Child Abuse & Neglect 76 (2018) 65–74

Contents lists available at ScienceDirect

Child Abuse & Neglect

journal homepage: www.elsevier.com/locate/chiabuneg

Research article

The influence of maltreatment history and out-of-home-care on MARK

children’s language and social skills
⁎
Jarrad A.G. Luma, , Martine Powellc, Pamela C. Snowb
a
School of Psychology, Deakin University, Australia
b
La Trobe Rural Health School, La Trobe University, Australia
c
Centre for Investigative Interviewing, Deakin University, Australia

AR TI CLE I NF O AB S T R A CT

Keywords: This study examined the extent to which maltreatment history and the characteristics of out-of-
Maltreatment home care correlated with the language and social skills of maltreated children. Participants in
Language this study were 82 maltreated children aged between 5 and 12 years of age. All children were
Social skills residing with state-designated carers in out-of-home-care. The children were presented with
Out-of-home-care
standardised tests assessing language and social skills. Results showed that the sample performed
Children
significantly below the normative mean on both tests. Correlation analyses showed social skills,
but not language skills were correlated with aspects of maltreatment history. The education level
of the state-designated carer/s was correlated with the children's language skills; higher educa-
tion level was associated with higher language skills. The study provides evidence that at the
group level, language and social skills are poor in maltreated children. However, gains in lan-
guage skills might be made via the out-of-home-care environment. Improvements in the social
skills of maltreated children may require additional support.

1. Introduction

The negative consequences of child maltreatment have been well documented (for reviews see Gilbert et al., 2009; Norman et al.,
2012). Children who have been maltreated are likely to experience social and behavioural problems (Hildyard & Wolfe, 2002), have
an increased risk of academic failure (Kendall-Tackett & Eckenrode, 1996; Romano, Babchishin, Marquis, & Fréchette, 2014) and
chronic health problems (Kerker et al., 2015; Kolko, Moser, & Weldy, 1990). Language and social skills (Bolger & Patterson, 2001a;
Darwish, Esquivel, Houtz, & Alfonso, 2001; Lum, Powell, Timms, & Snow, 2015) are also vulnerable to the effects of maltreatment.
This is a concern as both are important predictors of social and academic outcomes (Conti-Ramsden, Durkin, Simkin, & Knox, 2009;
Lonigan, Burgess, & Anthony, 2000; Malecki & Elliot, 2002; Snowling, Bishop, Stothard, Chipchase, & Kaplan, 2006). The current
study investigated the extent to which maltreatment history and out-of-home care characteristics relate to individual differences in
the language and social skills of maltreated children.
Children exposed to maltreatment have repeatedly been found to have poorer language skills compared to non-maltreated
children (for reviews see Law & Conway, 1992; Stacks, Beeghly, Partridge, & Dexter, 2011; Veltman & Browne, 2001). There is a
reliable association between maltreatment and language skills even after controlling for socio-economic factors that might otherwise
contribute to group differences. Lum, Powell, Timms and Snow (2015) used meta-analysis to summarise the data from 26 studies that
compared the language skills of maltreated children to a control group matched on socioeconomic background. The analyses revealed

⁎
Corresponding author at: School of Psychology, Deakin University, 221 Burwood Highway, Melbourne, VIC, Australia.
E-mail address: jarrad.lum@deakin.edu.au (J.A.G. Lum).

http://dx.doi.org/10.1016/j.chiabu.2017.10.008
Received 28 February 2017; Received in revised form 27 September 2017; Accepted 14 October 2017
0145-2134/ © 2017 Published by Elsevier Ltd.
J.A.G. Lum et al. Child Abuse & Neglect 76 (2018) 65–74

that, on average, maltreated children had significantly poorer expressive and receptive language skills compared to controls.
Another common finding in the field is that maltreated children have poorer social skills compared to non-maltreated children.
Social skills are a collection of abilities and behaviours that promote positive social interactions (Gresham & Elliott, 1984) and are
closely related to pragmatic language abilities (Snow & Douglas, 2017). There is evidence suggesting that social skill deficits are
common in maltreated children (for a recent review see Holosko, 2015). For instance, maltreated children have been found to show
lower levels of empathy and perspective taking (Burack et al., 2006; Kay & Green, 2016), are poorer at initiating social interactions
(Darwish et al., 2001), less cooperative and assertive (Kinard, 1999) and more withdrawn (Bolger & Patterson, 2001b). In instances
where maltreated children do interact with others, higher levels of aggression (Aber & Allen, 1987) and bullying have been observed
(Hong, Espelage, Grogan-Kaylor, & Allen-Meares, 2012).
However, close inspection of the language and social skills research in maltreated children reveals variability in findings. Some
studies report no differences between maltreated and non-maltreated children on measures of language functioning (e.g., Beers & De
Bellis, 2002; Gregory & Beveridge, 1984; McFadyen & Kitson,1996). Also, some subgroups of maltreated children do not differ from
controls on tasks assessing emotion recognition (Pollak, Cicchetti, Hornung, & Reed, 2000) and, aggression (Straker & Jacobson,
1981). Along similar lines, within a sample of maltreated children, it is not the case that language and social skills are universally
poorer. Stacks, Beeghly, Partridge, and Dexter (2011) examined the language skills of 963 infants with substantiated cases of mal-
treatment longitudinally over a five-year period. Replicating past research, at the group level, language skills were consistently below
age appropriate levels. However, within the maltreated sample, there were children who had normal to well above average language
skills (e.g., +3 SD above the normative mean). The social skills of the same sample of children were examined by Schultz, Tharp-
Taylor, Haviland, and Jaycox (2009). Again, analysis at the group level revealed poorer social skills. But within the sample, in-
spection of the standard deviation and range of scores indicates there were maltreated children with social skills that were age-
appropriate and in some cases in the superior range.
There has been considerable interest in identifying variables that might explain individual differences in the language and social
skill outcomes of maltreated children. Research undertaken to date has examined the extent to which maltreatment history (Culp
et al., 1991; Kaplow & Widom, 2007; Manly et al., 2001), home environment and, broader demographic factors (Bradley, Caldwell,
Fitzgerald, Morgan, & Rock, 1986; Chamberland, Lacharité, Clément, & Lessard, 2014; Howell, Miller, Lilly, & Graham-Bermann,
2013) explains individual differences in the social, cognitive and language functioning of maltreated children. In part, this research is
motivated by the need to determine which areas of functioning or case history should be investigated further with the view to
developing interventions and/or policies that reduce the developmental impact of maltreatment (Pears & Fisher, 2005a).
There is some evidence that type of maltreatment has different effects on language and social skills. Fox, Long, and Langlois
(1988), Allen and Oliver (1982) and Culp et al. (1991) found poorer language skills in children who had been neglected compared to
those who had been abused. Hoffman-Plotkin and Twentyman (1984) and Bousha and Twentyman (1984) found differences in the
social skills between abused and neglected children. Abused children were found to interact more with others compared to neglected
children. However, the abused children exhibited higher rates of aggression compared to neglected children. One factor that may
contribute to individual differences is the age at which maltreatment commences, with earlier onset being associated with greater
problems (De Bellis, Baum et al., 1999a; De Bellis, Keshavan et al., 1999b), although data in this area are currently lacking. Indeed,
the meta-analysis undertaken by Lum et al. (2015) found only one study, (conducted by Barnett, Vondra & Shonk, 1996) that in-
vestigated the relationship between onset of maltreatment and language and social skills. In that study, no association was found
between maltreatment onset and language or social skills. However, it is important to note that the sample size in the study was small
(n = 13) and further study is required to determine the replicability of the result.
Another aspect of maltreatment that may impact language and social skills is the influence of the out-of-home care environment.
It is becoming increasingly more common for maltreated children to be removed from their homes and placed in the care of others for
varying periods of time. For example, in Australia, there was a 67% increase in out-of-home placement from 2005 to 2012 (Australian
Institute of Health and Welfare, 2013). However, there is a paucity of research that specifically investigates the effects of the out-of-
home care environment on language and social skills (Byrne, 2016). Pears and Fisher (2005a) examined the relationship between out-
of-home care placement characteristics and language skills. In that study, no relationship was found between language and the age of
first placement or the number of days in foster care. In a subsequent study, Pears and Fisher (2005b) found no association between
indicators of social skills (emotion understanding, theory of mind) with the length of time in out-of-home care or the number of
transitions between out-of-home placements. Whether other aspects of out-of-home care impact on language and social skills is yet to
be examined.

1.1. Aim of the current study

The aim of the current study was to further examine the correlates of language and social skills in a sample of maltreated children
currently in out-of-home care. In this study, we sought to extend the current literature by examining the extent to which different
maltreatment types as well as, the age at which they were substantiated, correlated with language and social skills. Additionally, we
studied associations between these two variables and out-of-home placement characteristics. To our knowledge, research undertaken
to date in this area has only focused on the length of time in out-of-home care placements (Pears & Fisher, 2005a; Pears & Fisher,
2005b). In this study, we examined whether the educational levels and socioeconomic status of the carers were related to the
measures of language and social skills.

66
J.A.G. Lum et al. Child Abuse & Neglect 76 (2018) 65–74

Table 1
Maltreatment Characteristics of the Sample.

Maltreatment n of cases % of Cases Perpetrator (% of cases) Age Maltreatment First Substantiated (Years)
Type
Biological Biological Relative Non- M SD Range
Mother Father Relative

Physical Abuse 11 13.9% 90.9% 18.2% 9.1% 45.5% 4.0 2.2 1.17 – 9.3
Sexual Abuse 8 10.1% 0.0% 25.0% 37.5% 75.0% 4.1 3.2 < 1 year – 9.8
Emotional Abuse 20 25.3% 90.0% 50.0% 15.0% 10.0% 3.3 2.5 < 1 year – 8.4
Neglect 68 86.1% 86.8% 36.8% 0.0% 0.0% 2.7 2.0 < 1 year – 9.3

2. Method

2.1. Participants

A total of 82 children (40 females) aged between 5;3 (years; months) and 12;10 participated in this study (M = 8;2 SD = 2;1). All
children had at least one substantiated case of physical abuse, sexual abuse, emotional abuse and/or neglect as indicated by gov-
ernment records. Multiple instances of abuse were common; 31.7% of the sample had been exposed to more than one substantiated
type of maltreatment. Summary data showing maltreatment history for the sample are presented in Table 1. A case of maltreatment
was identified as substantiated by the government agency when an outcome of an investigation concluded that a child/young person
had either been harmed or at significant risk of harm from abuse or neglect.
All children who took part in this study were receiving services provided by a government child protection agency in a major
Australian city.1 At the time of this research, all children were currently in departmental supported out-of-home care, that is, they had
been removed from their biological parent/s and were residing with state-designated carer/s. This may involve the child being placed
with relatives, foster carers or in family group homes. Consent for the child’s participation in the study was provided by the carer. The
sample size of 82 children represents around 4% of cases handled by the child protection agency annually.
This study was completed with the approval of the Deakin University Human Research Ethics Committee and the relevant state
government authority.

2.2. Materials

Children underwent a range of standardised tests that measured their language, social skills and general intellectual functioning.
Children’s language skills were assessed using the Clinical Evaluation of Language Fundamentals-4th Edition, Australian
Standardisation (Semel, Wiig, & Secord, 2003a). Social skills were assessed with the Social Skills Improvement System Rating Scales
(Gresham & Elliott, 2008). General intelligence was assessed using the Raven’s Coloured Progressive Matrices (Raven & Court, 1998).
Data from the Raven’s Coloured Progressive Matrices were used as a covariate in the analyses. Each child’s maltreatment and out-of-
home care placement history was obtained from government records. Carers provided demographic information relating to their
education level and socio-economic status, via structured interview.

2.2.1. The clinical evaluation of language fundamentals-4th Edition, Australian Standardisation (CELF-4)
The Clinical Evaluation of Language Fundamentals-4th Edition, Australian Standardisation (CELF-4) (Semel et al., 2003a) assesses how
well children use vocabulary and grammar to understand and produce spoken language. Children were administered subtests from
the CELF-4 to permit calculation of the Core Language Score (CLS). The CLS is a composite variable derived from summing subtests
that assess children’s expressive and receptive language skills. The CLS is standardised to a mean of 100 and standard deviation of 15.
Scores of 85 or lower indicate that a language problem or impairment is highly likely (Sensitivity = 0.83, Sensitivity = 0.90; Semel,
Wiig, & Secord, 2003b). The internal consistency of the CLS is high, with an average reliability coefficient of 0.96 across the stan-
dardisation sample (Semel et al., 2003a).

2.2.2. The social skills improvement system’s (SSIS) rating scales

The Social Skills Improvement System’s (SSIS) Rating Scales (Gresham & Elliott, 2008) was used to assess social skills. This tool
comprises 83 statements describing different aspects of prosocial (e.g., shows kindness to others when they are upset) and problem
behaviour (e.g., acts sad or depressed). The respondent, who in this study was the child’s carer/s, indicates how frequently the
behaviour occurs on a four-point scale: ‘never’, ‘seldom’, ‘often’, or ‘almost always’.
In this study, two composite indices from the SSIS were used to summarise children’s socials skills. These were the ‘Social Skills’
and ‘Problem Behaviour’ composite scores. The Social Skills composite measures the extent to which the child engages in behaviour
that facilitates positive social interactions while also limiting negative social interactions (Gresham & Elliott, 2008). This composite
measures pro-social communication, cooperation, assertion, responsibility, empathy, engagement and self-control. The Problem

1
The city is not named to protect the identities of the participants.

67
J.A.G. Lum et al. Child Abuse & Neglect 76 (2018) 65–74

Behaviour composite evaluates the frequency of behaviours that negatively impact on social interactions. The SSIS examines this
construct in relation to bullying, externalising, internalising and also attentional problems. The Social Skills and Problem Behaviour
composites are standardised to a mean of 100 and standard deviation of 15. Higher scores on the Social Skills composite indicate
increasingly better social skills. A standard score of −1 SD below the normative mean indicates the presence of social skill problems.
On the Problem Behaviour composite, higher standard scores indicate increasing levels of problem behaviour. A standard score of +1
SD or greater indicates the presence of behaviour problems. Coefficient alpha for the Social Skills composite is reported to be 0.97 and
0.95 for Problem Behaviours composite (Gresham & Elliott, 2008).

2.2.3. Raven’s coloured progressive matrices (RCPM)

Raven’s Coloured Progressive Matrices (RCPM) (Raven & Court, 1998) was used to estimate general intelligence. This test was
selected because it places minimal demands on children’s oral language skills. On this test, children identify a missing piece from a
visuospatial pattern by selecting one of six possible options. The RCPM has been found to correlate positively with Performance IQ
(r = 0.52), Verbal IQ (r = 0.54) and Full Scale IQ (r = 0.55) from the Wechsler Intelligence Scale for Children (Chalmers & Carleton,
1955). Children’s performance on the RCPM was described using as a standard score (i.e., mean of 100, SD of 15) using conversion
tables provided by Sattler (2001).

2.2.4. Maltreatment and out of home care history

Each child’s statutory maltreatment history and details of his/her out-of-home placements were obtained from records held by
child protective services. Maltreatment history and out-of-home placement information was accessed by a social worker employed by
the agency. To ensure information was systematically recorded, a coding instrument was used. For each child, the following aspects
of maltreatment history were collected: (i) Type(s) of substantiated maltreatment to which the child was exposed, (ii) age of child
when maltreatment was first documented, (iii) age of child at most recent substantiated maltreatment instance, (iv) the relationship
of the perpetrator to the child. The specific types of maltreatment recorded were neglect, emotional abuse, physical abuse and sexual
abuse. The out-of-home placement information obtained was (i) length of time in current placement, (ii) total time in out-of-home
care and (iii) the total number of out-of-home placements.
Finally, we obtained demographic information about each carer, specifically, highest educational level and household income.
Education level was coded using a 4-point scale where 1 = Secondary/High School Education, 2 = Tertiary and Further Education
colleges (equivalent to Community College/Junior College education in the United States), 3 = Undergraduate/Bachelor Degree,
4 = Post-graduate Degree. An estimate of the extent of socioeconomic advantage was also obtained. This was achieved using the
Index of Relative Socio-Economic Advantage and Disadvantage computed by the Australian Bureau of Statistics (Australian Bureau of
Statistics, 2001) using the most recent census data. This index was calculated for each child using postcode (zipcode) information.
Higher values indicate higher levels of socio-economic advantage. Socio-economic advantage is defined as the extent to which people
in an area have access to “material and social resources, and their ability to participate in society” (Australian Bureau of Statistics,
2001, p. 6). For this study, the index was transformed to a z-score, with higher values indicating higher levels of socio-economic
advantage.

2.3. Procedures

The language and intelligence tests were individually administered to each child in a single session lasting around 50–60 min. All
tests were administered to the child by a speech-language pathologist. Testing took place in a meeting room located at the child
protection agencies’ offices. While testing took place, the child’s carer completed the social skills inventory and also a demographics
questionnaire.

3. Results

3.1. Summary data and preliminary analyses from standardised tests, maltreatment history and placement conditions

Tables 1 and 2 present summary data showing maltreatment history and a summary of the out-of-home placement arrangements.
Table 1 indicates neglect was the most frequent form of maltreatment in this sample. Emotional abuse was the second most common
form of maltreatment, with the biological mother identified as the perpetrator in nearly all substantiated cases. It should be noted
that the perpetrator data summarised in Table 1 exceeds 100% on each row. This occurred because in most cases there was more than
one perpetrator. There were only a small number of cases where physical and/or sexual abuse were substantiated. Subsequently, the
relationship between these two maltreatment types with language and social skills was not examined. At the group level, Table 1
shows a tendency for neglect to be substantiated earlier than other forms of maltreatment. However, inspection of the range of values
across all four maltreatment types shows substantial variability with respect to the age at which the abuse or neglect was first
substantiated. Summary data showing carers’ background and household information are presented in Table 2. This table also shows
the children’s placement histories. It is noted that within the sample the experience of children with respect to the total number of out
of home care placements varies considerably. Ninety percent of the sample experienced between 1 and 50 out of home placements.
The remaining 8% experienced between 51–79 placements. Two-percent of the sample experienced more than 100 out-of-home care
placements.
Data from the group’s performance on the standardised tests assessing language, social skills and intelligence are presented in

68
J.A.G. Lum et al. Child Abuse & Neglect 76 (2018) 65–74

Table 2
Background Characteristics of Out-of-Home Carers.

Variable n %

Education Level
Female Carer
% High School Level 47 60.30%
% Bachelor Level or Higher 41 39.70%

Male Carer
% High School Level 20 51.3
% Bachelor Level or Higher 19 48.7

Household Income
% Below Average Weekly Earning 13 16.7
% Average Weekly Earning 42 53.8
% Above Average Weekly Earning 19 24.2

Variable M SD Range

Socioeconomic Advantage and Disadvantage for Area 0.54 0.63 -0.65 to 1.97
Time in Current Placement (months) 34.0 26.4 1–109
Total Number of Out of Home Care Placements 19.5 27.6 1–185

Table 3
Summary Data from Language, Social Skills and General Intelligence Tests.

Test % of sample scoring < 1 SD below normative mean M SD Range

Language
Core Language Score (CELF-4) 31.6% 91.7 14.6 49–125

Social Skills
Social Skills Index (SSIS-RS) 39.2% 91.8 16.6 48–124
Problem Behaviour Index (SSIS-RS) 54.4% 118.9 17.0 85–160

General Intelligence
RCPM 11.4% 100.1 10.1 76–114

Abbreviations: CELF–4 = Clinical Evaluation of Language Fundamentals; SSIS-RS = Social Skills Improvement System-Rating Scales; RCPM = Raven's Coloured
Progressive Matrices.

Table 3. At the group level, language and social skills are, on average, below the normative mean. One sample t-tests were conducted
comparing the group’s average scores against the normative means of 100. These analyses revealed the CLS (t (82) = 5.013,
p < 0.001) and Social Skills Composite (t (82) = 4.314, p < 0.001) were significantly below the normative mean, indicating
poorer language and prosocial skills. The problem behaviour composite was significantly above the normative mean (t (82) = 9.841,
p < 0.001), indicating that the sample was exhibiting levels of problem behaviour that were above age-appropriate levels. The
group’s average intelligence was within normal limits (t (82) = 1.644, p = 0.104).
However, we also identified substantial individual differences on the language and social skills standardised tests. Just over 30%
of the children met the psychometric criterion for language impairment on the CELF-4. On the Social Skills composite, which
measures difficulties executing prosocial type behaviours, 39.2% showed below age-appropriate performance. The Problem
Behaviours composite, which assesses the presence of maladaptive behaviours, showed 54.4% of the sample was exhibiting in-
appropriate levels for their age. Thus, at the group level, problems were present with respect to language and social skill. However, at
the individual level, not all children presented with problems in these two domains.

3.1.1. The relationship between maltreatment history, language and social skills
The first set of analyses examined associations between maltreatment history and individual differences in language and social
skills. Two sets of analyses were conducted to investigate these associations. First, bivariate correlations were computed that mea-
sured the association between the maltreatment history variables reported in Table 1 and data from standardised tests presented in
Table 3. Second, partial correlations were then computed that controlled for individual differences in age and intellectual functioning
(assessed using RCPM). These two variables were controlled given the variability in age and intellectual functioning within the
sample. A dummy variable was used to examine the relationship between substantiated maltreatment on the one hand, and language
and social skills on the other. The presence of neglect or emotional abuse was coded as 1 and its absence as 0. Results from the
analyses are presented in Table 4.
The measure of children’s overall language functioning was not found to be related to the presence of neglect or emotional abuse.
However, significant negative associations were found between the presence of a substantiated report of neglect and the social skills
composite from the SSIS-RS. The direction of the association indicates that children with substantiated cases of neglect had lower

69
J.A.G. Lum et al. Child Abuse & Neglect 76 (2018) 65–74

Table 4
Associations between Maltreatment History and Measures of Language and Social Skills.

Maltreatment Language (CELF-4) Social Skills (SSIS-RS) Problem Behaviour (SSIS-RS)

History
No covariates Controlling:Age & RCPM No covariates Controlling:Age & RCPM No covariates Controlling:Age & RCPM
Scores Scores Scores

Substantiated report of:

Neglect −0.053 −0.095 −0.238* −0.293* 0.167 0.189
Emotional 0.139 0.165 −0.099 −0.121 −0.022 0.004
Abuse

Age of first substantiated report of:

Neglect −0.062 −0.076 0.138 0.144 0.012 0.003
Emotional −0.031 −0.083 −0.532* −0.648* 0.459 0.564*
Abuse

* p < 0.05.

social skills compared to other children in the sample. Specifically, the mean Social Skills composite score for children with a
substantiated report of neglect was 90.1 (SD = 17.0) and for children without a substantiated report of neglect 101.8 (SD = 11.3). In
this sample, a substantiated report of neglect was associated with, on average, an 11-point lower score on the Social Skills composite
compared to maltreated children without a substantiated report of neglect.
A significant negative association was also found between the age at which emotional abuse was substantiated and social skills. To
illustrate this association Fig. 1 presents a scatterplot reporting the age emotional abuse was substantiated and Social Skills composite
score. The direction of the association indicates children whose abuse was substantiated at an earlier age evidenced ‘better’ social
skills compared to children in which emotional abuse was substantiated at an older age. More specifically, Fig. 1 shows that seven out
of nine children whose abuse was substantiated between the ages of 0–3, scored within the average range on the Social Skills
composite score. That is a score of 85 or better. In contrast, only four out of eleven children whose abuse was substantiated at age 4-
years or older scored within the average range on the Social Skills composite score.
The Problem Behaviour composite from the SSIS-RS was not found to be significantly correlated with substantiated reports of any
maltreatment type once age and intellectual functioning were controlled. However, a significant positive association was found
between the age emotional abuse was substantiated and the Problem Behaviour Composite. Fig. 2 presents a scatterplot showing the
association between the age emotional abuse was substantiated and the Problem Behaviour composite. On the Problem Behaviour
composite, higher scores indicate higher levels of the behaviour. The direction of the association in these cases indicates that children
whose maltreatment was substantiated earlier, had lower levels of behaviour problems. Fig. 2 shows that seven out of nine children
whose abuse was substantiated between the ages of 0–3, scored within the average range on the Problem Behaviour composite. That
is, those children evidenced problem behaviour that was at least in the average range. For those children whose abuse was sub-
stantiated at age 4-years or older, four out of 11 scored within the average range on the Problem Behaviour composite.

3.1.2. The relationship between out-of-home placement characteristics and children’s social skills, language and intellectual functioning
The next set of analyses examined associations between the scores from each of the standardised tests and demographic char-
acteristics of the child’s current carer/s. As with the previous analyses, associations were examined using bivariate and partial

Fig. 1. The relationship between the age emotional abuse was first substantiated and children’s standardised scores from the Social Skills Composite.

70
J.A.G. Lum et al. Child Abuse & Neglect 76 (2018) 65–74

Fig. 2. The relationship between the age emotional abuse was first substantiated and children’s standardised scores from the Problem Behaviour Composite.

correlations. The partial correlations statistically controlled for age and intellectual functioning. Results from these analyses are
presented in Table 5.
Statistically significant associations were found between children’s language skills and female as well as male carers’ education
level. The direction of the correlation indicates that higher carer education levels were associated with higher language scores in the
children. To provide more information about the nature of this relationship Table 6 shows the mean CLS at each education level for
female and male carers. The main trend to emerge from this table is that CLS that were around 1 SD below the normative mean were
associated with carers whose highest education level was secondary/high school. At higher education levels the mean CLS was
within ± 1 SD of the normative mean.
All other correlations were not found to be significant after controlling for age and intellectual functioning. The Social Skills or
Problem Behaviour Composites were not found to be significantly correlated with out-of-home care placement characteristics.

4. Discussion

This study examined language skills, and social skills and their correlates in a sample of primary-school aged maltreated children.
At the group level, the children evidenced lower language and social skills compared to normative samples. However, there were
individual differences on both measures of language and social skills. The analyses revealed that individual differences in language
skills were not correlated with the aspects of maltreatment history examined, but were with respect to out-of-home placement
characteristics. The opposite pattern of results was observed for social skills. Individual differences in children’s social skills were
related to maltreatment history but not to the characteristics associated with the out-of-home placement. Collectively, these data
indicate that maltreatment elevates the risk of language and social skills problems. Promisingly, it seems that there are elements of
out-of-home placement that may have a positive effect on language skills. The non-significant relationship observed between out-of-
home care and social skills may indicate ongoing formal support is required to improve this aspect of development for this group.
At the group level, the sample was found to have language and social skills that were below age-expected levels. It is ac-
knowledged this result may reflect selection bias. That is, carers who were concerned about their child's language or social skills were

Table 5
Associations between Out of Home Care Characteristics and Measures of Language and Social Skills.

Placement Characteristics Language (CELF-4) Social Skills Problem Behaviour

No covariates Covariates: No covariates Covariates: No covariates Covariates:

Age & RCPM Scores Age & RCPM Scores Age & RCPM Scores

Carer is a biological relative −0.027 −0.028 0.048 0.037 −0.029 −0.013

Female Carer Education Level 0.361* 0.345* −0.004 0.034 0.058 0.016
Male Carer Education Level 0.335* 0.322* −0.142 −0.136 0.162 0.135
Household Income −0.130 −0.135 0.050 0.043 −0.112 −0.103
Socioeconomic Advantage and 0.059 0.097 0.088 0.123 −0.170 −0.210
Disadvantage for Area
Time in Current Placement −0.052 −0.046 −0.072 −0.064 0.016 0.007
(months)
Total Number of Out of Home 0.278 *
0.200 −0.218 −0.207 0.241*
0.220
Care Placements

71
J.A.G. Lum et al. Child Abuse & Neglect 76 (2018) 65–74

Table 6
Summary Statistics showing Core Language Scores reported by Education Level and Carer Gender.

Highest Education Level Female Carer Male Carer

M SD n M SD n

a
Secondary/High School Education 85.2 15.6 46 86.2 13.9 21
Tertiary and Further Education College 96.2 10.0 15 96.4 8.6 10
Bachelor Degree 96.6 13.9 11 104.6 9.7 5
Post-Graduate Degree 93.6 12.7 10 91.8 14.5 4

a
Equivalent to Community College/Junior College in North America.

more likely to participate in the study. While this possibility cannot be entirely discounted, it is noted that there were some children
with language and social skills that were well above the normative mean on the standardised tests. So, the presence of poor language/
social skills does not appear to have influenced participation for all carers. Also, it should also be noted that the group and child-level
data from the language and social skills measures are largely consistent with past research (e.g., Holosko, 2015; Lum et al., 2015).
These findings add to the growing body of evidence that childhood maltreatment negatively affects key skills needed to succeed
academically and socially. Also, consistent with past research (Schultz et al., 2009; Stacks et al., 2011), we observed significant
performance variability, with some children performing above the normative means.
This study extends the literature by identifying variables that covary with individual differences in language functioning and
social skills. Regarding maltreatment history, a substantiated case of neglect or emotional abuse was not associated with higher or
lower scores relative to the other maltreated children in the sample. The null result observed for neglect is inconsistent with some
past research. Several studies have found that neglected children have poorer language skills compared to children who have ex-
perienced other forms of maltreatment (e.g., Allen & Oliver, 1982; Culp et al., 1991; Fox et al., 1988). However, in the meta-analysis
by Lum et al. (2015) neglect was not found to predict lower language scores across studies. Given this, the non-significant association
between language and neglect appears to be more in line with the meta-analysis data. The results of the current study, along with the
meta-analysis, indicate that any type of maltreatment may negatively impact on language skills.
However, neglect might be associated with poorer social skills compared to other forms of maltreatment. In our sample, children
with a substantiated case of neglect obtained a lower score on the Social Skills composite from the SSIS-RS compared to other
children in the sample. This indicates these children had lower levels of prosocial behaviour, for example, lack of positive interac-
tions, more withdrawn behaviour, and internalising behaviour problems compared to others in the sample. However, children with
substantiated neglect did not score higher on the Problem Behaviour composite compared to others. This suggests comparable levels
of externalising behaviour problems (e.g., aggressive behaviour, bullying) between neglected and other children. The lower levels of
prosocial behaviour observed in the neglected children are consistent with past research (Bousha & Twentyman, 1984; Hoffman-
Plotkin & Twentyman, 1984). Thus, internalising problems (e.g., withdrawn behaviour), appears to be an outcome associated with
neglect. The finding that abused children did not have externalising-type behaviour problems is inconsistent with some past studies.
Children who have experienced abuse have been found to have significantly higher levels of externalising behaviour problems
(Bousha & Twentyman, 1984; Hoffman-Plotkin & Twentyman, 1984; Lansford et al., 2007). Our findings may be accounted for by the
small number of children in our sample with a substantiated case of abuse (n = 11). However, the findings of this study potentially
indicate that levels of problem behaviour may be comparable between children who have experienced neglect and emotional abuse.
Interestingly, children who had emotional abuse substantiated at an earlier age evidenced higher levels of social skills and lower
levels of problem behaviour. To our knowledge, this is the first study to examine the social skills in emotionally abused children.
Thus, explanations for this pattern of results are necessarily tentative. These findings may highlight the importance of early inter-
vention in maltreatment cases. Once maltreatment has been substantiated, the outcome of the investigation may lead to either
intervention by a government agency or at least awareness of the problem by carers or parents. Our data might be indicating that
early intervention may be reducing the long-term impact emotional abuse has on social skills. However, one limitation of our study
was that no information was available concerning the type of support or interventions that were afforded to the children following
confirmation of maltreatment, and we did not follow the sample up over time.
At a more general level, caution is required when interpreting the correlations concerning the earliest ages at which cases were
substantiated. Decisions to investigate cases of maltreatment depend on a range of variables including the ability to identify a
perpetrator, the child’s capacity to provide information, and resources of the child welfare agency (Cicchetti & Toth, 2005;
Eckenrode, Powers, Doris, Munsch, & Bolger, 1988). In the context of this study, the government records concerning the age of the
first substantiated case of maltreatment are unlikely to be error-free, an issue that has been discussed by Bromfield and Higgins
(2004). These factors may also explain the absence of correlations between age of first substantiated report of neglect and emotional
abuse with the language measures. Indeed, one study found that neglect was less likely to be reported to statutory authorities
(Nightingale & Walker, 1986). This might explain the non-significant correlations observed between the age of first reporting for
neglect and language skills.
Another finding of this study was the significant associations between language scores and out-of-home-care placement char-
acteristics. Before considering these results it should be noted that one limitation of our data is that carer information was available
only for the current placement. In our study significant correlations were observed between the education level of the carers and
maltreated children’s language scores. Higher language scores were found in maltreated children living with carers with higher

72
J.A.G. Lum et al. Child Abuse & Neglect 76 (2018) 65–74

education level. Why might carer education level influence language skills in maltreated children? Data from the language devel-
opment literature provides one explanation for this association. In this literature, a positive correlation between parental education
level and children’s language proficiency has repeatedly been found (e.g., Zambrana, Ystrom, & Pons, 2012). The explanation given
for this association is that parents with higher education levels tend to use a greater diversity in words, speak to their children using
complex syntactic structures and ask more open-ended questions compared to those with lower education levels (Hoff, 2003; Pan
et al., 2005). Thus, one aspect of the out-of-home environment that may be influencing the language skills of maltreated children is
the communicative characteristics of the carers. Importantly, research has shown that the communicative style used by parents with
high education levels can be trained, leading to improved language skills in children (Snow & Beals, 2006; Whitehurst et al., 1994).
Thus, with training, it seems plausible to suggest that all carers can provide a linguistic environment that promotes optimal language
development. Based on the data presented in Table 6, such training would be especially beneficial to carers whose highest education
level is secondary/high school. For this group, the average language scores of the maltreated children was lowest compared to other
education levels.
No significant correlations were observed between measures of social skills and out-of-home care placement characteristics. Pears
and Fisher (2005b) also found non-significant correlations between emotion understanding and two aspects of out-of-home care:
number of days in care and number of transitions. Additional research is required to better understand the impact of out-of-home care
on social skills. However, based on the data so far it seems that unlike language, social skills are less amenable to the influence of the
characteristics of the carer(s) and/or home environment. This finding potentially indicates that maltreated children with low social
skills may require formal support above that provided incidentally in out-of-home care.

5. Conclusions

This study examined language and social skills and their correlates in maltreated children. Replicating a large body of work, the
study further confirms that as a group, these children perform below the normative average on standardised tests of language and
social skills. However, there is substantial variability between children in these domains. The study demonstrated that maltreatment
characteristics correlated with social skills. Also, there are carer-related variables that correlate with language functioning. Thus,
while out-of-home-care may provide a means to improve language skills in this group, additional support appears to be needed to
foster social skills. Social skills allow children to use their oral language skills to make and maintain friendships, and reduce the
likelihood of interpersonal conflict, and so are critical to mental health in the developmental period and beyond (Snow & Douglas,
2017).

Acknowledgement

This research was supported by the Australian Research Council (Grant#: DP1095509).

References

Aber, J. L., & Allen, J. P. (1987). Effects of maltreatment on young children's socioemotional development: An attachment theory perspective. Developmental
Psychology, 23(3), 406.
Allen, R. E., & Oliver, J. M. (1982). The effects of child maltreatment on language development. Child Abuse & Neglect, 6(3), 299–305.
Australian Bureau of Statistics (2001). Socio-economic indexes for areas (SEIFA). Canberra: Australian Bureau of Statistics.
Australian Institute of Health and Welfare (2013). Child protection Australia 2011–12. Canberra: Australian Institute of Health and Welfare.
Barnett, D., Vondra, J. I., & Shonk, S. M. (1996). Self-perceptions, motivation, and school functioning of low-income maltreated and comparison children. Child
Abuse & Neglect, 20(5), 397–410.
Beers, S. R., & De Bellis, M. D. (2002). Neuropsychological function in children with maltreatment-related posttraumatic stress disorder. The American Journal of
Psychiatry, 159(3), 483–486.
Bolger, K. E., & Patterson, C. J. (2001a). Developmental pathways from child maltreatment to peer rejection. Child Development, 72(2), 549–568.
Bolger, K. E., & Patterson, C. J. (2001b). Pathways from child maltreatment to internalizing problems: Perceptions of control as mediators and moderators. Development
and Psychopathology, 13(04), 913–940.
Bousha, D. M., & Twentyman, C. T. (1984). Mother–child interactional style in abuse, neglect, and control groups: Naturalistic observations in the home. Journal of
Abnormal Psychology, 93(1), 106.
Bradley, R. H., Caldwell, B. M., Fitzgerald, J. A., Morgan, A. G., & Rock, S. L. (1986). Behavioral competence of maltreated children in child care. Child Psychiatry and
Human Development, 16(3), 171–193.
Bromfield, L. M., & Higgins, D. J. (2004). The limitations of using statutory child protection data for research into child maltreatment. Australian Social Work, 57(1),
19–30.
Burack, J. A., Flanagan, T., Peled, T., Sutton, H. M., Zygmuntowicz, C., & Manly, J. T. (2006). Social perspective-taking skills in maltreated children and adolescents.
Developmental Psychology, 42(2), 207–217.
Byrne, N. (2016). Systematic review of speech and language therapy outcomes for children who are in out of home care (OOHC). Speech, Language and Hearing, 20,
57–61.
Chalmers, S. L., & Carleton, F. O. (1955). The relationship between Raven's Colored Progressive Matrices and two tests of general intelligence. Journal of Clinical
Psychology, 11, 84–85.
Chamberland, C., Lacharité, C., Clément, M.-È., & Lessard, D. (2014). Predictors of development of vulnerable children receiving child welfare services. Journal of Child
and Family Studies.
Cicchetti, D., & Toth, S. L. (2005). Child maltreatment. Annual Review of Clinical Psychology, 1, 409–438.
Conti-Ramsden, G., Durkin, K., Simkin, Z., & Knox, E. (2009). Specific language impairment and school outcomes. I: Identifying and explaining variability at the end of
compulsory education. International Journal of Language & Communication Disorders, 44(1), 15–35.
Culp, R. E., Watkins, R. V., Lawrence, H., Letts, D., Kelly, D. J., & Rice, M. L. (1991). Maltreated children's language and speech development: abused, neglected, and
abused and neglected. First Language, 11(33, Pt 3), 377–389.
Darwish, D., Esquivel, G. B., Houtz, J. C., & Alfonso, V. C. (2001). Play and social skills in maltreated and non-maltreated preschoolers during peer interactions. Child
Abuse & Neglect, 25(1), 13–31.
De Bellis, M. D., Baum, A. S., Birmaher, B., Keshavan, M. S., Eccard, C. H., Boring, A. M., ... Ryan, N. D. (Baum et al., 1999a). Developmental traumatology part I:

73
J.A.G. Lum et al. Child Abuse & Neglect 76 (2018) 65–74

Biological stress systems. Biological Psychiatry, 45(10), 1259–1270.

De Bellis, M. D., Keshavan, M. S., Clark, D. B., Casey, B., Giedd, J. N., Boring, A. M., ... Ryan, N. D. (Keshavan et al., 1999b). Developmental traumatology part II: brain
development. Biological Psychiatry, 45(10), 1271–1284.
Eckenrode, J., Powers, J., Doris, J., Munsch, J., & Bolger, N. (1988). Substantiation of child abuse and neglect reports. Journal of Consulting and Clinical Psychology,
56(1), 9–16.
Fox, L., Long, S. H., & Langlois, A. (1988). Patterns of language comprehension deficit in abused and neglected children. Journal of Speech and Hearing Disorders, 53(3),
239–244.
Gilbert, R., Widom, C. S., Browne, K., Fergusson, D., Webb, E., & Janson, S. (2009). Burden and consequences of child maltreatment in high-income countries. The
Lancet, 373(9657), 68–81.
Gregory, H. M., & Beveridge, M. (1984). The social and educational adjustment of abused children. Child Abuse & Neglect, 8(4), 525–531.
Gresham, F. M., & Elliott, S. N. (1984). Assessment and classification of children's social skills: A review of methods and issues. School Psychology Review, 13(3),
292–301.
Gresham, F., & Elliott, S. N. (2008). Social skills improvement system (SSIS) rating scales. Bloomington: Pearson Assessments.
Hildyard, K. L., & Wolfe, D. A. (2002). Child neglect: Developmental issues and outcomes. Child Abuse & Neglect, 26(6), 679–695.
Hoff, E. (2003). Causes and consequences of SES-related differences in parent-to-child speech. In M. H. Bornstein, & B. H. Robert (Eds.). Socioeconomic status, parenting,
and child development (pp. 147–160). Mahwah, NJ, US: Lawrence Erlbaum Associates Publishers.
Hoffman-Plotkin, D., & Twentyman, C. T. (1984). A multimodal assessment of behavioral and cognitive deficits in abused and neglected preschoolers. Child
Development, 55(3), 794–802.
Holosko, M. J. (2015). The empirical base for the implementation of social skills training with maltreated children. In J. S. Wodarski, M. J. Holosko, & M. D. Feit (Eds.).
Evidence-informed assessment and practice in child welfare (pp. 261–278). New York: Springer.
Hong, J. S., Espelage, D. L., Grogan-Kaylor, A., & Allen-Meares, P. (2012). Identifying potential mediators and moderators of the association between child mal-
treatment and bullying perpetration and victimization in school. Educational Psychology Review, 24(2), 167–186.
Howell, K. H., Miller, L. E., Lilly, M. M., & Graham-Bermann, S. A. (2013). Fostering social competence in preschool children exposed to intimate partner violence:
Evaluating the preschool kids' club intervention. Journal of Aggression, Maltreatment & Trauma, 22(4), 425–445.
Kaplow, J. B., & Widom, C. S. (2007). Age of onset of child maltreatment predicts long-term mental health outcomes. Journal of Abnormal Psychology, 116(1), 176–187.
Kay, C. L., & Green, J. M. (2016). Social cognitive deficits and biases in maltreated adolescents in UK out-of-home care: Relation to disinhibited attachment disorder
and psychopathology. Development and Psychopathology, 28(01), 73–83.
Kendall-Tackett, K. A., & Eckenrode, J. (1996). The effects of neglect on academic achievement and disciplinary problems: A developmental perspective. Child
Abuse & Neglect, 20(3), 161–169.
Kerker, B. D., Zhang, J., Nadeem, E., Stein, R. E., Hurlburt, M. S., Heneghan, A., ... Horwitz, S. M. (2015). Adverse childhood experiences and mental health, chronic
medical conditions, and development in young children. Academic Pediatrics, 15(5), 510–517.
Kinard, E. M. (1999). Perceived social skills and social competence in maltreated children. American Journal of Orthopsychiatry, 69(4), 465–481.
Kolko, D. J., Moser, J. T., & Weldy, S. R. (1990). Medical/health histories and physical evaluation of physically and sexually abused child psychiatric patients: A
controlled study. Journal of Family Violence, 5(4), 249–267.
Lansford, J. E., Miller-Johnson, S., Berlin, L. J., Dodge, K. A., Bates, J. E., & Pettit, G. S. (2007). Early physical abuse and later violent delinquency: A prospective
longitudinal study. Child Maltreatment, 12(3), 233–245.
Law, J., & Conway, J. (1992). Effect of abuse and neglect on the development of children's speech and language. Developmental Medicine & Child Neurology, 34(11),
943–948.
Lonigan, C. J., Burgess, S. R., & Anthony, J. L. (2000). Development of emergent literacy and early reading skills in preschool children: evidence from a latent-variable
longitudinal study. Developmental Psychology, 36(5), 596.
Lum, J. A., Powell, M., Timms, L., & Snow, P. (2015). A meta-analysis of cross sectional studies investigating language in maltreated children. Journal of Speech,
Language, and Hearing Research, 58(3), 961–976.
Malecki, C. K., & Elliot, S. N. (2002). Children's social behaviors as predictors of academic achievement: A longitudinal analysis. School Psychology Quarterly, 17(1),
1–23.
Manly, J. T., Kim, J. E., Rogosch, F. A., & Cicchetti, D. (2001). Dimensions of child maltreatment and children's adjustment: Contributions of developmental timing and
subtype. Development and Psychopathology, 13(4), 759–782.
McFadyen, R. G., & Kitson, W. J. (1996). Language comprehension and expression among adolescents who have experienced childhood physical abuse. Journal of Child
Psychology and Psychiatry, 37(5), 551–562.
Nightingale, N. N., & Walker, E. F. (1986). Identification and reporting of child maltreatment by Head Start personnel: Attitudes and experiences. Child Abuse & Neglect,
10(2), 191–199.
Norman, R. E., Byambaa, M., De, R., Butchart, A., Scott, J., & Vos, T. (2012). The long-term health consequences of child physical abuse, emotional abuse, and neglect:
A systematic review and meta-analysis. PLOS Medicine, 9(11), e1001349.
Pan, B. A., Rowe, M. L., Singer, J. D., & Snow, C. E. (2005). Maternal correlates of growth in toddler vocabulary production in low income families. Child Development,
76(4), 763–782.
Pears, K., & Fisher, P. A. (2005a). Developmental, cognitive, and neuropsychological functioning in preschool-aged foster children: Associations with prior mal-
treatment and placement history. Journal of Developmental and Behavioral Pediatrics, 26(2), 112–122.
Pears, K. C., & Fisher, P. A. (2005b). Emotion understanding and theory of mind among maltreated children in foster care: Evidence of deficits. Development and
Psychopathology, 17(1), 47–65.
Pollak, S. D., Cicchetti, D., Hornung, K., & Reed, A. (2000). Recognizing emotion in faces: developmental effects of child abuse and neglect. Developmental Psychology,
36(5), 679.
Raven, J. C., & Court, J. H. (1998). Raven's progressive matrices and vocabulary scales. Oxford Psychologists Press.
Romano, E., Babchishin, L., Marquis, R., & Fréchette, S. (2014). Childhood maltreatment and educational outcomes. Trauma, Violence, & Abuse, 16(4), 418–437.
Sattler, J. M. (2001). Assessment of children: Cognitive applications (4th ed.). La Mesa: Jerome M. Sattler Publisher, Incorporated.
Schultz, D., Tharp-Taylor, S., Haviland, A., & Jaycox, L. (2009). The relationship between protective factors and outcomes for children investigated for maltreatment.
Child Abuse & Neglect, 33(10), 684–698.
Semel, E., Wiig, E. H., & Secord, W. A. (2003a). Clinical Evaluation of Language Fundamentals-4th Edication, Australian Standardisation (CELF-4). Marrickville: Harcourt
Assessment.
Semel, E., Wiig, E. H., & Secord, W. A. (2003b). Clinical evaluation of language fundamentals-4th edication, Australian standardisation (CELF-4) examiner's manual.
Marrickville: Harcourt Assessment.
Snow, C. E., & Beals, D. E. (2006). Mealtime talk that supports literacy development. New Directions for Child and Adolescent Development, 111, 51–66.
Snow, P. C., & Douglas, J. (2017). Psychosocial aspects of pragmatic language difficulties. In L. Cummings (Vol. Ed.), Research in clinical pragmatics, series: Perspectives
in pragmatics, philosophy & psychology: vol. 11, (pp. 617–649). Cham: Springer-Verlag.
Snowling, M. J., Bishop, D., Stothard, S. E., Chipchase, B., & Kaplan, C. (2006). Psychosocial outcomes at 15 years of children with a preschool history of speech-
language impairment. Journal of Child Psychology and Psychiatry, 47(8), 759–765.
Stacks, A. M., Beeghly, M., Partridge, T., & Dexter, C. (2011). Effects of placement type on the language developmental trajectories of maltreated children from infancy
to early childhood. Child Maltreatment, 16(4), 287–299.
Straker, G., & Jacobson, R. (1981). Aggression, emotional maladjustment, and empathy in the abused child. Developmental Psychology, 17(6), 762.
Veltman, M. W. M., & Browne, K. D. (2001). Three decades of child maltreatment research: Implications for the school years Trauma, Violence. & Abuse, 2(3),
215–239.
Whitehurst, G. J., Arnold, D. S., Epstein, J. N., Angell, A. L., Smith, M., & Fischel, J. E. (1994). A picture book reading intervention in day care and home for children
from low-income families. Developmental Psychology, 30(5), 679.
Zambrana, I. M., Ystrom, E., & Pons, F. (2012). Impact of gender, maternal education, and birth order on the development of language comprehension: A longitudinal
study from 18 to 36 months of age. Journal of Developmental & Behavioral Pediatrics, 33(2), 146–155.

74