Poor Kids? Economic Resources and Adverse Peer Relations in a Nationally Representative Sample of Swedish Adolescents
Poor Kids? Economic Resources and Adverse Peer Relations in a Nationally Representative Sample of Swedish Adolescents
Simon Hjalmarsson 1
0 Institute for Futures Studies , Holländargatan 13, SE-101 31, Stockholm , Sweden
1 The Swedish Institute for Social Research, Stockholm University , SE-106 91, Stockholm , Sweden
2 Simon Hjalmarsson
There is limited knowledge on the impact of economic resources on adverse peer relations during adolescence. This study used a nationally representative sample (n = 4725, 51% girls) of Swedish eighth-grade students (approximately age fourteen) to examine associations between economic resources and adverse peer relations in the form of peer rejection and bullying victimization. Adolescents from households in the lowest within-school household income quintile were found to be rejected by school class peers to a greater extent than more advantaged students, but an association was not found between relative household income and bullying victimization. In contrast, adolescents unable to participate in activities with peers for economic reasons experienced more rejection and were at higher risk of victimization. The results underline the multidimensionality of adverse peer relations and advance our knowledge on how economic resources relate to peer relations in youth.
Peer rejection ● Bullying victimization ●; Economic resources ● Material deprivation ● Poverty ●; Adolescence
Adolescence is characterized by an increased importance of
peers. At their best, peer relations serve as sources of
support at a time of transitions and developmental changes.
However, peer relations are far from equally rewarding to
all. Some adolescents experience rejection by peers, and
eight percent of fifteen-year-olds in the countries covered by
the Health Behaviour in School-Aged Children (HBSC)
survey report being victims of school bullying
. Additionally, the risk of suffering adverse
peer relations is not equal for all adolescents. The
identification of individual and behavioral characteristics
associated with higher risks of victimization has made
significant contributions to understanding bullying and the
development and targeting of intervention programs.
However, less is known about the extent to which
socioeconomic factors play a role.
If adolescents with worse economic conditions are at a
higher risk of experiencing adverse peer relations, such as
peer rejection or bullying victimization, this would
represent yet another disadvantage facing this group of
adolescents. Bullying victimization can have serious short- and
(Arseneault et al. 2010; Wolke and
. Adverse peer relations could thus be one
pathway through which economic disadvantage translates
into socioeconomic inequalities in other dimensions. For
Due et al. (2009
a) argued that social inequalities in
early adulthood depression are partly explained by the
differential exposure and impact of bullying victimization in
This article examines the association between economic
resources and adverse peer relations using data that are
uniquely well-suited to this question, namely the Swedish
part of the 2010 wave of the Children of Immigrants
Longitudinal Study in Four European Countries
(CILS4EU). This is a nationally representative sample of
Swedish eighth-grade students (approximately age
fourteen), containing a school-based survey directed to the
adolescents, and matched information on household income
and parental characteristics gathered from administrative
and taxation register data. While previous research has
predominantly examined the association between economic
resources and adverse peer relations through measures of
household economic resources measured in absolute terms,
the current study applies an adolescent-centered perspective
of economic resources: Economic resources are measured
on (1) the household level, in terms of household income
assessed relative to the household income of others
in the same school, and (2) the individual level, in terms
of adolescents’ self-reported economic and material
deprivation. In addition, adverse peer relations are measured
using two outcomes: Bullying victimization is measured
through self-reports, while peer rejection is reported by
school class peers and collected through sociometric
(network) procedures. The use of two distinct forms of adverse
peer relations, reported from different sources, allow
for a more comprehensive understanding of the potential
associations between economic resources and adverse
Peer Rejection and Bullying Victimization
There are several distinct forms of adverse peer relations, of
which peer rejection and bullying victimization are two
examples. Peer rejection describes a peer group’s attitude
towards a member, that is, the degree of social avoidance, or
reluctance to affiliate with, the individual student
discussion in Juvonen and Gross 2005)
. It should be noted that,
as a concept, peer rejection does not necessarily involve
negative treatment by peers
(cf. Schuster 2001)
. To be
bullied, on the other hand, is to repeatedly and over time be
the target of intentionally hurtful behaviors by peers when a
power imbalance is also present, making it hard to defend
Although peer rejection does not always involve
(e.g., Knack et al. 2012)
, peer rejection and
bullying victimization tend be fairly strongly correlated,
especially when bullying victimization is measured through
(Schuster 2001; Scholte et al. 2013)
rejection has also been found to predict victimization in
(e.g., Hodges and Perry 1999;
Salmivalli and Isaacs 2005)
. Here, however, the purpose is not to
examine links between peer rejection and bullying
victimization but to examine whether economic resources are
associated with either.
Economic Resources and Adverse Peer Relations
Lacking the economic resources needed to participate in
consumption and activities undertaken by peers could make
it harder to create and maintain friendship relations. Indeed,
poorer children and adolescents report having fewer friends
(Olsson 2007; Sletten 2010)
, receive less friendship
nominations from school peers
(Bolger et al. 1995; Hjalmarsson
and Mood 2015)
, and are at a higher risk of social isolation in
the school class
(Hjalmarsson and Mood 2015)
with few friends, and especially those with no friends, in
school are more likely to be victimized
(Juvonen and Graham
, perhaps since bystanders are likely less prone to
defend peers that they do not consider friends or with whom
they do not want to be affiliated
(cf. Hodges and Perry 1999)
Theories of relative poverty and deprivation suggest that
economic inability to participate in common or expected
activities or consumption generates feelings of shame and
inadequacy, sometimes resulting in social withdrawal
(e.g., Townsend 1979)
. Previous research has found both
household economic resources and adolescents’ own
experienced economic and material deprivation to be
associated with internalizing symptoms and self-rated
(e.g., Plenty and Mood 2016)
problems (e.g., anxiety, insecurity in social interactions, and
social withdrawal) are important risk factors for
(Juvonen and Graham 2014; Reijntjes et al. 2010)
Peers might also stigmatize adolescents from poorer
backgrounds. Experimental research has found that
persongroup differences increase the risk of victimization
et al. 1986; Boivin et al. 1995)
. When adolescents are asked
to explain why bullying occurs, a common claim is that the
victim deviates from group norms
Thornberg and Knutsen 2011)
. A consistent finding from
qualitative interviews with children in low-income families
is the perceived importance of maintaining economic
appearances in order to avoid being seen as different
2006; Ridge 2011)
. A lack of economic resources could
itself form the basis for perceived deviation, but having less
economic resources than others could also lead to deviation
from group norms in other dimensions. For instance, poorer
students may be construed as different based on “wrong” or
out of fashion clothing (e.g., Frisén et al. 2008). Not being
able to participate in normatively expected activities or
consumption may lead to being seen by peers as “boring,
cheap or weird”
Previous Research on Inequality in Rejection and
Scarce research attention has been given to whether
economic resources are related to peer rejection. An early
Patterson et al. (1990
, 1991), who found that
elementary school students from low-income families were
twice as likely as other students to be rejected by school
peers in a small American city. The association remained
statistically significant after adjustment for ethnicity, family
structure, and gender.
In comparison, more attention has been given to
socioeconomic inequalities in bullying victimization. A recent
(Tippett and Wolke 2014)
found that students
from lower socioeconomic backgrounds had increased odds
of victimization but that the overall association, while
statistically significant, was fairly weak. However,
also noted that a wide range of measures have
been used to conceptualize socioeconomic background.
Stronger associations have been found by studies using
measures of household wealth. For instance,
Due et al.
b) used HBSC-data on 11-, 13-, and 15-year-olds in
35 countries and found that lower scores on the Family
Affluence Scale (FAS) were associated with a higher risk of
victimization. Employing a similar measure, Chaux and
Castellanos (2015) used data from a large-scale survey of
Colombian students in the 5th and 9th grades and found that
students from households with fewer assets than others in
the same school class were at a higher risk of victimization.
Weaker associations have been found by studies using
measures of parental education or occupation
Nordhagen et al. 2005; Lemstra et al. 2012)
Only rarely have studies had access to information on
between household economic resources and a wide range of
social outcomes—including victimization—using a
nationally representative sample of Swedish 10–18-year-olds.
Household income was associated with victimization, as
was living in an economically vulnerable household
(households with a low disposable income and no cash
margin). However, after adjustment for parental and
household characteristics (education, health, family
structure, immigration background and type of region),
associations were substantially reduced and no longer
These results indicate that victimization might not be
affected by household economic resources per se but rather
result from other factors associated with poorer economic
conditions. The comparatively weak association between
parental education and bullying victimization has already
been mentioned, but other background factors, such as
family structure, immigration background, parental health
or unemployment, could also play a role. Students from
single-parent households are more likely to report being
victims of bullying in school
(Fransson et al. 2017)
Regarding immigration background, recent research in the
Swedish context has reported mixed results.
Hjern et al.
found that first-generation immigrant students have a
higher overall risk of victimization, while
Plenty and Jonsson
did not. However, both studies found the immigrant
density of the school/school class to be of importance,
indicating a more complex pattern of associations. Parental
health problems may be associated with adolescents’ own
health issues or with internalizing symptoms brought forth by
the situation of having a sick parent. A similar argument can
be made for parental unemployment.
What Economic Resources Matter?
In the cases where previous literature has specifically
studied the role of economic resources, the measures have
almost exclusively been household-centered, referring to the
household economic resources, and predominantly in
absolute terms. Theoretically, this is unlikely to be the most
appropriate choice when the outcome of interest is adverse
peer relations in adolescence. First, insofar as parents’
resources matter, they are more likely to matter in relative
than in absolute terms. During adolescence, school peers are
likely to constitute the most salient reference group, and the
economic conditions of peers are likely to determine what
level of consumption and material resources are needed for
participation and to avoid being perceived (or perceiving
oneself) as deviating from group norms
(cf. Bernburg et al.
. If adolescents from low-income households tend to
attend schools in which their economic conditions are
comparable to peers, associations between household
income and adverse peer relations may not be evident if
household income is measured in absolute terms.
Adolescents with fewer economic resources relative to others in the
same school could still be more likely to experience adverse
Second, defining the adolescents’ economic situation in
terms of the household economy renders the adolescents’
own economy invisible, and ignores the fact that the
withinhousehold distribution of resources may vary widely
between families. Individual experiences of material
deprivation can exist within more affluent households, and
children in low-income households can sometimes be kept
on material par with peers through parental self-sacrifice or
support from outside the household
(e.g., Kochuyt 2004)
Research using parental reports
(Main and Bradshaw 2014)
and household expenditure surveys
indicate that parents in low-income
households allocate larger shares of their income on the material
conditions of their children. The link between household
income poverty and child material deprivation has in fact
been found to be fairly weak, at least in the Swedish context
(Mood and Jonsson 2015)
. Household income could thus be
too indirect to capture adolescents’ experienced material and
A more direct measure can be achieved using indicators
of adolescent self-reported material and economic
deprivation. Adolescents’ own economic and material conditions is
what directly affects their opportunities for consumption
and participation. For instance, lacking an own room could
make it harder to bring home friends; an own cash margin
allows for participation in spontaneous activities, or to make
spending choices parents might not willingly provide
funding for (e.g., following trends or fashion, buying
snacks, lending money to a friend); and lacking access or
control over material assets (e.g., a computer, TV, or phone
with social functions), could bar from participation in
activities, or complicate communication with peers.
Adolescents’ own resources will also be observable to peers to a
much larger extent than the economic resources of parents.
Measures of own resources can be both absolute and
relative. Previous research suggests that children and
adolescents who lack an own cash margin (absolute measure)
or who are not able to afford what many others in the same
age group can afford (relative measure) are at a higher risk
of victimization, even after controlling for household
economic resources and parental characteristics
The Current Study
The research presented above leads to the prediction that
adolescents who are poorer than school peers are more
likely to experience adverse peer relations. Thus, the first
hypothesis states that adolescents from households with a
lower disposable income relative to others in the same
school are rejected by peers to a larger extent than their
more affluent school class peers (Hypothesis 1). Relative
household income is predicted to be associated with
bullying victimization in the same way: Adolescents from
households with a lower disposable income relative to
others in the same school are more likely to be victims of
bullying than their more affluent school class peers
Household income may be too indirect a measure to
capture adolescents’ experienced material deprivation,
because the link between parental and child economic
conditions has been found to be fairly weak. Therefore,
adolescents’ own economic and material conditions are
expected to be associated with adverse peer relations
independently of household income. The third hypothesis
thus states that at a given household income, economically
or materially deprived adolescents are rejected by peers to a
larger extent than school class peers not experiencing
deprivation (Hypothesis 3). The last hypothesis is that
economic or material deprivation is associated with bullying
victimization in a similar way: At a given household
income, economically or materially deprived adolescents
are more likely to be victims of bullying than school class
peers not experiencing deprivation (Hypothesis 4).
Procedure and Participants
Survey data were gathered from the Swedish part of the first
wave of the CILS4EU-project
(Kalter et al. 2013)
main goal of the CILS4EU-project was to collect
information on the social, cultural and structural integration of
adolescents of immigrant descent, the survey was conducted
in schools and used a two-step clustered sampling approach.
Schools were randomly selected within four strata based on
the proportion of students with immigration background,
oversampling schools with a higher proportion. Within the
selected schools two school classes were randomly chosen
for participation in the survey (
weights account for the sampling probabilities and allow
inferences about the 2010/2011 cohort of Swedish
eighthgrade students (approximately age fourteen).
Statistics Sweden (the Swedish government statistics
agency) collected the data in schools during the winter of
2010–2011. Respondents were asked to answer
questionnaires and to complete tests on language proficiency and
cognitive skills. In addition, sociometric (network) data
were collected: students were asked to nominate other
students in their class for questions of interpersonal
relationships (e.g., “Who do you not want to sit next to?”). After
ethical approval, Statistics Sweden linked the survey to
information drawn from register data on household income
and parental characteristics.
The original sample consisted of 5699 students in 251
classes and 129 schools. Students absent at the time of the
survey (n = 674) were removed. Twenty-six students were
omitted due to implausible responses on multiple items, as
were the sociometric nominations they made. To reliably
measure economic conditions in the school, schools where
less than 30 students had valid information on income were
excluded (affecting 14 schools, 18 school classes and
274 students), mostly affecting schools with fewer than 30
participating students. The analytical sample thus consisted
of 4725 students in 233 classes and 115 schools.
The analytical sample was evenly distributed in regards
to gender (51% female) and consisted of 69% students of
majority background (Swedish born students with at least
one Swedish born parent), 20% of students had a second
generation immigration background (born in Sweden to
1 For a small number of observations, responses from Wave 2 and
Wave 3 were used to replace missing data on stable characteristics
(e.g., date and country of birth, gender).
foreign born parents), and 11% of students had a first
generation immigration background (born outside Sweden
to foreign born parents). Among students with a first or
second generation immigration background 44% had a
Middle Eastern or North African background, 25% an
Eastern European background, 13% was of Sub-Saharan
African descent, 8% had an Asian background, 5% had a
Southern or Western European background, and 5%
descended from other regions or could not be classified.
For the analysis of peer rejection a sub-sample was used.
Since sociometric data are sensitive to nominator
nonresponse 33 school classes in which less than 70% of
students answered the question used to measure rejection were
excluded; a response rate lower than approximately this
level has been suggested as problematic
(e.g., Cillessen and
Marks 2011; but see also Marks et al. 2013)
subsample for peer rejection was thus reduced to 4106 students
in 200 classes and 112 schools.
Peer rejection The measure of peer rejection was based on
the number of received sociometric nominations on the
question, “Who do you not want to sit next to?”. Students
were allowed to make up to five nominations. Nominations
where the nominated student was simultaneously reported
as a friend were removed (185 students made 331 such
overlapping nominations, representing approximately 3% of
the total nominations given). To account for differences in
class size and for class differences in nomination
propensities, the number of received nominations were
standardized by subtracting the school class mean of received
nominations and dividing the remainder by the school class
(Newcomb and Bukowski 1983)
expressing the difference from the school class mean in
standard deviations. In order to facilitate the interpretation
of the results, the standardized measure was then multiplied
by the overall sample standard deviation. The resulting
measure can be interpreted as the number of nominations
received by the student relative to the school class mean.
Continuous sociometric rejection measures have been found
to have good test-retest reliability
(Jiang and Cillessen
Bullying victimization Bullying victimization was based
on three questions in which the respondents were asked to
think about the past month and report how often (every day,
one or several times a week, less often, never) they had been
afraid of, teased by, or bullied by other students.
Adolescents experiencing any of these events at least once a week
were considered victimized, as were students experiencing
all of the above events on a monthly basis.
Relative household income Household income quintiles
were created within schools and were based on taxation
register data on the disposable income of the household
(income from labor, capital, and social benefits available to
the household after taxation) in 2010. In cases where
custodial parents were registered as living in separate
households, the mean of the two households was used.
Adolescents in the middle within-school income quintile
was used as the reference category.
Misses out on activities This item was measured based on
the question: “How often do you miss out on activities your
friends do because you can’t afford it?” Students reporting
that they often or always experience this problem were
coded as 1, while all others were allocated to the reference
Lack of own cash margin Adolescents able to access 300
SEK (approximately $30) by the next day were seen as
having a cash margin and coded as the reference group,
while adolescents who were not able to do so or who
answered that they might be able to do so were seen as
lacking a cash margin.
Lack of own room Adolescents reporting to have their
own room for the question: “Do you have a room just for
yourself?” was coded as the reference category, while 1
refers to not having one’s own room.
Own material assets Material assets were measured as the
sum of material assets the adolescent reported to own (out
of own computer, TV, smartphone, and having a gaming
console in the household).
Parental education Parental education was coded as the
highest achieved educational degree of the parent(s) in
2010. It was based on register data and distinguishes
between parents with at most a junior high school degree
(högstadieexamen), a senior high school degree
(gymnasieexamen), and parents with a higher educational degree
Family structure Family structure was distinguished
between adolescents whose parents reside in the same
household (i.e., who have parents that are married or
cohabiting) and all other family forms. The variable was
based on register data from 2010.
Migration background In regards to migration
background, adolescents of majority background (i.e., having at
least one Swedish-born parent) formed the reference
category. Further distinctions were made between
secondgeneration immigrants (born in Sweden to foreign-born
parents), first-generation immigrants (born abroad to
foreign-born parents) having migrated to Sweden more than
2 years prior to the interview, and first-generation
immigrants having lived in Sweden for less than 2 years.
Adolescents’ place of birth was assessed using register data, as
was the time since migration. Parents’ country of birth was
determined based on information provided by the students.
Parental receipt of disability pension Parental receipt of
disability pension was collected from register data and
indicates that at least one parent received disability pension
during the year 2010. Disability pension compensates for
income reductions caused by a permanently reduced
working ability due to sickness, injury or disability.
Parental unemployment The adolescent was seen as
having an unemployed parent if register data indicated that at
least one parent had been unemployed for more than half of
the year (during 2010).
Gender Gender used boys as the reference category and
was based on self-reported survey data.
Age Age was measured in days but expressed in years and
centered on the sample mean. The information came from
the student questionnaires.
Cognitive test score Cognitive test score was coded as the
sum of correctly answered questions (0–30) on a cognitive
test that was carried out as part of the survey. This sum was
then centered on the school class mean. A non-centered
version of the variable was included as an auxiliary variable
in the multiple imputation models (see section on Item
Language test score The language test was exclusively
used as an auxiliary variable in the multiple imputation
models. The variable is based on a test of language
proficiency (in Swedish), where the scores ranged from 0–30.
Externalizing behavior Externalizing behavior (scale 0–3,
Cronbach’s alpha .70) was measured as the mean score of
eight items. The index was created based on four questions
asking whether the respondent, during the last 3 months,
had “Deliberately damaged things that were not yours”,
“Stolen something from a shop/from someone else”,
“Carried a knife or weapon”, or “Been very drunk”; and four
questions where respondents reported how often
(every day, once or several times a week, once or several
times a month, less often, never) they “argue with a teacher”,
“get a punishment in school”, “skip a class”, or “come late to
Internalizing problems Internalizing problems (scale 0–3,
Cronbach’s alpha .78) was measured as the mean score on
three items regarding how often the statements “I feel very
worried”, “I feel anxious”, and “I feel depressed” was true
about the student (never true, rarely true, sometimes true,
Table 1 presents (unweighted) descriptive statistics for the
analytical sample. While the use of administrative register
data means that few observations lack information on
household characteristics, approximately 16% (n = 738) of
observations in the analytical sample lack a response on at
least one of the questions concerning adolescents’ own
economic resources (see proportion for each analysis
variable in Table 1). For this reason, multiple imputation by
chained equations (MICE) was used to impute missing
values. The mi impute command in Stata 14.2 was used to
generate 20 imputed datasets. Imputation models included
all variables used in the respective analysis. Additionally,
since non-response was judged to mainly be a problem of
respondent fatigue (other questions late in the survey also
suffered low response rates), non-centered test scores from
the cognitive ability and language proficiency test were
included as auxiliary variables in both imputation models.
Imputation models were stratified by gender to allow the
analyses to assess interactions with gender. Cases with
missing information on bullying victimization (n = 123)
were included in the imputation model for bullying
victimization, but excluded from the analysis, consistent with
recommendations for missing data on outcome variables by
Von Hippel (2007
). Similarly, observations with missing
data on household income and parental education were used
in imputation models but excluded from analyses since the
imputation model was not developed to account for the
missing data mechanism of the missing register data. The
same approach was used for observations lacking
information on the cognitive test score used in the analysis of
peer rejection (see Analytical Approach below). Models
using casewise deletion are available in the Appendix
(Table 3). The results are substantively similar.
a Descriptive statistics for peer rejection refers to the rejection sub-sample. Missing refers to respondents in excluded school classes
Associations between economic resources and peer
rejection were analyzed using ordinary least squares regressions.
All models analyzing rejection included a control variable
for cognitive ability (centered on the school class mean) due
to the phrasing of the question used to construct the
rejection measure (“Who do you not want to sit next to?”).
The worry is that the measure may entail aspects of
academic ability, making the student more or less attractive to
sit next to (e.g., inability to help during class).
Associations between economic resources and
victimization were analyzed using logistic regression. Estimates
are presented as Average Marginal Effects (AME), which—
as opposed to odds ratios—allow for comparisons across
(e.g., Mood 2010)
. AME can be interpreted as the
average percentage point change in the dependent variable
associated with a one-unit increase in the independent
For both outcomes, three sequential models were used.
The first model examines relative household income. The
second model adjusts for parental characteristics. The third
model adds measures of own experienced economic and
material deprivation. All models also adjust for age
(expressed in years from the sample mean) and gender.
Analyses were made using Stata 14.2. Official survey
weights were used to account for the over-sampling of
schools with larger proportions of students with
immigration backgrounds. Standard errors are clustered (and
heteroscedasticity robust) on the school class level.
Although the employment of control variables here is
fairly extensive, it is unlikely to successfully adjust for all
parental characteristics that may lead to confounding. As a
robustness test, models including self-reported internalizing
and externalizing behavior were also performed. The
argument is that, after adjustment for the parental
characteristics included in the models, most of the remaining
unobserved parental characteristic should affect the risk of
rejection and victimization through characteristics and
behaviors of the adolescent (because peers are not very
likely to observe parental characteristics directly). Thus, if
an association remains after adjusting for internalizing and
externalizing behavior, this provides stronger support for
the hypothesis that there is an effect of household economy.
However, since internalizing and externalizing behaviors
are also likely to (1) be a mediating path for effects of
economic resources and (2) be outcomes of victimization
and rejection (i.e., they are endogenous to the model), these
analyses should be considered only as a robustness test.
Among responding students in the analytical sample, the
(weighted) proportion of students reporting to miss out on
activities with peers for economic reasons was 7%. A
substantially larger proportion (32%) lacked a cash margin,
8% reported to lack an own room, and the average number
of own material assets was 2.6. The measures of own
material deprivation were, at most, weakly correlated (r
ranging between −.17 and .15). Turning to the (weighted)
prevalence of adverse peer relations, 9.5% of students were
classified as victimized and the average number of received
rejection nominations (using the peer rejection sub-sample)
was 1.8. The measures of adverse peer relations were
weakly correlated (r = .14, p < .01). Bivariate correlations
between all outcome and predictor variables are available in
the Appendix (Table 4) and unweighted sample
characteristics are shown in Table 1 (above).
Results for Peer Rejection
We begin by analyzing associations between economic
resources and peer rejection, where the outcome is the
received number of rejection nominations relative to the
average number of nominations received by other students
in the same class. Table 2 (Model 1) shows the association
between relative household income and rejection,
controlling for age, gender, and cognitive ability. Students in the
lowest within-school household income quintile received
0.7 more rejection nominations (0.742, CI 0.451–1.032)
than students in the middle-within-school household
income quintile. Additionally, students in the second-lowest
within-school household income quintile received more
peer rejection nominations (0.370, CI 0.065–0.674) than
students in the middle-within-school household income
quintile. Adolescents from households in the highest and
second-highest within-school household income quintiles
received slightly fewer rejection nominations than the
reference category, but in contrast to the lower
withinschool household income quintiles, the differences were not
Adjusting for parental characteristics (Table 2, Model 2)
reduced the strength of the association between relative
household income and peer rejection. Being from a
household in the lowest within-school household income
quintile was associated with receiving 0.6 more rejection
nominations (0.551, CI 0.230–0.873) than being from
households in the middle quintile. Students in the
secondlowest within-school household income quintile received
0.3 more peer rejection nominations (0.283, CI
-0.028–0.595), no longer being a statistically significant
difference from the reference category. Confounding in the
form of parental characteristics thus explains at least part of
the association between relative household income and peer
rejection. It is noteworthy that adolescents who migrated to
Sweden less than 2 years prior to the interview received
large numbers of rejection nominations. The group,
however, is small (n = 45), and excluding them from the
analyses does not change results (not shown).
The inclusion of variables measuring adolescent
experienced economic and material deprivation (Table 2, Model
3), only slightly changed the relative household income
quintile coefficients. Adolescents missing out on activities
with friends because they could not afford to participate
received, on average, 0.7 (0.673, CI 0.174–1.172) more
rejection nominations than students who did not experience
this problem. Lacking an own cash margin, lacking an own
room, and the number of own material assets were not
significantly associated with the number of rejection
nominations received. Exploratory analyses in which
gender was interacted with economic resources showed that
economic and material resources appear to be somewhat
more important for girls than for boys (not shown).
In summary, adolescents in the lowest within-school
household income quintile were found to be rejected by
peers to a greater extent than more advantaged students,
even after adjustment for household and parental
characteristics (Hypothesis 1). This was also the case for
adolescents who missed out on activities with peers due to a
lack of economic resources. Since an association was not
found for all measures of own experienced economic and
material deprivation, the results provide only partial support
for Hypothesis 3. Nevertheless, relative household income
and at least some aspects of own experienced deprivation
are associated with peer rejection.
Results for Bullying Victimization
We now turn to analyses of bullying victimization. No
association between relative household income and
victimization was found (Table 2, Model 4), and this remained
the case after adjusting for parental characteristics (Table 2,
Model 5). Adding measures of adolescent experienced
economic and material deprivation (Table 2, Model 6) only
marginally affected the coefficients for relative household
income. However, adolescents reporting missing out on
activities for economic reasons were 11 percentage points
(0.109, CI 0.057–0.161) more likely on average to be
victimized in school than students who did not experience this
problem. Lacking a cash margin was associated with a
4percentage-point greater probability of victimization (0.040,
CI 0.018–0.062). Given that the average (weighted) risk in
the sample was 9.5%, this represents a large increase in the
probability of bullying victimization. In contrast, no
statistically significant association was observed for lacking an
own room or for the number of own material assets. No
association was found to be statistically significantly
moderated by gender (not shown).
Thus, no evidence of a relationship between relative
household income and victimization was found (Hypothesis
2), while measures of adolescents’ experienced economic
deprivation—in the form of lacking an own cash margin
and reporting missing out on activities with peers for
economic reasons—were associated with a substantially higher
risk of victimization. Since no association was found with
the number of own material assets, or with having one’s
own room, the hypothesis that own experienced economic
and material deprivation plays a role again finds only partial
support (Hypothesis 4).
The results are robust to alternative specifications of relative
household income (not shown). The results were
substantively similar when relative household income was
measured continuously (centered on the school mean), and
when non-linearity were examined using within-school
household income deciles instead of within-school
household income quintiles. Neither were conclusions altered
when models using equivalized household income (dividing
the income by the square root of the number of parents and
children in the household) were performed. As information
on the number of residents in the household was only
available through survey data, the use of equivalized
household income introduces additional non-response.
Nonequivalized versions of relative household income were thus
Alternative specifications of the outcome variables have
also been tried (not shown). For the measure of peer
rejection, versions where the received number of
nominations was top-coded to two or three standard deviations
from the within-school-class mean before standardization
led to similar conclusions. Non-standardized versions also
show substantively similar results. For the measure of
bullying victimization, not including students experiencing all
of the events on a monthly basis decreases the number of
victimized students, but does not alter conclusions.
As an additional robustness test, analyses adjusting for
adolescent reported problems and behaviors were
performed (see Appendix, Table 5), which, in general,
somewhat reduced the strength of associations, especially for
measures of adolescent experienced economic and material
deprivation. However, all previously observed associations
remained statistically significant, meaning that, even when
we compare adolescents with equal behavior (in the
measured respects), those with worse economic conditions still
have worse outcomes. This result reduces the number of
plausible confounding factors, as these factors would have
to have an impact through other pathways than the
externalizing or internalizing behaviors of the adolescent.
Experiences of adverse peer relations in adolescence, such
as peer rejection and bullying victimization, can have
serious short- and long-term consequences
(Arseneault et al.
2010; Wolke and Lereya 2015)
. While risk factors
connected to personal and behavioral characteristics have
received substantial research attention, less is known about
the extent to which socioeconomic factors play a role and,
more specifically, whether economic resources are
associated with the risk of experiencing adverse peer relations.
Previous research examining this association have
predominantly used household economic resources measured
in absolute terms. In contrast, this article takes an
adolescent-centered perspective, arguing that the economic
resources likely to have repercussions for adverse peer
relations are those that influence the ability to participate in
consumption and activities on par with peers. Thus,
household income was assessed relative to others in the
same school, and self-reported indicators of economic and
material deprivation were used to capture adolescents’ own
experienced economic conditions.
The results showed relative household income to be
associated with peer rejection. Adolescents from
households in the lowest within-school household income quintile
were—also after adjustment for potentially confounding
factors—rejected by school class peers to a greater extent
than more advantaged students. Previous findings have
shown an association also between relative household
income and the number of friends
(Hjalmarsson and Mood
, which suggests that relative household income may
have an effect on both positive and negative dimensions of
peer relations during adolescence—perhaps because
growing up in a household with less economic resources than
others could make it difficult to reach normative
expectations in one’s peer group.
On the other hand, relative household income does not
appear to matter for the risk of bullying victimization. This
finding is largely in line with previous studies using
absolute measures of household economic resources
, as well as with studies of various measures of
socioeconomic background, which, in general, have found
only weak associations with bullying victimization
and Wolke 2014)
. Overall, it seems unlikely for measures of
household economic resources—whether measured
absolutely or relative to others in the same school—to have more
than marginal effects on the risk of bullying victimization.
That relative household income was found to be
associated with peer rejection, but not with bullying
victimization, suggests that a low economic standing may yield lower
peer status, but that this lower status is not necessarily
translated into the more severe outcome of being bullied. It
has been suggested that peer rejection implies a distancing
that could be interpreted by potential perpetrators as a group
sanctioning to target the rejected student, and that it might
reduce the likelihood of bystanders to intervene in bullying
(Hodges and Perry 1999)
. However, although
peer rejection is likely to form part of the process leading up
to victimization, economic resources is only one factor out
of a larger set of risk factors for peer rejection, and the risk
for bullying victimization is, in turn, also affected by many
other circumstances (e.g.,
Kärnä et al. 2010
). More research
is needed on how risk factors relate to different forms of
adverse peer relations, and on how risk- and protective
factors combine to make students more or less vulnerable to
becoming targets of bullying behavior
(e.g., Knack et al.
, but household income is unlikely to be of any major
importance for the risk of experiencing bullying
It should however be added that this study examined
associations between relative household income and
adverse peer relations on the individual level and within
schools. There is an additional need for research focusing
on whether economic characteristics on the school level,
such as the economic conditions of the student body or the
within-school economic inequality
(e.g., Elgar et al. 2009)
are associated with the prevalence of bullying victimization
on the school level.
When it comes to own experienced economic and
material deprivation, adolescents missing out on activities
with peers for economic reasons were rejected to a larger
extent and had a substantially higher risk of victimization.
Adolescents lacking an own cash margin were also in
higher risk of victimization. These findings are in line with
previous research on links between children’s and
adolescents’ own experienced material deprivation and negative
outcomes ranging from psychosomatic complaints to social
isolation and bullying victimization
(Plenty and Mood
2016; Hjalmarsson and Mood 2015; Olsson 2007)
Not all aspects of own experienced economic and
material deprivation were associated with adverse peer
relations, which could indicate that the impact of economic
resources is more detrimental when it affects specific
aspects of social relations, such as the ability to participate
in activities, rather than when affecting ownership of
material assets. However, it should be acknowledged that
the measure of own material assets was fairly crude. In
addition to mere ownership, the brand, version, or specific
functions could have additional impact. It is also possible
that, for instance, lacking one’s own room could matter
more in areas where access to an own room is the norm,
than in areas where it is more common. More research from
a child-centered perspective of economic resources thus
appears to be needed. The links between household income
and child experienced economic and material deprivation
have been found to be fairly weak, at least in the Swedish
context, and this can also be deduced from the results in this
article as the coefficients for relative household income
were largely unchanged when introducing adolescent-level
economic variables. More research is needed on the
determinants of adolescents’ own economic and material
conditions, for instance on the distribution of resources
within families, their other income sources, consumption
habits, and whether these differ based on characteristics
such as gender, social class, or ethnicity.
This study is not without its limitations. First, the
questions used to measure bullying victimization were related
primarily to direct forms of bullying behavior. As the
CILS4EU survey did not include questions regarding
indirect- or cyber bullying victimization it remains for
future research to address whether these, more indirect
forms of bullying victimization, are associated with
economic resources. A second limitation stems from the nature
of cross-sectional data. While adverse peer relations are
unlikely to affect household income, it is not as unlikely for
peer rejection and bullying victimization to both be affected
by economic and material deprivation and to affect
adolescent perceptions of own experienced deprivation. More
research on measures of own economic and material
deprivation is needed to address this issue. Third, like all
research using observational data, this study cannot entirely
exclude the possibility that the observed patterns are
generated by some other, here unobserved, characteristic or
situation. Nevertheless, access to taxation and
administrative register data made it possible to adjust for an
extensive range of potential confounders using highly
reliable measures and without introducing additional bias
through selective non-response.
The link between economic resources and adolescents’ peer
relations has so far received limited research interest. This
paper examines the association between economic
resources and adverse peer relations using an adolescent-centered
perspective of economic resources, involving both
household income and own experienced economic and material
assets, as well as two distinct forms of adverse peer
relations: peer rejection and bullying victimization. The results
showed that adolescents from households with a lower
income than others in the same school were rejected by
school class peers to a greater extent than students from
more affluent households. Relative household income was
not, however, found to be associated with bullying
victimization; a finding largely in line with previous research
using absolute measures of household economic resources
. These results suggest that while low
relative household income is associated with unfavorable
relations with peers, this does not necessarily translate into
the more severe outcome of being bullied. In contrast,
adolescents’ own experienced economic deprivation—
when related to problems of participation—was found to be
associated with peer rejection and bullying victimization.
Therefore, as far as peer rejection and bullying
victimization is concerned, policies targeting the household
economy are likely to be less efficient than policies directly
Acknowledgements The author is grateful for comments on earlier
versions of the present study from participants at the 1st International
CILS4EU user conference and at the Annual meeting of the Swedish
part of the European Social Policy Analysis Network (ESPAnet); from
participants at internal seminars within the Level-of-Living unit at the
Swedish Institute for Social Research and within the Swedish
CILS4EU-group. I am particularly grateful for comments from Carina
Mood, Jenny Torssander, Stephanie Plenty, Peter Fallesen, Roujman
Shahbazian, Jan O. Jonsson, and Johan Westerman.
Funding The present study was funded by grants from the Swedish
Research Council for Health, Working Life, and Welfare (FORTE)
(grant no. 2012–1741 and grant no. 2016-07099). The
CILS4EUproject was funded by the New Opportunities for Research Funding
Agency Co-operation in Europe (NORFACE).
Compliance with Ethical Standards
Conflict of Interest The author declare that he has no competing
Ethical Approval The study received ethical approval from the
Regional Ethics Committee, Stockholm. Approval reference number
Informed Consent Informed consent was obtained from all
participating students and from their parents.
Open Access This article is distributed under the terms of the
Creative Commons Attribution 4.0 International License (http://crea
tivecommons.org/licenses/by/4.0/), which permits unrestricted use,
distribution, and reproduction in any medium, provided you give
appropriate credit to the original author(s) and the source, provide a
link to the Creative Commons license, and indicate if changes were
N = 3919; Correlations with peer rejection uses the rejection sub-sample (N = 3443)
*p < .05, **p < .01
Arseneault , L. , Bowes , L. , & Shakoor , S. ( 2010 ). Bullying victimization in youths and mental health problems: 'Much ado about nothing'? Psychological Medicine , 40 ( 5 ), 717 - 729 .
Attree , P. ( 2006 ). The social costs of child poverty: A systematic review of the qualitative evidence . Children & Society , 20 ( 1 ), 54 - 66 .
Bernburg , J. G. , Thorlindsson , T. , & Sigfusdottir , I. D. ( 2009 ). Relative deprivation and adolescent outcomes in Iceland: A multilevel test . Social Forces , 87 ( 3 ), 1223 - 1250 .
Boivin , M. , Dodge , K. A. , & Coie , J. D. ( 1995 ). Individual-group behavioral similarity and peer status in experimental play groups of boys: The social misfit revisited . Journal of Personality and Social Psychology , 69 ( 2 ), 269 .
Bolger , K. E. , Patterson , C. J. , Thompson , W. W. , & Kupersmidt , J. B. ( 1995 ). Psychosocial adjustment among children experiencing persistent and intermittent family economic hardship . Child Development , 66 ( 4 ), 1107 - 1129 .
Chaux , E. , & Castellanos , M. ( 2015 ). Money and age in schools: Bullying and power imbalances . Aggressive Behavior , 41 ( 3 ), 280 - 293 .
Cillessen , A. H. N. , & Marks , P. E. L. ( 2011 ). Conceptualizing and measuring popularity . In A. H. N. Cillessen , D. Schwartz & L. Mayeux (Eds.), Popularity in the Peer System (pp. 25 - 56 ). New York; London: Guilford Press.
CILS4EU. ( 2016 ). Children of Immigrants Longitudinal Survey in Four European Countries . Technical Report. Wave 1 - 2010/ 2011, v1.2 .0. Mannheim: Mannheim University.
Due , P. , Damsgaard , M. T. , Lund , R. , & Holstein , B. E. ( 2009a ). Is bullying equally harmful for rich and poor children?: A study of bullying and depression from age 15 to 27 . The European Journal of Public Health , 19 ( 5 ), 464 - 469 .
Due , P. , Merlo , J. , Harel-Fisch , Y. , Damsgaard , M. T. , Holstein , B. E. , Hetland , J. , Currie , C. , Gabhainn , S. N. , de Matos , M. G. , & Lynch , J. ( 2009b ). Socioeconomic inequality in exposure to bullying during adolescence: A comparative, cross-sectional, multilevel study in 35 countries . American Journal of Public Health , 99 ( 5 ), 907 - 914 .
Elgar , F. J. , Craig , W. , Boyce , W. , Morgan, A. , & Vella-Zarb , R. ( 2009 ). Income inequality and school bullying: Multilevel study of adolescents in 37 countries . Journal of Adolescent Health Care , 45 ( 4 ), 351 - 359 .
Fernqvist , S. ( 2013 ). En erfarenhet rikare?: En kvalitativ studie av barns strategier och barnfattigdomens villkor i välfärdsstaten (Doctoral dissertation) . Acta Universitatis Upsaliensis . http:// www.diva-portal.org/smash/record.jsf? pid=diva2%3A571574& dswid=238
Fransson , E. , Låftman , S. B. , Östberg , V. , Hjern , A. , & Bergström , M. ( 2017 ). The Living Conditions of Children with Shared Residence - the Swedish Example . Child Indicators Research. Advance online publication . https://doi.org/10.1007/s12187-017- 9443-1
Frisén , A. , Holmqvist , K. , & Oscarsson , D. ( 2008 ). 13-year-olds' perception of bullying: Definitions, reasons for victimisation and experience of adults' response . Educational Studies , 34 ( 2 ), 105 - 117 .
Hjalmarsson , S. , & Mood , C. ( 2015 ). Do poorer youth have fewer friends? The role of household and child economic resources in adolescent school-class friendships . Children and Youth Services Review , 57 , 201 - 211 .
Hjern , A. , Rajmil , L. , Bergström , M. , Berlin, M. , Gustafsson , P. A. , & Modin , B. ( 2013 ). Migrant density and well-being-A national school survey of 15-year-olds in Sweden . The European Journal of Public Health , 23 ( 5 ), 823 - 828 .
Hodges , E. V. , & Perry , D. G. ( 1999 ). Personal and interpersonal antecedents and consequences of victimization by peers . Journal of Personality and Social Psychology , 76 ( 4 ), 677 .
Inchley , J. , & Currie , D. ( 2016 ). Health Policy for Children and Adolescents No. 7. Growing up unequal: gender and socioeconomic differences in young people's health and well-being . Health Behaviour in School-aged Children Study: International Report from the 2013 /2014 Survey. Geneva: World Health Organization.
Jiang , X. L. , & Cillessen , A. H. N. ( 2005 ). Stability of continuous measures of sociometric status: A meta-analysis . Developmental Review , 25 ( 1 ), 1 - 25 .
Juvonen , J. , & Graham , S. ( 2014 ). Bullying in schools: The power of bullies and the plight of victims . Annual Review of Psychology , 65 , 159 - 185 .
Juvonen , J. , & Gross , E. F. ( 2005 ). The rejected and the bullied: Lessons about social misfits from developmental psychology . In D. Kipling (Ed.), The social outcast: Ostracism, social exclusion, rejection, and bullying . New York: Psychology Press. ( 2005 ).
Kalter , F. , Heath , A. F. , Hewstone , M. , Jonsson , J. O. , Kalmijn , M. , Kogan , I. , & Van Tubergen , F. ( 2013 ). Children of Immigrants Longitudinal Survey in Four European Countries (CILS4EU) . GESIS Data Archive, Cologne. ZA5353 Data file Version , 1 .
Kärnä , A. , Voeten , M. , Poskiparta , E. , & Salmivalli , C. ( 2010 ). Vulnerable children in varying classroom contexts: Bystanders' behaviors moderate the effects of risk factors on victimization . Merrill-Palmer Quarterly , 56 ( 3 ), 261 - 282 .
Knack , J. M. , Tsar , V. , Vaillancourt , T. , Hymel , S. , & McDougall , P. ( 2012 ). What protects rejected adolescents from also being bullied by their peers? The moderating role of peer‐valued characteristics . Journal of Research on Adolescence , 22 ( 3 ), 467 - 479 .
Kochuyt , T. ( 2004 ). Giving away one's poverty. On the consumption of scarce resources within the family . The Sociological Review , 52 ( 2 ), 139 - 161 .
Kornrich , S. , & Furstenberg , F. ( 2013 ). Investing in children: Changes in parental spending on children, 1972 - 2007 . Demography, 50 ( 1 ), 1 - 23 .
Lemstra , M. E. , Nielsen , G. , Rogers , M. R. , Thompson , A. T. , & Moraros , J. S. ( 2012 ). Risk indicators and outcomes associated with bullying in youth aged 9-15 years . Canadian Journal of Public Health , 103 ( 1 ), 9 - 13 .
Main , G. , & Bradshaw , J. ( 2014 ). Children's necessities: Trends over time in perceptions and ownership . Journal of Poverty and Social Justice , 22 ( 3 ), 193 - 208 .
Marks , P. E. , Babcock , B. , Cillessen , A. H. , & Crick , N. R. ( 2013 ). The effects of participation rate on the internal reliability of peer nomination measures . Social Development , 22 ( 3 ), 609 - 622 .
Mood , C. ( 2010 ). Logistic regression: Why we cannot do what we think we can do, and what we can do about it . European Sociological Review , 26 ( 1 ), 67 - 82 .
Mood , C. , & Jonsson , J. O. ( 2015 ). Trends in child poverty in Sweden: parental and child reports . Child Indicators Research , 9 , 825 - 854 .
Newcomb , A. F. , & Bukowski , W. M. ( 1983 ). Social impact and social preference as determinants of children's peer group status . Developmental Psychology , 19 ( 6 ), 856 .
Nordhagen , R. , Nielsen , A. , Stigum , H. , & Köhler , L. ( 2005 ). Parental reported bullying among nordic children: a population‐based study . Child: Care, Health and Development , 31 ( 6 ), 693 - 701 .
Olsson , E. ( 2007 ). The economic side of social relations: Household poverty, adolescents' own resources and peer relations . European Sociological Review , 23 ( 4 ), 471 - 485 .
Olweus , Dan ( 1993 ). Bullying at School: What we know and what we can do . Oxford: Blackwell.
Patterson , C. J. , Kupersmidt , J. B. , & Vaden , N. A. ( 1990 ). Income level, gender, ethnicity, and household composition as predictors of children's school‐based competence . Child Development , 61 ( 2 ), 485 - 494 .
Patterson , C. J. , Vaden , N. A. , & Kupersmidt , J. B. ( 1991 ). Family background, recent life events and peer rejection during childhood . Journal of Social and Personal Relationships , 8 ( 3 ), 347 - 361 .
Plenty , S. , & Jonsson , J. O. ( 2017 ). Social exclusion among peers: The role of immigrant status and classroom immigrant density . Journal of Youth and Adolescence , 46 ( 6 ), 1275 - 1288 .
Plenty , S. , & Mood , C. ( 2016 ). Money, peers and parents: social and economic aspects of inequality in youth wellbeing . Journal of Youth and Adolescence , 45 ( 7 ), 1294 - 1308 .
Reijntjes , A. , Kamphuis , J. H. , Prinzie , P. , & Telch , M. J. ( 2010 ). Peer victimization and internalizing problems in children: A metaanalysis of longitudinal studies . Child Abuse & Neglect , 34 ( 4 ), 244 - 252 .
Ridge , T. ( 2011 ). The everyday costs of poverty in childhood: A review of qualitative research exploring the lives and experiences of low-income children in the UK . Children & Society , 25 ( 1 ), 73 - 84 .
Salmivalli , C. , & Isaacs , J. ( 2005 ). Prospective relations among victimization, rejection, friendlessness, and children's self and peer perceptions . Child Development , 76 ( 6 ), 1161 - 1171 .
Scholte , R. H. J. , Burk , W. J. , & Overbeek , G. ( 2013 ). Divergence in self- and peer-reported victimization and its association to concurrent and prospective adjustment . Journal of Youth and Adolescence , 42 ( 12 ), 1789 - 1800 .
Schuster , B. ( 2001 ). Rejection and victimization by peers . In J. Juvonen & S. Graham (Eds.) , Peer harassment in school: The plight of the vulnerable and victimized . New York: Guilford Press.
Sletten , M. A. ( 2010 ). Social costs of poverty; leisure time socializing and the subjective experience of social isolation among 13-16- year-old Norwegians . Journal of Youth Studies , 13 ( 3 ), 291 - 315 .
Thornberg , R. ( 2011 ). 'She's weird!'-The social construction of bullying in school: A review of qualitative research . Children & Society , 25 ( 4 ), 258 - 267 .
Thornberg , R. , & Knutsen , S. ( 2011 ). Teenagers' explanations of bullying . Child & Youth Care Forum , 40 ( 3 ), 177 - 192 .
Tippett , N. , & Wolke , D. ( 2014 ). Socioeconomic status and bullying: a meta-analysis . American Journal of Public Health , 104 ( 6 ), 48 - 59 .
Townsend , P. ( 1979 ). Poverty in the United Kingdom: A survey of household resources and standards of living. Harmondsworth: Penguin books .
Von Hippel , P. T. ( 2007 ). Regression with missing Ys: An improved strategy for analyzing multiply imputed data . Sociological Methodology , 37 ( 1 ), 83 - 117 .
Wolke , D. , & Lereya , S. T. ( 2015 ). Long-term effects of bullying . Archives of Disease in Childhood , 100 , 879 - 885 .
Wright , J. C. , Giammarino , M. , & Parad , H. W. ( 1986 ). Social status in small groups: Individual-group similarity and the social “misfit . Journal of Personality and Social Psychology , 50 ( 3 ), 523 - 536 .