Content uploaded by Stephen Joseph
Author content
All content in this area was uploaded by Stephen Joseph on Aug 18, 2018
Content may be subject to copyright.
The Multidimensional Peer Victimization Scale: A Systematic Review
Stephen Joseph*
School of Education, University of Nottingham, UK.
Hannah Stockton
School of Education, University of Nottingham, UK
*Corresponding author:
Stephen Joseph
School of Education,
Jubilee Campus,
University of Nottingham, UK
Conflicts of interest: None
This research did not receive any specific grant from funding agencies in the public,
commercial, or not-for-profit sectors.
Abstract
Developing bullying interventions and testing their success depends on the valid and
reliable measurement of peer victimization. The objective of this study was to examine the
psychometric properties of the Multidimensional Peer Victimization Scale (MPVS, Mynard
& Joseph, 2000). This systematic review examined 34 published studies demonstrating that
the MPVS is a reliable, valid, and psychometrically sound measure for capturing multiple
facets of peer victimization across a variety of samples. Results also highlighted that there are
relatively stable sex differences in the rates and pattern of peer victimization, with males
experiencing more direct forms of victimization and females experiencing more indirect
forms of victimization. Recommendations for further research are discussed, alongside new
ways to further advance the assessment of peer victimization.
Keywords: Peer victimization; bullying; Multidimensional Peer Victimization Scale;
systematic review; psychometric properties.
1. Introduction
Peer victimization involves the repeated and systematic abuse of power by one or
more peers over a period of time in purposeful attempts to injure or inflict discomfort
(Olweus, 1993). Peer victimization is a relatively frequent experience among young people:
estimates vary depending on age and gender, but research has suggested that between 5% and
30% of children and adolescents are victims (Eslea et al., 2004; Stassen Berger, 2007). Other
estimates have suggested that rates of victimization may reach as high as 32% in high-income
countries and 60% in low- to middle-income countries (Currie et al., 2012; Fleming &
Jacobsen, 2010).
Peer victimization experiences are associated with a range of physical, emotional,
academic and behavioural problems. Several systematic reviews and meta-analyses have
demonstrated that victims generally have a lower quality of life and experience poor self-
esteem (Hawker & Boulton, 2000); experience loneliness and isolation (Storch & Masia-
Warner, 2004); increased psychosomatic complaints (Gini & Pozzoli, 2009); greater anxiety
and depression (Hawker & Boulton, 2000); are at greater risk for suicidal ideation and
behaviours (van Geel et al. 2014); greater externalising problems such as aggression,
delinquency and misconduct (Reijntjes et al., 2011); and perform less well academically
(Nakamoto & Schwartz 2010) than those who are not victimized. The psychological
difficulties experienced through peer victimization in childhood and adolescence may
produce negative outcomes well into adulthood (see McDougall & Vaillancourt, 2015). As
such, peer-victimization and how to provide helpful interventions for young people is a topic
of much interest to educationalists and other professionals (Crothers & Levinson, 2004).
In order to develop interventions and assess their success it is necessary to accurately,
reliably, and comprehensively assess the construct of peer-victimization. As such, researchers
have developed numerous self-report measures. A recent review identified 41 unique
measures of peer victimization (Vivolo-Kantor, Martell, Holland, & Westby, 2014). While
this number has the advantage of permitting choice over instrument selection, it has
simultaneously resulted in significant inconsistencies in measurement that can contribute to
conflicting prevalence estimates and research results (Vivolo-Kantor et al., 2014). No one
measure is universally recognised as the instrument of choice, although some measures are
used more frequently than others.
One commonly used measure is the Multidimensional Peer-Victimization Scale
(MPVS; Mynard & Joseph, 2000). The MPVS is a 16-item self-report instrument that
contains four subscales: physical victimization, comprising items examining how often the
child has been subject to physical harm such as being punched or kicked; verbal
victimization, comprising items examining behaviours such as name calling or being made
fun of; social manipulation, comprising items concerned with negative social behaviours by
some children to turn others against the child; and attacks on property, comprising items
relating to the damage or theft of possessions. Each item is scored on a three point Likert-
scale of 0 = not at all, 1 = once and 2 = more than once, with participants indicating how
often during the school year they had experienced each of the 16 victimization experiences.
Total victimization scores range from a possible 0 to 32, with subscale scores ranging from 0
to 8. Higher scores indicate that a child has been subjected to more incidents of peer
victimization.
The MPVS was developed with a sample of 812 children aged 11-16 years who
completed an initial survey of 45 items, reduced using factor analysis to the final 16 items
representing the four distinct factors. When developed, the MPVS provided a new,
empirically derived, and broader conceptualisation of peer victimization than instruments
available at the time, and uniquely provided convergent validity with self-reports of being
bullied (Vivolo-Kantor et al., 2014).
Although two relatively recent reviews of bullying scales have been conducted
(Vessey, Strout, DiFazio, & Walker, 2014; Vivolo-Kantor et al., 2014), these reviews focused
on the range of measures available and commented on the psychometric properties of each
measure as reported in their original development and validation studies. As such, the
psychometric data on the MPVS presented in both of these reviews was limited to the
original study. In the 18 years since its publication the MPVS has become a popular measure
and the evidence concerning its psychometric properties has accumulated. Despite the
widespread application of the MPVS in the bullying literature, and the relevance of this
literature in the wider context of child and adolescent well-being, a comprehensive literature
review regarding its use has not been conducted.
Given this gap in the literature, we undertook a systematic review of studies that have
employed the MPVS and reported data on its psychometric properties, including findings
relating to its factor structure, internal consistency reliability, construct validity and
associations with outcome variables. The aims of this paper were to review and summarise
the use of the MPVS in peer-reviewed published studies and to evaluate the available
evidence for its psychometric properties and applicability to a range of sample types and age
groups.
2. Method
2.1 Search and Selection Strategy
During July 2017, four electronic databases (ISI Web of Science, PsycINFO, Wiley
Online and GoogleScholar1) were searched for empirical papers citing the original MPVS
paper (Mynard & Joseph, 2000). These databases were also searched using the search term
‘Multidimensional Peer Victimization Scale.’ Reference lists from relevant studies were also
reviewed to ensure that we had identified all eligible studies that presented empirical results
for the MPVS. Studies were selected for inclusion in the review if the authors: (1) published
the paper in English; (2) published the paper in a peer-reviewed scientific journal; (3)
reported that the full 16-item MPVS had been administered; (4) used the correct scoring
procedure for the MPVS items (0 = not at all; 1 = once; 2 = more than once); (5) provided
information regarding psychometric properties such as factor analysis, internal consistency,
construct validity, and/or provided mean total scores. Studies were excluded if they were
qualitative studies, meta-analyses, literature reviews or did not present original empirical
results (e.g. if they provided a summary of ongoing research or studies still in progress).
2.2 Review Strategy
There were three main steps to the review. In Step One, all citations generated by the
database searches were reviewed. After eliminating duplicates, a comprehensive abstract
screening was conducted whereby information relating to inclusion and exclusion criteria was
extracted. This information included basic descriptive data such as the nature of the paper
(i.e. empirical study, literature review, book chapter, conference paper, doctoral thesis), the
language it was written in, and whether or not the MPVS had been used. Papers that did not
meet these criteria were excluded. In cases where it was not discernible from the abstract, the
paper was retained for step two.
In Step Two, a list of eligible studies was compiled and full-text articles extracted. Each
article was subjected to a thorough review and further descriptive data was documented,
including the size and general characteristics of the sample, whether or not the full 16-item
MPVS had been used, the scoring system that had been adopted, and whether or not data
regarding psychometric properties of the MPVS had been reported. This list was used to
finalise the studies to be included in the review.
In Step Three, for the studies that met the inclusion criteria, abstraction of results
focused on indicators of scale reliability and validity; results of factor analytic procedures;
and key study findings such as correlational and longitudinal relationships with other
variables of interest.
3. Results
3.1 Search Results
The search strategy identified 324 original articles published between the initial
publication of the MPVS in April 2000 and July 2017. Screening resulted in the exclusion of
290 papers. The flow diagram (Figure 1) details the study selection procedure. The main
reasons for exclusion were that the paper was a book chapter, doctoral dissertation or
conference paper (99 papers); the paper was not published in English (61 papers); the paper
was a review of existing research or summarised results of other published studies (21
papers); the paper was a qualitative study (2 papers); the original MPVS article was cited but
the MPVS was not administered (65 papers); the studies used a modified version of the
MPVS that did not include all 16 items (10 papers); the studies used the MPVS but did not
use the original scoring system (10 papers), or studies modified the MPVS by adding new
items to produce an idiosyncratic modified form of the MPVS making comparisons and
generalisations about the reliability and validity of the original MPVS impossible (14 papers).
Studies were however included if they added new items to produce additional subscales
alongside the original MPVS subscales. Morrow, Hubbard and Swift (214) added four items
to assess social rebuff. Betts, Houston and Steer (2015) added four items to assess electronic
victimization. In these two studies, results for the original MPVS were reported alongside the
new subscales. Finally, an additional 8 papers could not be located despite requests to authors
and extensive searches. Therefore, of the 324 papers identified, 34 fulfilled the strict
inclusion and exclusion criteria and were included in the final review (see Table 1 for a
summary of each paper).
- Insert Table 1 about here –
Figure 1. Flow diagram depicting study selection.
3.2 Description of Studies Included
The majority of studies included in this review involved either primary
school students (Andreou et al., 2005;
Azeredo et al., 2017; Balogun & Olapegba,
2007; Balogun et al., 2006; Defeyter et al.,
2015; Litman et al., 2015; Morrow, Hubbard & Swift, 2014; Morrow, et al., 2014;
!
"
Not written in English (n = 61)
Not peer reviewed (n = 99)
Review or meta-analysis (n =
21)
Qualitative study (n = 2)
#""
$
Did not use MPVS (n = 65)
Did not use all 16 items (n = 10)
Used additional items (n = 14)
Used incorrect scoring system (n
= 10)
Full text could not be accessed
% !&
'
#"
(%
)(%
(
*$
'
Piek et al., 2005;), secondary school students (Akram & Munawar, 2016; Anderson et al.,
2010; Betts et al., 2015; Betts et al., 2017; Betts & Spenser, 2017; Biebl et al., 2011; Bird et
al., 2017; Candel & Iacob, 2015; Fontaine et al., 2016; Kaiser & Malik, 2015; McFarlane et
al., 2017; Murphy et al., 2015; Mynard et al., 2000; Popoola, 2005; Rao & Kishore, 2013;
Scarpa et al., 2012; Shakoor et al., 2015; Waytowich et al., 2011), or both primary and
secondary school students (Fung & Raine, 2012; Law & Fung, 2013; Raine, Fung & Lam,
2011). Three studies included samples of university students (Cosgrove, Nickerson &
DeLucia, 2017; Lam, Raine & Lee, 2016; Lee, Abell & Holmes, 2015) and one study
included a community based sample of adults with or without schizophrenia (McGuire,
Barbanel, Brune & Langdon, 2015).
Mean participant ages ranged from a low of 8.4 years (range 5.3 to 10.11 years) in
Defeyter, Graham and Russo (2015) to a high of 22.14 years (range 18 to 60 years) in
Cosgrove, Nickerson and DeLucia (2017), with the majority of studies reporting a mean
participant age in the range of 11 to 15 years old. Of the five studies using adult samples, four
studies focused on recent experiences of peer victimization as an adult, while Cosgrove et al.
(2017) asked participants to respond to the MPVS with respect to their experiences of
victimization during schooling.
A number of studies involved samples with specific characteristics, including
adolescents with hearing impairment (Akram & Munawar, 2016), adolescents seeking
treatment for paranoid ideation (Bird et al., 2017), participants with schizophrenia (McGuire,
Barbanel, Brune & Langdon, 2015), children at risk of Developmental Co-ordination
Disorder (DCD; Piek, Barratt, Allen, Jones & Louise, 2005), obese adolescents (Rao &
Kishore, 2013), and juvenile delinquents (Waytowich et al., 2011). Two studies involved
participants that were enrolled in the Twins Early Development Study (TEDS; Fontaine,
Hanscombe, Berg, McCrory & Viding, 2016; and Shakoor, McGuire, Cardno, Freeman,
*
Plomin & Ronald, 2015), one study involved adolescents that were part of the longitudinal
Southern Illinois Twins and Siblings Study (SITSS; Biebl, DiLalla, Davis, Lynch & Shinn,
2011), and one study involved individuals from the Pelotas Cohort Study (Azeredo et al.,
2017).
Participants were from a variety of countries including Pakistan (Akram & Munawar,
2016; Kaiser & Malik, 2015; McFarlane et al., 2017), Brazil (Azeredo et al., 2017), Greece
(Andreou, Vlachou & Didaskalou, 2005), Nigeria (Balogun & Olapegba, 2007; Balogun,
Olapegbe & Opayemi, 2006; Popoola, 2005), Romania (Candel & Iacob, 2015), Hong Kong
(Fung & Raine, 2012; Lam, Raine & Lee, 2016; Law & Fung, 2013; Raine, Fung & Lam,
2011), Australia (Piek, Barratt, Allen, Jones & Louise, 2005) and Italy (Scarpa, Carraro,
Gobbi & Nart, 2012). Sample sizes varied from a low of 34 in Bird et al. (2017) to a high of
4,972 in Shakoor et al. (2015).
3.3 Examination of Scores
14 out of 34 studies (41%) provided mean scores for the MPVS total; 7 additional
studies also provided mean scores for each of the four MPVS subscales. With respect to mean
MPVS total, scores ranged from a low of 3.41 in Lam, Raine and Lee (2016) to a high of
23.16 in Popoola (2005), with most studies reporting means between 8 and 11. A notable
finding was that the studies with the three highest average scores were conducted with
Nigerian participants (Balogun & Olapegba, 2007; Balogun, Olapegba & Opayemi, 2006; &
Popoola, 2005).
Looking across studies, we observed that there was a trend for studies with samples of
younger participants to report higher mean scores than studies with older participants. Four
studies tested the impact of age on peer victimization: Andreou, Vlachou and Didaskalou
(2005) reported that children in 6th grade experienced significantly less attacks on property
$
than children in 4th grade, but no other significant differences by age were observed in this
study. Balogun, Olapegba and Opayemi (2006) reported that children aged 9 years or older
experienced more social manipulation than children aged below 9 years, but there were no
other significant differences by age for the other subscales or total score. Candel and Iacob
(2015) reported that MPVS total scores were significantly correlated with age (r = -.31), and
that participants aged between 11 and 13 years reported significantly more peer victimization
than participants aged between 17 and 19 years old. Lam, Raine and Lee (2016) reported a
significant positive correlation between age and peer victimization.
With respect to the average subscale scores, these ranged from a low of 0.18 for
physical victimization for females in Cosgrove, Nickerson and DeLucia (2017) to a high of
6.50 for attacks on property in Popoola (2005). More generally, verbal victimization showed
the highest subscale scores compared to the other subtypes of victimization, with 6 of the 7
studies that reported subscale means showing the highest mean scores for verbal
victimization (Andreou et al., 2005; Fontaine et al., 2016; Fung & Raine, 2012; Kaiser &
Malik, 2015; Mynard et al., 2000; Scarpa et al., 2012). Similarly, verbal victimization was
reported to be the most prevalent type of bullying in Azeredo et al. (2017) and Morrow,
Hubbard and Swift (2014), with 37.9% and 29% of participants endorsing verbal
victimization items, respectively. Physical victimization and attacks on property showed the
lowest subscale scores.
3.4 Sex Differences
16 out of 34 studies (47%) reported significant sex differences in MPVS total or
subscale scores. Overall, boys reported significantly more peer victimization than females,
with 6 studies reporting significantly higher MPVS total scores for boys than girls (Azeredo
et al., 2017; Kaiser & Malik, 2015; Lam, Raine & Lee, 2016; Litman et al., 2015; McFarlane
et al., 2017; Shakoor et al., 2015). An additional 9 studies reported that boys experienced
significantly more physical victimization than girls (Akram & Munawar, 2016; Anderson et
al., 2010; Andreou et al., 2005; Balogun & Olapegba, 2007; Betts et al., 2015; Cosgrove et
al., 2017; Fontaine et al., 2016; Litman et al., 2015; Popoola, 2005) and 5 studies reported
that boys experienced significantly more attacks on property than girls (Balogun et al., 2006;
Betts et al., 2015; Cosgrove et al., 2017; Fontaine et al., 2016; Litman et al., 2015). Five
studies reported that girls experienced more social manipulation than boys (Andreou et al.,
2005; Betts et al., 2015; Fontaine et al., 2018; Piek et al., 2005; Popoola, 2005).
3.5 Internal Consistency Reliability and Split Half Reliability
Cronbach’s alpha coefficient was reported in 25 studies (74%). The alpha coefficients
for the 16-item total score ranged from good to excellent across samples, with the lowest
reported as α = .74 in Lam, Raine and Lee (2016) and the highest α = .96 in Candel and Iacob
(2015). For the subscales, alpha coefficients ranged from .60 to .93, again representing good
internal consistency reliability. Kaiser and Malik (2015) reported the lowest range of alpha
scores (from .62 for physical victimization to .73 for social manipulation) while Morrow,
Hubbard and Swift (2014) reported the highest range (from .84 for their newly developed
‘social rebuff’ subscale to .93 for both verbal and social victimization). Only one study
reported Split half reliability (Balogun & Olapegba, 2007), which was found to be acceptable
(r = .76). No studies reported test re-test reliability.
3.6 Tests of Validity
3.6.1 Concurrent Validity. Evidence bearing on the concurrent validity of the MPVS –
that is, the extent to which the MPVS is correlated with other measures of peer victimization
– was reported in 4 studies. Balogun and Opalegba (2007) reported a correlation of r = .54 for
the MPVS and the Aggression Scale (Buss & Durkee, 1975); Betts and Spenser (2017)
reported significant positive correlations ranging from r = .21 to r = .62 between all four
MPVS subscales and three cyber-victimization subscales in two separate studies; Law and
Fung (2013) reported a correlation of r = .31 for the MPVS and the Online Victimization
Scale; and Lee, Abell and Holmes (2015) demonstrated a significant positive correlations of r
= .31 between the MPVS and the Cyberbullying Victimization scale (CBV) and r = .21 to r
= .30 with the CBV subscales.
3.6.2 Convergent Validity. Evidence for the convergent validity of the MPVS – that is,
the degree to which the MPVS correlated with measures of conceptually related constructs –
was reported in 24 studies. Peer victimization was positively associated with physical and
psychological health problems (Akram & Munawar, 2016), rumination (Candel & Iacob,
2015), poor attachment quality (Cosgrove, Nickerson & DeLucia, 2017), conduct problems,
emotional problems and negative parental discipline (Fontaine et al., 2016), negative emotion
including sadness, anger, embarrassment and nervousness (Morrow, Hubbard, Barhight &
Thomson, 2014), schizotypal personality / schizotypy (Fung & Raine, 2012; Lam, Raine &
Lee, 2016; Raine, Fung & Lam, 2011), paranoid ideation (Bird et al., 2017), depression,
anxiety and stress (Kaiser & Malik, 2015), general aggression, reactive aggression and
proactive aggression (Lam, Raine & Lee, 2016; Law & Fung, 2013; Raine, Fung & Lam,
2011), violence attribution errors (Waytowich et al., 2011), PTSD symptoms (Litman et al.,
2015; Mynard et al., 2000), posttraumatic cognitions, loneliness, and feelings of inferiority,
incompetence and being disliked (as assessed by the Social Comparison Scale; Murphy,
Murphy & Shevlin, 2015); and behavioural problems (Rao & Kishore, 2013). Similarly, the
MPVS was negatively associated with global self-worth (Mynard et al., 2000; Piek et al.,
2005) self-esteem (Betts et al., 2015; Rao & Kishore, 2013); positive interactions with peers
(Andreou et al., 2005); and academic achievement (Morrow, Hubbard & Swift, 2015).
Other study findings provide further support for the convergent validity of the MPVS.
Biebl et al. (2011) reported that chronic victims of bullying (those that experienced
victimization at age 5, 14 and 16 years) showed significantly higher rates of conduct
problems, physical health problems and headaches than non-victims (those that did not
experience victimization at any time). Fontaine et al. (2016) assessed Callous-Unemotional
(CU) traits at 7, 9 and 12 years and found that youths with high CU traits at both 7 and 12
years reported the highest levels of all four subtypes of peer victimization while youths with
low CU traits at both 7 and 12 years reported the lowest levels of all forms of peer
victimization. Using multi-group path analysis, Betts, Houston, Steer and Gardner (2017)
showed that for males, more frequent attacks on property predicted higher levels of loneliness
and depressive symptoms and lower levels of social confidence; and higher levels of verbal
victimization predicted lower global self-worth and higher levels of loneliness. For females,
the only significant path showed that higher levels of verbal victimization predicted lower
levels of global self-worth. Longitudinally, bullying victimization at age 12 was positively
associated with paranoia, hallucinations and cognitive distortion at age 16 (Shakoor et al.,
2015). Finally, maternal mood symptoms during pregnancy were associated with subsequent
physical and verbal victimization in their 11-year old offspring (Azeredo et al., 2017).
No studies reported evidence for divergent or discriminant validity.
3.7 Subscale inter-correlations
Nine studies reported significant positive inter-correlations between the four
subscales. These ranged from a low of r = .24 for physical and social manipulation
victimization in Cosgrove et al. (2017) to a high of r = .65 for verbal and social manipulation
victimization in Kaiser and Malik (2015). Overall, four studies reported that the lowest
subscale inter-correlations were between physical and social manipulation (Akram &
Munawar, 2016; Cosgrove et al., 2017; Fontaine et al., 2016; and Fung & Raine, 2012) and
four studies reported that the highest subscale inter-correlations were between verbal and
social manipulation (Anderson et al., 2010; Cosgrove et al., 2017; Fontaine et al., 2018; and
Kaiser & Malik, 2015). These findings are in contrast to the subscale inter-correlations
reported in the original Mynard and Joseph (2002) study, where physical victimization and
social manipulation were actually found to be the most strongly associated, and verbal
victimization and social manipulation were the second least strongly associated subscales.
3.8 Factor Structure
Five studies reported on factor analysis of the MPVS. Balogun and Opalegba (2007)
performed Principal Components Analysis with varimax rotation and Kaiser normalisation;
the results revealed four factors which showed a degree of agreement with the original factor
structure although there were some notable differences in item loadings. Items 5 and 9 from
the physical victimization subscale, item 4 from the attacks on property subscale and item 2
from the social manipulation subscale loaded on the verbal victimization factor; and item 15
from the verbal victimization factor loaded on the physical victimization factor. This resulted
in a 6-item factor that the authors named Provocative Victimization (to replace verbal
victimization), a 4-item factor that the authors named Confrontational Victimization (to
replace Attacks on property), and two 3-item factors which remained as Physical
Victimization and Social Manipulation. As this study was conducted with Nigerian primary
school children, the authors suggested that these differences in the factor structure may be
due to cultural and value differences.
Law and Fung (2013) employed maximum likelihood estimation Confirmatory Factor
Analysis to test a four-factor structure of the MPVS. The high CFI value (0.940), RMSEA =
0.08 and high factor loadings for all items indicated a good fitting four-factor model.
Two studies used all 16 items of the MPVS but included additional items. First, Betts,
Houston and Steer (2015) added 4-items to assess electronic victimization (e.g., “Sent you a
nasty text”) and used Confirmatory Factor Analysis to examine the factor structure. The
proposed 5-factor model (comprising the original 4 subscales plus the electronic
victimization subscale) was compared to a 2-factor model (overt and covert aggression) and a
4-factor model (comprising physical, social & electronic, verbal, and attacks on property).
The 5-factor model was the best fitting and met many of the requirements needed for good fit
– RMSEA was acceptable, CFI and GFI both exceeded an acceptable value of .90; and all
items exceeded or approached the minimum acceptable loading of .60.
Second, Morrow, Hubbard and Swift (2014) added four items designed to capture
social rebuff and used Confirmatory Factor Analysis to investigate the factor structure of the
revised MPVS. Results provided modest support for the proposed 5-factor model: χ2 (160) =
506.23, p = .00; RMSEA = .11; CFI = .85; SRMR = .08. All standardised factor loadings
were significant and greater than .55. This model provided a better fit than any of the 6
competing models that were tested, including a one-factor model and four different four-
factor models.
A subsequent study by Morrow, Hubbard, Barhight and Thomson (2014) further
investigated the factor structure of this adapted MPVS by performing several Confirmatory
Factor Analyses. The first model to be tested was the five factor model comprising the
original four factors plus a social rebuff factor; this model fit the data relatively well, χ2 (160)
= 338.81, p < .001; RMSEA = 0.03; CFI = 0.88; SRMR = 0.06. All standardized factor
loadings were significant and greater than 0.40. Additionally, all factor correlations were
positive and significant, yet did not indicate excessive overlap (0.18– 0.64). They then tested
two competing models: a single-factor model that did not fit the data better than the
hypothesised 5-factor model; and a four-factor model where social manipulation and social
rebuff were merged into one factor due to their conceptual similarity. Although this model fit
the data relatively well, the hypothesised five-factor model was a significantly better fit. In
summary, evidence supports the separate assessment of the four factors of the MPVS but
there may be contexts in which researchers wish to include items that include both electronic
victimization and social rebuff.
4. Discussion
We identified 34 articles published between April 2000 and July 2017 that reported
results on the Multidimensional Peer Victimization Scale. These studies reflect a broad range
of sample sizes of primary school, secondary school and adult populations from a number of
diverse backgrounds. The discussion that follows will summarise the salient findings of this
review: namely, that the MPVS was found to be a reliable and valid measure with good
evidence to support the four-factor structure; and that there are relatively stable sex
differences in the rates and pattern of peer victimization when assessed using the MPVS. We
will also identify research gaps and provide recommendations for future research.
4.1 Psychometric Properties
Reliability of the MPVS was assessed in terms of internal consistency reliability, with
25 studies reporting Cronbach’s alpha coefficient. Based on recommendations that
Cronbach’s alpha coefficients be ≥ .80 in order to be acceptable for basic research tools
(Streiner, 2003), the literature reviewed here supports the reliability of the MPVS. Eight
studies reported Cronbach’s alpha greater than .80 for the MPVS total score, with an
additional 11 studies reporting acceptable internal consistency reliability for the MPVS
subscales. One further study reported acceptable split-half reliability (Balogun & Olapegba,
2007). No studies reported test-retest reliability. These additional tests of reliability should be
investigated further in future research.
This review revealed evidence to support the validity of the MPVS, with four studies
providing evidence for its concurrent validity by demonstrating the expected associations
with related measures of similar constructs. With respect to convergent validity, 24 studies
reported associations between the MPVS and conceptually related constructs, including
measures of physical, psychological and behavioural problems that have previously been
shown to be associated with peer victimization. However, it has been argued that in order to
establish construct validity it is important to demonstrate both discriminant and convergent
validity (Campbell & Fiske, 1959), yet no studies reported on the discriminant validity of the
MPVS. Future studies employing the MPVS should seek to include measures that examine
discriminant validity.
Overall, research on the factor structure of the MPVS supported the original 4 factor
structure reported by Mynard and Joseph (2000). The only study that did not adequately
support the original four-factor was conducted by Balogun and Olapegba (2007). Although a
four-factor solution emerged and there was a degree of agreement with respect to item
loadings on some of the factors, the resulting factor structure was not similar enough to the
original to be considered comparable. Nevertheless, these divergent results from Balogun and
Olapegba (2007) may be attributed to a number of variables, particularly cultural differences
since this study was conducted with a Nigerian population. Other evidence from this review
also indicated that cultural factors may play a role in the pattern and extent of bullying
reported: the studies with three highest average scores were conducted with Nigerian
participants (Balogun & Olapegba, 2007, Balogun et al., 2006, and Popoola, 2005).
Together, these findings concerning the psychometric properties of the MPVS
demonstrate the reliability and validity of this scale. The factor analytic studies supported the
division of peer victimization into distinct but related subtypes, strengthening the argument
that peer victimization is best characterised as a multidimensional rather than singular
construct. This review has also reported on the reliability of the MPVS across a range of
samples including school children, university students and adult populations, demonstrating
that the MPVS can be used with a wide variety of age groups in a range of settings.
4.2 Sex Differences
Results from this review revealed consistent findings regarding sex differences in peer
victimization across numerous studies. Overall, males reported significantly more
victimization than females. Findings for the victimization subscales showed that direct forms
of victimization, namely physical victimization and attacks on property, were more likely to
be experienced by boys, while indirect victimization, particularly social manipulation, was
more likely to be experienced by girls. This pattern of victimization by gender replicates both
that reported in the original MPVS study (Mynard & Joseph, 2000), and that reported in the
wider literature(Andreou & Metallidou, 2004; Bjorkqvist et al., 1992; Crick & Grotpeter,
1995; Olweus, 1993; Smith et al., 2002). These relatively stable gender differences in peer
victimization have implications for bullying interventions. They suggest that schools could
tackle bullying most effectively by tailoring intervention programs in a way that targets
specific gender-related behaviours and victimization experiences. However, these now need
to be conducted in such a way that recognises greater diversity and fluidity in constructions
of gender than in previous research.
4.3 Future Research Recommendations
There are three broad areas for future development that we wish to highlight. First,
notably absent in the reviewed literature were studies testing the MPVS longitudinally. This
*
finding mirrors that of the wider bullying literature, which is largely cross-sectional and
presents simple associations between peer victimization and various outcomes. Longitudinal
studies would allow examination of the MPVS as a predictive measure, particularly with
respect to its efficacy in predicting future behaviours such as aggression, intimacy and self-
esteem. Longitudinal studies would enable examination of how victimization is related to
subsequent adjustment and how patterns and rates of victimization unfold over time,
particularly across the transition from primary to secondary school and from childhood
through puberty and into late adolescence.
Second, also notably absent was the use of the MPVS as a tool to evaluate
interventions. The prevention of bullying is becoming more of a priority among educators
given its widespread short- and long-term deleterious effects (Crothers, Kolbert & Barker,
2006). Numerous intervention and prevention programs have been suggested, including
interventions focused on the victim (such as counselling or conflict resolution, social skills
and assertiveness training); interventions focused on teachers and other adults (such as
encouraging teachers to identify and discipline bullies, and including parents in this process);
interventions focused on peers (including teaching bystanders to intervene and peer support
methods such as befriending), and interventions focused on the whole school community
(including workshops designed to modify the overall culture and climate of the school, and
integrating anti-bullying messages within the curriculum). The MPVS provides a suitable
outcome measure to test the efficacy of these types of interventions.
Third, one final issue that became apparent when conducting this review was the
number of studies that had used a different scoring system to that recommended in the
original validation study (Mynard & Joseph, 2000), or made other amendments. Adopting
alternative scoring systems compromises our ability to compare prevalence rates across
studies and in this instance, precluded their inclusion in this review. A total of 10 papers were
$
excluded for this reason alone and it is possible that these excluded papers may have
contributed relevant information concerning the psychometric properties of the MPVS had
they used the original rating scale. Our review also noted that since the development of the
MPVS there had been interest in social rebuff and electronic victimization as additional forms
of peer-victimization and it may be that in some contexts researchers will also wish to include
additional items for both of these dimensions. As such, researchers are encouraged to use this
full 24-item version (See Appendix).
4.4 Conclusions
The purpose of this paper was to review the growing literature pertaining to the
psychometric properties of the MPVS. Through a synthesis of research findings, the current
review establishes the MPVS as a reliable, valid, and psychometrically sound tool for
capturing multiple facets of peer victimization across a variety of samples, including primary
school and secondary school age children, as well as university students and adults. This
exhaustive review has also demonstrated the importance of assessing subtypes of
victimization and has highlighted new ways to further refine and advance the assessment of
peer victimization.
References
* Akram, B., & Munawar, A. (2016). Bullying victimization: a risk factor of health problems
among adolescents with hearing impairment. JPMA: Journal of the Pakistan
Medical Association, 66(1), 13-17.
* Anderson, C. G., Rawana, E. P., Brownlee, K., & Whitley, J. (2010). An investigation of the
relationship between psychological strengths and the perception of bullying in
early adolescents in schools. Alberta Journal of Educational Research, 56(4),
470.
Andreou, E., & Metallidou, P. (2004). The relationship of academic and social cognition to
behaviour in bullying situations among Greek primary school
children. Educational Psychology, 24(1), 27-41.
* Andreou, E., Vlachou, A., & Didaskalou, E. (2005). The roles of self-efficacy, peer
interactions and attitudes in bully-victim incidents: Implications for intervention
policy-practices. School Psychology International, 26(5), 545-562.
* Azeredo, C. M., Santos, I. S., Barros, A. J., Barros, F. C., & Matijasevich, A. (2017).
Maternal depression and bullying victimization among adolescents: results from
the 2004 Pelotas Cohort Study. Depression and Anxiety, 34(10), 897-907.
* Balogun, S. K., & Olapegba, P. O. (2007). Cultural validation of the multidimensional peer
victimization scale in Nigerian children. Journal of Cross-Cultural
Psychology, 38(5), 573-580.
* Balogun, S. K., Olapegba, P. O., & Opayemi, A. S. (2006). Influence of Gender, Age,
Religion, and Ethnicity on Peer-Victimization among Primary Four Pupils in
Ibadan, Nigeria. Studies of Tribes and Tribals, 4(2), 109-112.
Berne, S., Frisén, A., Schultze-Krumbholz, A., Scheithauer, H., Naruskov, K., Luik, P., &
Zukauskiene, R. (2013). Cyberbullying assessment instruments: A systematic
review. Aggression and Violent Behavior, 18(2), 320-334.
* Betts, L. R., Houston, J. E., & Steer, O. L. (2015). Development of the multidimensional
peer victimization scale–revised (MPVS-R) and the multidimensional peer
bullying scale (MPVS-RB). The Journal of Genetic Psychology, 176(2), 93-109.
* Betts, L. R., Houston, J. E., Steer, O. L., & Gardner, S. E. (2017). Adolescents’ experiences
of victimization: The role of attribution style and generalized trust. Journal of
School Violence, 16(1), 25-48.
* Betts, L. R., & Spenser, K. A. (2017). Developing the cyber victimization experiences and
cyberbullying behaviors scales. The Journal of Genetic Psychology, 1-18.
* Biebl, S. J., DiLalla, L. F., Davis, E. K., Lynch, K. A., & Shinn, S. O. (2011). Longitudinal
associations among peer victimization and physical and mental health
problems. Journal of Pediatric Psychology, 36(8), 868-877.
* Bird, J. C., Waite, F., Rowsell, E., Fergusson, E. C., & Freeman, D. (2017). Cognitive,
affective, and social factors maintaining paranoia in adolescents with mental
health problems: A longitudinal study. Psychiatry Research, 257, 34-39.
Björkqvist, K., Lagerspetz, K., & Kaukiainen, A. (1992). Do girls manipulate and boys fight?
Developmental trends in regard to direct and indirect aggression. Aggressive
Behavior, 18, 117-127.
Campell, D. T., & Fiske, D. W. (1959). Convergent and discriminant validation by the
multitrait-multimethod matrix. Psychological Bulletin, 56, 81-105.
* Candel, O. S., & Iacob, L. M. (2015). Rumination, co-rumination and peer victimization in
Romanian adolescents. Annals of AII Cuza University. Psychology Series, 24(2),
25.
* Cosgrove, H. E., Nickerson, A. B., & DeLucia, J. (2017). Past Peer Victimization and
Current Adult Attachment in College Students. Journal of College
Counseling, 20(1), 22-36.
Craig, W., Harel-Fisch, Y., Fogel-Grinvald, H., Dostaler, S., Hetland, J., Simons-Morton, B.,
et al. (2009). A cross-national profile of bullying and victimization among
adolescents in 40 countries. International Journal of Public Health, 54, 216–224.
Crick, N. R., & Grotpeter, J. K. (1995). Relational aggression, gender, and social‐
psychological adjustment. Child Development, 66(3), 710-722.
Crothers, L. M., Kolbert, J. B., & Barker, W. F. (2006). Middle school students’ preferences
for anti-bullying interventions. School Psychology International, 27(4), 475-487.
Crothers, L. M., & Levinson, E. M. (2004). Assessment of bullying: a review of
methods. Journal of Counselling and Development, 82, 496-503.
Currie, C., Zanotti, C., Morgan, A., Currie, D., DeLooze, M., Roberts, C., & Barnekow, V.
(2012). Social determinants of health and well-being among young people.
Health Behaviour in School-aged Children (HBSC) study: International report
from the 2009/2010 survey. Health Policy for Children and Adolescents, No. 6.
Copenhagen, Denmark: WHO Regional Office for Europe.
* Defeyter, M. A., Graham, P. L., & Russo, R. (2015). More than just a meal: breakfast club
attendance and children’s social relationships. Frontiers in Public Health, 3.
Fleming, L. C., & Jacobsen, K. H. (2009). Bullying among middle-school students in low and
middle income countries. Health Promotion International, 25(1), 73-84.
* Fontaine, N. M., Hanscombe, K. B., Berg, M. T., McCrory, E. J., & Viding, E. (2018).
Trajectories of callous-unemotional traits in childhood predict different forms of
peer victimization in adolescence. Journal of Clinical Child & Adolescent
Psychology, 47(3), 458-466.
* Fung, A. L. C., & Raine, A. (2012). Peer victimization as a risk factor for schizotypal
personality in childhood and adolescence. Journal of Personality
Disorders, 26(3), 428-434.
Gini, G., & Pozzoli, T. (2009). Association between bullying and psychosomatic problems: a
meta-analysis. Pediatrics. 123, 1059-1065.
Gladstone, G. L., Parker, G. B. & Malhi, G. S. (2006). Do bullied children become anxious
and depressed adults? A cross-sectional investigation of the correlates of bullying
and anxious depression. Journal of Nervous and Mental Disease, 194, 201-208.
Haddaway, N. R., Collins, A. M., Coughlin, D. & Kirk, S. (2015). The role of Google Scholar
in evidence reviews and its applicability to grey literature searching. PLoS ONE,
10, e0138237. https://doi.org/10.1371/journal.pone.0138237
Hawker, D. S. J. & Boulton, M. J. (2000). Twenty years’ research on peer victimization and
psychosocial maladjustment: a meta-analytic review of cross-sectional studies.
Journal of Child Psychology and Psychiatry, 41, 441-455.
* Kaiser, A., & Malik, S. (2015). Peer Victimization and Psychiatric Symptoms among
Adolescents. Pakistan Journal of Medical Research, 54(4), 113.
* Lam, B. Y., Raine, A., & Lee, T. M. (2016). Effect of theory of mind and peer victimization
on the schizotypy–aggression relationship. NPJ Schizophrenia, 2, 16001.
* Law, A. K. Y., & Fung, A. L. C. (2013). Different forms of online and face-to-face
victimization among schoolchildren with pure and co-occurring dimensions of
reactive and proactive aggression. Computers in Human Behavior, 29(3), 1224-
1233.
* Lee, J., Abell, N., & Holmes, J. L. (2017). Validation of measures of cyberbullying
perpetration and victimization in emerging adulthood. Research on Social Work
Practice, 27(4), 456-467.
* Litman, L., Costantino, G., Waxman, R., Sanabria‐Velez, C., Rodriguez‐Guzman, V. M.,
Lampon‐Velez, A., & Cruz, T. (2015). Relationship between peer victimization
and posttraumatic stress among primary school children. Journal of Traumatic
Stress, 28(4), 348-354.
McDougall, P., & Vaillancourt, T. (2015). Long-term adult outcomes of peer victimization in
childhood and adolescence. American Psychologist, 70, 300-310.
* McFarlane, J., Karmaliani, R., Khuwaja, H. M. A., Gulzar, S., Somani, R., Ali, T. S., &
Muhammad, A. (2017). Preventing peer violence against children: methods and
baseline data of a cluster randomized controlled trial in Pakistan. Global Health:
Science and Practice, 5(1), 115-137.
* McGuire, J., Barbanel, L., Brüne, M., & Langdon, R. (2015). Re-examining Kohlberg's
conception of morality in schizophrenia. Cognitive neuropsychiatry, 20(5), 377-
381.
* Morrow, M. T., Hubbard, J. A., Barhight, L. J., & Thomson, A. K. (2014). Fifth-grade
children’s daily experiences of peer victimization and negative emotions:
Moderating effects of sex and peer rejection. Journal of Abnormal Child
Psychology, 42(7), 1089-1102.
* Morrow, M. T., Hubbard, J. A., & Swift, L. E. (2014). Relations among multiple types of
peer victimization, reactivity to peer victimization, and academic achievement in
fifth-grade boys and girls. Merrill-Palmer Quarterly, 60(3), 302-327.
* Murphy, S., Murphy, J., & Shevlin, M. (2015). Negative evaluations of self and others, and
peer victimization as mediators of the relationship between childhood adversity
and psychotic experiences in adolescence: the moderating role of
loneliness. British Journal of Clinical Psychology, 54(3), 326-344.
Mynard, H., & Joseph, S. (2000). Development of the Multidimensional Peer Victimization
Scale. Aggressive Behaviour, 26, 169-178.
* Mynard, H., Joseph, S., & Alexander, J. (2000). Peer-victimisation and posttraumatic stress
in adolescents. Personality and Individual Differences, 29(5), 815-821.
Nakamoto, J., & Schwartz, D. (2009). Is peer victimization associated with academic
achievement? A meta-analytic review. Social Development, 19, 221-242.
Olweus, D. (1993). Victimization by peers: Antecedents and long term outcomes. In K. H.
Rubin & J. B. Asendorpf (Eds.), Social withdrawal, inhibition, and shyness in
childhood (pp. 315–341). London: Psychology Press.
* Piek, J. P., Barrett, N. C., Allen, L. S. R., Jones, A., & Louise, M. (2005). The relationship
between bullying and self‐worth in children with movement coordination
problems. British Journal of Educational Psychology, 75(3), 453-463.
* Popoola, B. I. (2005). Prevalence of Peer Victimisation among Secondary School Students
in Nigeria. International Education Journal, 6(5), 598-606.
* Raine, A., Fung, A. L. C., & Lam, B. Y. H. (2011). Peer victimization partially mediates the
schizotypy-aggression relationship in children and adolescents. Schizophrenia
Bulletin, 37(5), 937-945.
* Rao, P. C. S., & Kishore, M. T. (2013). Effect of obesity in self-esteem, peer victimization
and behavior problems in adolescents. Indian Journal of Clinical
Psychology, 40(2), 137-141.
Reijntjes, A., Kamphuis, J. H., Prinzie, P., Boelen, P. A., van der Schoot, M., & Telch, M. J.
(2011). Prospective linkages between peer victimization and externalizing
problems in children: A meta-analysis. Aggressive Behavior, 37, 215–222.
http://dx.doi.org/10.1002/ab.20374
* Scarpa, S., Carraro, A., Gobbi, E., & Nart, A. (2012). Peer-victimization during physical
education and enjoyment of physical activity. Perceptual and Motor
Skills, 115(1), 319-324.
* Shakoor, S., McGuire, P., Cardno, A. G., Freeman, D., Plomin, R., & Ronald, A. (2014). A
shared genetic propensity underlies experiences of bullying victimization in late
childhood and self-rated paranoid thinking in adolescence. Schizophrenia
Bulletin, 41(3), 754-763.
Smith, P. K., Cowie, H., Olafsson, R., & Liefooghe, A.P.D. (2002). Definitions of bullying: A
comparison of terms used, and age and sex differences, in a 14-country
international comparison. Child Development, 73, 1119–1133.
Solberg, M. E., & Olweus, D. (2003). Prevalence estimation of school bullying with the
Olweus Bully/Victim Questionnaire. Aggressive Behavior, 29(3), 239-268.
Storch, E. A., & Masia-Warner, C. (2004). The relationship of peer victimization to social
anxiety and loneliness in adolescent females. Journal of Adolescence, 27(3), 351-
362. doi: 10.1016/j.adolescence.2004.03.003
Streiner, D. L. (2003). Starting at the beginning: An introduction to coefficient alpha and
internal consistency. Journal of Personality Assessment, 80, 99-103.
van Geel, M., Vedder, P., & Tanilon, J. (2014). Relationship between peer victimization,
cyberbullying, and suicide in children and adolescents: A meta-analysis. JAMA
Pediatrics, 168, 435–442.
Vessey, J., Strout, T. D., DiFazio, R. L., & Walker, A. (2014). Measuring the youth bullying
experience: A systematic review of the psychometric properties of available
instruments. Journal of School Health, 84(12), 819-843.
Vivolo-Kantor, A. M., Martell, B. N., Holland, K. M., & Westby, R. (2014). A systematic
review and content analysis of bullying and cyber-bullying measurement
strategies. Aggression and Violent Behavior, 19(4), 423-434.
* Waytowich, V. L., Onwuegbuzie, A. J., & Elbedour, S. (2011). Violence and attribution
error in adolescent male and female delinquents. International Journal of
Education, 3(1), 1.
References marked with an asterisk (*) are included in the review.
Note: Fontaine et al (2018) was available for inclusion in the review as an online first
publication in 2016.
Appendix 1. The Multidimensional Peer Victimization Scale – 24 (MPVS-24).
Subtype Number Item
Physical Victimization 1 Punched me
5 Kicked me
9 Hurt me physically in some way
13 Beat me up
Verbal Victimization 3 Called me names
7 Made fun of me because of my appearance
11 Made fun of me for some reason
15 Swore at me
Social Manipulation 2 Tried to get me into trouble with my friends
6 Tried to make my friends turn against me
10 When I tried to play with one person, another person would
not let me
14 Made other people not talk to me
Attacks on Property 4 Took something of mine without permission
8 Tried to break something of mine
12 Stole something from me
16 Deliberately damaged some property of mine
Electronic Victimization217 Sent me a nasty text
19 Said something mean about me on a social networking site
21 Wrote spiteful things about me in a chatroom
23 Wrote nasty things to me using instant messenger
Social Rebuff 18 Ignored me
20 Refused to talk to me
22 Would not let me join in their game
24 Had a secret and would not tell me
NB: The first 16 items are the original MPVS and the final 8 items are new subscales adapted
from Betts et al (2015) and Morrow et al (2014), with the exception that item 10 is not an
original MPVS item, but was added to the Social Manipulation subscale by Morrow et al
(2014) to replace the original MPVS ‘Refused to talk to me’ item which they moved from the
Social Manipulation subscale to the Social Rebuff subscale, now here as item 20.
For a copy of the MPVS-24 see supplementary materials. The MPVS-24 is free to use with
permission from the author.
*
\
Supplementary Material:
Multidimensional Peer-Victimization Scale-24 (MPVS-24)
+& ,-&
%%./
&(%0 1,
2
3
4
, /
, 5 (& %
, 6
, 50 &
, 70
, 5 0 %
, 4 ( %
, 5(0
*, - %% &%
$,8'%&9
&
,4
,
$
,+
,40
,&
,:(% %
, %"
,'
*, ( &0
$,;0
,8(
,8 <
,8%
,-&
Scoring key for the MPVS-24:
2$
3
4
!($9(
$(,
Subscales
' ==*=%!
' ==$=
' ===!(! >
' ===0%
' =*==)! >
' =$==(?
Table 1 Summary of Published Studies Using the Multidimensional Peer-Victimisation Scale
Study Sample Type Age N M(SD) Sex difference Reliability Correlation with outcome variables Longitudinal Results Comments
Akram & Munawar
(2016)
Adolescents with
hearing impairment
attending 2 large
schools in Gujrat
district of Pakistan’s
Punjab province. 64%
boys
12-15 years 286 - Boys experienced
more physical
victimisation than
girls (p < .05), but
there was no
significant
difference between
girls and boys in
social manipulation
(p > .05).
α not reported but
subscale inter-
correlations ranged
from r = .38 (for
physical and social) to
r = . 56 (for physical
and verbal)
All four subtypes were correlated with
physical health problems (r’s = .36 to .
41) and psychological health problems
(r’s = .35 to .42).
Multiple regression analyses showed
peer victimisation was a risk factor for
physical health problems such as
headache, abdominal pain, cough, cold,
skin problems and nausea; as well as
being positive and significant predictors
of psychological problems such as
disturbed appetite, nightmares, bed
wetting and worrying about going to
school.
- MPVS was
translated into Urdu
using lexicon
equivalence method
of translation
(translation detail
provided in paper).
Anderson, Rawana,
Brownlee & Whitley
(2010)
7th and 8th grade
students attending
public schools in a
small urban city in
North-western
Ontario.
Boys mean
age =
12.96(.74)
and girls
mean age =
12.92(.68)
85 - A sex difference was
found for physical
victimisation: boys
emerged as
significantly more
likely to be
physically
victimized than
girls, t(76.87)=–
Not reported but all
subtypes of
victimisation were
positively correlated;
lowest was between
physical and verbal
victimisation r = .495
p < .01, and strongest
was between verbal
- -
1.404, p<.01. and social
manipulation r = .629
p < .01
Study Sample Type Age N M(SD) Sex difference Reliability Correlation with outcome variables Longitudinal Results Comments
Andreou, Vlachou
& Didaskalou
(2005)
Primary education
pupils drawn from 10
primary schools in
central Greece
Age range
9-12 years
(M = 10.21;
SD = 0.86)
448 Physical
2.29(1.9)
Verbal
3.09(2.29)
Social
2.77(2.36)
Attack
2.33(2.20)
Boys scored
significantly higher
than girls on
Physical and Verbal
Victimization and
significantly lower
on Social
Manipulation.
Alphas range from .67
to .85 for four
subscales
Children in 6th grade had experienced
significant less attacks on property than
had children in 4th grade (F = 3.15, p > .
05). No significant age difference was
observed for any other subscale.
Total peer victimisation scores were
negatively associated with positive
interactions with peers (r = -.21, p < .01)
-
Azeredo, Santos,
Barros, Barros &
Matijasevich (2017)
Participants were part
of the Pelotas Cohort
Study (Santos et al.,
2014), a study of
mothers and infants in
Pelotas, Brazil
M = 11.0
years, SD =
0.3 years
3841 Mean scores
not reported
but verbal
victimisation
was the most
prevalent type
of bullying
(37.9%)
Males reported
significantly more
victimisation than
females
- Severe current maternal depression was
significantly associated with physical
victimisation, social manipulation and
attacks on property in their 11 year old
offspring.
Maternal mood
symptoms during
pregnancy were
significantly
associated with
physical and verbal
victimisation in their
11 year old offspring.
Balogun &
Olapegba (2007)
Grade 4 pupils
attending primary
schools in Ibadan,
Nigeria.
Age range
7-12 years
(M = 8.90;
SD = .94)
240 Total for boys
M = 16.21;
SD = 6.85
Total for girls
No significant
difference by gender
for total or subscales
except Physical
α = .78
Split half reliability
of .76
Concurrent validity test with the Buss &
Durkee (1975) Aggression Scale yielded
a correlation of .54
- This study
attempted cultural
validation of the
MPVS with
M = 15.7 SD
= 6.36.
Victimisation, which
is significantly
higher for boys
Nigerian children.
Item 3 (“called me
names”) was
slightly modified to
“Abused and called
me bad/ugly names”
so as to be culturally
relevant. All other
items remained the
same.
Study Sample Type Age N M(SD) Sex difference Reliability Correlation with outcome variables Longitudinal Results Comments
Balogun, Olapegba
& Opayemi (2006)
Primary school pupils
from Ibadan
metropolis in Nigeria.
Age range
7 -12 years;
(M = 8.9;
SD = 0.94)
240 For boys: M =
16.21 (6.85)
For girls: M =
15.7 (6.36)
No significant
gender differences
for total score but
boys experienced
more attacks on
property than girls
(3.44 vs 2.89; t =
2.38; df = 238, p < .
05)
- Religion and ethnicity were found not to
have any significant effect on peer-
victimization f(2, 237) = 0.93 p >.05,
f(3, 239) = 0.47 p > .05. No significant
difference for age on total score, but
children aged 9 or over experienced
more social manipulation than children
aged below 9. No age differences for the
other subscales.
-
Betts, Houston &
Steer (2015)
Students attending
urban secondary
schools in a city in the
East Midlands of the
UK
Age range
11-15 years
(M = 13.4;
SD = 1.2)
371 - Boys reported
experiencing higher
levels of physical
victimization and
greater attacks on
property than girls,
whereas girls
reported
Physical α = .91
Social α = .87
Verbal α = .84
Property α = .90
Electronic α = .91
Subscale inter-
correlations ranged
from .37 to .60
All subscales showed significant
negative correlation with self-esteem (r
ranged from -.18 to -.33).
-Created the MPVS-
R by using the
MPVS alongside an
additional 4 items to
assess electronic
victimisation
experiencing greater
levels of social and
electronic
victimization than
boys.
Study Sample Type Age N M(SD) Sex difference Reliability Correlation with outcome variables Longitudinal Results Comments
Betts, Houston,
Steer & Gardner
(2017)
Students attending two
urban secondary
schools in a city in the
East Midlands of the
UK
Age range
11-15 years
(M = 13
years 4
months, SD
= 1 year 2
months)
280 - - Physical α = .78
Verbal α = .78
Social α = .81
Attacks α = .79
Electronic α = .81
Used multi-group path analysis. For
males, more frequent attacks on
property predicted higher levels of
loneliness and depressive symptoms and
lower levels of social confidence.
Higher levels of verbal victimisation
predicted lower global self-worth and
higher levels of loneliness. For females,
the only significant path showed that
higher levels of verbal victimisation
predicted lower levels of global self-
worth.
Used the 20-item
MPVS-R which is
the MPVS plus 4
items assessing
electronic
victimisation (see
Betts, Houston &
Steer, 2015 above).
Betts & Spenser
(2017) – Study 1
Students attending a
secondary school in
Age range
11-15 years
393 - - Alpha’s ranged from .
62 to .86
All four victimisation subtypes were
positively correlated with all three
- Study aimed to
develop a measure
Study 2
the East Midlands of
the UK.
Students attending a
(different) secondary
school in the East
Midlands of the UK
(M = 12.81,
SD = 1.32)
Age range
11-15 years
(M = 12.12;
SD = 0.98)
345 - - Alpha’s ranged from .
60 to .88
subtypes of cyber victimisation: for
Threats r’s = .21 to .36; for Sharing
Images r’s = .23 to .49; for Personal
Attack r’s = .21 to .55; all p’s < .001
All four victimisation subtypes were
positively correlated with all three
subtypes of cyber victimisation: for
Threats r’s = .29 to .42; for Sharing
Images r’s = .27 to .36; for Personal
Attack r’s = .38 to .62; all p’s < .001
-
of Cyber bullying
and cyber
victimisation and
used the MPVS to
examine convergent
validity.
Study Sample Type Age N M(SD) Sex difference Reliability Correlation with outcome variables Longitudinal Results Comments
Biebl, DiLalla,
Davis, Lynch &
Shinn (2011)
Participants were a
subset of youth who
participated in the
longitudinal Southern
Illinois Twins and
Siblings Study
(SITSS; DiLalla,
2002).
T1 M =
5.00; SD =
0.00.
T2 age
range 10-18
years, M =
14.00; SD =
2.52.
T3 age
range 12-20
years, M =
16.24; SD =
T1: 283
T2: 85
T3: 70
- - Inter-rater reliability at
T1 ranged from .80
to .84.
At T2, physical
victimisation α = .89;
verbal victimisation α
= .74; social
manipulation α = .82;
and attacks on
property α = .77.
At T3, Relational
victimisation α = .89;
- - At T1 when
participants were
aged 5, a modified
version of the
MPSV was used to
create a coding
scheme for use
during a 20 minute
play session. At T2
the full MPVS was
used. At T3 a
slightly amended
2.61. Physical victimisation
α = .83; and Overall
victimisation α = .89
version of the
MPVS was used
(minor adjustments
to rating scale and
time frame reported
on)
Bird, Waite,
Rowsell, Fergussen
& Freeman (2017)
Clinical sample of
adolescents seeking
treatment for paranoid
ideation. 82% female
sample.
Age range
11-16 years
(M = 14.9;
SD = 1.25)
34 M = 16.0
SD = 8.60
- - MPVS total was significantly positively
correlated with paranoia at baseline (r
= .56; p < .001).
The partial correlation
between baseline
MPVS and paranoia at
3 month follow-up,
controlling for
baseline paranoia,
approached
significance (r = .33, p
= .06).
Study Sample Type Age N M(SD) Sex difference Reliability Correlation with outcome variables Longitudinal Results Comments
Candel & Iacob
(2015)
121 Romanian
students aged between
11 and 13 years old;
and 95 students aged
between 17 and 19
years old.
11-13 years
and 17-19
years
216 M = 9.44
SD = 7.07
Total α = .96 MPVS total was significantly correlated
with age (r = -.30; p < .01) and
rumination (r = .16; p < .05). High
ruminators reported significantly more
peer victimisation than low ruminators;
t = -2.24; p = 0.02. Participants aged
between 11 and 13 years reported
significantly more peer victimisation
than participants aged between 17 and
*
19 years; t = 4.67; p < .001.
Cosgrove,
Nickerson &
DeLucia (2017)
Undergraduate and
graduate students
attending 2
universities in the
North-eastern US.
Sample was 77.7%
female
Age range
18-60 years
(M = 22.14
SD = 5.57)
386 Only reported
for physical
and attacks on
property
subscales. For
men: Physical
M = 0.60
(0.68);
Attacks M =
0.84 (0.73)
For women:
Physical M =
0.18 (0.39);
Attacks M =
0.54 (0.57).
Men experienced
more frequent
physical and attacks
on property
victimisations than
women: Physical
victimisation F(1,
385) = 50.51, p < .
001, partial n2 = .12
and Attacks on
Property F(1, 385) =
16.60, p < .001,
partial n2 = .04.
Total α = .89
Inter-correlations
between subscales
ranged from .24 (for
physical and social
manipulation) and .56
(for social
manipulation and
verbal)
MPVS and attachment quality (Revised
Adult Attachment Scale; RAAS) r = .37
p < .01
No significant correlation between
MPVS and number of current
friendships r = -.09, p > .05
Previous verbal victimisation was the
most significant predictor of poor
attachment quality during young
adulthood (β = .19), t(355) = 3.12, p < .
01. It was also found that previous
relational victimisation significantly
predicted less stable attachments above
physical or property damage
victimisation (β = .16, t(355) = 2.53 p
< .05).
- Because this study
was primarily
concerned with
investigating
recalled experiences
of peer
victimisation,
instructions were
modified to
encourage
participants to think
back to their
experiences in
elementary, middle
and high school
rather than their
current experiences.
Study Sample Type Age N M(SD) Sex difference Reliability Correlation with outcome variables Longitudinal Results Comments
Defeyter, Graham &
Russo (2015)
Participants were
recruited from 8 inner-
city mixed-gender
primary schools in the
UK.
Age range
5.3-10.11
years
(M = 8.4;
SD = 1.69)
268 - - - Children attending Breakfast Club (BC)
and After School Club (ASC) reported
lower levels of physical victimisation
than students attending no clubs. In
addition, a reduction in social
victimisation and attacks on property
Overall, levels of
physical, verbal and
social victimisation
decreased over time,
while the level of
attacks on property
To check that
children understood
the questions and to
make sure that
incidents were not
just examples of
$
was observed in children attending BC
and ASC
remained constant. rough and tumble
play, children were
asked to provide
examples to each
question.
Fontaine,
Hanscombe, Berg,
McCrory & Viding
(2018)
Participants were
drawn from a larger
sample of 9,462
families enrolled in the
Twins Early
Development Study
(TEDS). For this
study, data from
assessments conducted
at age 7, 12 and 14
years were analysed.
7, 12 and
14 years
4156 Physical M =
0.76 (1.29)
Verbal M =
2.13 (1.74)
Social M =
1.45 (1.61)
Attacks M =
0.99 (1.35)
Compared with
girls, boys had
higher mean levels
of physical
victimization, verbal
victimization, and
attacks on property,
whereas girls had
higher mean levels
of social
manipulation.
Physical α = .80
Verbal α = .84
Social α = .82
Attacks α = .83
Subscale inter-
correlations ranged
from .40 (p < .001) for
physical and social to .
62 (p < .001) for
verbal and social.
Youths on the stable high trajectory had
the highest levels of all forms of peer
victimization while youths on the stable
low trajectory reported the lowest levels
of all forms of victimisation.
All four subtypes of victimisation were
positively correlated with conduct
problems, emotional problems and
negative parental discipline.
- Callous-
Unemotional (CU)
traits were assessed
at 7, 9 and 12 years
old. Four trajectories
of CU traits were
identified: Stable
High (CU traits
remained high
between 7 and 12
years); Increasing
(CU traits increased
from 7 to 12 years);
Decreasing (CU
traits decreased from
7 to 12 years) and
Stable Low (CU
traits remained low
between 7 and 12
years)
Study Sample Type Age N M(SD) Sex difference Reliability Correlation with outcome variables Longitudinal Results Comments
Fung & Raine
(2012)
Participants were
drawn from 10
Age range
9-15 years
3508 Physical M =
1.57 (3.3)
Physical α = .87
Verbal α = .78
MPVS total was significantly correlated
with SPQ-C total r = .39; p < .001. All
- SPQ-C (Schizotypal
Personality
primary and 10
secondary schools in
Hong Kong.
(male mean
= 11.76
(1.84);
female
mean =
12.04
(1.75)
Verbal M =
3.56 (4.23)
Social M =
2.02 (3.58)
Attacks M =
1.67 (2.86)
Total M =
8.82 (11.2)
Social α = .85
Attacks α = .73
Total α = .90
Subscale inter-
correlations ranged
from .43 (p < .001) for
physical and social to .
57 (p < .001) for
verbal and attacks on
property.
MPVS subscales were significantly
positively correlated with all SPQ-C
subscales (r’s = .20 to .31; p’s < .001).
Children in the high victimisation group
(scoring 1SD above MPVS mean)
scored significantly higher on the SPQ-
C total and all subscales than children in
the low victimisation group (scoring
1SD below MPVS mean).
Questionnaire –
Child) is a measure
of schizotypal
personality adapted
for use with
children.
Kaiser & Malik
(2015)
Participants were
recruited from schools
and colleges in
Sargodha city,
Pakistan.
Age range
14-18 years
(M = 16.14)
400 Physical M =
2.68 (2.32)
Verbal M =
3.25 (2.75)
Social M =
3.02 (2.63)
Attacks M =
3.09 (2.51)
Male adolescents
reported
significantly more
peer victimisation
than females,
scoring significantly
higher on all four
subscales.
Physical α = .62
Verbal α = .65
Social α = .73
Attacks α = .65
Subscale inter-
correlations ranged
from .54 (p < .001) for
physical and verbal
to .65 (p < .001) for
verbal and social.
All four subtypes of victimisation
showed positive correlated with
depression, anxiety and stress. Multiple
regression analyses showed that all
components of peer victimisation
positively predicted anxiety (22% of
variance), depression (19% of variance)
and stress (17% of variance).
Lam, Raine & Lee
(2016)
Bilingual
undergraduate students
recruited in Hong
Kong. 68.6% female
sample.
Age range
18-25 years
(M = 18.92;
SD = 1.16).
237 Total MPVS
M = 3.41 (SD
= 3.51)
Males experienced
significantly more
victimisation than
females (p < .05).
Total scale α = .74 Peer victimisation was positively
correlated with Schizotypy (r = .29),
General aggression (r = .42), Reactive
aggression (r = .38) and Proactive
aggression (r = .33) (all p’s < .001).
Age was positively associated with
victimisation (p < .05)
- MPVS was
translated and back
translated from
English to Chinese.
Study Sample Type Age N M(SD) Sex difference Reliability Correlation with outcome variables Longitudinal Results Comments
Law & Fung (2013) Schoolchildren
recruited from four
middle schools and
one elementary school
located in wide-
ranging areas of Hong
Kong. Sample was
60.6% male.
Age range
9-20 years,
(M = 13.91;
SD = 2.52)
1122 - - Physical α = .89
Verbal α = .82
Social α = .89
Property α = .82
(although not entirely
clear from the paper
whether these were
based on study sample
or are just reporting
previously established
reliability alphas)
The MPVS total and subscale scores
were all significantly higher for children
who were categorised as proactive
aggressors, reactive aggressors or co-
occurring aggressors than for non-
aggressive school children.
MPVS was significantly correlated with
OVS (online victimisation scale) r = .
311 p <.001
- MPVS was put
through thorough
back translations to
arrive at Chinese
version.
Lee, Abell &
Holmes (2015)
Undergraduate
students enrolled in
social science
disciplines at a large
public university in the
south eastern US.
Sample was 61.9%
female.
Age range
18-25 years
(M = 20.92,
SD = 1.54)
286 - - Physical α = .81
Verbal α = .79
Social α = .76
Property α = .77
The MPVS was positively correlated
with the CBV global and subscales (r = .
31 for the global, r = .30 for
verbal/written victimization, r = .28 for
visual/sexual victimization, and r = .21
for social exclusion victimization).
Effect sizes were generally small,
ranging from .04 to .10.
- Study reports on the
development and
validation of 2 new
cyberbullying
scales:
Cyberbullying
Perpetration (CBP)
and Cyberbullying
Victimisation
(CBV). The MPVS
was used to test the
construct convergent
validity of the CBV.
Study Sample Type Age N M(SD) Sex difference Reliability Correlation with outcome variables Longitudinal Results Comments
Litman, Costantino,
Waxman, Sanabria@
Velez, Rodriguez@
Guzman, Lampon@
Velez & Cruz (2015)
Hispanic/Latino
children from three
public schools in New
York City.
Age range
6-11 years,
(M = 8.51,
SD = 1.23)
358 Percentage
reporting
having
experienced
at least one
victimisation
event more
than once
during the
school year:
physical
22.8%; verbal
38.1%, social
38.7%,
attacks on
property
36.8%.
Boys were more
likely to be
victimised than
girls. Physical
victimisation and
attacks on property
higher for boys than
girls. See paper for
detailed breakdown
of means for each
subscale presented
by gender.
Physical α = .73
Verbal α = .77
Social α = .71
Property α = .76
Subscale inter-
correlations ranged
from .54 to .67
Correlation between MPVS total and
PTSD symptoms for boys: r = .33 p < .
001 and for girls: r = .29 p < .001 (see
paper for these correlations broken
down for each age group 7 to 10 years).
For boys, Attacks on Property most
strongly correlated with PTSD
symptoms (r = .36). For girls, Social
Manipulation most strongly correlated
with PTSD symptoms (r = .29).
- Note participants
were pre-screened
for trauma
experience using the
Child Trauma
Screening
Questionnaire
(CTSQ; Constantino
et al., 2014).
Also note that
assessments were
conducted by
bilingual coauthors
(English/Spanish
speaking) in face to
face sessions with
the children in the
school setting.
McFarlane,
Karmaliani,
Khuwaja, Gulzar,
Sumani, Ali, Sumani
et al. (2017)
6th grade students
attending single-
gender public schools
in Sindh province,
Pakistan
Age range
11-13
years, boys
M = 12.53
(0.06); girls
M = 12.16
(0.11)
1752 For boys M =
12.32 (0.50)
For girls M =
7.89 (0.47).
94% of boys
and 85% of
girls reported
Boys reported
significantly more
peer victimisation
than girls.
Study reports
frequencies and
percentages of every
Total α = .87
Physical α = .67
Verbal α = .64
Social α = .70
Attacks α = .66
No associations with outcomes reported
(even though they assessed depression)
- MPVS was forward
translated from
English into Urdu
and Sindhi.
Independent back-
translation was then
performed; any
one or more
episode of
victimisation
in the
preceding 4
weeks.
MPVS item by
gender.
discrepancies
between translators
were discussed and
resolved until
language agreement
was reached.
Study Sample Type Age N M(SD) Sex difference Reliability Correlation with outcome variables Longitudinal Results Comments
McGuire, Barbanel,
Brüne & Langdon
(2015)
24 participants with
Schizophrenia (M =
45.65 years; SD = 9.6)
and 20 control
participants (M =
38.60 years; SD =
14.7).
44 - - - Interpersonal conflict, as measured by
the MPVS, was not significantly
associated with scores on the Moral
Judgements Interview (MJI; an
assessment of ‘moral symptoms’ in
people with schizophrenia).
There were no significant differences in
MPVS scores for participants with and
without schizophrenia.
-
Morrow, Hubbard,
Barhight &
Thomson (2014)
Participants were
recruited from eight 5th
grade public schools in
Age range
10-11 years
181 - No significant sex
differences for any
type of victimisation
Physical α = .71
Verbal α = .84
Social α = .82
Peer rejection was significantly
positively correlated with verbal
victimisation (r = .16, p < .05).
Used the MPVS but
added 4 additional
items to capture
a Mid-Atlantic state.
were found. Attacks α = .78
Social rebuff α = .74
Subscale inter-
correlations ranged
from .18 for physical
victimisation and
social rebuff, to .64 for
social manipulation
and social rebuff.
Each peer victimisation variable
positively predicted each negative
emotion (sadness, anger, embarrassment
and nervousness). Results further
showed physical victimization positively
predicted all four negative emotions,
verbal victimization positively predicted
anger and embarrassment, and social
rebuff positively predicted nervousness.
Social Rebuff,
which refers to the
experience of being
ignored, left out or
excluded by peers
and is regarded as
distinct from social
manipulation.
Study Sample Type Age N M(SD) Sex difference Reliability Correlation with outcome variables Longitudinal Results Comments
Morrow, Hubbard &
Swift (2014)
Participants were
recruited from 5th
grade public schools
within one school
district in a Mid-
Atlantic state.
Age range
10-11 years
179 Verbal
victimisation
was most
frequent
(29%)
followed by
social rebuff
(22%)
There were no
significant sex
differences in rates
of victimisation.
Physical α = .85
Verbal α = .93
Social α = .93
Property α = .90
Social rebuff α = .84
Subscale inter-
correlations ranged
from .60 to .78
Social manipulation was negatively
correlated with academic achievement
(r = -.20, p < .01) but no other
victimisation subscales were.
Used the MPVS but
added 4 additional
items to capture
Social Rebuff (see
Morrow, Hubbard,
Barhight &
Thomson 2014
above).
Murphy, Murphy &
Shevlin (2015)
Recruited from 10
secondary schools in
N Ireland. 56.1%
female.
Age range
15-18 years
(M = 16.20,
SD = 1.06)
785 For total
score M =
10.35, SD =
7.80.
Subscales not
reported
α = .89. Subscales not
reported.
ELES (Early Life Experiences Scale;
assesses memories of familial threat and
subordination) r = .396
SCS (Social Comparison Scale; assesses
feelings of inferiority, incompetence and
being disliked) r = .265
PTCI (Posttraumatic Cognitions
Inventory; assesses negative cognitions
about self, world and self-blame) r = .
- -
445
APSS (Adolescent Psychotic-Like
Symptom Screener; assesses
hallucinatory and delusional
experiences) r = .380
UCLA Loneliness Scale r = .366.
T-tests showed participants who were
lonely reported significantly higher
MPVS scores (M = 16.61; SD = 8.37)
than participants who were not lonely
(M = 9.07; SD = 3.04)
Study Sample Type Age N M(SD) Sex difference Reliability Correlation with outcome variables Longitudinal Results Comments
Mynard, Joseph &
Alexander (2000)
Children and
adolescents in years 8
to 11 in secondary
schools in Essex, UK.
12-16 years 331 Physical M =
3.68 (2.83)
Verbal M =
5.47 (2.55)
Social M =
3.28 (2.60)
Attacks M =
2.78 (2.81)
- - MPVS total score was positively
associated with IES total (r = .24, p < .
02), but when examining MPVS
subscales only Social Manipulation was
significantly associated with IES.
MPVS total score was negatively
associated with Global Self-Worth (r =
-.27, p < .001). When examining the
subscales, only Verbal Victimisation was
significantly negatively associated with
Global Self-Worth.
-IES is the Impact of
Event Scale, a
measure of PTSD
Piek, Barrett, Allen,
Jones & Louise
(2005)
Children attending
primary schools in
Western Australia.
7-11 years 86 DCD (boys)
11.7(7.76)
DCD (girls)
There was a gender
effect for the social
manipulation
α for total score = .87;
for four subscales α
ranged from .66 to .76
Global self-worth r = -.326; p = .002 - The wording of 6
items was adapted to
cater for the younger
Separated into a
control group and a
group ‘at risk’ of
Developmental
Coordination Disorder
(DCD)
10.8(7.53)
Control
(boys)
8.78(4.93)
Control (girls)
12.10(8.66)
subscale (F(1, 82) =
5.41, p = .023)
where girls scored
significantly higher
(M = 3.35; SD =
2.55) than boys (M
= 2.26; SD = 1.81)
age range.
Study Sample Type Age N M(SD) Sex difference Reliability Correlation with outcome variables Longitudinal Results Comments
Popoola (2005) Secondary school
students (Male = 204,
Female = 181)
selected from ten
secondary schools
across 10 local
government areas in
Osun State, Nigeria.
Age range
10-19 years
385 Total M =
23.16 (3.15)
Physical M =
6.18 (1.46)
Verbal M =
5.48 (1.86)
Social M =
4.99 (1.73)
Attacks M =
6.50 (1.50)
Low level of
victimisation
Results showed
significant
differences between
males and females
on all forms of
victimisation, with
female participants
reporting higher
social, verbal and
attacks on property
than male students.
Male students
- - - Score of 0 to 16 =
Low level of
victimisation
Score of 17 to 21 =
Moderate level
Score of 22 to 32 =
High level of
victimisation
= 2.1%
Moderate
level = 27.3%
High level =
70.6%
reported
significantly higher
physical
victimisation than
female students.
Raine, Fung & Lam
(2011)
Participants consisted
of schoolchildren
(2112 males and 1678
females) drawn from
10 primary and 10
seoncdary schools in
Hong Kong.
Age range
8-16 years.
Male M =
11.7; SD =
2.0
Female M =
12.04; SD =
2.0)
3804 Total score M
= 8.9; SD =
11.27
α for total scale = .90 Total MPVS score was significantly
positively associated with reactive
aggression (r = .38), proactive
aggression (r = .29), Total SPQ (r = .39)
and the SPQ subscales: interpersonal (r
= .29), disorganised (r = .30) and
cognitive-perceptual (r = .35). Peer
victimisation mediated the association
between schizotypal personality and
aggression.
Note SPQ is a
measure of
Schizotypal
personality
Study Sample Type Age N M(SD) Sex difference Reliability Correlation with outcome variables Longitudinal Results Comments
Rao & Kishore
(2013)
54 obese and 54
normal weight school-
going adolescents.
Age range
11-16 years
108 For obese:
M = 10.85
(6.49)
For normal
weight:
M = 10.78
(6.03)
Alpha’s not reported
but subscale inter-
correlations ranged
from r = .36 for verbal
and attacks on
property, and r = .56
for physical and verbal
(p’s < .01)
Peer victimisation was negatively
correlated with self-esteem (r = -.42, p <
.01) and positively correlated with
behavioural problems (r = .24, p < .01)
in obese adolescents.
There was no significant difference in
MPVS scores for obese and normal
weight adolescents.
*
Scarpa, Carraro,
Gobbi & Nart
(2012)
Pupils attending a
middle school (grade
7) in a north-eastern
region of Italy.
Age range
12-13 years
M = 12.2
395 Total
victimisation
M = 5.02
(5.33)
Physical M
= .23 (.36)
Verbal M = .
54 (.59)
Social M = .
28 (.41)
Attacks M = .
21 (.39)
Physical α = .74
Verbal α = .75
Social manipulation α
= .68
Attacks on property α
= .76
Negative associations between peer-
victimisation during sport practice and
enjoyment of physical activity were
noted (r = -.14, p < .01). Verbal
victimisation and total victimisation
were both negatively associated with
enjoyment of sport (note that the MPVS
was completed only with reference to
victimisation during physical activity
and sport practice at school)
The Italian version
of the MPVS, given
in this study, was
validated by Carraro
et al. (2011) with the
following CFA fit
statistics: GFI = .94,
AGFI = .92, and
RMSEA= .052;
Cronbach’s alpha
values ranged from .
70 to .80.
Study Sample Type Age N M(SD) Sex difference Reliability Correlation with outcome variables Longitudinal Results Comments
Shakoor, McGuire,
Cardno, Freeman,
Plomin & Ronald
(2015)
Participants were
members of the Twins
Early Development
Study (TEDS) of twins
born in England and
Wales between 1994
Participants
were tested
at age 12
(M = 11.56)
and age 16
(M = 16.32)
4972
pairs
4826
pairs
Total = 7.55
(7.24)
Males = 8.40
(7.63)
Females =
6.82 (6.79)
Males reported
significantly more
victimisation than
females (p < .01)
α = .91 - Bullying victimisation
at age 12 was
associated with
paranoia at age 16 (r =
.26, p < .01).
Associations were
At age 12 bullying
victimisation was
assessed using the
full MPVS. At age
16, bullying
victimisation was
$
and 1997. lower but still
significant for
Hallucinations (r = .
18, p < .01), Cognitive
Disorganisation (r = .
20, p < .01) and
parent-rated negative
symptoms (r = .12, p
<.01)
assessed using a
shortened 6 item
version, so only
results for full
version are included
in review.
Waytowich,
Onwuegbuzie &
Elbedour (2011)
Juvenile delinquents
participating in two
delinquency
intervention programs
in Florida, US. 28.2%
female sample
Age range
12-16 years
(M = 14.6;
SD = 1.05)
181 - - Physical α = .80
Verbal α = .78
Social α = .76
Property α = .83
Verbal victimisation and attacks on
property significantly predicted violence
attribution errors.
Google Scholar is a commonly used web-based academic search engine, cataloguing between 2 and 100 million
records of both academic and grey literature (articles not formally published by commercial academic publishers). It has
received considerable attention as a method for searching for literature, particularly in searches for grey literature, as
required by systematic reviews. The reliance on GS as a standalone resource has been greatly debated, but recent
evidence has suggested that although it should not be used alone for systematic review searches, it forms a powerful
addition to other traditional search methods (Haddaway, Collins, Coughlin & Kirk, 2015)
We have reworded the items from the original Betts et al. (2015) paper from the second person pronoun (“sent you a
nasty text”) to the first person pronoun (sent me a nasty text; said something mean about me on a social networking site)
in line with the rest of the MPVS-24 items.