Journal of Clinical Child and Adolescent Psychology
Copyright © 2005 by
2005, Vol. 34, No. 2, 326–332
Lawrence Erlbaum Associates, Inc.
The Relation Between Gender Role Orientation and Fear and Anxiety
in Nonclinic-Referred Children
Peter Muris, Cor Meesters, and Miranda Knoops
Department of Medical, Clinical, and Experimental Psychology
Maastricht University, The Netherlands
This study examined the relation between gender role orientation and fear and anxi-
ety in a sample of nonclinic-referred children (N = 209) ages 10 to 13 years. Children
and their parents completed questionnaires assessing children’s gender role orienta-
tion, toy and activity preferences, and fear and anxiety. Results generally indicated
that femininity and a preference for girls’ toys and activities were positively associ-
ated with fear and anxiety, whereas masculinity and a preference for boys’ toys and
activities were negatively related to these emotions. Furthermore, gender role orien-
tation accounted for more of the variance in fear and anxiety scores than the
child’s sex.
It is well established that girls display higher levels
lated behaviors (e.g., avoidance behavior). Conversely,
of fear and anxiety than boys (e.g., Ollendick, King, &
fearfulness and anxiety are inconsistent with the mas-
Muris, 2002). For example, Lewinsohn, Gotlib, Lew-
culine gender role. Such emotions are less accepted in
insohn, Seeley, and Allen (1998), who examined gen-
boys, who are expected to behave bravely and to dis-
der differences in the prevalence of anxiety disorders in
play active and purposeful coping behavior.
a community sample of 1,079 adolescents, concluded
Few studies have examined the connection between
that girls more frequently met the diagnostic criteria
gender role orientation and fear and anxiety in child-
for current and lifetime diagnosis of phobic and other
hood samples. One exception is a study by Brody, Hay,
anxiety disorders than boys. Interestingly, the results
and Vandewater (1990), who investigated the relations
of this study indicated that this gender difference
between gender and gender role orientation and chil-
emerged early in life: At age 6, girls were twice as
dren’s feelings toward peers as indexed by an Emo-
likely to have experienced an anxiety disorder than
tional Story Task in 120 nonclinic-referred children
were boys. Similar gender differences have been con-
ages 6 to 12 years. Results showed that girls reported
sistently reported for youths’ fear and anxiety levels as
higher levels of fear toward peers than boys. Most im-
measured with standardized rating scales (Muris,
portant, gender role orientation accounted for more of
Merckelbach, Ollendick, King, & Bogie, 2002, Muris
the variance in predicting fear than did the child’s sex.
& Ollendick, 2002; Spence, 1998).
That is, biological gender was no longer associated
One explanation for the finding that girls are more
with fear toward peers, after the influence of gender
fearful and anxious than boys has to do with individual
role orientation was partialled out. In general, boys and
differences in gender role orientation (Ollendick, Yang,
girls who scored higher on feminine gender role traits
Dong, Xia, & Lin, 1995). Briefly, this explanation im-
were more prone to report higher levels of fears toward
plies that girls and boys are socialized to develop gen-
peers.
der-linked feminine and masculine behaviors, traits,
Another investigation by Ginsburg and Silverman
and skills. According to theories on the development of
(2000) addressed the relation between gender role ori-
gender roles (e.g., Bem, 1981), the expression of fear
entation and the intensity of fears in a sample of clini-
and anxiety is in agreement with the feminine gender
cally referred children with anxiety disorders (N = 66)
role, which is generally acquired by girls and which
ranging in age between 6 and 11 years. Children com-
tolerates the expression of negative emotions and re-
pleted a questionnaire measuring masculinity and fem-
ininity and the Fear Survey Schedule for Children–
We thank the children, parents, and staff of primary schools
Revised (FSSC–R; Ollendick, 1983) as an index of
“Aquamarijn,” “De Muze,” “Michiel de Ruijter,” and “De Vosse-
childhood fear. Results indicated that masculinity was
burcht” in Nijmegen, The Netherlands, for their participation in this
negatively related to fearfulness. Unexpectedly, how-
study.
ever, no relation was found between femininity and
Requests for reprints should be sent to Peter Muris, Institute of
children’s fearfulness. To account for these mixed find-
Psychology, Erasmus University Rotterdam, Burgemeester Oudlaan
50, Suite J5–31, P.O. Box 1738, 3000 DR Rotterdam, The Nether-
ings, Ginsburg and Silverman (2000) pointed out that
lands. E-mail: muris@fsw.eur.nl
the children in their study were on average relatively
326

---

GENDER ROLE AND FEAR–ANXIETY
young (i.e., 8.9 years) and that it may well have been
der role orientation. The CSRI contains masculinity
the case that gender roles were insufficiently crystal-
items (e.g., “When I play games, I really like to win”)
lized in their sample.
and femininity items (e.g., “I am a kind and caring per-
Thus, although previous research has not yielded
son”). Previous research has shown that the CSRI has
fully consistent results, available evidence demon-
good internal consistency (with ?s well above .70) and
strates that gender role orientation is linked in a theoreti-
test–retest stability (with an average 1-year test–retest
cally meaningful way to fearfulness in youths. This
correlation of .60), and reasonable validity as estab-
study further examined the connection between gender
lished through relations with gender-typed toy and ac-
role orientation and fear and anxiety in children. With
tivity preferences (Boldizar, 1991).
Ginsburg and Silverman’s (2000) comments in mind,
The Toy and Activity Preference Questionnaire
we selected children in the last two grades of primary
(TAPQ) was based on Perry, Perry, and Hynes’ (1990)
school (ages 10 to 13 years), for which it has been shown
measure and consisted of a list of 53 toys and activi-
that by then gender roles have become more clearly
ties.2 Children were asked to indicate for each TAPQ
manifested (Boldizar, 1991). It was investigated wheth-
item to what extent they liked to play with the toy or to
er (a) girls display higher levels of fear and anxiety than
engage in the pertinent activity based on a scale of 1
boys and (b) masculinity is negatively associated with
(not at all), 2 (not), 3 (a little), 4 (much) and 5 (very
fear and anxiety, whereas femininity is positively re-
much). For each TAPQ item, 12 psychologists indi-
lated to such negative emotions. As there is some evi-
cated whether the pertinent toy or activity was most ap-
dence to suggest that gender role orientation may be
propriate for boys or girls, or suitable to both genders.
equal in importance or even more important than sex in
In this way, a weight was assigned to each item that in-
mediating children’s fear and anxiety (e.g., Brody et al.,
dicated to what extent each toy and activity was con-
1990), we also examined whether (c) biological sex and
sidered as appropriate for one of the genders. Toys and
gender role orientation each account for a unique contri-
activities such as “playing football” and “cars” (boys)
bution in the variance of these emotions.
and “playing hopscotch” and “dolls” (girls) were unan-
imously considered as typical for one of the genders,
and these items received the maximum weight of 12.
Method
Other toys and activities were less clearly linked to one
of both sexes and thus were given a lower weight rang-
Participants
ing between 1 and 11. TAPQ scores were derived by
Three hundred children and their parents were in-
multiplying children’s ratings on the 5-point scale with
vited to participate in this study. Child participants
the weights as derived from the psychologists’ judg-
were told that the study was concerned with the feel-
ments and then combining relevant items to “prefer-
ings, behaviors, and activities of older primary school
ence for boys’ toys and activities” and ”preference for
children. Almost three-quarters of the youths and their
girls’ toys and activities” scores. Boldizar (1991) dem-
parents (N = 220; i.e., 73.3%) responded favorably to
onstrated that this type of questionnaire yields reliable
our mailed invitation. Eventually, complete data sets
(with ?s higher than .80) and valid indications of the
were obtained for 209 children (103 boys and 106
male and female gender roles.
girls), with a mean age of 11.07 years (SD = 0.85, range
A shortened version of the FSSC–R (Ollendick,
10 to 13 years). The percentages of children with a low,
1983) was used to assess children’s general fear levels.
middle, and upper socioeconomic background were
The FSSC–R is a reliable (with an ? in the .90 range
37.8%, 31.6%, and 30.6%, respectively, and the vast
and a 1-week test–rest correlation of .82) and widely
majority of the children (more than 90%) were White.
used instrument for assessing childhood fear, and there
The remaining children were Mediterranean, North
is evidence indicating that this is also true for the short-
African, and Asian. For most children, parent ques-
ened version of the scale (Muris, Bodden, Merckel-
tionnaires were completed by the mother (71.3%). In
bach, Ollendick, & King, 2003).
remaining cases, the father (13.9%) or both parents
The 25-item version of the Revised Child Anxiety
(14.8%) filled in the measures.1
and Depression Scale (RCADS; Chorpita, Yim, Mof-
fitt, Umemoto, & Francis, 2000; Muris, Meesters, &
Questionnaires
Schouten, 2002) measures symptoms of anxiety disor-
ders and major depression as defined in the Diagnostic
The Children’s Sex Role Inventory (CSRI; Boldizar,
1991) is a self-report questionnaire for measuring gen-
2Inspection of the ratings on individual TAPQ items revealed that
1A series of 3 (informant: mother, father, both parents) × 2 (gen-
only eight items of the child version and two items of the parent ver-
der) analyses of variance performed on the parent report scores of the
sion received relatively low scores (i.e., >75% of the respondents en-
TAPQ, FSSC–R, and RCADS revealed neither main effects of infor-
dorsing the not at all or not options). This indicates that most activi-
mant nor interaction effects of informant and gender. This indicates
ties and plays listed in the questionnaire were relevant for the
that parent scores were not influenced by the type of informant.
children in this study.
327

---

MURIS, MEESTERS, KNOOPS
and Statistical Manual of Mental Disorders (4th ed.
FSSC–R, and RCADS and to return the materials in a
[DSM–IV], American Psychiatric Association, 1994).
sealed envelope. Children of these parents completed
In this study, anxiety-relevant items were used to com-
questionnaires at school. This was done during class-
pute a total anxiety score; the major depressive disor-
room sessions with a research assistant and the teacher
der items were discarded. The shortened RCADS has
always present to ensure confidential and independent
been demonstrated to be reliable in terms of internal
responding. Children received a small gift as a reward
consistency (average ? = .74) and temporal stability
for their participation in the study.
(average 4-week test–rest correlation = .83), and dis-
plays good validity as evidenced by positive associ-
ations with other childhood anxiety questionnaires
Results
(Muris et al., 2002).
In an attempt to cross-validate children’s self-report
General Results
data, parents completed questionnaires measuring
Statistically significant gender differences were
their child’s toys and activities preferences, fear, and
found for the CSRI and the TAPQ. As expected, boys
anxiety. These parent versions of TAPQ, FSSC–R, and
scored higher on masculinity and displayed a stronger
RCADS were identical to the child versions, except
preference for boys’ toys and activities (as indexed by
that instructions and items were rephrased in terms of
child and parent report) than girls, t(207) = 3.97, 10.59,
the parents’ perspective.
and 11.58, respectively, all ps < .001. Conversely, girls
As can be seen in Table 1, all questionnaires were
scored higher on femininity and had a stronger prefer-
reliable in terms of internal consistency, with Cron-
ence for girls’ toys and activities, t(207) = 3.10, 9.76,
bach’s alphas varying between .80 and .88 for child
and 11.23, all ps < .01 (see Table 1). Furthermore, cor-
scales and between .86 and .88 for parent question-
relations computed between CSRI and TAPQ revealed
naires. Furthermore, acceptable parent–child agree-
that masculinity was significantly associated with
ment was found for the questionnaires that were ob-
child- and parent-reported preference for boys’ toys
tained in both children and parents. For example,
and activities (rs = .43 and .39, respectively, both ps <
parent–child correlations were .74 for preference for
.001), whereas femininity was significantly related to
boys’ toys and activities, .72 for preference for girls’
preference for girls’ toys and activities (rs = .41 and
toys and activities, .62 for the FSSC–R total score, and
.30, ps < .001).
.57 for RCADS total anxiety (all ps < .001).
Gender Differences in Fear
Procedure
and Anxiety
Parents were approached by mail (via school). In
As shown in Table 1, girls displayed significantly
the letter, parents received information about the study
higher scores on the FSSC–R and the RCADS than
and were asked to give their signed informed consent
boys, t(207) ? 3.59, ps < .001. In other words, girls re-
for their child’s participation. In addition, parents
ported higher levels of fearfulness and anxiety symp-
were invited to complete parent versions of the TAPQ,
toms than boys. Parents also rated girls as more fearful
Table 1. Descriptive Statistics (Means, Standard Deviations, Sex Differences, and Reliability Coefficients) of Measures
Total Sample
(N = 209)
Boys (n = 103)
Girls (n = 106)
M
SD
M
SD
M
SD
Cronbach’s ?
Child Report
CSRI Masculinity
2.46
0.40
2.57a
0.43
2.35b
0.34
.84
CSRI Femininity
2.61
0.38
2.53a
0.38
2.68b
0.36
.80
TAPQ Preference for boys’ TA
26.71
5.17
29.83a
4.22
23.71b
4.12
.86
TAPQ Preference for girls’ TA
18.48
4.24
16.06a
2.95
20.82b
3.98
.86
FSSC–R Total fear
39.66
7.92
39.97a
6.90
42.27b
8.00
.88
RCADS Total anxiety
11.04
6.95
9.34a
6.19
12.70b
7.27
.86
Parent Report
TAPQ Preference for boys’ TA
27.21
5.19
30.51a
3.90
24.01b
4.20
.86
TAPQ Preference for girls’ TA
19.56
4.70
16.63a
3.26
22.40b
4.11
.87
FSSC–R Total fear
38.96
8.13
36.84a
7.20
41.01b
8.49
.88
RCADS Total anxiety
8.66
6.04
7.85a
6.05
9.43a
5.94
.86
Note: CSRI = Children’s Sex Role Inventory; TAPQ = Toy and Activity Preference Questionnaire; TA = Toys and Activities; FSSC–R = Fear
Survey Schedule for Children–Revised; RCADS = Revised Child Anxiety and Depression Scale. Means with different subscripts differ at p <
.05/10 (i.e., Bonferroni correction).
328

---

GENDER ROLE AND FEAR–ANXIETY
(FSSC–R) than boys, t(207) = 3.82, p < .001. No sig-
A similar pattern of results emerged when comput-
nificant difference between boys and girls was found
ing correlations between parent-reported toy and activ-
with regard to scores on the parent version of the
ity preferences and parent ratings of children’s fear and
RCADS, t(207) = 1.90, p > .05.
anxiety. That is, in particular, a preference for girls’
toys and activities correlated positively with these neg-
ative emotions. Note also that these correlations were
Gender Role Orientation and Fear
generally weaker than for the child self-report data
and Anxiety
(tests for comparing correlation coefficients indicated
significant differences for five out of six comparisons,
The relation between gender role orientation and
with Zs ? 2.08, ps < .05, and that significant correla-
fear and anxiety was examined in two ways. First, cor-
tions were predominantly found for fear.
relations were computed between CSRI masculinity
and femininity, on the one hand, and FSSC–R and
RCADS scores, on the other hand. As can be seen in
Unique Contribution of Sex
Table 2, a number of small but significant associations
and Gender Role Orientation to Fear
between a feminine gender role orientation as mea-
and Anxiety
sured by the CSRI and fear and anxiety were found.
When analyzing the data for both genders separately, it
To examine the relative contributions of sex and
appeared that, in boys, gender role orientation was not
gender role orientation to children’s fear and anxiety, a
linked to fear and anxiety. In girls, however, femininity
series of regression analyses was carried out. In these
was still significantly associated with fear scores.
regression analyses, fear and anxiety, as indexed by the
Second, correlations were computed between toys
total scores on the child and parent versions of the
and activities preferences (TAPQ) and fear and anxi-
FSSC–R and RCADS, were the dependent variables,
ety. Results indicated that, in particular, a preference
whereas the child’s sex (i.e., a dummy variable with 0 =
for girls’ toys and activities was positively associated
boy and 1 = girl) and gender role orientation as mea-
with levels of fear and anxiety. Thus, the more children
sured by CSRI and TAPQ were the predictors. Table 3
preferred stereotyped girl toys and activities, the
summarizes the results of the regression analyses in
higher their level of fear and anxiety. Note that this link
which the predictor variables accounted for a signifi-
also emerged when analyzing the data of both genders
cant proportion of the variance in fear and anxiety
separately, although it should be noted that the correla-
scores.
tion between a preference for girls’ toys and activities
In the regression analysis predicting children’s
and fear was significantly larger in girls than in boys, rs
self-reported fear from biological gender and CSRI
= .49 versus .20; t(207) = 2.37, p < .05. A preference
scores, sex, masculinity, and femininity all accounted
for boys’ toys and activities was less clearly connected
for a significant and independent proportion of the
to fear and anxiety. That is, only when computing cor-
variance. The direction of the ?s was as expected: That
relations for the total sample were significant negative
is, female sex and a feminine gender role orientation
correlations between a preference for boys’ toys and
were positively associated with fear, whereas a mascu-
activities and fear and anxiety found (Table 2), but still
line gender role was negatively associated with fear. In
the magnitude of these correlations was significantly
the regression analysis predicting children’s self-re-
smaller than those obtained between a preference for
ported fear from biological gender and TAPQ scores,
girls’ toys and activities and these emotions (Zs = 2.38
only toy and activity preferences accounted for a sig-
and 2.01, respectively, ps < .05.
nificant proportion of the variance. A preference for
Table 2. Correlations Between Gender Role Orientation and Fear and Anxiety
FSSC–R Total Fear
RCADS Total Anxiety
Total Sample
Boys
Girls
Total Sample
Boys
Girls
Child Report
CSRI Masculinity
–.12
–.10
.03
–.05
.01
.02
CSRI Femininity
.22**
.05
.25**
.19**
.15
.14
TAPQ Preference for boys’ TA
–.26***
–.04
–.13
–.15*
.06
–.07
TAPQ Preference for girls’ TA
.48***
.20*
.49***
.34***
.24*
.27**
Parent Report
TAPQ Preference for boys’ TA
–.22**
–.05
–.11
–.08
.06
–.05
TAPQ Preference for girls’ TA
.38***
.27**
.30**
.15*
.05
.12
Note: CRSI = Children’s Sex Role Inventory; TAPQ = Toy and Activity Preference Questionnaire; TA = Toys and Activities; FSSC–R = Fear
Survey Schedule for Children–Revised; RCADS = Revised Child Anxiety and Depression Scale.
*p < .05. **p < .01. ***p < .001.
329

---

MURIS, MEESTERS, KNOOPS
Table 3. Main Results of Regression Analyses Predicting
Discussion
Fear and Anxiety Levels From Sex and Gender Role
Orientation Scores
The results confirm previous research showing that
gender role orientation is associated with fear and anx-
B
SE B
?
iety in children. With respect to this finding, a number
FSSC–R Total fear (child report)a
of remarks are in order. First, a comparison with the
Sex
3.72
1.15
.24*
Ginsburg and Silverman (2000) study in which a simi-
CSRI Masculinity
–3.50
1.60
–.18*
CSRI Femininity
5.25
1.67
.25*
lar method was employed to assess gender role orienta-
FSSC–R Total fear (child report)b
tion (CSRI) and fear (FSSC–R) reveals a remarkable
Sex
–1.72
1.46
–.11
difference in the results. That is, whereas our data indi-
TAPQ Preference for boys’ TA
–.39
.12
–.26*
cated that, in particular, CSRI femininity was (posi-
TAPQ Preference for girls’ TA
.95
.14
.51*
tively) linked to fear, the Ginsburg and Silverman
FSSC-R Total fear (child report)c
Sex
–2.07
1.48
–.13
study found that especially CSRI masculinity was
CSRI Masculinity
–2.01
1.54
–.10
(negatively) associated with this negative emotion.
CSRI Femininity
–2.46
1.63
.12
It seems most plausible that sample differences ac-
TAPQ Preference for boys’ TA
–.37
.13
–.24*
count for these diverging findings. For example,
TAPQ Preference for girls’ TA
.88
.15
.47*
whereas this study relied on a sample of nonclinical
RCADS Total anxiety (child report)d
Sex
2.38
1.05
.17*
children, Ginsburg and Silverman studied children
CSRI Masculinity
–1.79
1.46
–.10
with anxiety disorders. It should be noted that feminin-
CSRI Femininity
3.74
1.53
.20*
ity scores were fairly high in Ginsburg and Silverman’s
RCADS Total anxiety (child report)e
clinical sample, and it is possible that a ceiling effect
Sex
–.29
1.42
–.02
occurred that hindered the finding of a link between
TAPQ Preference for boys’ TA
–.16
.11
–.12
TAPQ Preference for girls’ TA
.55
.14
.34*
femininity and fear. Further, participants resided in dif-
RCADS Total anxiety (child report)f
ferent countries (i.e., the Netherlands vs. the United
Sex
–.55
1.43
–.04
States) and, most important, had a different cultural
CSRI Masculinity
–1.13
1.49
–.07
background. That is, in this study more than 90% of the
CSRI Femininity
2.12
1.58
.12
children were White. In the Ginsburg and Silverman
TAPQ Preference for boys’ TA
–.15
.12
–.11
TAPQ Preference for girls’ TA
.49
.14
.30*
study, this percentage was less than 50% as their sam-
FSSC–R Total fear (parent report)g
ple contained a large number of Hispanic American
Sex
–2.09
1.71
–.13
children.
TAPQ Preference for boys’ TA
–.33
.13
–.21*
Second, results indicated that toy and activity pref-
TAPQ Preference for girls’ TA
.71
.15
.41*
erence as an index of gender role orientation was a
Note: CSRI = Children’s Sex Role Inventory, TAPQ = Toy and Ac-
better predictor of fear and anxiety than femininity and
tivity Preference Questionnaire, TA = Toys and Activities, FSSC-R =
masculinity as measured with the CSRI. Although both
Fear Survey Schedule for Children-Revised, RCADS = Revised
instruments were reliable in terms of internal consis-
Child Anxiety and Depression Scale. All predictor variables were
tency, it may well be the case that the TAPQ is a more
entered simultaneously. Sex was entered in the regression equation
as a dummy variable with 0 = boy and 1 = girl.
straightforward index of gender role orientation.3 In-
aR2 = .15. bR2 = .28. cR2 = .29. dR2 = .09. eR2 = .13. fR2 = .14. gR2 = .17
deed, attitudinal questionnaires for measuring gender
(all ps < .001).
role orientation in children, such as the CSRI, have
*p < .05.
been criticized because items are not comparable in
terms of desirability (e.g., feminine items may be per-
boys’ toys and activities was negatively associated with
ceived as less desirable than masculine items) or rather
fear; a preference for girls’ toys and activities was posi-
difficult to comprehend and hence lead to different in-
tively linked to fear.
terpretations (Bigler, 1997).
To examine which gender role variable was most
Third, significant correlations between gender role
predictive of fear and anxiety, a further regression anal-
orientation and fear and anxiety were predominantly
ysis was conducted in which biological sex, CSRI, and
present in girls. This suggests that gender role orienta-
TAPQ scores were entered simultaneously into the
tion plays a more important role in the expression of
equation. This analysis showed that only toy and activ-
ity preferences accounted for a significant proportion
3Additional analyses performed on this data yielded some evi-
of the variance. Other regression analyses predicting
dence for this idea. That is, when controlling for relevant TAPQ
children’s self-reported anxiety and parent reports of
scores, boys and girls no longer differed with regard to CSRI mascu-
children’s fear yielded highly similar results. Toy and
linity and femininity scores, both Fs(1, 205) ? .12. In contrast, when
activity preferences, in particular a preference for girls’
controlling for CSRI scores, boys and girls were still significantly
toys and activities, appeared to be the most important
distinct with respect to TAPQ preference for boys’ toys and activi-
ties, F(1, 205) = 88.88, p < .001, and preference for girls’ toys and ac-
predictor of children’s fear and anxiety levels (see
tivities, F(1, 205) = 81.72, p < .001. This indicates that the TAPQ has
Table 3).
better discriminant validity than the CSRI.
330

---

GENDER ROLE AND FEAR–ANXIETY
fear and anxiety in girls than in boys. Finally, gender
contribution of gender and gender role orientation to
role orientation was more substantially connected to
children’s fear and anxiety levels was rather modest
fear than to anxiety. This probably has to do with the
(with percentages of explained variance varying be-
fact that avoidance behavior plays a key role in the con-
tween 9% and 29%), and so additional factors are
tinuation of fears (e.g., Muris & Merckelbach, 2001).
likely involved in the manifestation of childhood fear
Obviously, a high feminine and a low masculine gen-
and anxiety. Fifth, about one quarter of the children
der role orientation have a sustaining effect on avoid-
and parents in the initial sample decided not to partici-
ance behavior and, in its wake, fears. Although avoid-
pate in the study. As we did not obtain data for these
ance behavior is also involved in anxiety, cognitive
nonparticipants, it is unclear to what extent the results
phenomena (i.e., catastrophic interpretations, worry)
were affected by selection bias. Finally, with an aver-
play a more prominent role in the maintenance of such
age age of 11 years, one could argue that the children in
symptoms (Eysenck, 1992), and it is plausible that
the study were still relatively young. The inclusion of
gender role orientation has less impact on such covert
adolescents would have made it possible to test Gins-
processes.
burg and Silverman’s (2000) developmental hypothe-
Interestingly, this study found that the child’s sex
sis that identification with gender roles intensifies as a
had minimal influence on fear and anxiety scores,
function of pubertal status, which results in a different
whereas gender role orientation (in particular toy pref-
pattern of relations between gender role and fear and
erence) was relatively important in determining these
anxiety for older compared to younger youth. Clearly,
negative emotions. More specifically, regression anal-
this issue warrants further research.
yses showed that the direct connection between child’s
sex, on the one hand, and fear and anxiety, on the other
hand, disappeared when all gender role orientation
variables were entered in the equation. This pattern of
References
results suggests a mediational model in which biologi-
American Psychiatric Association. (1994). Diagnostic and statisti-
cal gender (at least to some extent) is related to the ac-
cal manual of mental disorders (4th ed.). Washington, DC: Au-
companying gender role orientation, which in turn is
thor.
associated with levels of fear and anxiety. Thus, the
Bem, S. L. (1981). Gender schema theory: A cognitive account of
data seem to indicate that the gender role orientation—
sex typing. Psychological Review, 88, 354–364.
not the child’s sex—is involved in the manifestation of
Bigler, R. S. (1997). Conceptual and methodological issues in the
measurement of children’s sex typing. Psychology of Women
these negative emotions (see, for a similar conclusion,
Quarterly, 21, 53–69.
Brody et al., 1990).
Boldizar, J. P. (1991). Assessing sex typing and androgyny in chil-
Parent-report data largely confirmed the results ob-
dren: The Children’s Sex Role Inventory. Developmental Psy-
tained with the child-report data. This was not surpris-
chology, 27, 505–515.
ing, given that parent–child agreement appeared to be
Brody, L. R., Hay, D. H., & Vandewater, E. (1990). Gender, gender
role identity, and children’s reported feelings toward the same
acceptable for various measures, with rs ranging from
and opposite sex. Sex Roles, 23, 363–387.
.57 to .74. Yet, results were generally weaker for the
Chorpita, B. F., Yim, L., Moffitt, C., Umemoto, L. A., & Francis, S.
parent-report data. Moreover, effects of sex and gender
E. (2000). Assessment of symptoms of DSM–IV anxiety and
role orientation were only present for fear and not for
depression in children: A revised child anxiety and depression
anxiety. This has probably to do with the fact that anxi-
scale. Behaviour Research and Therapy, 38, 835–855.
Eysenck, M. W. (1992). Anxiety: The cognitive perspective. Hills-
ety is more difficult to observe than fear (e.g., Stallings
dale, NJ: Lawrence Erlbaum Associates, Inc.
& March, 1995). More precisely, whereas children’s
Ginsburg, G. S., & Silverman, W. K. (2000). Gender role orientation
fear is, at least to some extent, visible to parents, anxi-
and fearfulness in children with anxiety disorders. Journal of
ety symptoms such as worry and somatic sensations re-
Anxiety Disorders, 14, 57–67.
main largely hidden when the child does not communi-
Lewinsohn, P. M., Gotlib, I. H., Lewinsohn, M., Seeley, J. R., & Al-
len, N. B. (1998). Gender differences in anxiety disorders and
cate about them.
anxiety symptoms in adolescents. Journal of Abnormal Psy-
This study suffers from a number of limitations.
chology, 107, 109–117.
First, the study solely relied on questionnaires for mea-
Muris, P., Bodden, D., Merckelbach, H., Ollendick, T. H., & King, N.
suring fear and anxiety. Second, the study was cor-
J. (2003). Fear of the beast: A prospective study on the effects of
relational in nature, and hence no conclusions about
negative information on childhood fear. Behaviour Research
and Therapy, 41, 195–208.
cause–effect relations can be drawn. Third, the study
Muris, P., Meesters, C., & Schouten, E. (2002). A brief questionnaire
relied on a sample of nonclinical children.4 Fourth, the
of DSM–IV defined anxiety and depression symptoms among
results of the regression analyses indicated that the
children. Clinical Psychology and Psychotherapy, 9, 430–442.
Muris, P., & Merckelbach, H. (2001). The etiology of childhood spe-
cific phobias: A multifactorial model. In M. W. Vasey & M.
4A comparison with normative means as provided by Muris et al.
Dadds (Eds.), The developmental psychopathology of anxiety
(2002) indicated that 10.0% of the children (6.8% of the boys, 13.2%
(pp. 355–385). New York: Oxford University Press.
of the girls) in this sample displayed RCADS scores in the clinical
Muris, P., Merckelbach, H., Ollendick, T. H., King, N. J., & Bogie,
range.
N. (2002). Three traditional and three new childhood anxiety
331

---

MURIS, MEESTERS, KNOOPS
questionnaires: Their reliability and validity in a normal adoles-
gender and friendship status. Journal of Abnormal Child Psy-
cent sample. Behaviour Research and Therapy, 40, 753–772.
chology, 23, 439–452.
Muris, P., & Ollendick, T. H. (2002). The assessment of contempo-
Perry, L. C., Perry, D. G., & Hynes, J. (1990). The relation of sex-typ-
rary fears in adolescents using a modified version of the Fear
ing to the outcomes children expect for sex-typed behavior. Un-
Survey Schedule for Children–Revised. Journal of Anxiety Dis-
published manuscript, Florida Atlantic University, Boca Raton.
orders, 16, 567–584.
Spence, S. H. (1998). A measure of anxiety symptoms among chil-
Ollendick, T. H. (1983). Reliability and validity of the Revised Fear
dren. Behaviour Research and Therapy, 36, 545–566.
Survey Schedule for Children (FSSC–R). Behaviour Research
Stallings, P., & March, J. S. (1995). Assessment. In J. S. March (Ed.),
and Therapy, 21, 685–692.
Anxiety disorders in children and adolescents (pp. 125–147).
Ollendick, T. H., King, N. J., & Muris, P. (2002). Fears and phobi-
New York: Guilford.
as in children: Phenomenology, epidemiology, and aetiology.
Child and Adolescent Mental Health, 7, 98–106.
Ollendick, T. H., Yang, B, Dong, Q., Xia, Y., & Lin, L. (1995). Per-
Received November 25, 2003
ceptions of fear in other children and adolescents: The role of
Accepted July 14, 2004
332

---