Introduction
Gender differences for poor reading are not a new phenomenon but
date back more than 100 years. As early as 1910 it was found that up to
85 per cent of children struggling with reading were boys (Pickle,
1998), although the cause remained unknown. In more recent times,
researchers have identified a number of potential explanations for a
greater prevalence of boys who are poor readers, including
neurobiological (Clements et al., 2006; Shaywitz et al., 1995), genetic
(Hawke, Wadsworth & DeFries, 2005), environmental (Olson, 2002) and
motivational (Martin, 2004). Despite the wealth of existing research on
gender ratio differences in poor reading, accurately gauging the degree
to which there are more boys than girls who are struggling readers
continues to be a matter of debate. Moreover, variations in the
prevalence of poor reading per se continue in the absence of a
universally accepted definition of what it means to be a poor reader
(Pererra-Larrd, Deane & Bunnell, 1999; Siegel & Smythe, 2005;
Wheldall & Limbrick, 2010).
Limbrick, Wheldall and Madelaine (2008) recently reviewed a number
of studies reporting gender ratios for poor reading, and found
considerable differences among studies in terminology, methodology,
severity of selection, and samples. Other researchers have reported
similar inconsistencies, and subsequently reported gender ratios for
poor reading based on large population samples (Flannery et al., 2000;
Katusic et al., 2001; Rutter et al., 2004).While the use of population,
or large representative samples, effectively reduces sample bias,
differences in reported gender ratios have still arisen across these
studies in terms of measurement and severity of selection. Additionally,
studies using populations are often subject to specific inclusion and
exclusion criteria that are not only unique to that study but also
reduce the actual sample. Even when studies use populations or large
representative samples, then, without a clear and stable definition of
poor reading, it is difficult to ascertain the relative prevalence of
boys and girls who are poor readers, except to say that the majority of
research studies suggest that there are generally more boys.
Are there really more boys than girls who are poor readers?
Despite the plethora of research indicating a greater prevalence of
boys who are poor readers, there is evidence to the contrary. Several
researchers have reported non-significant or very small gender
differences for poor reading (Prior, Sanson, Smart & Oberklaid,
1995; Siegel & Smythe, 2005). In a large meta-analysis of studies
reporting gender differences on psychological variables, Hyde (2005)
proposed the 'gender similarities hypothesis', where there are
more similarities than differences in the abilities of boys and girls.
Hyde found that effect sizes in studies reporting on gender differences
for reading and reading-related skills were relatively small.
Other researchers have also suggested that there is little
difference between the proportion of boys and girls who are poor
readers, but gender ratios can be inflated by confounding variables,
such as behaviour and attention. Shaywitz, Shawywitz, Fletcher and
Escobar (1990), for example, were among the first to identify the role
of attentional factors in sample selection. They compared the number of
students classified as having a reading disability by a
research-identified method (discrepancy formula) versus a
school-identified method (recommendation by teachers and eligibility for
special education services), and found that there were almost four times
as many boys identified as having a reading disability using the
school-identified criterion.They concluded that more boys were
identified as poor readers due to referral bias, where behavioural and
attentional problems are often factored in, and are more commonly seen
in boys. Conversely, in the research-identified sample, there were no
significant gender differences in poor reading. Others have also found
that inattentive or troublesome behaviour can lead to higher referral
rates for boys (Beaman, Wheldall & Kemp, 2006), thereby increasing
gender ratios for poor reading.
Another reason for a greater proportion of boys being identified as
poor readers is related to differences in the distribution of reading
scores for boys and girls. Boys have a greater variability in reading
scores than girls (Hawke et al., 2009), and as a result more boys score
in the tail of the distribution. Consequently, more boys are identified
as poor readers. Machin and Pekkarinen (2008) recently demonstrated this
phenomenon on the Programme for International Student Assessment (PISA),
a large-scale assessment administered to secondary students across 41
countries. Boys had greater variance in scores than girls for reading,
resulting in a greater preponderance of boys in the bottom 5 per cent.
Similarly, Share and Silva (2003) found that when separate distributions
were used for boys and girls, as opposed to a combined distribution for
all students, gender ratios decreased from 1.79:1 to 1.02:1. Evidence
also suggests that variability in boys' scores is not limited to
reading but is evident in other areas. Machin and Pekkarinen (2008)
found that there was a greater variance in boys' scores for
mathematics, which led to a greater proportion of boys scoring in the
top 5 per cent. Boys also show greater variability on psychological
variables such as verbal (sequential reasoning), quantitative (inductive
reasoning with quantitative concepts), and non-verbal (inductive
reasoning) (Lohman & Lakin, 2009).
Large-scale assessments
Given the sizeable inconsistencies between studies reporting gender
ratios for poor reading, particularly in the definition, measurement,
severity of selection and sample, one approach to estimate the
proportion of boys who are poor readers may be to examine performance on
a large-scale assessment by gender. Although the use of large-scale
assessments is a topic of controversy (see, for example, Mills, 2008),
there are considerable advantages. For example, there is the opportunity
to assess a large population sample on a single measure, using a
consistent severity of selection criterion. The assessment is
administered at exactly the same time for all students in common grades,
and participation rates are usually very high. Additionally, although no
assessment is completely unbiased (Soodak, 2000), test bias is often
reduced in large-scale assessments because they are typically
norm-referenced and objective (Sloane & Kelly, 2003). Authentic or
teacher-made assessments, on the other hand, are often more difficult to
design, administer and mark, and often fail reliability standards
(Sloane & Kelly, 2003). If large-scale assessments with high
reliability and reduced bias are administered to large population
samples, then they may be a valid method of identifying gender ratios
for poor reading.
There have been several international studies reporting gender
differences for poor reading using large-scale assessments. Recently,
Lynn and Mikk (2009) found that girls consistently outperformed boys on
the PISA. In the years 2000, 2003 and 2006, girls obtained higher
average reading scores than boys, with an effect size of 0.42. Machin
and Pekkarinen (2008) reported similar findings on the PISA for 2003,
but further concluded that variance in boys' reading scores
accounted for the greater proportion of boys identified in the bottom 5
per cent.
In Australia, large-scale assessments for reading have
traditionally been state-or Territory-based. In New South Wales, for
instance, the Basic Skills Test (BST) was a state-wide assessment for
all state school students in Years 3 and 5. A recent study by Wheldall
and Limbrick (2010) analysed reading performance on the New South Wales
BST from 1997 to 2006, with a sample of more than 1 million students.
Severity of selection was firstly defined by scoring in the lowest BST
band for the reading component, and by this definition average gender
ratios for poor reading were 1.66:1 for Year 3 and 2.26:1 for Year 5. A
less-stringent severity of selection measure (the lowest two BST bands
combined) revealed gender ratios of 1.44:1 and 1.99:1 for Years 3 and 5
respectively. By using a single large-scale assessment on an entire
population sample, over a 10-year period, Wheldall and Limbrick
concluded that gender ratios were not as variable, inconsistent or large
as previously reported. Gender ratios were stable over time, varying
only with grade and severity of selection. They also found a slightly
greater variance in boys' scores.
Large-scale assessments and reading comprehension
Although large-scale assessments often vary in how
'reading' is defined and measured (Johnson, 1996), assessments
such as the BST, for instance, are effectively measures of reading
comprehension (Wheldall & Limbrick, 2010). (It should be
acknowledged that not all tests of reading comprehension are measuring
the same skills: Bowyer-Crane & Snowling, 2005; Keenan, Betjemann
& Olson, 2008.) Cutting and Scarborough (2006) compared the Wechsler
Individual Achievement Test (WIAT), the Gray Oral Reading Test (GORT)
and the Gates-MacGinitie Reading Test--Revised, and concluded that
variances in sentence and passage length, language, word frequencies,
and test format led to variances in the skills measured. Similarly,
researchers have also indicated that cloze tests measure word
recognition (Nation & Snowling, 1997), and that there is a stronger
link between cloze tests and decoding, than for multiple-choice tests
(Francis et al., 2005). In a recent study, Keenan, Betjemann and Olson
(2008) compared the Gray Oral Reading Test (GORT), the Qualitative
Reading Inventory (QRI), the Woodcock-Johnson Passage Comprehension
subtest (WJPC), and the Peabody Individual Achievement Test (PIAT)
reading comprehension subtest. They reported only modest correlations
among these tests, indicating that these tests, which are all reading
comprehension measures, were not directly comparable. The PIAT and WJPC,
for instance, focused more on decoding than other measures.
Additionally, reading comprehension tests measure different skills,
depending on the developmental level of the reader.
Identifying poor readers by performance on large-scale assessments
As discussed by Wheldall and Limbrick (2010), the use of
large-scale assessments provides an opportunity for a consistent
definition and measure of poor reading. In past years, studies have
indicated that the incidence of reading disability varies anywhere from
4 per cent to 6 per cent (Sofie & Riccio, 2002) to 38 per cent
(Aaron et al., 2008). Wheldall and Limbrick, analysing student
performance on the BST over a 10-year period, found a prevalence rate of
14 per cent for Year 3 students and 1.43 per cent to 7.65 per cent for
Year 5 students.
The BST was limited to students in New South Wales but in 2008,
along with similar measures used in other annual state and
Territory-based literacy and numeracy benchmark assessments, it was
replaced by a single large-scale assessment across Australia: the
National Assessment Program-Literacy and Numeracy (NAPLAN) (Ministerial
Council on Education, Employment, Training and Youth
Affairs--MCEETYA--2008a). The NAPLAN was first administered in May 2008
to all government and non-government students in Years 3, 5, 7 and 9.
All students in these grades, across the nation, were assessed in
reading, writing, spelling, punctuation and grammar, and numeracy.
The introduction of the NAPLAN marks the first instance where all
Australian students in common grades are assessed on the same measure,
at the same time (May of every year) (MCEETYA, 2008a). Previous studies
have varied in reported gender ratios due to a lack of consensus in
defining and measuring poor reading, as well as variations in the grades
assessed. The introduction of the NAPLAN, therefore, provides the first
opportunity to gauge gender ratios of poor reading using a single
definition and measure nationwide, across four grades.
Wheldall and Limbrick (2010) analysed the performance of Years 3
and 5 students on the reading component of the BST, defining poor
reading as either a score in the lowest reading band or a score in the
lowest two reading bands. Using two definitions takes into account the
essentially arbitrary definition of reading disability (Siegel &
Smythe, 2005). The purpose of this study is to extend the findings
reported by Wheldall and Limbrick (2010) to report gender ratios of poor
reading for a nationwide population, using two definitions of poor
reading (a score in the lowest reading band or a score in the lowest two
reading bands) (discussed below). This study will analyse student
performance on the NAPLAN in 2008 specifically for reading but will also
consider the other components of NAPLAN, including spelling, writing,
punctuation and grammar, and numeracy. This paper considers the
following research questions:
* Is there any difference in the incidence of reading problems
between boys and girls?
* Are there similar differences between boys and girls in spelling,
writing, punctuation and grammar, and numeracy?
* Are there differences between boys and girls (in all five
components: reading, spelling, writing, punctuation and grammar, and
numeracy) across the various Australian states and territories?
* Are these differences consistent over grades (Years 3, 5, 7 and
9)?
Method
Participants
More than 1 million students in Years 3, 5, 7 and 9, attending
government and non-government schools across Australia, participated in
the NAPLAN in May 2008.
The data analysed in this study are derived from the secondary data
presented in the document entitled 2008 National Assessment Program
Literacy and Numeracy: Achievement in Reading, Writing, Language
Conventions and Numeracy, publicly released by the Ministerial Council
on Education, Employment, Training and Youth Affairs (MCEETYA, 2008a).
While the total number of students participating in the NAPLAN was
reported for each year across states and Territories, the number of
students scoring in each band, by gender, was not reported; only
percentages were provided. A request for the actual numbers was met with
the reply that the release of data had been deferred indefinitely until
all policies and procedures for data release had been finalised.
Consequently, some of the calculations in this study are based on the
percentages reported by the Ministerial Council on Education,
Employment, Training and Youth Affairs, and not on raw data.
For reading, there was a total of 262,372 Year 3 students
Australia-wide (85,682 in New South Wales, 62,230 in Victoria, 55,770 in
Queensland, 26,635 in Western Australia, 18,717 in South Australia,
6,377 in Tasmania, 4,174 in the ACT, and 2,787 in the Northern
Territory). Similar numbers were reported for writing (262,010),
spelling (262,612), grammar and punctuation (262,612), and numeracy
(261,597).
In Year 5, there were a total of 262,872 students who participated
in reading (85,775 in New South Wales, 62,954 in Victoria, 55,459 in
Queensland, 26,630 in Western Australia, 18,664 in South Australia,
6,158 in Tasmania, 4,341 in the ACT, and 2,891 in the Northern
Territory). Similar numbers were reported for writing (262,600),
spelling (263,126), grammar and punctuation (263,126), and numeracy
(262,268).
In Year 7, there were a total of 265,627 students who participated
in reading (85,350 in New South Wales, 63,760 in Victoria, 56,296 in
Queensland, 27,379 in Western Australia, 19,222 in South Australia,
6,422 in Tasmania, 4,527 in the ACT, and 2,671 in the Northern
Territory). Similar numbers were reported for writing (265,507),
spelling (266,083), grammar and punctuation (266,083), and numeracy
(265,275).
In Year 9, there were a total of 262,549 students who participated
in reading (84,520 in New South Wales, 62,853 in Victoria, 56,133 in
Queensland, 27,392 in Western Australia, 18,647 in South Australia,
6,179 in Tasmania, 4,439 in the ACT, and 2,386 in the Northern
Territory). Similar numbers were reported for writing (262,841),
spelling (263,297), grammar and punctuation (263,297), and numeracy
(262,122).
Measure
The NAPLAN is a large-scale assessment first implemented in 2008 as
a replacement for individual state and territory literacy and numeracy
benchmark assessments, including the BST. Components of reading,
writing, language conventions (including spelling, grammar and
punctuation) and numeracy were administered to all students in Years 3,
5, 7 and 9, across Australia. The NAPLAN was developed in consultation
with all Australian states and territories, to ensure that it covered
the curriculum (MCEETYA, 2008a).
The NAPLAN results are reported across five scales (reading,
writing, spelling, punctuation and grammar, and numeracy), for each of
the four grades (Years 3, 5, 7, and 9). Scores for each scale (reading,
writing, spelling, punctuation and grammar, and numeracy) are each
reported on a continuous scale of 0 to 1000, which is divided into 10
bands (Year 3 reports bands 1 to 6, where Band 1 is the lowest and Band
6 is the highest; Year 5 reports bands 3 to 8;Year 7 reports bands 4 to
9; and Year 9 reports bands 5 to 10). The scale is designed so that
student performance can be measured longitudinally as students move up
through the grades. The lowest band for each year level is considered as
not meeting the national standard (see MCEETYA, 2008c).
The reading scale is a test of reading comprehension, comprising
both multiple-choice questions and open-ended questions. Students in
each year level are presented with two texts (for example, a story,
poem, letter or an advertisement) and are required to answer five
questions per text. Year 3 students were presented with all
multiple-choice questions (10 in total);Year 5, 7 and 9 students were
presented with a combination of multiple-choice questions and open-ended
questions (between 8 and 10 questions in total).All texts presented were
similar to those used in Year 3, 5, 7 and 9 classrooms, and questions
varied in difficulty to demonstrate student understanding of the texts
(MCEETYA, 2008d). The complexity of reading skills measured were
appropriate for each year level, ranging from the ability to find
information in texts; connecting ideas; making decisions on a
character's thoughts and actions; and recognising the main idea of
the text (MCEETYA, 2008e), to interpreting the primary message in the
text; reflecting on character motivations and writing strategies;
and interpreting symbolic language (MCEETYA, 2008e).
The writing scale required students to write in response to a
presented text, to arrange ideas and demonstrate proficiency in
vocabulary and sentence structure (MCEETYA, 2008e). Writing skills
assessed varied between year levels but included demonstrating story
structure and the development of ideas, and competently using language
conventions.
Language conventions (spelling, and grammar and punctuation)
required students to recognise and amend spelling mistakes, and answer
multiple-choice questions. Language convention skills varied between
year levels (for in-depth sample questions and reporting for all year
levels, see MCEETYA, 2008e).
Numeracy required students to apply mathematical knowledge to a
range of questions and included number, measurement, time and shapes,
varying according to Year (MCEETYA, 2008e).
For the purpose of this study, poor reading was first defined as
scoring in the lowest band. For Year 3, this is defined as Band 1; for
Year 5, it is defined as Band 3; for Year 7, it is defined as Band 4
and, for Year 9, it is defined as Band 5. A second definition of poor
reading was defined as scoring in the lowest two bands. For Year 3, this
meant combined bands 1 and 2; for Year 5, combined bands 3 and 4; for
Year 7, combined bands 4 and 5; and for Year 9, combined bands 5 and 6.
Poor performance in writing, spelling, punctuation and grammar, and
numeracy were also defined by scoring in the lowest band or the lowest
two bands. The small percentages of students who were exempted from
participating in the NAPLAN, and were classified as not meeting the
minimum national standard separately from students scoring in the lowest
band, were not included in this study. Australia-wide, average
participation rates for the NAPLAN in 2008 ranged between 93.5 per cent
and 96.9 per cent.
Results
Gender differences in means
Means and standard deviations for reading, by location and gender,
are presented in Table 1. This secondary data was produced by MCEETYA
(2008a). Standard deviations were provided for overall reading means,
but not for reading means by gender. Due to the density of the
information provided, means and standard deviations are provided for
reading only. For means and standard deviations for writing, spelling,
punctuation and grammar, and numeracy, refer to Appendices 1 to 4.
To ascertain whether the means for boys and girls for reading,
writing, spelling, punctuation and grammar, and numeracy, were
significantly different, we first had to make several statistical
decisions, given the nature of the data reported by MCEETYA (2008a).
First, the MCEETYA (2008) report provided the actual total number of
students by state and scale, but not actual numbers by gender. We
therefore had to calculate approximate numbers of boys and girls in each
year level. We assumed a prevalence of 1.05 boys for every 1 girl,
consistent with birth rates published by the appropriate state
authorities. Given the extremely large overall sample sizes involved,
any minor inaccuracies in determining sub-sample size are unlikely to
have any effect on final outcomes. Second, while standard deviations
were reported for all students per year level, they were not reported
separately for boys and girls. For this reason, the calculations in this
study are based on the overall standard deviations reported by MCEETYA
(2008a).
In all year levels, boys had significantly lower means for reading
than girls. Girls obtained higher mean scores in reading than boys in
Year 3 (t = 45.7, p < 0.0001),Year 5 (t = 41.2, p < 0.0001),Year 7
(t = 35.9, p < 0.0001), and Year 9 (t = 22.9,p < 0.0001).While
reading means varied among states and territories, the reading means for
boys were significantly lower than the reading means for girls
throughout Australia.
Australia-wide, boys had significantly lower means than girls for
writing in Year 3 (t = 99.2, p < 0.0001),Year 5 (t = 101.5, p <
0.0001),Year 7 (t = 107.4, p < 0.0001) and Year 9 (t = 113.2, p <
0.0001). For spelling, average means were also significantly lower for
boys than for girls in all year levels: Year 3 (t = 61.6, p <
0.0001),Year 5 (t = 66.6,p < 0.0001),Year 7 (t = 72.7,p < 0.0001)
and Year 9 (t = 74.87,p < 0.0001).
Average punctuation and grammar means were again significantly
lower for boys than for girls, across Year 3 (t = 65.5,p <
0.0001),Year 5 (t = 76.3,p < 0.0001), Year 7 (t = 84.7,p <
0.0001), and Year 9 (t = 75.3,p < 0.0001).Variability among states
and territories was again evident. Highest mean scores were observed in
New South Wales, and Victoria for writing; New South Wales for spelling;
Victoria and the ACT for punctuation and grammar and numeracy. The
lowest mean scores were observed in the Northern Territory on all
components.
For numeracy, the average Australian means were in favour of boys.
For all year levels boys had significantly higher means than girls: Year
3 (t = 27.2, p < 0.0001),Year 5 (t = 43.5, p < 0.0001), Year 7 (t
= 52.7, p < 0.0001), and Year 9 (t = 32.4,p < 0.0001).
Effect sizes
Because the sample sizes are extremely large in this study, and
therefore even slight differences between boys and girls may be
statistically significant, effect sizes were calculated for gender
differences for each scale, by year level (see Table 2). Effect size is
a determination of the power or strength of the relationship between two
variables, and is typically considered small (up to .3), medium
(approximately .5) or large (more than .8). For reading, effect sizes
for gender were generally small. Across Australia, reading effect sizes
decreased with the year level of schooling and were, on average, .18
(Year 3), .16 (Year 5), .14 (Year 7), and .09 (Year 9). Differences in
reading between boys and girls appear to be negligible, and reduce over
time. Effect sizes for reading also varied among states and Territories,
with the largest effect sizes observed in Western Australia (Years 3,7
and 9), the ACT (Year 5), and Queensland (also Years 7 and 9). The
lowest effect sizes were observed in Tasmania (Years 3, 5 and 9) and
jointly the ACT and Northern Territory (Year 7).
For writing, effect sizes for gender bordered on moderate in some
instances and increased with year level: .39 (Year 3), .40 (Year 5), .42
(Year 7), and .44 (Year 9). There was some variability between states
and territories. The Northern Territory had the lowest effect sizes for
all year levels (ranging from .26 to .34), whereas Tasmania had the
largest effect sizes for all year levels (ranging from .44 to .50).The
effect sizes for writing were the highest reported in this study.
Average effect sizes for spelling were small and increased slightly
with year level, ranging from .24 to .29. Similarly, grammar and
punctuation average effect sizes were also small, ranging from .26 to
.33, indicating that gender differences in spelling and grammar and
punctuation are, like reading, negligible. Effect sizes varied among
states and territories.
Numeracy also had small effect sizes for all year levels, but in
favour of boys. Average effect sizes increased from Year 3 (.11) to Year
7 (.20), but decreased in Year 9 (.13).
Percentage of boys and girls in the lowest two bands
Across Australia, the percentages of students scoring in the lowest
band were 6.1 per cent (Year 3), 7.5 per cent (Year 5), 4.6 per cent
(Year 7) and 5.9 per cent (Year 9). When the lowest two bands were
combined, the percentages of students were 18.1 per cent (Year 3), 21.0
per cent (Year 5), 18.7 per cent (Year 7), and 23.5 per cent (Year 9).
Table 3 presents the percentages of boys and girls scoring in each
band for reading. In all year levels, there were more boys than girls
scoring in the lowest band. Across all states and territories,
consistently more boys than girls scored in the lowest band.
For writing, spelling, and punctuation and grammar, there were also
consistently more boys than girls who scored in the lowest bands, and
this was evident in all states and territories. For numeracy, in the
majority of instances there were more girls than boys in the lowest
band, with the exception of Year 3 (New South Wales, Western Australia,
ACT), Year 5 (Tasmania), Year 7 (Tasmania, ACT) and Year 9 (ACT). But
these differences in the percentages of boys and girls were, overall,
minimal.
Gender ratios of poor readers
To ascertain the prevalence of boys and girls identified as poor
readers, gender ratios were calculated for all scales, by year level. To
take into account the arbitrary nature of poor reading, two definitions
of poor reading (severity of selection) were applied:
* scoring in the lowest band for each year level
* scoring in the lowest two bands for each year level.
Gender ratios for poor reading are presented in Table 4.
There was little variation in gender ratios for reading across year
levels. Using the first definition of poor reading (scoring in the
lowest band), the average gender ratios for poor reading were 1.57:1
(Year 3), 1.44:1 (Year 5), 1.68:1 (Year 7), and 1.48:1 (Year 9).The ACT
consistently had the highest gender ratios across all year levels
(ranging from 1.71:1 to 2.67:1). The lowest gender ratios were reported
in the Northern Territory for all year levels (ranging from 1.08:1 to
1.16:1).
The second definition of poor reading was scoring in the lowest two
bands for reading. By this definition, average gender ratios were 1.44:1
(Year 3), 1.32:1 (Year 5), 1.47:1 (Year 7), and 1.30:1 (Year 9). Gender
ratios are lower when using a less stringent definition.
For writing, the average gender ratios for the lowest band were
considerably higher than for reading (ranging from 2.35:1 to 2.63:1).
The largest gender ratios were reported in the ACT across all year
levels (ranging from 3.24:1 to 6.75:1), while the lowest were observed
in the Northern Territory (ranging from 1.29:1 to 1.43:1). By the second
definition (lowest two bands), writing gender ratios ranged from 1.99:1
to 2.24:1. Similar to reading, gender ratios for poor writing were
consistently lower when using a broader definition.
Average gender ratios for poor spelling and grammar and punctuation
were 1.92:1 and 1.97:1 respectively. Gender ratios decreased to 1.68:1
and 1.7:1 respectively when defined by the two lowest bands. Gender
ratios varied among states and territories: the largest ratios were
observed in ACT, while the smallest were observed in the Northern
Territory.
Gender ratios for poor numeracy indicated that there were more
girls than boys in the lowest bands. For Band 1 alone, gender ratios
ranged from 0.83:1 to 0.97:1, and for bands 1 and 2 combined gender
ratios ranged from 0.84:1 to 0.96:1. For Years 3, 5 and 7 the gender
ratios were only slightly lower for the combined two bands, compared to
the lowest band. For Year 9, there was no difference at all.
Differences in poor numeracy gender ratios were also evident across
states and Territories. For Year 3, while Victoria, Queensland, South
Australia, Tasmania and the Northern Territory showed more girls than
boys, New South Wales, Western Australia and the ACT showed more boys
than girls. For Years 5 and 7, only Tasmania had more boys than girls,
and in Year 9 there were consistently more girls than boys.
Conversely, excepting numeracy, gender ratios in the top bands, for
all year levels and scales, were in favour of girls. For reading, the
gender ratio of superior reading was 1.23:1 (Year 3), 1.27:1 (Year 5),
1.13:1 (Year 7), and 1.04:1 (Year 9). For superior writing, gender
ratios were 1.72:1 (Year 3), 1.93:1 (Year 5), 1.92:1 (Year 7), and
2.02:1 (Year 9). For superior spelling, gender ratios were 1.33:1 (Year
3), 1.35:1 (Year 5), 1.37:1 (Year 7), and 1.48:1 (Year 9). For superior
punctuation and grammar performance, again in favour of girls, gender
ratios were 1.39:1 (Year 3), 1.56:1 (Year 5), 1.59:1 (Year 7), and
1.53:1 (Year 9). But for numeracy, there were more boys than girls in
the top band. Average gender ratios (in favour of boys) were 1.39:1
(Year 3), 1.62:1 (Year 5), 1.65:1 (Year 7), and 1.43:1 (Year 9).
Discussion
The purpose of this study was to identify gender ratios for poor
reading by performance on a new large-scale assessment, the NAPLAN,
which is the first Australian assessment to test all students in common
grades, on the same measure at the same time (MCEETYA, 2008a). Defining
poor reading by performance on the NAPLAN is therefore the first
opportunity in Australia where a single definition of poor reading could
be agreed upon, nationwide. The same definition of poor performance was
also applied to spelling, writing, grammar and punctuation, and
numeracy. This study also sought to identify whether there were any
gender differences across states and territories in all components of
the NAPLAN, and across year levels of schooling.
Gender ratios for poor readers
In this study, poor reading on the NAPLAN was defined by two
severities of selection: a score in the lowest band; and a score in
either of the lowest two bands. Using the first definition, average
gender ratios for poor reading, across Australia, were 1.57:1 (Year 3),
1.44:1 (Year 5), 1.68:1 (Year 7) and 1.48:1 (Year 9). By the second
definition of poor reading, gender ratios were 1.44:1 (Year 3), 1.32:1
(Year 5), 1.47:1 (Year 7), and 1.30:1 (Year 9). These gender ratios are
consistent with a large body of evidence indicating that there are more
boys than girls who are poor readers (Liederman, Kantrowitz &
Flannery, 2005;Miles, Haslum & Wheeler, 1998; Rutter et al., 2004;
Stevenson, 1992).Although previous studies have varied significantly in
defining, measuring and interpreting poor reading, and as a result
gender ratios have also varied significantly (Limbrick, Wheldall &
Madelaine, 2008), by using a large-scale reading assessment and applying
two definitions of poor reading, the findings in this study support
previous research that suggests that differences between boys and girls
in reading are not particularly large (Prior et al., 1995; Siegel &
Smythe, 2005). According to Hyde's (2005) gender similarities
hypothesis, boys and girls are more alike than not in reading and
reading-related abilities. In a large meta-analysis, Hyde reported very
small effect sizes in gender differences for studies examining reading
comprehension and vocabulary. Lohman and Lakin (2009) also recently
found that, despite significant gender differences in mean scores, only
small effect sizes were observed. In this study small effect sizes for
gender differences in reading were also reported, adding weight to the
body of evidence suggesting mean gender differences for reading are
negligible.
The findings in this study are also consistent with previous
studies using large-scale assessments. For example, girls have been
shown to consistently outperform boys in reading performance on the PISA
(Lynn & Mikk, 2009; Machin & Pekkarinen, 2008). In
Australia,Wheldall and Limbrick (2010) reported gender ratios for poor
reading by performance on the BST, administered to New South Wales
students attending state schools. By defining poor reading as scoring in
the lowest bands on the BST, gender ratios were 1.66:1 (Band 1) and
1.44:1 (combined bands 1 and 2) for Year 3, and 2.26:1 (Band 1) and
1.99:1 (combined bands 1 and 2) for Year 5. By comparison, gender ratios
of poor reading on the NAPLAN, for New South Wales students, were 1.82:1
(Band 1) and 1.60:1 (combined bands 1 and 2) for Year 3, and 1.57:1
(Band 1) and 1.40:1 (combined bands 1 and 2) for Year 5, which are only
slightly higher than the national average. As indicated by Wheldall and
Limbrick (2010), there are more boys than girls who are poor readers,
but these gender differences are not great.
The findings in this study also demonstrate that gender ratios vary
with severity of selection. Across all year levels, states and
territories, gender ratios for reading were higher for the lowest band
than for the combined lowest two bands, suggesting that there are more
boys in the very tail of the distribution. This also supports Wheldall
and Limbrick's (2010) study, where gender ratios for poor reading
varied with severity of selection on the BST. Others have also reported
that gender ratios vary with severity of selection, and that there are
more boys than girls in the tail of the distribution (Flannery et al.,
2000; Stevenson, 1992).
The percentage of students scoring in the lowest two bands for
reading is also consistent with previous research. In this study, 4.6
per cent to 7.5 per cent of students scored in the lowest band,
representing the very tail of the distribution. Using the less stringent
definition of poor reading (lowest two bands) accounted for 18.1 per
cent to 23.5 per cent of students. While there have been mixed findings
as to what percentage of students actually constitute 'poor
readers', ranging anywhere from 4 per cent (Sofie & Riccio,
2002) to 38 per cent (Aaron et al., 2008), it has been generally
accepted that there is a small percentage of students (4-6 per cent) who
are truly struggling readers (Sofie & Riccio, 2002). In terms of
performance on the NAPLAN, the lowest band might reasonably constitute
that small percentage of students who are experiencing particularly
severe difficulties in learning to read.
The findings reported in this build on the findings reported by
Wheldall and Limbrick (2010): that is, there are more boys than girls
who are poor readers, but the degree to which there are more boys is not
large. This conclusion not only relates to New South Wales students but
extends to the entire Australian population. Although there are more
boys than girls who are poor readers, when using a single definition and
measure of poor reading, gender ratios are not as large or as
inconsistent as previously reported. This conclusion is strengthened by
the fact that more than one million students participated in the NAPLAN
in 2008, and as indicated by Wheldall and Limbrick, more than one
million students participated in the BST across a 10-year period.Taken
together, these studies add further weight to existing research
indicating that there is overall little difference between boys and
girls in reading and that gender ratios for poor reading are much lower
than originally believed (Prior et al., 1995; Siegel & Smythe, 2005;
Smart et al., 2001).
Although the findings in this study are consistent with previous
research, it is also worth noting that the gender differences reported
could possibly be a result of other factors. For example, research
suggests that a greater variance in boys' scores, compared to
girls, can lead to a higher proportion of boys in the tail of the
distribution (Hawke,Wadsworth & DeFries, 2009). Research by Wheldall
and Limbrick (2010) and Machin and Pekkarinen (2008), both of which
examine the results of large-scale assessments, supports these findings.
While it is clear in this study that there are more boys in the lowest
reading bands, and that the proportion of boys increases with severity
of selection, without the standard deviations for both boys and girls it
is difficult to ascertain whether the results in this study are in fact
due to more variability in boys' scores.
A second potential reason for a greater preponderance of boys
scoring in the lowest bands may be attentional. Findings reported by
Shaywitz, Shaywitz, Fletcher and Escobar (1990) suggest that boys may
have performed more poorly due to lower attention levels at the time of
assessment in a group testing context. Similar to other studies
reporting gender ratios for poor reading using large-scale assessments,
data on attentional factors were not collected.
Additionally, it has been argued that the greater preponderance of
boys identified as poor readers is not due to differences in reading,
but rather due to referral bias. For instance, Shaywitz, Shaywitz,
Fletcher & Escobar (1990) found that boys were up to four times more
prevalent than girls in samples referred by schools compared to samples
that were research-identified. Others have reached similar conclusions
(Smart et al., 2001). But in this study the findings are based on a
population with high participation rates, and therefore are not
confounded by referral bias.
Gender differences in writing, spelling, grammar and punctuation,
and numeracy
Gender differences in writing, spelling, grammar and punctuation,
and numeracy were also evident. For writing, spelling, and grammar and
punctuation, girls significantly outperformed boys. Girls consistently
obtained higher means than boys Australia-wide. For numeracy, boys
obtained significantly higher means than girls. These findings are
complementary to previous research indicating that girls tend to perform
better in writing and language conventions (Dodd et al., 2003; Halpern
& LaMay, 2000) whereas boys tend to perform better in mathematics
(Machin & Pekkarinen, 2008; Marks, 2008).
Gender ratios for poor spelling and grammar and punctuation were
very small, and similar to those for poor reading but the gender ratios
for writing were notably higher than for the other scales, evident
across all year levels. In Year 3, for example, the national gender
ratio for poor writing was 4:1, compared with reading (1.57:1), spelling
(1.85:1), and punctuation and grammar (1.74:1). Indeed, the highest
gender ratio reported in this study was for writing in Year 3: the ACT
had a gender ratio of 6.75:1; New South Wales was 4:1; and for Victoria
it was 4.25:1. These gender ratios for poor writing appear to decrease
with age: while there was an average gender ratio of 4:1 for Year 3
students, the gender ratio for Year 9 students declined to 2.47:1.
Despite this decline, however, even in Year 9 the gender ratios for poor
writing were higher than for the other scales.
Writing also had the largest effect sizes in this study. While
there were significant differences between the means for boys and girls
on all scales, only the effect sizes for writing were close to medium.
Effect sizes for reading, spelling, and punctuation and grammar were
very small, but for writing the difference between boys and girls was
more substantial. These findings suggest that while more boys than girls
struggle with poor reading, writing is also a skill with which boys, at
least in Year 3, experience particular difficulties.
Gender ratios for poor writing, spelling, grammar and punctuation,
and numeracy were defined by two severities of selection. Like poor
reading, gender ratios also varied by severity of selection in poor
writing, spelling, and punctuation and grammar. Defining poor
performance by scoring in the lowest band consistently yielded higher
gender ratios than defining poor performance by scoring in the combined
lowest two bands. Similar to reading, it appears that there are more
boys in the very tail of the distribution for other aspects of literacy.
Again, these findings are not dissimilar to previous research. With
regard to poor writing, for example, Berninger and colleagues (2008)
explored the co-morbidity of reading and writing disabilities, and found
that writing disabilities, particularly for boys, are liable to be more
severe than reading disabilities. Similarly, in poor spelling, some
studies have found that poor phonological decoding abilities are not
only correlated with poor reading, but also with poor spelling (Friend
& Olson, 2008). Others have found that a disability in reading also
affects spelling (Hudson, High & Al Otaiba, 2007; Smart et al.,
2001).The fact that there are more boys than girls in the tail of the
distribution for both reading and spelling in this study, and that
gender ratios for both reading and spelling decreased when using a
broader definition of poor performance, would seem to support this
link.These findings also suggest that poor reading is not an isolated
event, but rather has implications for other aspects of literacy.
Conversely, boys had, on average, slightly higher numeracy means
than girls, and gender ratios for poor numeracy were very small in
favour of boys. There was little difference in gender ratios for either
severity of selection for Years 3, 5 and 7, and in Year 9 there was no
gender difference at all when combining the two lowest bands. In the
upper end of the distribution, there were more boys than girls, which
support previous findings that boys demonstrate superior mathematics
skills to girls (Machin & Pekkainen, 2009; Marks, 2008). Previous
studies have also found that, similar to reading scores, boys have a
greater variance in mathematics scores than girls (Machin &
Pekkainen, 2009). It is possible, then, that the greater proportion of
boys scoring in the top band in this study may be due to greater
variance in scores. As indicated earlier, this is difficult to ascertain
without standard deviations for boys and girls.
Gender ratios across Australian states and territories
Across states and territories, there were consistently more boys
than girls who scored in the lowest two bands for reading, writing,
spelling, and grammar and punctuation, and gender ratios consistently
varied with severity of selection. Throughout the country, there were
more boys in the lowest band than in the lowest two bands combined. For
numeracy, there were on average more girls than boys in the lowest two
bands, but differences were minimal.
Gender ratios for poor performance varied considerably across
states and territories. For poor reading, the largest gender ratios were
observed in the ACT, while the smallest gender ratios were observed in
the Northern Territory. Similar patterns were also evident for writing
and spelling. For poor grammar and punctuation, while the largest gender
ratios were observed in the ACT, the smallest gender ratios were
observed in both the Northern Territory and Tasmania. For poor numeracy,
gender ratios varied considerably between states and territories.
Highest ratios were observed in Western Australia, Tasmania, and the
ACT, whereas the lowest ratios were observed in Victoria and South
Australia, depending on year level of schooling.
There were also noticeable differences across states and
territories in the percentages of students scoring in the lowest bands.
For example, an overall average 7.4 per cent of boys and 4.7 per cent of
girls scored in the lowest band for reading in Year 3, compared to 38.2
per cent of boys and 33.0 per cent of girls scored in the lowest band in
the Northern Territory. Similar results for the Northern Territory were
also in evidence for Years 5, 7 and 9 in reading. The highest proportion
of poor readers was observed in the Northern Territory, even though the
smallest gender ratios were also observed in the Northern Territory.
There are several reasons to account for this pattern. First, it
may be that gender ratios are smaller in the Northern Territory because
there are a higher percentage of students scoring in the lowest band.
Previous research indicates that there are more boys than girls in the
very tail of the reading distribution (Hawke et al., 2009; Lohman &
Lakin, 2009). As evidenced in this study, gender ratios for poor reading
decrease as the definition of poor reading becomes less stringent. This
would indicate that a tail of the distribution which accounts for
approximately 35 per cent of students would result in lower gender
ratios for poor reading in the Northern Territory, compared to a small
tail which accounts for approximately 5 per cent of students in the
other states and the ACT.
Second, given that approximately 35 per cent of children in the
Northern Territory scored in the lowest reading band, and that there are
almost as many girls identified as poor readers as boys, poor reading
may result from factors other than gender. Given the fact that gender
ratios in this study are more moderate than previously reported, and
that effect sizes for gender differences in reading are very small, it
does not appear that gender is a strong predictor of reading. Indeed, a
number of studies have previously reported that gender is not a strong
predictor of poor reading (Limbrick, Wheldall, & Madelaine, in
press; Strand, Deary & Smith, 2006), but rather accounts for only a
small percentage of variance in reading performance (Fluss et al., 2009;
Strand, Deary & Smith, 2006). On the other hand, studies have shown
that factors such as socioeconomic status can account for up to 24 per
cent of variance in reading performance (Fluss et al., 2009). Burt, Holm
and Dodd (1999) found that socioeconomic status significantly correlated
with performance on reading tasks, whereas gender did not. Consequently,
it may be possible that the higher proportion of students observed
scoring in the lowest reading bands in the Northern Territory are due to
reasons related to socioeconomic status.
One aspect of socioeconomic status is differences in geographical
location of students. For instance, although the ACT and the Northern
Territory both contain very small populations compared to New South
Wales, Victoria and Queensland, the distributions of students in
metropolitan and remote areas differ considerably.
In the MCEETYA report (2008a), student location was broken down
into four locations: metro, provincial, remote, and very remote. In all
instances, the percentages of students scoring in the lowest bands were
higher in very remote locations, compared to results for metro,
provincial, and remote locations. In the ACT, the majority of students
were located in metro areas with no remote or very remote students. In
the Northern Territory, the majority of students are located in
provincial, remote and very remote areas, with very few in metro areas.
These findings indicate that students in metro areas outperform students
in remote areas. In the Northern Territory, for instance, the reading
means for provincial (366.5) and for remote (329.6) were substantially
larger than for very remote (195.9).
Additionally, studies consistently demonstrate that minority
groups, such as Indigenous Australians, are severely disadvantaged in
terms of educational and socioeconomic factors (Bradley et al., 2007).
Many Indigenous Australians live in remote and very remote areas and
therefore do not have access to well-resourced schools, as well as
having English as a second language (Bradley et al., 2007). Other
factors include attendance rates (for example, difficulties due to
living in remote areas) and a lack of skilled teaching being available
(Prior, 2009). As indicated by Prior (2009), a substantial number of
Indigenous students do not reach national literacy standards on
assessments such as the NAPLAN because of these disadvantages.
The findings in this study appear consistent with this. As
indicated above, approximately 35 per cent of students in the Northern
Territory were identified as poor readers by scoring in the lowest band.
Although the student population in the Northern Territory is the
smallest across the country, it had the highest percentage of Indigenous
students. For instance, in the Northern Territory approximately 36 per
cent of Year 3 students who participated in Year 3 reading were
Indigenous. By comparison, in New South Wales, Victoria, Queensland,
Western Australia, South Australia, Tasmania and the Australian Capital
Territory, the number of Indigenous students ranged from 0.01 per cent
to 0.07 per cent of the population.
For many Indigenous Australians, English is not the first language,
nor is it the language spoken at home. As a result, language is another
factor which should be taken into consideration when analysing
performance on large-scale assessments such as the NAPLAN. For instance,
in the Northern Territory, 61.9 per cent of Year 3 students who scored
in the lowest reading band indicated that English was not their first
language. This is considerably higher than for other states and the ACT,
as well as higher than the Australian average of 6.6 per cent.
The impact of sociodemographic factors on academic performance has
repeatedly been documented (Baker, Goesling & Letendre, 2002;
Coutinho, Oswald & Best, 2002); differences arise according to
socioeconomic status, geographical location, ethnicity and having
English as a second language (Coutinho, Oswald & Best, 2002).
Additionally, previous research has shown that regardless of whether an
assessment is large-scale or small-scale, these factors are consistently
shown to have an effect (Baker, Goesling & Letendre, 2002; Bradley
& Corwyn, 2002). The advantage of the NAPLAN is that it is a
standardised assessment, administered Australia-wide. Although gender
ratios for poor reading therefore have the same meaning across the
country, they should be analysed with reference to such factors that are
known to affect performance.
Finally, while there were consistently more boys than girls
identified as poor readers, there were consistently more girls than boys
identified as superior readers, across all states and Territories.
Gender ratios in the top performing bands were consistently in favour of
girls: 0.81:1 (Year 3), 0.79:1 (Year 5), 0.89:1 (Year 7) and 0.96:1
(Year 9). Girls were also shown to have performed better than boys in
writing, spelling, and punctuation and grammar. This result adds weight
to the findings reported in Wheldall and Limbrick (2010), suggesting
that while more boys are identified as poor readers, more girls are
identified as superior readers.
Gender differences across years of schooling
For reading, writing, spelling, and grammar and punctuation, girls
obtained significantly higher means than boys, and this was evident in
all year levels. Across Years 3, 5, 7, and 9, girls outperformed boys.
Effect sizes for these components of the NAPLAN varied little by the
year level of schooling. Greater variation in effect sizes was observed
among states and territories, than among year levels of schooling. For
numeracy, boys consistently obtained higher means than girls, evident
across all year levels but effect sizes were very small, ranging from
.11 to .20 across Years 3 to 9.
Gender ratios for poor reading varied little between year levels of
schooling, ranging from 1.44:1 to 1.68:1. Gender ratios fluctuated with
the year level of schooling, but did not steadily increase or decrease
over time. For writing, spelling, and grammar and punctuation, gender
ratios for the lowest band appeared to steadily increase from Year 3 to
Year 7, but decreased in Year 9.
It is worth noting that the average percentage of students scoring
in the lowest reading band fluctuated across the year levels of
schooling: 6.1 per cent (Year 3), 7.5 per cent (Year 5), 4.6 per cent
(Year 7) and 5.9 per cent (Year 9). For New South Wales alone, the
percentage of students was 4.0 per cent (Year 3), 5.7 per cent (Year 5),
4.0 per cent (Year 7), and 5.1 per cent (Year 9).The increase in New
South Wales students from Year 3 to Year 5 is not consistent with
Wheldall and Limbrick's (2010) findings, where there were
considerably fewer students scoring in the lowest band in Year 5,
compared to Year 3. By performance on the BST, there were approximately
14 per cent ofYear 3 students scoring in the lowest band, but only 1.43
per cent in Year 5.Wheldall and Limbrick reasoned that because the BST
was a continuous scale for Years 3 and 5, it was anticipated that there
would be fewer Year 5 students scoring in the lowest band.Year 5
students would be two years older than Year 3 students, and would have
made some reading progress. Others have also found that the percentage
of students who are poor readers can decline with age (Badian, 1999;
Smart et al., 2001;Wright, Fields & Newman, 1996).Although the BST
and the NAPLAN are not identical, they are very similar in format and
content. It is still possible, however, that differences in the
percentage of students scoring in the lowest bands for Years 3 and 5 may
be a result of differences in test format and content, or timing and
administration, between the BST and the NAPLAN. Given that 2008 marks
the first year for the NAPLAN, future research on subsequent years is
warranted to establish consistency in the percentage of students scoring
in the lowest band across all year levels, as well as gender ratios for
poor reading.
Implications of the findings
Like Wheldall and Limbrick (2010), the findings in this study
demonstrate that while there are more boys than girls who are poor
readers, the preponderance of boys is not as great as previously
reported (Liederman, Kantrowitz & Flannery, 2005). Additionally, the
small effect sizes indicate that mean gender differences are negligible.
As previous studies have shown, gender is not a strong or consistent
predictor of reading, but rather, there are almost as many girls who
also struggle with reading. Although the reasons for poor reading are
beyond the scope of this article, other studies have demonstrated that
factors such as low socioeconomic status (Burt, Holm & Dodd, 1999),
poor phonological awareness (Savage & Carless, 2004), troublesome
behaviour (Smart et al., 2001), and even lack of motivation (Mucherah
& Yoder, 2008), can affect reading outcomes, irrespective of gender.
While such factors account for a considerable variance in reading
ability (for instance, phonological awareness can account for
approximately 50 per cent of variance in reading--see Fluss et al.,
2009), gender has not been consistently shown to account for large
variances in reading outcomes (Savage & Carless, 2004; Strand,
Dreary & Smith, 2006).The findings in this study appear to be
consistent with this. It may be reasonable, then, that interventions for
struggling readers identified by performance on the NAPLAN should focus
on tackling reading difficulties regardless of gender. Although this
study found that gender ratios for poor reading are not large, there
nevertheless exists a large body of research indicating a preponderance
of boys who are poor readers (Liederman, Kantowitz & Flannery,
2005). As indicated by some researchers (Shaywitz, Shaywitz, Fletcher
& Escobar, 1990; Smart et al., 2001;Willcutt & Pennington,
2000), it is plausible that such a preponderance is not the result of
gender differences in reading, but rather that more boys have been
identified as poor readers because they are more likely to display
external troublesome behaviour. Girls, on the other hand, are less
likely to display external problem behaviour and are more difficult to
identify (Levy et al., 2005), and are therefore less likely to receive
intervention. But identification of poor readers by performance on the
NAPLAN, however, presents a nationwide opportunity to identify all
struggling readers, regardless of gender, and unimpeded by referral bias
or sample selection. As indicated by Wheldall and Limbrick (2010), if
gender ratios for poor reading are lower than is commonly believed,
there should be a greater focus on identifying struggling girls as well
as boys.
Limitations
One of the major limitations of this study was the unavailability
of raw data required to perform more complex statistical computations.
The results in this study are based on statistical assumptions on a
secondary data set where standard deviations for boys and girls, for
example, were not provided. It could not be ascertained, therefore,
whether there was a greater proportion of boys scoring in the lowest
bands because of a greater variance in boys' scores. Future
researchers may be able to access the required data to establish whether
there are any differences in the distribution of reading scores for an
entire population sample in Australia, particularly in light of previous
research indicating boys demonstrate greater variance in scores than
girls.
Despite the fact that there are more boys than girls identified as
poor readers by performance on the NAPLAN, this identification is by a
single score on a NAPLAN band. Although the NAPLAN is longitudinal,
where all four year levels are graded on a single scale to allow student
progress to be monitored across time (MCEETYA, 2008a), it does not
differentiate whether boys struggle with particular aspects of reading
(for example, decoding, sight words, etc). Future studies might consider
examining the specific skills required for reading by gender, and
whether these differences vary with grade.
Conclusion
Variations in the gender ratios for poor reading throughout the
literature have arisen due to a lack of consensus in defining and
measuring poor reading. The introduction of the NAPLAN marks the first
large-scale assessment administered Australia-wide (MCEETYA, 2008a), and
with it marks the first opportunity to gauge gender ratios of poor
reading across the country using a single, standardised measure of
reading, with a consistent definition of poor reading. By analysing
student performance on the NAPLAN, the findings in this study
demonstrate that there are more boys than girls who are identified as
poor readers. But the effect sizes for mean differences reported in this
study are comparatively small, indicating that the overall mean
differences between boys and girls for reading are minimal. Gender
ratios for poor reading were also shown to vary with severity of
selection. Similar gender differences were reported for spelling, and
punctuation and grammar, indicating the possibility that the underlying
skills required for reading may also affect other components of
literacy. The largest gender ratios and effect sizes reported in this
study were for writing, indicating that this may be an area of
particular difficulty for boys. While no clear pattern emerged with
respect to gender ratios across grades, there were considerable
differences in gender ratios between states and territories.
Previous studies using population samples have been subject to
specific inclusion and exclusion criteria, which can substantially
reduce the sample. Large-scale assessments, on the other hand, are
administered to all students in common grades, with the exception of a
very small minority of students with severe disabilities. Participation
rates for large-scale assessments, therefore, are extremely high. This
study demonstrates that the introduction of the NAPLAN has provided an
opportunity to establish a stable and consistent definition of poor
reading across the country and to determine accurate estimates of the
gender ratio for reading problems in Australia.
References
Aaron, P. G., Joshi, R. M., Gooden, R., & Bentum, K. E. (2008).
Diagnosis and treatment of reading disabilities based on the component
model of reading: An alternative to the discrepancy model of LD. Journal
of Learning Disabilities, 41, 67-84.
Badian, N. A. (1999). Reading disability defined as a discrepancy
between listening and reading comprehension: A longitudinal study of
stability, gender differences and prevalence. Journal of Learning
Disabilities, 32, 138-148.
Baker, D. P., Goesling, B., & Letendre, G. K. (2002).
Socioeconomic status, school quality, and national economic
development: Across-national analysis of the 'Heyneman--Loxley
effect' on mathematics and science achievement. Comparative
Education Review, 46, 291-312.
Beaman, R., Wheldall, K., & Kemp, C. (2006). Differential
teacher attention to boys and girls in the classroom. Educational
Review, 58, 339-366.
Berninger, V. W., Nielsen, K. H., Abbott, R. D., Wijsman, E., &
Raskind, W. (2008). Gender differences in severity of writing and
reading disabilities. Journal of School Psychology, 46, 151-172.
Bowyer-Crane, C., & Snowling, J. (2005). Assessing
children's inference generation: What do tests of reading
comprehension measure? British Journal of Educational Psychology,
75(13), 189-201.
Bradley, R. H., & Corwyn, R. F. (2002). Socioeconomic status
and child development. Annual Review of Psychology, 53, 371-399.
Bradley, S., Draca, M., Green, C., & Leeves, G. (2007). The
magnitude of educational disadvantage of indigenous minority groups in
Australia. Journal of Population Economics, 20, 547-569.
Burt, L., Holm, A., & Dodd, B. (1999). Phonological awareness
skills of 4-year-old British children: An assessment and developmental
data. International Journal of Language & Communication Disorders,
34, 311-335.
Clements, A. M., Rimrodt, S. L., Abel, J. R., Blankner, J. G.,
Mostofsky, S. H., Pekar, J. J., Denckla, M. B., & Cutting, L. E.
(2006). Sex differences in cerebral laterality of language and
visuospatial processing. Brain and Language, 98, 150-158.
Coutinho, M. J., Oswald, D. P., & Best, A. M. (2002). The
influence of sociodemographics and gender on the disproportionate
identification of minority students as having learning disabilities.
Remedial and Special Education, 23, 49-59.
Cutting, L. E., & Scarborough, H. S. (2006). Prediction of
reading comprehension: Relative contributions of word recognition,
language proficiency, and other cognitive skills can depend on how
comprehension is measured. Scientific Studies of Reading, 10, 277-300.
Dodd, B., Holm, A., Hua, Z., & Crosbie, S. (2003). Phonological
development: A normative study of British English-speaking children.
Clinical Linguistics & Phonetics, 17, 617-643.
Flannery, K. A., Liederman, J., Daly, L., & Schultz, J. (2000).
Male prevalence for reading disability is found in a large sample of
Black and White children free from ascertainment bias. Journal of the
International Neuropsychological Society, 6, 433-442.
Fluss, J., Ziegler, J. C., Warszawski, J., Ducot, B., Richard, G.,
& Billard, C. (2009). Poor reading in French elementary school: The
interplay of cognitive, behavioral, and socioeconomic factors. Journal
of Developmental & Behavioral Pediatrics, 30, 206-216.
Francis, D. J., Fletcher, J. M., Stuebing, K. K., Lyon, G. R.,
Shaywitz, B. A., & Shaywitz, S. E. (2005). Psychometric approaches
to the identification of LD: IQ and achievement scores are not
sufficient. Journal of Learning Disabilities, 38, 98-108.
Friend, A., & Olson, R. K. (2008). Phonological spelling and
reading deficits in children with spelling disabilities. Scientific
Studies of Reading, 12, 90-105.
Halpern, D. F., & LaMay, M. I. (2000). The smarter sex: A
critical review of sex differences in intelligence. Educational
Psychology Review, 12, 229-246.
Hawke, J. L., Olson, R. K., Willcut, E. G., Wadsworth, S. J., &
DeFries, J. C. (2009). Gender ratios for reading difficulties. Dyslexia,
15, 239-242.
Hawke, J. L., Wadsworth, S. J., & DeFries, J. C. (2005).
Genetic influences on reading difficulties in boys and girls: The
Colorado twin study. Dyslexia, 12, 21-29.
Hudson, R. F., High, L., & Al Otaiba, S. (2007). Dyslexia and
the brain: What does current research tell us? The Reading Teacher, 60,
506-515.
Hyde, J. S. (2005). The gender similarities hypothesis. American
Psychologist, 60, 581-592.
Johnson, S. (1996). The contribution of large-scale assessment
programmes to research on gender differences. Educational Research and
Evaluation, 2, 25-49.
Katusic, S. K., Colligan, R. C., Barbaresi, W. J., Schaid, D. J.,
& Jacobsen, S. J. (2001). Incidence of reading disability in a
population-based birth cohort, 1976-1982, Rochester, MN. Mayo Clinic
Proceedings, 76, 1081-1092.
Keenan, J. M., Betjemann, R. S., & Olson, R. K. (2008). Reading
comprehension tests vary in the skills they assess: Differential
dependence on decoding and oral comprehension. Scientific Studies of
Reading, 12, 281-300.
Levy, F., Hay, D. A., Bennett, K. S., & McStephen, M. (2005).
Gender differences in ADHD subtype comorbidity. Journal of the American
Academy of Child Adolescent Psychiatry, 44, 368-376.
Liederman, J., Kantrowitz, L., & Flannery, K. (2005). Male
vulnerability to reading disability is not likely to be a myth: A call
for new data. Journal of Learning Disabilities, 38, 109-129.
Limbrick, L., Wheldall, K., & Madelaine, A. (2008). Gender
ratios of reading disability: Are there really more boys than girls who
are low-progress readers? Australian Journal of Learning Difficulties,
13, 161-179.
Limbrick, L., Wheldall, K., & Madelaine, A. (in press). Why do
more boys than girls have a reading disability? A review of the
evidence. Australian Journal of Special Education.
Lohman, D. F., & Lakin, J. M. (2009). Consistencies in sex
differences on the Cognitive Abilities Test across countries, grades,
test forms, and cohorts. British Journal of Educational Psychology, 79,
389--07.
Lynn, R., & Mikk, J. (2009). Sex differences in reading
achievement. Trames, 13, 3-13.
Machin, S., & Pekkarinen, T. (2008). Global sex differences in
test score variability. Science, 322, 1331-1332.
Marks, G. N. (2008). Accounting for the gender gaps in student
performance in reading and mathematics: Evidence from 31 countries.
Oxford Review of Education, 34, 89-109.
Martin, A. J. (2004). School motivation of boys and girls:
Differences of degree, differences of kind, or both? Australian Journal
of Psychology, 56, 133-146.
Miles, T. R., Haslum, M. N., & Wheeler, T. J. (1998). Gender
ratio in dyslexia. Annals of Dyslexia, 48, 27-55.
Mills, K. A. (2008). Will large-scale assessments raise literacy
standards in Australian schools? Australian Journal of Language and
Literacy, 31 , 211--225.
Ministerial Council on Education, Employment, Training and Youth
Affairs. (MCEETYA) (2008a). 2008 National Assessment Program Literacy
and Numeracy: Achievement in Reading, Writing, Language Conventions and
Numeracy. Canberra: Author. Retrieved 2 February 2009 from
http://www.naplan.edu.au/verve/
_resources/2ndStageNationalReport_18Dec_v2.pdf
Ministerial Council on Education, Employment, Training and Youth
Affairs. (MCEETYA) (2008b). NAPLAN Frequently asked questions. Canberra:
Author. Retrieved 2 February 2009 from
http://www.naplan.edu.au/frequently_asked_
questions/frequently_asked_questions.html
Ministerial Council on Education, Employment, Training and Youth
Affairs. (MCEETYA) (2008c). National Literacy and Numeracy Assessments
May 2008: NAPLAN Reporting Scales. Canberra: Author. Retrieved 2
February 2009 from http://www.naplan.edu.au/verve/_resources/NAPLAN_Reporting_Scales.pdf
Ministerial Council on Education, Employment, Training and Youth
Affairs. (MCEETYA) (2008d). NAPLAN Sample Student Reports. Canberra:
Author. Retrieved 2 February 2009 from
http://www.naplan.edu.au/naplan_2008_
reporting/naplan_2008_reporting.html
Ministerial Council on Education, Employment, Training and Youth
Affairs. (MCEETYA) (2008e). NAPLAN Sample Questions. Canberra: Author.
Retrieved 2 February 2009 from
http://www.naplan.edu.au/test_samples/test_samples.html
Mucherah, W., & Yoder, A. (2008). Motivation for reading and
middle school students' performance on standardized testing in
reading. Reading Psychology, 29, 214-235.
Nation, K., & Snowling, M. (1997). Assessing reading
difficulties: The validity and utility of current measures of reading
skill. British Journal of Educational Psychology, 67, 359-370.
Olson, R. K. (2002). Dyslexia: Nature and nurture. Dyslexia, 8,
143-159.
Pereira-Laird, J., Deane, F. P., & Bunnell, J. (1999). Defining
reading disability using a multifaceted approach. Learning Disability
Quarterly, 22, 59-71.
Pickle, M. J. (1998). Historical trends in biological and medical
investigations of reading disabilities: 1850--1915. Journal of Learning
Disabilities, 31, 625-635.
Prior, M. (2009). Early language and reading skills in Indigenous
children in Australia. Learning Difficulties Australia Bulletin, March,
15-16.
Prior, M., Sanson, A., Smart, D., & Oberklaid, F. (1995).
Reading disability in an Australian community sample. Australian Journal
of Psychology, 47, 32-37.
Rutter, M., Caspi, A., Fergusson, D., Horwood, L. J., Goodman, R.,
Maughan, B., Moffitt, T. E., Meltzer, H., & Carroll, J. (2004). Sex
differences in developmental reading disability: New findings from 4
epidemiological studies. Journal of the American Medical Association,
291 , 2007-2012.
Savage, R., & Carless, S. (2004). Predicting curriculum and
test performance at age 7 years from pupil background, baseline skills
and phonological awareness at age 5. British Journal of Educational
Psychology, 74, 155-171.
Share, D. L., & Silva, P. A. (2003). Gender bias in
IQ-discrepancy and post--discrepancy definitions of reading disability.
Journal of Learning Disabilities, 36, 4-14.
Shaywitz, B. A., Shaywitz, S. E., Pugh, K. R., Constable, R. T.,
Skudlarski, P., Fulbright, R. K., Bronen, R. A., Fletcher, J. M.,
Shankweller, D. P., Katz, L., & Gore, J. C. (1995). Sex differences
in the functional organization of the brain for language. Nature, 373,
607-609.
Shaywitz, S. E., Shaywitz, B. A., Fletcher, J. M., & Escobar,
M. D. (1990). Prevalence of reading disability in boys and girls:
Results of the Connecticut Longitudinal Study. Journal of the American
Medical Association, 22, 998-1002.
Siegel, L. S., & Smythe, I. S. (2005). Reflections on research
on reading disability with special attention to gender issues. Journal
of Learning Disabilities, 38, 473-477.
Sloane, F. C., & Kelly, A. E. (2003). Issues in high-stakes
testing programs. Theory into Practice, 42, 12-17.
Smart, D., Prior, M., Sanson, A., & Oberklaid, F. (2001). A
six-year follow-up from early primary school to secondary school.
Australian Journal of Psychology, 53, 45-53.
Sofie, C. A., & Riccio, C. A. (2002). A comparison of multiple
methods for the identification of children with reading disabilities.
Journal of Learning Disabilities, 35, 234-244.
Soodak, L. C. (2000). Performance assessments and students with
learning problems: Promising practice or reform rhetoric? Reading &
Writing Quarterly, 16, 257-280.
Stevenson, J. (1992). Identifying sex differences in reading
disability. Reading and Writing: An Interdisciplinary Journal, 4,
307-326.
Strand, S., Deary, I. J., & Smith, P. (2006). Sex differences
in Cognitive Abilities Test scores: A UK national picture. British
Journal of Educational Psychology, 76, 463-480.
Wheldall, K., & Limbrick, L. (2010). Do more boys than girls
have reading problems? Journal of Learning Disabilities, 43(2).
Willcutt, E. G., & Pennington, B. F. (2000). Comorbidity of
reading disability and Attention-Deficit/Hyperactivity Disorder:
Differences by gender and subtype. Journal of Learning Disabilities, 33,
179-191.
Wright, S. F., Fields, H., & Newman, S. P. (1996). Dyslexia:
Stability of definition over a 5 year period. Journal of Research in
Reading, 19, 46-60.
Lisa Limbrick
Kevin Wheldall
Alison Madelaine
Macquarie University Special Education Centre
Lisa Limbrick is undertaking a PhD in Special Education at the
Macquarie University Special Education Centre, Macquarie University.
Email: lisa.limbrick@speced.sed.mq.edu.au
Kevin Wheldall is Professor of Education and Director of Macquarie
University Special Education Centre, Macquarie University.
Alison Madelaine is Lecturer at Macquarie University Special
Education Centre, Macquarie University.
Appendix 1 Writing means, standard deviations * and sample sizes
NSW VIC QLD WA SA TAS
Year 3 Boys 414.9 412.1 377.3 383.6 401.9 401.7
Girls 440.8 439.7 407.1 413.2 429.0 430.2
All 427.6 425.8 391.8 398.1 415.1 415.7
(66.4) (65.7) (77.1) (69.6) (65.4) (64.1)
N 85 684 62 071 55 671 26 668 18 607 6 380
Year 5 Boys 482.0 487.4 454.0 457.3 467.4 461.8
Girls 509.4 518.1 484.4 487.2 494.3 493.5
All 495.4 502.4 468.9 471.7 480.8 477.3
(69.1) (71.4) (75.3) (72.6) (67.8) (69.8)
N 85 765 62 825 55 400 26 645 18 605 6 149
Year 7 Boys 520.8 532.3 506.6 507.1 522.8 502.1
Girls 550.6 567.8 539.4 538.9 553.6 540.2
All 535.3 549.7 522.7 522.5 538.1 520.6
(74.1) (77.1) (78.8) (76.1) (72.1) (76.3)
N 85 497 63 648 56 271 27 367 19 165 6 391
Year 9 Boys 551.8 570.8 536.3 542.4 553.2 538.4
Girls 587.5 607.7 575.4 580.2 588.7 577.6
All 569.4 588.9 555.3 560.8 571.2 557.2
(79.6) (83.4) (85.9) (83.0) (78.9) (83.8)
N 84 696 63 009 56 218 27 400 18 560 6 163
ACT NT All
Year 3 Boys 409.9 323.7 400.6
Girls 436.6 351.5 428.4
All 423.2 337.4 414.2
(64.1) (108.7) (71.6)
N 4 168 2 761 262 010
Year 5 Boys 477.4 396.0 472.2
Girls 504.1 427.6 501.4
All 490.7 410.9 486.5
(64.4) (114.5) (73.6)
N 4 339 2 872 262 600
Year 7 Boys 518.0 439.4 517.8
Girls 551.4 473.0 550.3
All 534.3 455.0 533.7
(70.7) (126.7) (77.9)
N 4 521 2 647 265 507
Year 9 Boys 550.4 485.9 551.2
Girls 591.1 528.9 588.4
All 571.0 506.9 569.4
(81.8) (127.1) (84.1)
N 4 449 2 346 262 841
The information reported in this table is derived from the MCEETYA
Report (2008a).
* Standard deviations were only reported for overall students, and not
by gender.
Appendix 2 Spelling means, standard deviations* and sample sizes
NSW VIC QLD WA SA TAS
Year 3 Boys 410.1 407.2 357.0 370.3 385.7 388.6
Girls 428.8 423.6 376.8 393.8 408.2 401.3
All 419.2 415.3 366.7 381.8 396.7 394.9
(75.4) (70.5) (76.2) (81.3) (76.9) (78.7)
N 85 778 62 209 55 861 26 697 18 734 6 385
Year 5 Boys 490.0 485.1 451.9 461.3 470.7 465.9
Girls 509.2 502.3 472.3 481.9 488.3 477.9
All 499.4 493.5 462.0 471.3 479.5 471.7
(72.7) (64.8) (68.9) (72.5) (69.9) (70.8)
N 85 868 62 952 55 535 26 697 18 677 6 173
Year 7 Boys 540.3 533.2 517.0 517.1 529.6 521.1
Girls 560.4 551.8 539.4 538.3 549.8 534.0
All 550.1 542.3 528.0 527.4 539.7 527.4
(72.1) (65.9) (71.1) (72.0) (68.8) (70.5)
N 85 600 63 790 56 389 27 459 19 225 6 424
Year 9 Boys 576.0 570.9 556.1 555.3 565.6 562.8
Girls 597.5 590.1 580.1 578.4 584.9 570.7
All 586.6 580.3 567.8 566.5 575.4 566.6
(71.7) (69.7) (72.6) (73.3) (71.4) (74.3)
N 84 757 63 071 56 292 27 448 18 707 6 185
ACT NT All
Year 3 Boys 396.0 287.7 390.1
Girls 417.9 312.2 409.3
All 406.9 299.8 399.5
(75.3) (125.6) (79.8)
N 4 175 2 773 262 612
Year 5 Boys 478.8 387.1 474.5
Girls 497.0 412.5 493.4
All 487.8 399.1 483.8
(68.7) (115.7) (72.7)
N 4 343 2 881 263 126
Year 7 Boys 535.4 446.7 528.8
Girls 553.7 468.5 549.1
All 544.3 456.8 538.7
(66.9) (113.2) (71.9)
N 4 544 2 652 266 083
Year 9 Boys 575.5 498.3 566.5
Girls 597.6 522.2 587.8
All 586.7 510.0 576.9
(66.6) (112.9) (72.9)
N 4 480 2 357 263 297
NB. The information reported in this table is derived from the MCEETYA
Report (2008a).
* Standard deviations were only reported for overall students, and not
by gender.
Appendix 3 Punctuation and grammar means, standard deviations * and
sample sizes
NSW VIC QLD WA SA TAS
Year 3 Boys 406.7 417.4 359.9 369.4 385.8 396.0
Girls 428.1 439.7 381.5 397.5 408.2 409.6
All 417.2 428.4 370.4 383.2 396.7 402.7
(80.8) (76.9) (86.9) (91.4) (79.9) (88.5)
N 85 778 62 209 55 861 26 697 18 734 6 385
Year 5 Boys 492.8 501.8 465.1 470.5 477.7 486.8
Girls 517.4 525.7 488.5 496.9 498.9 500.2
All 504.9 513.4 476.6 483.2 488.3 493.4
(79.5) (71.8) (78.9) (83.5) (73.5) (80.8)
N 85 868 62 952 55 535 26 697 18 677 6 173
Year 7 Boys 524.8 525.5 506.6 503.1 517.6 520.2
Girls 548.9 550.4 530.0 527.3 540.2 534.4
All 536.6 537.7 518.0 514.9 528.8 527.1
(72.3) (68.2) (68.7) (74.0) (69.9) (72.1)
N 85 600 63 790 56 389 27 459 19 225 6 424
Year 9 Boys 565.3 565.4 552.7 545.4 555.3 552.9
Girls 587.3 584.4 574.2 567.1 573.9 561.9
All 576.1 574.7 563.2 555.9 564.7 557.2
(71.4) (65.6) (71.0) (67.7) (65.8) (65.5)
N 84 757 63 071 56 292 27 448 18 707 6 185
ACT NT All
Year 3 Boys 408.2 279.6 392.2
Girls 431.2 302.7 414.6
All 419.6 291.0 403.2
(83.1) (150.1) (87.5)
N 4 175 2 773 262 612
Year 5 Boys 499.5 386.8 484.4
Girls 527.1 414.8 508.4
All 513.2 400.0 496.2
(72.5) (142.0) (80.6)
N 4 343 2 881 263 126
Year 7 Boys 535.7 433.0 517.3
Girls 557.9 457.8 541.2
All 546.6 444.5 529.0
(73.1) (126.1) (72.7)
N 4 544 2 652 266 083
Year 9 Boys 575.7 490.4 558.9
Girls 601.5 509.3 579.6
All 588.7 499.6 569.1
(69.0) (113.8) (70.4)
N 4 480 2 357 263 297
NB. The information reported in this table is derived from the MCEETYA
Report (2008a).
* Standard deviations were only reported for overall students, and not
by gender.
Appendix 4 Numeracy means, standard deviations * and sample sizes
NSW VIC QLD WA SA TAS
Year 3 Boys 412.6 421.9 371.3 383.5 392.8 401.3
Girls 405.0 411.7 364.4 380.1 384.6 398.5
All 408.9 416.9 367.9 381.9 388.8 399.9
(70.6) (63.8) (67.0) (66.4) (64.9) (67.7)
N 85 364 62 133 55 507 26 591 18 698 6 356
Year 5 Boys 493.4 496.2 463.6 465.7 467.6 466.1
Girls 482.0 482.8 452.7 455.3 453.2 463.0
All 487.8 489.7 458.2 460.7 460.4 464.6
(72.4) (65.8) (62.7) (63.4) (60.7) (62.9)
N 85 496 62 906 55 284 26 594 18 654 6 126
Year 7 Boys 558.7 560.8 545.5 541.1 544.3 535.7
Girls 543.6 543.4 532.1 525.8 528.1 531.8
All 551.3 552.3 539.0 533.7 536.2 533.8
(78.3) (69.4) (70.4) (68.7) (67.7) (67.5)
N 85 110 63 880 56 191 27 293 19 171 6 401
Year 9 Boys 595.1 596.3 574.3 575.3 577.9 570.3
Girls 587.7 584.8 566.9 565.9 564.6 565.4
All 591.4 590.7 570.7 570.7 571.1 568.0
(75.1) (66.6) (66.2) (66.6) (62.8) (65.1)
N 84 129 63 021 55 952 27 371 18 652 6 176
ACT NT All
Year 3 Boys 416.0 342.2 400.6
Girls 407.0 334.5 393.1
All 411.5 338.4 396.9
(66.8) (86.3) (70.4)
N 4 148 2 800 261 597
Year 5 Boys 490.5 420.6 481.6
Girls 477.0 411.4 469.9
All 483.8 416.3 475.9
(64.1) (81.0) (68.8)
N 4 313 2 895 262 268
Year 7 Boys 565.9 491.3 552.3
Girls 546.1 484.5 537.3
All 556.2 488.1 545.0
(71.0) (84.0) (73.2)
N 4 523 2 706 265 275
Year 9 Boys 598.5 537.5 586.5
Girls 591.4 527.5 577.6
All 594.9 532.6 582.2
(68.0) (83.5) (70.2)
N 4 452 2 369 262 122
NB. The information reported in this table is derived from the MCEETYA
Report (2008a).
* Standard deviations were only reported for overall students, and not
by gender.Table 1 Reading means, standard deviations* and sample sizes, by state
and Territory
NSW VIC QLD WA SA TAS
Year 3 Boys 405.2 413.9 363.1 377.0 392.2 396.0
Girls 419.7 426.0 379.5 396.8 409.2 406.5
All 412.3 419.9 371.1 386.7 400.5 401.2
(SD) (80.1) (74.9) (84.9) (87.7) (80.5) (84.2)
N 85 682 62 230 55 770 26 635 18 717 6 377
Year 5 Boys 488.3 491.3 459.6 467.2 472.5 473.4
Girls 501.3 502.3 472.7 480.5 483.2 479.5
All 494.7 496.7 466.1 473.6 477.9 476.4
(SD) (74.9) (69.3) (77.5) (77.2) (71.3) (75.8)
N 85 775 62 954 55 459 26 630 18 664 6 158
Year 7 Boys 538.1 538.6 522.9 521.9 528.8 530.5
Girls 547.1 547.6 533.5 532.4 538.2 538.1
All 542.5 543.0 528.1 527.0 533.5 534.2
(SD) (69.0) (63.1) (67.1) (67.0) (65.2) (68.5)
N 85 350 63 760 56 296 27 379 19 222 6 422
Year 9 Boys 579.9 582.7 564.5 566.2 572.9 577.6
Girls 586.5 586.5 572.2 573.7 576.8 580.2
All 583.1 584.6 568.2 569.8 574.9 578.8
(SD) (66.9) (62.6) (68.0) (65.6) (64.1) (67.9)
N 84 520 62 853 56 133 27 392 18 647 6 179
ACT NT All
Year 3 Boys 414.1 297.4 393.1
Girls 428.0 316.0 408.2
All 421.0 306.6 400.5
(SD) (81.5) (134.1) (84.5)
N 4 174 2 787 262 372
Year 5 Boys 495.5 397.6 478.4
Girls 511.2 413.6 490.7
All 503.3 405.1 484.4
(SD) (72.2) (123.3) (76.5)
N 4 341 2 891 262 872
Year 7 Boys 554.6 463.6 531.9
Girls 561.9 473.9 541.4
All 558.2 468.4 536.5
(SD) (70.2) (107.7) (68.2)
N 4 527 2 671 265 627
Year 9 Boys 597.0 521.9 575.0
Girls 606.6 526.7 581.0
All 601.9 524.2 578.0
(SD) (68.4) (101.8) (67.0)
N 4 439 2 386 262 549
The information reported in this table is derived from the MCEETYA
Report (2008a).
* Standard deviations were only reported for overall students, and not
by gender.
Table 2 NAPLAN effect sizes for boys and girls on all scales
Reading Writing Spelling G&P Numeracy
Year 3 NSW 0.18 0.39 0.25 0.26 0.11
VIC 0.16 0.42 0.23 0.29 0.16
QLD 0.19 0.39 0.26 0.25 0.10
WA 0.23 0.43 0.29 0.31 0.05
SA 0.21 0.41 0.29 0.28 0.13
TAS 0.12 0.44 0.16 0.15 0.04
ACT 0.17 0.42 0.29 0.28 0.13
NT 0.14 0.26 0.20 0.15 0.09
Australia 0.18 0.39 0.24 0.26 0.11
Year 5 NSW 0.17 0.40 0.26 0.31 0.16
VIC 0.16 0.43 0.27 0.33 0.20
QLD 0.17 0.40 0.30 0.30 0.17
WA 0.17 0.41 0.28 0.32 0.16
SA 0.15 0.40 0.25 0.29 0.24
TAS 0.08 0.45 0.17 0.17 0.05
ACT 0.22 0.41 0.26 0.38 0.21
NT 0.13 0.28 0.22 0.20 0.11
Australia 0.16 0.40 0.26 0.30 0.17
Year 7 NSW 0.13 0.40 0.28 0.33 0.19
VIC 0.14 0.46 0.28 0.37 0.25
QLD 0.16 0.42 0.32 0.34 0.19
WA 0.16 0.42 0.29 0.33 0.22
SA 0.14 0.43 0.29 0.32 0.24
TAS 0.11 0.50 0.18 0.20 0.06
ACT 0.10 0.47 0.27 0.30 0.28
NT 0.10 0.27 0.02 0.20 0.08
Australia 0.14 0.42 0.28 0.33 0.20
Year 9 NSW 0.10 0.45 0.30 0.31 0.10
VIC 0.06 0.44 0.28 0.29 0.17
QLD 0.11 0.46 0.33 0.30 0.11
WA 0.11 0.46 0.32 0.32 0.14
SA 0.06 0.45 0.27 0.28 0.21
TAS 0.04 0.47 0.11 0.14 0.08
ACT 0.14 0.50 0.33 0.37 0.10
NT 0.05 0.34 0.21 0.17 0.12
Australia 0.09 0.44 0.29 0.29 0.13
Table 3 Percentages of students per reading band, by year level and
gender
Band 1 & below Band 2 Band 3
Boys Girls Boys Girls Boys Girls
Year 3 NSW 5.1 2.8 11.8 8.5 17.8 15.9
VIC 2.7 1.4 9.2 6.7 17.6 15.3
QLD 13.1 8.8 19.4 16.2 20.6 20.5
WA 11.7 7.4 16.0 12.8 18.9 17.8
SA 6.6 4.3 13.9 10.2 18.4 16.6
TAS 6.8 5.6 14.5 11.4 18.0 17.0
ACT 4.9 2.2 10.2 7.2 14.1 14.5
NT 38.2 33.0 15.1 15.0 13.7 14.6
ALL 7.4 4.7 13.5 10.4 18.4 16.9
Band 4 Band 5 Band 6 & above
Boys Girls Boys Girls Boys Girls
Year 3 NSW 23.1 24.3 21.9 24.9 19.1 22.8
VIC 23.5 24.4 23.8 26.3 19.7 24.1
QLD 20.4 23.5 15.2 18.9 8.7 10.9
WA 21.3 22.7 18.5 21.6 12.3 17.1
SA 23.2 24.2 19.9 23.4 14.2 19.0
TAS 21.7 23.2 20.7 22.3 17.0 19.9
ACT 21.5 22.9 23.7 25.8 22.6 26.2
NT 13.0 15.0 10.2 12.1 8.1 8.8
ALL 22.3 23.8 20.3 23.3 15.9 19.6
Band 3 & below Band 4 Band 5
Boys Girls Boys Girls Boys Girls
Year 5 NSW 6.9 4.4 13.4 10.8 23.6 22.1
VIC 4.7 3.0 12.4 9.9 24.5 22.8
QLD 13.6 9.3 18.2 16.2 25.0 25.7
WA 11.9 8.2 16.3 14.3 25.0 24.2
SA 8.5 6.4 16.0 13.5 26.5 25.4
TAS 10.4 8.3 15.9 15.2 24.4 24.4
ACT 4.8 2.8 11.8 8.2 22.5 20.4
NT 38.4 33.4 14.1 13.0 16.9 17.6
ALL 8.8 6.1 14.7 12.3 24.4 23.5
Band 6 Band 7 Band 8 & above
Boys Girls Boys Girls Boys Girls
Year 5 NSW 25.8 27.1 18.6 21.5 10.7 13.4
VIC 26.8 28.5 19.1 22.2 9.4 11.8
QLD 23.2 25.4 12.8 15.7 5.1 6.6
WA 24.8 26.7 14.9 18.0 6.2 7.9
SA 25.0 28.1 15.0 17.0 5.7 7.7
TAS 25.2 25.5 15.8 17.3 7.4 8.3
ACT 27.4 28.5 20.5 22.7 11.3 16.2
NT 16.2 18.8 9.1 10.9 3.9 4.8
ALL 25.2 27.0 16.7 19.6 8.2 10.4
Band 4 & below Band 5 Band 6
Boys Girls Boys Girls Boys Girls
Year 7 NSW 5.1 2.8 14.9 12.0 26.4 26.3
VIC 3.4 1.7 13.8 10.5 28.2 27.5
QLD 6.8 4.1 17.9 14.3 29.0 29.4
WA 7.9 4.7 17.7 14.2 28.9 29.6
SA 5.4 3.7 16.5 13.0 28.8 27.9
TAS 6.1 4.5 16.2 14.1 28.1 27.4
ACT 4.0 1.5 11.4 8.3 21.8 23.0
NT 33.3 29.6 16.6 15.3 19.5 21.7
ALL 5.7 3.4 15.7 12.5 27.7 27.6
Band 7 Band 8 Band 9 & above
Boys Girls Boys Girls Boys Girls
Year 7 NSW 26.3 29.3 17.3 18.9 9.2 10.3
VIC 28.0 31.0 17.2 19.7 7.4 8.3
QLD 25.6 28.9 13.5 16.0 5.2 6.0
WA 25.3 28.9 14.2 16.1 5.0 5.6
SA 25.8 30.0 15.3 16.9 5.7 6.9
TAS 26.1 28.2 15.3 16.9 7.2 8.4
ACT 25.6 29.4 21.7 22.2 14.6 14.7
NT 15.8 17.6 9.7 9.7 3.9 4.7
ALL 26.3 29.5 16.0 17.9 7.2 8.1
Band 5 & below Band 6 Band 7
Boys Girls Boys Girls Boys Girls
Year 9 NSW 6.3 3.8 18.3 15.9 27.4 28.8
VIC 4.2 2.9 16.6 14.8 29.3 30.5
QLD 9.8 6.4 20.8 18.5 28.7 31.1
WA 9.1 6.0 20.6 18.5 29.3 31.2
SA 6.9 5.3 18.7 16.9 28.8 31.3
TAS 6.6 6.1 18.0 17.6 29.2 28.0
ACT 4.3 1.8 13.5 11.1 24.0 23.5
NT 29.2 27.0 17.9 17.5 21.4 22.4
ALL 7.1 4.8 18.6 16.5 28.4 30.0
Band 8 Band 9 Band 10 & above
Boys Girls Boys Girls Boys Girls
Year 9 NSW 25.4 27.8 15.2 16.3 6.7 6.9
VIC 26.5 28.8 15.3 15.8 5.8 5.8
QLD 23.3 25.9 11.9 13.0 3.9 4.0
WA 24.2 26.7 12.3 13.1 3.7 4.0
SA 25.6 27.0 12.9 13.5 4.4 4.3
TAS 25.0 25.7 14.6 15.7 6.0 6.2
ACT 27.7 29.0 19.3 21.9 10.9 12.4
NT 16.6 19.3 9.5 9.6 3.2 2.6
ALL 25.1 27.3 14.1 15.0 5.4 5.6
The percentages in this table are reported in the MCEETYA Report on
NAPLAN, pages 7, 57, 107 and 157.
Table 4 Gender ratios of poor performance on NAPLAN scales by two
definitions, by year level and state and territory
Reading Writing Spelling
Lowest Lowest Lowest Lowest Lowest Lowest
Band 2 Bands Band 2 Bands Band 2 Bands
Year 3 NSW 1.82 1.60 4.00 3.20 2.63 2.15
VIC 1.93 1.65 4.25 3.38 2.25 1.91
QLD 1.49 1.34 2.20 2.03 1.73 1.50
WA 1.58 1.42 2.29 2.16 1.85 1.60
SA 1.53 1.45 2.45 2.41 1.92 1.70
TAS 1.21 1.24 3.75 3.14 1.39 1.29
ACT 2.23 1.82 6.75 4.59 2.73 2.15
NT 1.16 1.08 1.29 1.22 1.22 1.13
All 1.57 1.44 2.35 2.24 1.85 1.64
Year 5 NSW 1.57 1.40 2.95 2.41 2.41 1.99
VIC 1.57 1.41 3.18 2.60 2.40 1.98
QLD 1.46 1.29 2.32 1.96 2.05 1.68
WA 1.45 1.30 2.35 2.00 1.97 1.66
SA 1.33 1.26 2.26 2.01 1.84 1.61
TAS 1.25 1.15 3.00 2.43 1.33 1.30
ACT 1.71 1.58 3.24 2.57 2.65 2.03
NT 1.15 1.12 1.31 1.28 1.24 1.18
All 1.44 1.32 2.44 2.10 2.02 1.73
Year 7 NSW 1.82 1.53 2.90 2.30 2.32 1.92
VIC 2.00 1.66 3.45 2.74 2.32 2.00
QLD 1.66 1.46 2.48 2.07 2.14 1.83
WA 1.68 1.46 2.37 2.00 1.93 1.70
SA 1.46 1.36 2.39 2.11 2.03 1.81
TAS 1.36 1.25 3.23 2.42 1.47 1.36
ACT 2.67 2.02 3.78 2.86 2.82 2.24
NT 1.13 1.10 1.30 1.29 1.22 1.19
All 1.68 1.47 2.63 2.19 2.07 1.81
Year 9 NSW 1.66 1.40 2.60 2.07 2.15 1.79
VIC 1.45 1.28 2.90 2.29 1.96 1.67
QLD 1.53 1.33 2.34 1.87 1.92 1.67
WA 1.52 1.32 2.34 1.88 1.94 1.62
SA 1.30 1.20 2.46 2.03 1.73 1.50
TAS 1.08 1.05 2.30 1.85 1.08 1.12
ACT 2.39 1.80 3.41 2.49 2.50 2.00
NT 1.08 1.05 1.43 1.32 1.30 1.21
All 1.48 1.30 2.47 1.99 1.74 1.56
Grammar &
punctuation Numeracy
Lowest Lowest Lowest Lowest
Band 2 Bands Band 2 Bands
Year 3 NSW 2.25 1.88 1.05 1.03
VIC 1.93 1.87 0.64 0.73
QLD 1.61 1.43 0.97 0.94
WA 1.73 1.52 1.09 1.06
SA 1.81 1.61 0.91 0.91
TAS 1.28 1.23 0.92 0.99
ACT 2.77 2.16 1.07 1.07
NT 1.16 1.10 1.00 0.97
All 1.74 1.58 0.97 0.96
Year 5 NSW 2.09 1.83 0.87 0.87
VIC 2.44 2.10 0.69 0.76
QLD 1.86 1.61 0.83 0.85
WA 1.87 1.64 0.88 0.88
SA 1.78 1.59 0.68 0.74
TAS 1.35 1.28 1.07 1.03
ACT 2.69 2.42 0.97 0.85
NT 1.19 1.20 0.93 0.93
All 1.89 1.69 0.83 0.84
Year 7 NSW 2.53 2.05 0.86 0.86
VIC 2.88 2.29 0.68 0.74
QLD 2.22 1.85 0.83 0.85
WA 2.01 1.69 0.87 0.86
SA 2.02 1.75 0.67 0.75
TAS 1.45 1.35 1.28 1.15
ACT 3.26 2.45 1.11 0.95
NT 1.22 1.17 0.99 0.97
All 2.23 1.88 0.84 0.85
Year 9 NSW 2.27 1.81 0.92 0.93
VIC 2.20 1.79 0.74 0.79
QLD 2.00 1.65 0.95 0.93
WA 2.01 1.62 0.92 0.90
SA 1.90 1.59 0.74 0.78
TAS 1.10 1.14 0.93 0.94
ACT 3.39 2.47 1.03 0.97
NT 1.20 1.22 0.94 0.93
All 2.02 1.67 0.89 0.89