed thereafter
covering the period 2005 to 2007, including over 176,000 children. In addition to these
normative gold standard samples were groups of children with identified special needs,
numbering over 3000 and 7000 respectively. Thus the EDI has the rigour to measure children’s
developmental health on a wide scale, and the capacity to provide substantive numbers of low-
frequency, heterogeneous groups such as linguistically diverse children and those with special
needs.
The analysis of children’s school readiness in this report is driven by the belief that these
are the outcomes of their earliest experiences fostered through relationships within the family,
community and school settings. While measuring children’s competencies in various domains of
functioning, the EDI at its core most truly reflects an interactionist and developmental systems
approach. This recognizes the embeddedness of children in multiple social systems as they
negotiate their external world as agents and recipients (Mashburn & Pianta, 2006). Since child
development is fairly open and fluid during these formative years, early intervention programs

Patterns of school readiness among selected subgroups of Canadian children: Children with
special needs and children with diverse language backgrounds
12
may favourably impact multiple domains of school competencies (Barnett, Young, &
Schweinhart, 1998).
Measures
The Early Development Instrument
The Early Development Instrument (EDI) is a teacher-completed 104-question checklist
measuring children’s readiness to learn at school before entry to Grade 1 in five domains. The
EDI was developed by Dan Offord and Magdalena Janus at the Offord Centre for Child Studies,
McMaster University in 1998, and finalised in 2000 (Janus & Offord, 2007).
The Physical Health and Well-being domain refers to children’s physical preparedness
for the school day, fine and gross motor skills, energy level throughout the day, and physical
independence. The Social Competence domain covers the following areas: competence and
cooperation in working together with others, ability to remember and follow rules, curiosity and
eagerness, approaches to learning and problem-solving. The Emotional Maturity domain covers
prosocial behaviour, aggression, inattention and hyperactivity, and anxious behaviours. The
domain of Language and Cognitive Development refers to the child’s ability to use language
correctly and covers cognitive aspects of language and numeracy, in several areas: basic literacy
and numeracy skills, interest and memory, and more complex literacy. The Communication
Skills and General Knowledge domain covers the child’s ability to clearly communicate one’s
own needs and thoughts in a way that is understandable to both adults and other children, the
ability to understand others, clear articulation, as well as aspects of general knowledge. In

Patterns of school readiness among selected subgroups of Canadian children: Children with
special needs and children with diverse language backgrounds
13
contrast to the previous domain, this one is about effective communication regardless of the
grammatical correctness
2
.
The EDI is completed by teachers in the second half of the kindergarten year, usually
over a period of 2-3 weeks between February and March. Thus, all children are at the same
grade level, but since the age of entry to kindergarten is based on the child’s birthday by
December of the preceding year, in any given classroom the variation in age can be as large as
one year (i.e., between children born in January and December).
The currently accepted definition of school readiness encompasses all areas of child
development: socio-emotional, cognitive, approaches to learning, and physical well-being.
However, there are few instruments available to provide an opportunity to assess a child along all
these dimensions in a feasible and economical format. From that point of view, the existing EDI
database of Canadian children offers unparalleled advantage of exploring empirical levels of
children’s readiness to learn at school. First, the instrument covers the relevant developmental
areas. Second, due to its implementation for whole populations of children, the database
includes data for all children in the community, and not only a sample of them. Thirdly, on
account of such wide-scale implementation, it allows aggregation of information for populations
of children with specific characteristics, who otherwise may contribute only very few scores to
community-level data. For instance, in the majority of communities there are subgroups of
children with specific characteristics, such as special needs or minority languages, who often
constitute less than five percent of the whole kindergarten community.
The EDI is completed by primary education teachers who receive board-wide instruction
in the tool. This provides consistent region by region and year over year reporting. It also
2
More information on the EDI, its implementation in Canada, results, and bibliography, can be
found at www.offordcentre.com/readiness.

Patterns of school readiness among selected subgroups of Canadian children: Children with
special needs and children with diverse language backgrounds
14
prevents differing interpretations and misunderstandings of questions, language challenges,
uneven educational and literacy backgrounds and the low return rates of parent-completed
questionnaires.
Procedure
The Normative Databases
Since 1999 EDI data has been collected for over 520,000 4-5 year old children in Canada
and more than ten other countries. Normative data on the EDI domains are based on a
community-representative subgroup of the current database collected in Canada over the years
1999/2000 to 2003/2004 for Normative I, and 2004/2005 to 2006/2007 for Normative II (Janus
& Duku, 2004; 2008). All ten Canadian provinces participated in the first normative database.
The second normative database, while significantly larger in number, comprised data from seven
of the provinces. The three smaller missing Maritime provinces have just recently completed
another EDI implementation in their scheduled rollout.
A small modification in coding of eighteen items was introduced in 2004 based on
outcomes of psychometric analysis (Janus & Offord, 2007). While overall these modifications
did not cause any substantive changes, it is a standard practice with the EDI data to consider
databases collected prior to 2004 separately from those collected after. We include the overall
results for both databases; however, for clarity of presentation the detailed analyses are presented
only for the more recent database, Normative II.
Several criteria in terms of data quality had to be applied in order to identify the subgroup
for the normative, “gold standard”, sample. First, only children attending the Senior
Kindergarten year (entry at 5 years of age) were included. Second, for communities that

Patterns of school readiness among selected subgroups of Canadian children: Children with
special needs and children with diverse language backgrounds
15
participated more than once over the respective data collection period the fullest or the most
recent implementation was included in that normative database. The next step in ensuring the
reliability of data was to restrict the sample by excluding the following: 1) missing or out of
appropriate range age, 2) missing gender, and 3) missing two or more domains. Children with
special needs status (or missing this information) were filtered out of this normative sample.
This resulted in the final sample of 116,860 children for Normative I, and 176,621 for Normative
II. This represents 88.4% and 91.7% respectively of the initial samples. Children with identified
special needs comprised a group numbering 3061 over the years 2000 to 2004 and 7089 from
2005 to 2007.
There are a number of subpopulations that could be identified within such a large and
comprehensive database as the EDI, due to the demographic data collected concurrently. They
are listed below.
- children with special needs: children reported on the first page of the EDI as
“having Special Needs”, based on school board special education classification
procedures. Note: this is the only subpopulation that is routinely excluded from
normative databases due to the fact that it includes children that are likely not
developing typically. Their data are not included in community-level school (or
neighbourhood) reports.
- children with first language other than English (in Anglophone schools) or
French (in Francophone schools): children reported by teachers as having a
different language than the language of instruction; they are termed Second
Language Learners; including those reported as bilingual, and those reported as
monolingual in a language other than English or French,

Patterns of school readiness among selected subgroups of Canadian children: Children with
special needs and children with diverse language backgrounds
16
- children with special problems: these are children for whom teachers endorse a
statement “child has a problem that influences his/her ability to do school work
in a regular classroom” (Question D1 on the EDI), based on parent information.
Note: these are endorsed for children who do not have “Special Needs” as well
as for those that do. They are separated in analyses.
- children with special skills: these are children for whom teachers endorse
one or more of the special skills listed in Section B,
- children who require additional assessment: these are children for whom
teachers endorse a statement “child requires a further assessment” (Question
D3 on the EDI) Note: these are endorsed for children who do not have
“Special Needs” as well as for those that do. They are separated in analyses.
- children with Aboriginal status: these are children for whom teachers
endorse a statement “child has Aboriginal status”.
The percentages of each of these subgroups in both Normative databases are on Figure 1.
In this report, we focus on the first two of these crucial, yet not very numerous,
subpopulations: children with special needs and children with first language other than English or
French (language of instruction). Our goal was to establish levels of EDI-measured school readiness
within subgroups and to compare these to established levels for children outside these populations;
that is, a control group of children with none of the six subgroup characteristics. The control groups
consisted of children with no special needs, monolingual in the language of instruction, no special
problems, no special skills, no need for further assessment, and no Aboriginal status. These control
groups consisted of 84,393 children for Normative I, and 127,559 for the Normative II period.

Patterns of school readiness among selected subgroups of Canadian children: Children with
special needs and children with diverse language backgrounds
17
Figure 1
Percentages of children comprising various subpopulations in the normative
databases
Subpopulation of Children with Special Needs
According to the 2006 Statistics Canada Participation and Activity Limitation Survey
(PALS), the percentage of 5 to 9 year old Canadian children with disabilities is 4.2%, representing
some 74,580 individuals (Statistics Canada, 2008a). The gender breakdown for this age group is
5.3% boys and 3.0% girls.

Patterns of school readiness among selected subgroups of Canadian children: Children with
special needs and children with diverse language backgrounds
18
Table 2
Composite results of PALS for childhood disabilities
Age in
Years
2001 Total
Population
Disabilities
#
Disabilities
%
Boys
%
Girls
%
2006
Total
Populati
on
Disabilities
#
Disabilities
%
Boys
%
Girls
%
0 -14
5,546,020
180,930
3.3
4.0
2.5
5,471,36
0
202,350
3.7
4.6
2.7
0-4
1,641,680
26,210
1.6
1.9
1.3
1,656,04
0
27,540
1.7
2.1
1.2
5-14
3,904,330
154,720
4.0
4.9
3.0
3,815,31
0
174,810
4.6
5.7
3.4
5-9
1,914,220
70,370
3.7
4.6
2.7
NA
74,580
4.2
5.3
3.0
10-14
1,990,110
84,350
4.2
5.1
3.3
NA
100,230
4.9
6.0
3.7
Sources: Statistics Canada, 2002a,b, 2008a
From the compilation of the Statistics Canada data, it is evident that the overall rate of
disabilities increases with age. For example, in 2006 only 1.7% of Canadian preschool children
(age 0 to 4) have disabilities. This rate increases to 4.6% for elementary school children (ages 5
through 14) (Statistics Canada, 2008b). In large measure this reflects the time taken for some
disabilities to become apparent, to be diagnosed, or to occur. PALS recognizes that some types
of disabilities are not identified before age five. For example, learning and psychological
disabilities are not specified for the preschool group in the PALS survey. As well, the process of
identification itself takes time to unfold. Additionally, disease and accidents over the course of
years contribute to the rates. The EDI Normative II kindergarten rate of 3.8% is a reasonable
reflection of the overall PALS rate of 4.2% for children ages 5 to 9 given that the disability rate
rises with age. The EDI is capturing a particular transition year between the preschool and
school-aged groups.

Patterns of school readiness among selected subgroups of Canadian children: Children with
special needs and children with diverse language backgrounds
19
The percentage of children with special needs as a component of the total population
grew between the two normative datasets (Figure 2). Both the PALS survey and the EDI have
shown increasing rates of disabilities among Canadian children over the last decade. In 2001,
the disability rate for Canadian children aged 5 to 9 according to PALS was 3.7% (Statistics
Canada, 2002a), growing to 4.2% by 2006 (Statistics Canada, 2008a). The EDI rate reflects this
pattern increasing from 2.55% to 3.8%. Of particular note is that while disability rates are rising,
overall school enrollment numbers are marginally declining. This has important implications for
both the health care and educational systems.
Figure 2
Percentage of children with special needs in the two databases
Percentage of Children with Special Needs
0
0.5
1
1.5
2
2.5
3
3.5
4
Norm I
Norm II
Database Period
Pe
rc
e
n
t
Norm I
Norm II
Boys significantly outnumber girls in the population with special needs. The EDI reflects a
rate of approximately 70% boys to 30% girls, a differential which is consistent over both sets of
data periods (Figure 3). The rates of boys with special needs, as well as boys who have one or
more special problems, or boys who need further assessment are strikingly higher than the girls’

Patterns of school readiness among selected subgroups of Canadian children: Children with
special needs and children with diverse language backgrounds
20
rates. Regarding special needs, boys have more biological health problems than girls, and are
more likely to have developmental delays, intellectual disabilities and behavioural disorders
(Gissler, Järvelin, Louhiala, & Hemminki, 1999). Consequently they make up the majority of
the children in special educational programmes.
Figure 3
Gender patterns in subpopulation of children with special needs in the two
databases
Gender Patterns over Database Periods
0
10
20
30
40
50
60
70
80
Boys
Girls
Gender and Database Period
P
e
rc
ent
Norm I
Norm II
Because the EDI is a developmental measure, it taps the various domains of functioning.
Physical Health and Well-Being, Social Competence, Emotional Maturity, Language and
Cognitive Development, and Communication Skills and General Knowledge are each scored on
a 10 point scale. These domains are not mutually exclusive: some combinations such as physical
health and emotional maturity have weak correlations (0.42) while social competence and
emotional maturity show strong interrelationships (0.77).

Patterns of school readiness among selected subgroups of Canadian children: Children with
special needs and children with diverse language backgrounds
21
In order to provide a consistent method of group comparison, independent of the sample
size, effect sizes for each of the five EDI domains were calculated in the following way:
)
.(.
)
(
)
(
control
ds
control
mean
attribute
mean
-
. Since it is independent of the measurement or sample size,
effect size of a difference between two groups is the best indicator of how meaningful this
difference is. It is customary to interpret the effect sizes of 0 to 0.3 as small, 0.3 to 0.8 as
moderate, and greater than 0.8 as large. An effect size of 0.8 means that 79% of the comparison
group is higher than the control group. An effect size of 1.6 indicates that 95% of the
comparison group is higher. Negative effect sizes mean the comparison group is lower than the
control group.
Children with identified special needs function significantly below the control group of
children in all domains. The EDI means in the group with special needs were statistically
significantly lower than those in the control group (Table 3).
Table 3
EDI Domain means, standard deviation and effect sizes for the difference
between children with Special Needs and Control group in the Normative II
database
Special Needs
Control
Domain
N
Mean
SD
N
Mean
SD
Effect Size
Physical Health and
Well-Being
7072
6.85
2.03
100820
8.77
1.26
-1.52
Social Competence
7070
5.39
2.52
101028
8.33
1.73
-1.70
Emotional Maturity
7006
5.93
1.94
100238
8.06
1.45
-1.47
Language and
Cognitive
Development
7043
5.65
2.97
100620
8.45
1.68
-1.67
Communication Skills
and General
Knowledge
7077
3.82
3.08
101010
7.86
2.31
-1.75

Patterns of school readiness among selected subgroups of Canadian children: Children with
special needs and children with diverse language backgrounds
22
The widest divergence was in the communication skills and general knowledge domain.
This domain taps the child’s ability to effectively communicate thoughts and needs to both adults
and children, to understand others’ communication, to articulate clearly, and to answer general
knowledge questions (Janus & Offord, 2007). On the face of it, this domain may seem to be
particularly impacted by the high rates of children with speech and language problems. Within
both the EDI results and PALS 2006, speech problems figure prominently among the rates of
childhood disabilities. For instance, PALS reports that 45% of elementary school aged children
with disabilities have speech problems.
For children with special needs all effect sizes for the five EDI domains in the Normative
II database were large (-1.71 to -2.1). These were all significant when controlling for age and
gender. Children with special needs are as a group negatively impacted in all areas of
functioning. This, however, should not be surprising. It is difficult to think of a condition that
would be limited to only a few areas of functioning. For example, a physical impairment may
have implications for one’s ability to engage socially with peers or to possess sufficient motor
skills for school work. A behavioural problem will invariably impact one’s social interactions as
well as one’s self-regulation in areas such as impulse control and focusing attentional processes.
The EDI also allows for a more nuanced consideration of difficulties which children may
experience. We further examined various disabilities in relation to the control group (Figure 4),
as recorded in Section D of the EDI. These included the following seven categories: physical
disabilities, visual, hearing and speech impairments, learning disabilities, emotional and
behavioural problems. All children for whom any of these disabilities were endorsed, were
considered, regardless whether the child was classified by the school board as requiring special
education or not. Over four percent of children (4.5%) fell into the category of having one or

Patterns of school readiness among selected subgroups of Canadian children: Children with
special needs and children with diverse language backgrounds
23
more special problems. Even though the impairments are not mutually exclusive: (for example, a
child may have both a speech and hearing problem or emotional and behavioural difficulties),
each of these disabilities was compared against the control group.
Figure 4
Percentages of various special problems in the Normative II database
Types of Special Problems
0
0.5
1
1.5
2
2.5
3
3.5
4
4.5
5
Physical
Visual
Hearing
Speech
Learning Emotional Behavioural
Problem by Type
P
e
rc
ent
Norm II %
Several features stand out in the descriptive analysis. Speech problems, as noted above,
are by far the greatest problem type. As well, the mean age of children with special problems in
each of these conditions is above that of the control group. Children with physical and learning
disabilities were almost a month older on average than their control counterparts (mean of 5.79
years compared with 5.70 years of age). The other categories of impairment showed
approximately a half month age increase. This may reflect the desire on the part of some parents
to register their children later, or the school’s recommendation to hold some children back or
follow a combination of half day kindergarten/half day grade 1. Finally in all problem areas, the

Patterns of school readiness among selected subgroups of Canadian children: Children with
special needs and children with diverse language backgrounds
24
communication skills and general knowledge domain seemed to be the one most affected. While
the control group had a mean of 8.13, the specific impairments had a range of 3.33 (learning
disabilities) and 3.78 (speech impairment) to a high of 6.12 (visual impairment). It must be
remembered though that a child may have co-morbidities across problem areas. Finally, and not
surprisingly, children with emotional or behavioural problems scored low on the social and
emotional domains.
Effect sizes were large in all domains in all non-sensory problem areas (i.e., physical
disability, speech and learning impairments, emotional and behavioural problems, see Table 4).
These had a range of -1.05 to -2.87 (i.e. 84 – 99% of control children scored better). Thus in all
domains, in all these problems areas, children were functioning one to several standard
deviations below their control group peers. And, it must be remembered that this group
encompasses more than just children with diagnosed special needs. Many of these students will
find themselves struggling in multiple domains without the benefit of formal recognition by the
school board, possibly hampering their ability to access valuable resources.
Table 4 Effect sizes of differences in comparison of groups with specific problems with the
control group
Physical
Disability
Visual
Impairment
Hearing
Impairment
Speech
Impairment
Learning
Disability
Emotional
Problem
Behaviour
Problem
Physical Health
& Well-Being
-2.13
-1.05
-0.78
-1.34
-1.96
-1.62
-1.42
Social
Competence
-1.47
-0.88
-0.85
-1.30
-2.19
-2.38
-2.57
Emotional
Maturity
-1.12
-0.69
-0.65
-1.05
-1.73
-2.44
-2.58
Language &
Cognitive
Development
-1.75
-1.04
-0.97
-1.76
-2.87
-1.63
-1.63
Communication
& General
Knowledge
-1.57
-0.95
-1.24
-2.05
-2.26
-1.49
-1.37

Patterns of school readiness among selected subgroups of Canadian children: Children with
special needs and children with diverse language backgrounds
25
Students with visual and hearing impairments, while functioning below their control
group peers, did not fare as poorly as those children having other types of problems. While most
domain effect sizes were large, effect sizes for emotional maturity were only moderate and for
hearing impairment moderate in the physical health and well-being domain (i.e., less than 0.8).
All effect sizes were significant when controlling for age and gender.
In conclusion, children with special educational needs and with problems affecting their
ability to learn at school are disadvantaged in all five domains of development in comparison to
their typically-developing peers. However, our analyses indicate that these children do not
constitute a uniform group, and much can be learned by examining the patterns of their school
readiness among categories of impairment. Since the frequencies of these impairments are
generally low, databases such as the EDI one, collected on a population level, can be particularly
helpful.
Subpopulation of Children with Diverse Language Backgrounds
Immigration patterns and demographics
Canada has historically been a country of immigrants. As of the 2006 Canadian census,
the following demographics now characterize our population (Chui, Tran & Maheux, 2007). The
foreign-born population now accounts for 19.8% of our total populace, reaching the highest
proportion in 75 years. These individuals have come from over 200 countries and speak almost
150 different languages. In the five years between the last census enumerations (2001-2006),

Patterns of school readiness among selected subgroups of Canadian children: Children with
special needs and children with diverse language backgrounds
26
over a million people have made their way to Canada, accounting for two-thirds of our
population growth during those years. Of these newcomers, one in five is under the age of 14.
Historically, most immigrants to Canada have been of European origin. For example in
1971, Europeans accounted for just over 60% of new immigrants. Beginning in the 1980s
though, people from Asia and the Middle East began to arrive in substantial numbers. As of the
2006 census, the proportion of those from Asia (including the Middle East) had surpassed that of
European-born. Of the million-plus newcomers who had arrived in Canada in the period from
2001 to 2006, 58% of them were from Asia. China, India, the Philippines and Pakistan topped
the list for country of origin. In total, these four comprised almost 38% of all recent immigrants.
Notably, recent arrivals from Europe accounted for only 16%, while those from Central and
South America and the Caribbean accounted for approximately a further 11%, with the African
continent just marginally lower. Figures 5 and 6, based on Census outcomes charts from the
Immigration in Canada, 2001 and 2006 Census reports, illustrate the changing immigration
patterns in Canada over the last thirty years.

Patterns of school readiness among selected subgroups of Canadian children: Children with
special needs and children with diverse language backgrounds
27
Figure 5
Region of birth of recent immigrants to Canada, 1971 to 2006
Notes:
1. 'Recent immigrants' refers to landed immigrants who arrived in Canada within five years prior to a given census.
2. "Other" includes Greenland, St Pierre and Miquelon, the category 'other country', as well as a small number of
immigrants born in Canada.
Sources: Statistics Canada, censuses of population, 1971 to 2006; Chui, Tran & Maheux (2007).
Figure 6
Immigrants by period of immigration—world regions of birth, Canada, 2001
(percentage distribution)
Source: Citizenship & Immigration (2005)

Patterns of school readiness among selected subgroups of Canadian children: Children with
special needs and children with diverse language backgrounds
28
Due to the increasing diversity of sending countries, Canada’s language landscape has
become correspondingly diverse (Statistics Canada, 2007a). For the first time, fully 20% of the
Canadian population reports a mother tongue which is neither English nor French (Statistics
Canada, 2007b). This record setting rate is in large measure driven by recent immigration
patterns in which four out of five people who now come to Canada are Allophones – those
whose mother tongue is neither of Canada’s official languages (Statistics Canada, 2007a).
Nevertheless, most of the recent immigrants stated that they were able to converse in either
English or French (Chui, Tran & Maheux, 2007).
Overall, the top ten non-official languages within Canada are, in order: Chinese
languages, Italian, German, Punjabi, Spanish, Arabic, Tagalog, Portuguese, Polish, and Urdu
(Statistics Canada, 2007a). However, in the intercensal period of 2001 to 2006, the languages
from Asian/Middle Eastern and Latin American regions recorded the largest gains. These
included Chinese, Punjabi, Arabic, Urdu, Tagalog, Tamil and Spanish. One out of eight
Canadians speaks a non-official language most often at home (Statistics Canada, 2007a,b). The
following figure illustrates the extent to which Canadian cities are facing substantial percentages
of immigrant children among their school-aged population.

Patterns of school readiness among selected subgroups of Canadian children: Children with
special needs and children with diverse language backgrounds
29
Figure 7
Proportion of immigrants among the school-age population (ages 5-24),
selected census metropolitan areas, 1991, 1996, 2001
Source: Statistics Canada (2004)
Definitions of language groups for EDI data analyses
Language information on the EDI has been collected since the 1999/2000 school year.
On the first page of the EDI, Question 12 asks about the child’s first language, giving the teacher
seven answer options: English only, French only, Other only, English and French, English and
Other, French and Other, and Other and Other (“Other” is always accompanied with a space for
specifying the language).
Children in Anglophone school boards with English only as the first language, and
children in Francophone school boards with French only as the first language were combined to
form a “Language-control” group. Children in any answer where “Other” language was listed
were combined to form a “Diverse-language” category. The only group entirely omitted from
the analyses in this report consisted of children who were reported to have both English and

Patterns of school readiness among selected subgroups of Canadian children: Children with
special needs and children with diverse language backgrounds
30
French as the first language. (Children with special needs do not form part of the normative
samples and were not included in any language analysis).
Question 8 on the first page of the EDI requests the teacher to record whether the child is
classified as having English as a Second Language (in Anglophone school boards) or French as a
Second Language (in Francophone School Boards). Children in the “diverse-language” category
AND endorsed as having the language of instruction as the second language are defined as
“Second Language Learners” (SLL) for the purpose of analyses. Children in the “diverse-
language” category AND NOT endorsed as having the language of instruction as the second
language are defined as “Non-Second Language Learners” (non-SLL or “Bilingual” (BIL)
(Figure 8).
Figure 8
Selection of the samples from the database (Normative II).
176621
169727
4220
2674
127559
37920
4248
19743
18177
2
3
4
5
6
7
8
9
1: Total population
2: Bilingual in English and French
3: Sample with excluded bilingual and
missing First Language variable
4: Missing First Language variable
5: Language-controls (French only &
English only)
6: EFSL classification valid (Yes or No)
7: Missing EFSL status variable
8: EFSL=Yes (SLL)
9: EFSL=No (Not SLL=Bilingual)

Patterns of school readiness among selected subgroups of Canadian children: Children with
special needs and children with diverse language backgrounds
31
Table 5
Frequencies of Language-control, and Diverse-language groups
Normative I
Normative II
N=113,642
N=169,727
N
%
N
%
Language-
control
84393
74.3
127,559
75.2
Diverse-
language
23,800
20.9
37,920
22.3
SLL
13,044
11.5
19,743
11.6
Bilingual 10,756
9.5
18,177
10.7
Missing EFSL
classification
5,449
4.8
4,248
2.5
Therefore, the following comparisons will be analyzed in this section of the report: first,
the Language-control (box 5) with Diverse-language group as a whole (boxes 8 & 9); second, the
Language-control with Second Language Learners (box 8), and Bilingual groups (box 9)
separately.
EDI outcomes
As can be seen from Table 5, there was a small increase over time in the frequency of
kindergarten children coming from diverse language background (20.9% to 22.3%). The group
of bilingual children who are fluent in the school’s prevailing language of instruction and
therefore are not considered to require school language supports has grown slightly, from 9.5%
to 10.7%. As the proportion of Second Language Learner group remained relatively steady, this
may be due to the sharp drop in the percentage of children with diverse language for whom the
English/French learner status was not specified (from almost 5% to 2.5%). While it is possible

Patterns of school readiness among selected subgroups of Canadian children: Children with
special needs and children with diverse language backgrounds
32
that teachers became more accurate in their reporting, it could, however, also suggest a greater
accuracy in identifying children who need SLL supports. While the increased rates identified by
the EDI reflect only kindergarten children, they are indicative of greater numbers of children
who are spread throughout the school system. As noted earlier, 20% of newcomers to Canada
are children of elementary school and preschool age.
Group comparisons
When Diverse-language group was compared with Language-controls, there were only
trivial effect sizes on four of the five domains (Table 6). Only in the Communication Skills and
General Knowledge domain there was a moderate effect size (-0.63) However, most mean
scores were higher among the Language-controls.
Table 6
Comparison of Diverse-language and Language-control groups: Domain
scores, standard deviations (SD), and effect sizes
Diverse-language
Language-control
Effect Size
Domain
N
Mean
SD
N
Mean
SD
Physical Health and
Well-Being
37804
8.79
1.28 127276
8.78
1.31
0.001
Social Competence
37906
8.21
1.81 127525
8.35
1.78
-0.08
Emotional Maturity
37566
8.01
1.42 126600
8.06
1.51
-0.03
Language & Cognitive
Development
37738
8.23
1.91 127034
8.53
1.73
-0.17
Communication Skills
& General Knowledge
37880
6.55
2.93 127503
8.00
2.32
-0.63

Patterns of school readiness among selected subgroups of Canadian children: Children with
special needs and children with diverse language backgrounds
33
Further, Second Language Learners scored lower than Language-controls in all EDI
domains (Table 7). The effect size of the difference was moderate for the Language and
Cognitive Development Domain (-0.39), but large for the Communication Skills one (-1.09).
Table 7
Comparison of Second Language Learner and Language-control groups:
Domain scores, standard deviations (SD), and effect sizes.
Second Language
Learners
Language-control
Effect Size
N
Mean
SD
N
Mean
SD
Physical Health and
Well-Being
19686 8.62
1.35
127276
8.78
1.31
-0.12
Social Competence
19731 7.96
1.89
127525
8.35
1.78
-0.22
Emotional Maturity
19509 7.84
1.45
126600
8.06
1.51
-0.15
Language and
Cognitive
Development
19626 7.85
2.06
127034
8.53
1.73
-0.39
Communication Skills
and General
Knowledge
19723 5.47
2.93
127503
8.00
2.32
-1.09
Bilingual children, as a whole, had slightly higher, but largely similar scores as the
Language-controls (Table 8). Effect sizes clearly indicate that there were no meaningful
differences in the scores between these two groups.

Patterns of school readiness among selected subgroups of Canadian children: Children with
special needs and children with diverse language backgrounds
34
Table 8
Comparison of Bilingual and Language-control groups: Domain scores,
standard deviations (SD), and effect sizes
Bilingual
Language-control
Domain
N
Mean
SD
N
Mean
SD
Effect Size
Physical Health and
Well-Being
18118
8.97
1.18 127276
8.78
1.31
0.15
Social Competence
18175
8.48
1.68 127525
8.35
1.78
0.07
Emotional Maturity
18057
8.18
1.37 126600
8.06
1.51
0.08
Language and
Cognitive Development
18112
8.65
1.63 127034
8.53
1.73
0.07
Communication Skills
and General
Knowledge
18157
7.71
2.44 127503
8.00
2.32
-0.13
The difference in magnitude between the two language-oriented domains when exploring
the SLL children’s outcomes (Table 7) is likely a result of the conceptual underpinnings of each
of them. The Language and Cognitive Development domain on the EDI is intended to cover
areas of literacy and numeracy, and the basic use of the language of instruction in a correct
manner. A child who came to school with limited understanding of the language, can still
benefit from classroom reading activities, can learn letters and numbers, and demonstrate
memory skills or time concepts. Such children will likely be able to express themselves with
less fluency than a child with full command of the language, but if these Language Learners are
normally developing, besides this one disadvantage, there is a potential for these children to
catch up. In contrast, the Communication Skills and General Knowledge domain covers areas of
functional knowledge of the language of instruction and the reality of the surroundings. This
domain reflects a child’s ability to act using language in ways that require quick reaction to

Patterns of school readiness among selected subgroups of Canadian children: Children with
special needs and children with diverse language backgrounds
35
spoken commands and full comprehension of phrases. Moreover, this domain reflects the
amount of exposure the child would have had to the language of instruction and its culture prior
to school entry. It is therefore not surprising that the magnitude of a gap between the Language
Learners and children fluent in the language of instruction will be bigger in the Communication
Skills and General Knowledge domain than in the Language and Cognitive Development
domain.
There were no particular gender differences in percentages of bilingual or SLL children.
All comparisons between the control group and the bilingual and SLL children remained
significant when controlling for age and gender.
Specific language groups
Since 2001/2 school year, the EDI has asked teachers to specify the child’s first
language(s). The EDI-identified top languages are virtually identical to those indicated by the
Statistics Canada reports. The EDI indicates that Punjabi, Spanish, Cantonese, Arabic,
Filipino/Tagalog, Mandarin, Urdu, Vietnamese, Tamil, Portuguese, Italian and Polish were the
most frequent non-official mother tongue
3
.
Both normative EDI periods reflect the same most frequent twelve languages. These
languages represent 7.7% (2001/2 to 2003/4) and 10.3% (2004/5 to 2006/7) of the entire EDI
populations for those years (8,781 and 17,428 respectively). These group sizes had a range of
312 to 2140 children in the earlier collection period and 663 to 3929 in the latter time. These top
twelve groups represent a mix of both Second Language Learners (SLL) and bilingual (BIL)
3
Only German, which was on the Statistics Canada list, was not among the top EDI languages. Statistics Canada
information includes the child and adult population and thus reflects immigration patterns over longer periods.

Patterns of school readiness among selected subgroups of Canadian children: Children with
special needs and children with diverse language backgrounds
36
speakers. The percentages with which these dozen languages occurred vary somewhat over the
two databases.
From the EDI databases, we selected children with these twelve most frequent mother
tongues. To enable more detailed comparisons, we used the cut-off of at least 0.4% of the
sample in the Normative II database,
4
as these analyses were only carried out for this database.
Table 9
Frequency of the twelve top languages in each Normative database
2001/2-2003/4
2004/5-2006/7
Language
Percent
Language
Percent
Punjabi
1.9%
Punjabi
2.3%
Cantonese
0.8%
Cantonese
1.3%
Spanish
0.7%
Spanish
1.0%
Urdu
0.7%
Urdu
0.7%
Filipino/Tagalog
0.6%
Filipino/Tagalog
1.0%
Vietnamese
0.6%
Vietnamese
0.6%
Arabic
0.6%
Arabic
0.9%
Polish
0.5%
Polish
0.4%
Portuguese
0.4%
Portuguese
0.4%
Tamil
0.4%
Tamil
0.4%
Mandarin
0.3%
Mandarin
0.8%
Italian
0.3%
Italian
0.4%
Punjabi, Spanish, Arabic, Filipino/Tagalog and Mandarin showed the greatest percentage
increase. This reflects the large gains in immigration from Asia and Latin America as described
in a Statistics Canada report (2007a). In particular China, India, the Philippines and Pakistan
were the greatest source countries for immigration. By language, Spanish ranked second in
terms of rate of growth.
4
The Mandarin and Italian language groups constituted only 0.3% in Normative I, but were retained for
consistency.

Patterns of school readiness among selected subgroups of Canadian children: Children with
special needs and children with diverse language backgrounds
37
These language groups are not mutually exclusive since there may be some situations in
which children were taught from infancy both Portuguese and Spanish, or Cantonese and
Mandarin, for example. Furthermore, each language represents a mixed group of mono-,
bilingual or even multilingual children: the criterion for inclusion being their first language(s)
assignation in a top twelve language group. Within each group, children were further classified
as Second Language Learners or Bilingual.
From this point, each of the twelve groups was compared directly with the Language-
control group defined earlier. Statistically significant differences (p<0.05) were noted in all of
the domains and in most of the languages when compared with the controls. In most cases, the
Language-controls had higher EDI scores than the Diverse-language comparison groups as a
whole. For six out of the 12 specific groups, the Language-control group always had better
scores than the Diverse-language group: Punjabi, Spanish, Arabic, Urdu, Vietnamese, and
Portuguese. However, once the Diverse-language groups were split into Second Language
Learners and Bilingual groups (see Table 9 for frequencies), the Bilingual group in almost any
language had the highest scores, with the exception of Language and Cognitive, and
Communication Skills domains.

Patterns of school readiness among selected subgroups of Canadian children: Children with
special needs and children with diverse language backgrounds
38
Table 10 Proportions of SLL and Bilingual students in each of the 12 diverse language
groups in each database
Normative I
Normative II
Language
Second
Language
Learners
Bilingual
Second
Language
Learners
Bilingual
%
%
%
%
Punjabi
67.9
32.1
75.5
24.5
Cantonese
57.1
42.9
71.2
28.8
Spanish
44.1
55.9
34.4
65.6
Urdu
62.7
37.3
65.7
34.3
Filipino/Tagalog
40.4
59.6
42.5
57.5
Vietnamese
63.2
36.8
72.6
27.4
Arabic
51.5
48.5
45.3
54.7
Polish
37.2
62.8
38.0
62.0
Portuguese
25.2
74.8
13.5
86.5
Tamil
53.7
46.3
49.4
50.6
Mandarin
56.3
34.7
73.3
26.7
Italian
8.8
91.2
4.4
95.6
The variance between the language groups appears to reflect the bigger historical picture.
The group’s historical migration pattern is likely to have influenced the balance of the mono- and
bilingual proportions within any language. Differences between groups in their migration
history provide for the varied pattern evident in Table 10. For example, whereas the European
waves of immigration occurred during the course of the 20
th
century, the greatest proportions
happened pre-1970 (see Figure 5). The EDI data indicate that most of the children in the Italian,
Portuguese and Polish groups are bilingual. Italian is the most marked of these groups with only
seven to nine percent being SLL. The remaining more than ninety percent have exposure to both
English and Italian in their early years. This is indicative of a migration timeframe in which the
parents of these kindergarten children, themselves either immigrants or children of immigrants,
continue to pass on the heritage language. Spanish and Filipino/Tagalog languages are more

Patterns of school readiness among selected subgroups of Canadian children: Children with
special needs and children with diverse language backgrounds
39
evenly divided between bilingual and monolingual speakers. Although the Philippines is a
recent and significant source country for immigration, Filipinos enjoy the benefit of English
being the national language of commerce and the language of instruction in most preschools and
schools. Thus a large number of parents and children have functional knowledge of English
upon arrival in Canada (personal communication, March, 2009). Explanations are not as readily
apparent for Spanish, but factors such as length of time spent in the United States (either as a
source country or en route to Canada) or the SES of the immigrant population which may
correlate with access to English educational instruction may play into these language balances.
More recently arrived groups (i.e., those from Asia such as China, India and Pakistan) have the
highest rates of SLL children, with percentages typically around 70%.
Most consistency was observed in the Communication and General Knowledge domain,
in which the Control group scored significantly better in eleven out of the twelve groups in both
time periods. The only exception were children speaking Italian, a group with relatively low
frequency and also with low SLL percentage, for whom there was either no difference or a better
score compared with the control group.
All differences noted above remained significant when controlling for age and gender.
To summarize, developmental outcomes among groups of non-official language speakers
vary across domains. Along with their language, immigrants bring cultural beliefs, values and
practices. These ways of thinking, relating and behaving may have a protective effect on
children’s developmental physical health and well-being outcomes, which may positively
contribute to social competence even in the face of linguistic challenges.

Patterns of school readiness among selected subgroups of Canadian children: Children with
special needs and children with diverse language backgrounds
40
Second Language Learners and Bilingual children in specific language groups
The next step in the analyses was to investigate, for each of the twelve most frequent
language groupings, whether the children who were Second Language Learners had different
EDI outcomes than children who were Bilingual (speaking the language of instruction and
another language), and whether either of these groups had different outcomes than the Language-
control group. This investigation may shed more light on the diversity of outcomes in relation to
specific language/cultural background. Arguably, the SLL group can be perceived as a proxy for
families with a history of quite recent immigration, while the BIL group may be perceived as a
group of bilingual children raised in families whose cultural and linguistic identity remains
salient to them, yet the children have been in a Canadian context a sufficient length of time to
become acculturated and conversant. However, we have no additional data to support this
perception for this particular sample.
Tables 11 and 12 show summary results of comparing EDI scores between Second
Language Learners (SLL), , Bilingual children (BIL) and Language-controls for each language
grouping in each database, for the SLL and BIL groups separately. Outcomes could be: no
statistical difference (=), significantly better scores (>), or significantly lower scores (<).
For the top twelve language groups, N varied from 31 to 2965 children in the SLL group
and 277 to 1166 in the BIL group.
To establish these differences, analyses of variance (ANOVA) and post hoc Tukey’s
Honestly Significant Differences tests (HSD) were conducted. An ANOVA reports if there are
any statistically significant differences among the three groups. A Tukey’s HSD indicates which
of the combinations or pairs are different from each other. Only differences statistically
significant at the p<0.05 level when controlling for age and gender among the groups were

Patterns of school readiness among selected subgroups of Canadian children: Children with
special needs and children with diverse language backgrounds
41
considered. The variety of outcomes for each subpopulation was tallied. These include: no
difference between Bilingual and Language-control (BIL = L-Ctrl) or between SLL and
Language-Control (SLL = L-Ctrl); Bilingual or SLL significantly greater than Language-controls
(BIL > L-Ctrl or SLL > L-Ctrl), or less than Language-controls (BIL < L-Ctrl or SLL < Ctrl) and
comparisons between the SLL and Bilingual groups (BIL > SLL, BIL < SLL, or BIL= SLL).
These analyses present a significantly more nuanced understanding of the impact of language
(and culture) on children’s developmental outcomes.
Second Language Learners versus Language-control Group
As explained before, the Language-control Group consisted of children who spoke only
the language of instruction (English or French, depending on the school board), but no other
languages. Thus, effectively, SLL vs L-Ctrl are largely comparisons of two monolingual groups,
of which one has the advantage of attending school in the language they know, and the other has
the disadvantage of attending in the language they do not know. Despite this obvious disparity,
the SLL groups did not do worse than the Language-control group in all school readiness
domains all the time (Table 10).

Patterns of school readiness among selected subgroups of Canadian children: Children with
special needs and children with diverse language backgrounds
42
Table 11 SLL groups among 12 language groupings compared with controls (L-ctrl).
Numbers for each domain indicate the number of languages in which there were
statistical differences (< or >), or no differences (=). Language groupings in each
category are listed in the alphabetical order
The SLL groups had the best outcomes in the Physical Health and Well-being domain. In
half of the language groupings, the SLL children are at the same level or better than the
Language-control group.
SLL>L-ctrl
SLL=L-ctrl
SLL<L-ctrl
Physical Health &
Well-Being
1
Mandarin
5
Cantonese, Italian,
Polish, Portuguese,
Tamil
6
Arabic, Filipino,
Punjabi, Spanish,
Urdu, Vietnamese
Social Competence
2
Filipino, Italian
10
Arabic, Cantonese,
Mandarin, Polish,
Portuguese, Punjabi,
Spanish, Tamil,
Urdu, Vietnamese
Emotional Maturity
4
Italian, Polish,
Portuguese, Tamil
8
Arabic, Cantonese,
Filipino, Mandarin,
Punjabi, Spanish,
Urdu, Vietnamese
Language & Cognitive
Development
3
Italian, Mandarin,
Tamil
9
Arabic, Cantonese,
Filipino, Polish,
Portuguese, Punjabi,
Spanish, Urdu,
Vietnamese
Communication &
General Knowledge
12
Arabic, Cantonese,
Filipino, Italian,
Mandarin, Polish,
Portuguese, Punjabi,
Spanish, Tamil,
Urdu, Vietnamese

Patterns of school readiness among selected subgroups of Canadian children: Children with
special needs and children with diverse language backgrounds
43
In the remaining four domains, the SLL groups were scoring poorer than the Language-
control group in the overwhelming majority of language groups. A few groups scored on par
with the Control group in the social, emotional and language-cognitive domains. The only
consistent group through these three domains were the Italian-only speaking children. As
expected, all the SLL groups scored lower than the Language-control group in the
Communication and General Knowledge domain.
Second Language Learners vs Bilingual groups
Outcomes of domain comparisons between the SLL and Bilingual groups clearly
demonstrated the advantage Bilingual children have (Table 12).

Patterns of school readiness among selected subgroups of Canadian children: Children with
special needs and children with diverse language backgrounds
44
Table 12
SLL groups among 12 language groupings compared with corresponding
bilingual (Bil) group. Numbers for each domain represent the number of
languages in which there were statistical differences (< or >), or no
differences (=). Note that there were no domains in which the SLL group
scored significantly better than the BIL group. Language groupings in each
category are listed in the alphabetical order.
The SLL group never scored better on any domain, and only occasionally at par with the
Bilingual group. The Italian-only and Portuguese-only speaking children were at least at par
with their Bilingual counterparts in three of the EDI domains. The Polish-only speaking group
was no different than the Bilingual Polish in Physical Health and Well-being, and the Tamil-only
SLL=BIL
SLL<BIL
Physical Health &
Well-Being
3
Italian, Polish, Portuguese,
9
Arabic, Cantonese,
Filipino, Mandarin,
Punjabi, Spanish, Tamil,
Urdu, Vietnamese
Social Competence
2
Italian, Portuguese,
10
Arabic, Cantonese,
Filipino, Mandarin, Polish,
Punjabi, Spanish, Tamil,
Urdu, Vietnamese
Emotional Maturity
3
Italian, Portuguese, Tamil
9
Arabic, Cantonese,
Filipino, Mandarin, Polish,
Punjabi, Spanish, Urdu,
Vietnamese
Language & Cognitive
Development
12
Arabic, Cantonese,
Filipino, Italian, Mandarin,
Polish, Portuguese,
Punjabi, Spanish, Tamil,
Urdu, Vietnamese
Communication &
General Knowledge
12
Arabic, Cantonese,
Filipino, Italian, Mandarin,
Polish, Portuguese,
Punjabi, Spanish, Tamil,
Urdu, Vietnamese

Patterns of school readiness among selected subgroups of Canadian children: Children with
special needs and children with diverse language backgrounds
45
speaking group was no different than the Bilingual Tamil in the Emotional Maturity domain.
Otherwise, for the overwhelming majority of the language groups, the Bilingual groups scored
better than the SLL groups.
While, as mentioned previously, having more than language, and one of them the
language of instruction, conveys obvious advantages over not being able to function in the
language of instruction, it would be intriguing to explore possible mechanisms through which
such powerful differences exist in kindergarten in all domains of development. It is possible that
parental and process factors such as parents’ expectations of what skills are necessary to develop
for school (and who is responsible to teach these), exposure to pre-academic experiences such as
daycare, or the process of adjustment to a new culture, climate and routine play a very important
role. We were unable to control for socioeconomic status in these analyses, or time of
immigration. It is possible that children who are functional in both languages have parents who
are more likely to also speak more than one language and more likely to be employed, thus
providing, on top of bilingualism, economic advantage. All these factors need to be considered
in more detail in future research.
Bilingual vs Language-control groups
The comparisons between children who are functionally fluent in two languages, one of
them the language of instruction (Bilingual), and children who are fluent in the language of
instruction only (Language-control) are of particular interest, as they may reveal where there are
potential advantages or disadvantages, as early as the kindergarten year, in having mastered more
than one language. The results are in Table 13.

Patterns of school readiness among selected subgroups of Canadian children: Children with
special needs and children with diverse language backgrounds
46
Table 13
Bilingual (BIL) groups among 12 language groupings compared with
Language control group (L-ctrl). Numbers for each domain represent the
number of languages in which there were statistical differences (< or >), or
no differences (=). Language groupings in each category are listed in the
alphabetical order
BIL>L-ctrl
BIL=L-ctrl
BIL<L-ctrl
Physical Health &
Well-Being
11
Arabic, Cantonese, Filipino,
Italian, Mandarin, Polish,
Punjabi, Spanish, Tamil,
Urdu, Vietnamese
1
Portuguese,
Social Competence
8
Cantonese, Filipino, Italian,
Mandarin, Polish, Punjabi,
Urdu, Vietnamese
3
Arabic, Spanish,
Tamil
1
Portuguese
Emotional Maturity
7
Cantonese, Filipino, Italian,
Punjabi, Tamil, Urdu,
Vietnamese
5
Arabic, Mandarin,
Polish, Portuguese,
Spanish
Language &
Cognitive
Development
6
Cantonese, Filipino, Italian,
Mandarin, Polish, Urdu
4
Arabic, Punjabi,
Tamil, Vietnamese
2
Portuguese,
Spanish,
Communication &
General Knowledge
1
Italian,
4
Arabic, Cantonese,
Mandarin, Tamil
7
Filipino, Polish,
Portuguese,
Punjabi, Spanish,
Urdu, Vietnamese
Research indicates that children who have relatively strong and equal abilities in two
languages demonstrate advantages in many areas of cognitive and linguistic functioning,
including metacognitive, metalinguistic, and sociolinguistic, when compared with monolingual
children (Winsler, Díaz, Espinosa & Rodríguez, 1999). EDI results showed that the Bilingual
group scored very well across the domains in comparison with the Language-control group. All
twelve groups had mean scores at or above the Language-control group on Physical Health and
Well-being and Emotional Maturity. In the Social Competence domain, eleven of twelve groups

Patterns of school readiness among selected subgroups of Canadian children: Children with
special needs and children with diverse language backgrounds
47
scored at or above the Language-control group. Even in the Language and Cognitive domain,
ten of twelve groups were at or above the Language-control group. Clearly, in these four
domains there were obvious strengths and advantages to being bilingual. Part of the benefit in
the social and emotional domains may be the result of additional roles that these children may
take on by being bridges between their monolingual peers on either side or by being coaches to
their counterparts who have difficulty with the language of instruction.
In the Communication and General Knowledge domain, there were more uneven results
in terms of the distribution of groups. Almost all Bilingual groups scored at the same level or
worse than the Controls. As mentioned previously, while describing the overall results (page
23), this result, while somewhat counterintuitive, may be a consequence of the child’s lesser
familiarity with the everyday language in the school’s English or French in comparison with
their first language. Our data does not allow distinguishing between bilingual children who were
exposed to their first language as well as English/French approximately simultaneously (that is,
from birth) versus those who learned the first language before they acquired fluency in
English/French (sequentially). Recent research indicates that the simultaneous bilinguals show
less proficiency in either language than those who acquired both languages early, but
simultaneously (Sebastian-Gallés, Echeverria, & Bosch, 2005; Werker, Maurer, & Yoshida,
2009). These findings are ascribed to the interference between the two languages.
The Bilingual language groups who were not significantly different than the controls in
the Communication and General Knowledge domain were: Arabic, Cantonese, Mandarin, and
Tamil. Italian Bilingual children scored better than the Control group. There does not seem to be
an obvious pattern to which language groups offer these best-case outcomes. Most of them
would historically be from relatively new immigrant source countries. It would seem then that

Patterns of school readiness among selected subgroups of Canadian children: Children with
special needs and children with diverse language backgrounds
48
pooling differences which exist on a language by language basis can obliterate these real and
salient differences which can be observed when comparing them in the case-by-case manner as
done earlier against the Language-control group. This is a critical factor in research focused on
second language learners and bilingual children, since averaging the scores of children from
different language backgrounds may well result in non-significant differences or quite variable
results from study to study.
Once again, the limitations of our comparisons have to be considered. Research from
both Canada and the US indicates that ethnicity, single parenting and SES are often interrelated
(Pagani, Jalbert, Lapointe, & Hebert, 2006; Patterson, Kupersmidt, & Vaden, 1990). Thus, our
results can be confounded by the impact of poverty. In Canada, large national studies indicate
that immigrant children (those most likely to contribute to our SLL group) are likely to catch up
to their peers by age 10 to 15 (Georgiades, Boyle & Duku, 2007; Statistics Canada 2002c, 2004;
Worsick, 2001). These results, however, are subject to two major limitations: first, they do not
distinguish among language groups, second, they tend to underrepresent children from families
where parents do not speak English or French, thus possibly skewing the outcomes towards the
groups that are likely to be more successful.
Overall Conclusions
This report summarized and interpreted results of an investigation of patterns of school
readiness in two important, yet infrequent at a community level, kindergarten populations:
children with special needs, and children with diverse language backgrounds. These analyses
were made possible on the basis of the population-based data collected with the Early
Development Instrument.

Patterns of school readiness among selected subgroups of Canadian children: Children with
special needs and children with diverse language backgrounds
49
Two main questions guided the analyses in the report: 1) are patterns of school readiness
different among children with special needs than those without, and in particular in respect to the
type of need?, 2) are patterns of school readiness different among children with diverse language
backgrounds, and in particular in respect to the specific language and fluency in the language of
instruction? The answer to both questions is yes, these special groups are different in their
school readiness patterns than children without these characteristics.
The analyses revealed many expected patterns, in particular among the children with
special needs, who have lower levels of school readiness than other children. However, we were
also able to explore differences and similarities among various special needs categories – a task
impossible with data from single sites – and found that students with visual and hearing
impairments were the least disadvantaged in school readiness outcomes.
Somewhat in parallel, while there is evidence of school-based outcomes for children with
diverse language backgrounds, data are rarely available in sufficient numbers to allow for
detailed description of specific language groups. The analyses for these categories revealed that
while there are overall patterns, often along expected lines when children are broadly categorized
regardless of their language of origin, a more intricate picture can emerge when different
language subgroups are compared. For example, we found that children fluent in English and
Italian had consistently very good school readiness outcomes; and that the Physical Health and
Well-being domain was the one where second language learners were the least likely to be
disadvantaged.
While these analyses posed more questions than they answered, our report makes it very
clear that it is crucial to collect and explore population level data. The results can guide practical

Patterns of school readiness among selected subgroups of Canadian children: Children with
special needs and children with diverse language backgrounds
50
intervention and prevention strategies at school and before, as well as provide broad-based
background for more detailed, in depth investigations of specific populations.
References
Barnett, W., Young, J., & Schweinhart, L. (1998). How preschool education influences long-
term cognitive development and school success. In W. S. Barnett & S. S. Boocock (Eds.),
Early care and education for children in poverty (pp. 167-184). Albany: State University
of New York Press.
Chui, T., Tran, K., & Maheux, H. (2007). Immigration in Canada: A portrait of the foreign-born
population, 2006 census. (Rep. No. 97-557-XIE) [online]. Ottawa, Ontario, Canada:
Statistics Canada, Social and Aboriginal Statistics Division. Retrieved March 18, 2009
from http://www12.statcan.ca/census-recensement/2006/as-sa/97-557/pdf/97-557-
XIE2006001.pdf
Citizenship and Immigration Canada (2005). Recent Immigrants in Metropolitan Areas:
Canada—A Comparative Profile Based on the 2001 Census [online]. Retrieved June 11,
2009 from http://www.cic.gc.ca/EnGLIsh/resources/research/census2001/canada/intro.asp
Doherty, G. (1997). Zero to six: The basis for school readiness. (Rep. No. R-97-8E). [online].
Ottawa, Ontario, Canada: Human Resources Development Canada. Retrieved March 5,
2009 from http://www.hrsdc.gc.ca/eng/cs/sp/sdc/pkrf/publications/1997-002557/SP-241-
02-01E.pdf
Entwisle, D.R. & Alexander, K.L. (1999). Early schooling and social stratification. In R.C.
Pianta & M.J. Cox (Eds.), The transition to kindergarten (pp. 13-18). Baltimore: Paul H.
Brookes Publishing.
Forget-Dubois, N., Lemelin, J., Boivin, M., Dionne, G., Séguin, J. R., Vitaro, F., & Tremblay,
R.E. (2007). Predicting early school achievement with the EDI: A longitudinal
population-based study. Early Education and Development, 18(3), 405-426.
Georgiades, K., Boyle, M. H., & Duku, E. (2007). Contextual influences on children’s mental
health and school performance: The moderating effects of family immigrant status. Child
Development, 78(5), 1572-1591.
Gissler M., Järvelin, M-R., Louhiala, P., & Hemminki, E. (1999). Boys have more health
problems in childhood than girls: follow-up of the 1987 Finnish birth cohort. Acta
Paediatrica, 88(3), 310-314.

Patterns of school readiness among selected subgroups of Canadian children: Children with
special needs and children with diverse language backgrounds
51
Janus, M., Lefort, J, Cameron R., & Kopechanski, L. (2007). Starting kindergarten: Transition
issues for children with special needs. Canadian Journal of Education, 30(3), 628-648.
Janus, M. & Duku, E.K. (2004). Normative data for the EDI: Part 1. School Readiness to Learn
(SRL) Project: Results. Hamilton, Ontario: McMaster University.
http://www.offordcentre.com/readiness/files/RESULTS.Normative_Data.pdf
Janus, M. & Duku, E. (2007). The school entry gap: Socioeconomic, family, and health factors
associated with children’s school readiness to learn. Early Education and Development,
18(3), 375-404.
Janus, M. & Duku, E. (2008). Gold Standard Normative II data. School Readiness to Learn
(SRL) Project: Results. Hamilton, Ontario: McMaster University.
http://www.offordcentre.com/readiness/files/Reports_2008_05_27_Gold_Data.pdf
Janus, M. & Offord, D.R. (2000). Readiness to learn at school. ISUMA, 1, 71- 75.
Janus, M. & Offord, D.R. (2007). Development and psychometric properties of the Early
Development Instrument (EDI): A measure of children’s school readiness. Canadian
Journal of Behavioural Science, 39, 1-22.
Jimerson, S., Egeland, B & Teo, A. (1999). A longitudinal study of achievement trajectories:
Factors associated with change. Journal of Educational Psychology, 91, 116-126.
Kagan, S.L., Moore, R., & Bredekamp, S. (1995). Reconsidering children’s early development
and learning: toward common views and vocabulary. Washington, DC: National
Education Goals Panel.
Keating, D.P. (2007). Formative evaluation of the Early Development Instrument: Progress and
prospects. Early Education and Development, 18(3), 561-570.
LaParo, K., & Pianta, R. (2000). Predicting children’s competence in the early school years: A
meta-analytic review. Review of Educational Research, 70(4), 443-484.
Love, J.M. & Raikes, H.H. (2004). Transition begin early. In R.E. Tremblay, R.G. Barr, and R.
DeV. Peters (Eds.). Encyclopedia on early childhood development [online]. Montreal,
Quebec: Centre of Excellence for Early Childhood Development, 1-5. Retrieved February
10, 2009 from http://www.childho-enencyclopedia.com/documents/Love-
RaikesANGxp.pdf
Mashburn, A., & Pianta, R. (2006). Social relationships and school readiness. Early Education
and Development, 17(1), 151-176.
Meisels, S.J. (1999). Assessing readiness. In R.C. Pianta & M.J. Cox (Eds.), The transition to
kindergarten (pp. 39-66). Baltimore: Paul H. Brookes Publishing.

Patterns of school readiness among selected subgroups of Canadian children: Children with
special needs and children with diverse language backgrounds
52
Pagani, L. S., Jalbert, J., Lapointe, P., & Hebert, M. (2006). Effects of Junior Kindergarten on
emerging literacy in children from low-income and linguistic-minority families. Early
Childhood Education Journal, 33(4), 209-215.
Patterson, C., Kupersmidt, J., & Vaden, N. (1990). Income level, gender, ethnicity and
household composition as predictors of children’s school-based competence. Child
Development, 61(2), 485-494.
Ramey, C., & Ramey, S. (2004). Early learning and school readiness: Can early intervention
make a difference? Merrill-Palmer Quarterly, 50(4), 471-491.
Sebastian-Gallés, N., Escheverria, S., & Bosch, L. (2005). The influence of lexical exposure on
lexical representation: Comparing early and simultaneous bilinguals. Journal of Memory
and Language, 52, 240-255.
Shepard, L., Kagan, S., & Wurtz, E. (Eds.). (1998). Principles and recommendations for early
childhood assessments. Washington, DC: National Education Goals Panel.
http://govinfo.library.unt.edu/negp/reports/prinrec.pdf
Statistics Canada. (2002a). Participation and Limitations Survey: 2001 Census. (Rep. No. 89-
579-XIE) [online]. Retrieved March 17, 2009 from http://www.statcan.gc.ca/pub/89-579-
x/89-579-x2002001-eng.pdf
Statistics Canada. (2002b). Participation and Limitations Survey: 2001 Census [online].
Retrieved March 17, 2009 from http://www.statcan.gc.ca/daily-
quotidien/021203/dq021203a-eng.htm.
Statistics Canada. (2002c). School performance of children from immigrant families [online].
Update on Family and Labour Studies. Vol 1. Retrieved March 16, 2009 from
http://www.statcan.gc.ca/pub/89-001-x/2002001/4064943-eng.htm
Statistics Canada (2004). Education Matters: Children of immigrants: how well do they do in
school? (Rep. No. 81-004-XIE) [online]. Retrieved June 11, 2009 from
http://www.statcan.gc.ca/pub/81-004-x/200410/7422-eng.htm
Statistics Canada. (2007a). The Evolving Linguistic Portrait, 2006 Census. (Rep. No. 97-555-
XIE) [online]. Ottawa, Ontario, Canada. Retrieved March 18, 2009 from
http://www12.statcan.ca/census-recensement/2006/as-sa/97-555/pdf/97-555-
XIE2006001.pdf
Statistics Canada (2007b). 2006 Census: Immigration, citizenship, language, mobility and
migration. The Daily, December 4, 2007. Retrieved March 16, 2009 from
http://www.statscan.gc.ca/daily-quotidien/071204/dq071204a-eng.htm

Patterns of school readiness among selected subgroups of Canadian children: Children with
special needs and children with diverse language backgrounds
53
Statistics Canada. (2008a). Participation and Limitations Survey: 2006 Census. (Rep. No. 89-
628-XWE) [online]. Retrieved March 17, 2009 from http://www.statcan.gc.ca/bsolc/olc-
cel/olc-cel?catno=89-628-X&lang=eng
Statistics Canada. (2008b). Back to school – September 2008 [online]. Retrieved March 16,
2009 from http://www.statcan.gc.ca/pub/81-004-x/2008003/article/5203237-eng.htm#a
Werker, J. F., Maurer, D. M., & Yoshida, K. M. (2009). Perception. In M. Bornstein (Ed).
Handbook of Cultural Developmental Science, (pp 89-126). New York: Psychology Press.
Winsler, A., Díaz, R., Espinosa, L., & Rodríguez, J. (1999). When learning a second language
does not mean losing the first : Bilingual language development in low-income, Spanish-
speaking children attending bilingual preschool. Child Development, 70(2), 349-362.
Worsick, C. (2001). School performance of the children of immigrants in Canada, 1994-98.
(Rep. No. 11F0019MIE2001178) [online]. Ottawa, Ontario, Canada: Statistics Canada.
Retrieved March 16, 2009 from
http://www.statcan.gc.ca/pub/11f0019m/11f0019m2001178-eng.pdf
Zuckerman, B., & Halfon, N. (2003). School readiness: An idea whose time has arrived.
Pediatrics, 111, 1433-1436.