Faculty of Social and Behavioural Sciences
Graduate School of Childhood Development and Education
Role models are overrated?
Masculinity and femininity personality characteristics, and
psychological well-being of adolescents in single-mother and
traditional father-mother families
Research Master Child Development and Education Thesis 1
Joske Spetter
dr. H.M.W. Bos and dr. B.J.H. Zijlstra 02-07-14
2 Abstract
This study explored the importance of feminization of society in explaining adolescents’ problem behaviour for those growing up in single-mother families with the following research questions: (1) Are there differences between adolescents in father-mother and single-mother families on gender role traits and psychological well-being, and (2) Are there differences between family types in (possible) associations between gender role traits and psychological well-being. Data of 1,337 adolescents (single-mother families: 121, traditional father-mother families: 1,216) from Amsterdam, the Netherlands, (agemean = 14.6 years)(male: 685) were collected by means of self-reports. The results do not corroborate the theory of feminization of the society. Associations between some of the psychosocial well-being variables and gender role traits were found. Interaction between family type and gender role traits was only significant for hashish use. This means that the focus, which is currently on the unfairly divided gender roles and the harm it has on boys, can be turned to other reasons why boys do not as well as girls in high school.
Keywords: adolescence, masculinity/femininity characteristics, psychosocial well-being,
3 Masculinity and femininity personality characteristics, and psychosocial well-being of
adolescents in single-mother and traditional father-mother families
“[I]t’s a bad time to be a boy in America.”1 “[…] we have a genuinely worrisome gap, with boys well behind girls. It is this gap that should concern educators, parents, school boards, and legislators. Engagement with school is perhaps the single most important predictor of academic success.”2 In her book, Hoff Sommers presents her concerns on feminization of society and the effects it has on boys. The feminization of society is the concern that there is a decrease in the number of male teachers in primary and secondary education, and that more and more children grow up in female headed, single-parent households. According to Timmerman and Van Essen (2004), feminization of society is a recurrent view among school board members, school and university directors, researchers and the media, and they see it as a threat for the well-being and gender development of boys (Timmerman & Van Essen, 2004).
Hoff Sommers states that there are two signs in modern society for the first concern of the feminization of society: more female teachers and the preference of the female teachers for quiet doll play and girls’ behaviour over the rough-and-tumble play and boys’ behaviour. This will stop boys from being boys and thus threaten their well-being and gender
development (Hoff Sommers, 2000). This concern for the rapid increase of female teachers and the decrease of male teachers and its consequences was already present in the nineteenth century (Timmerman & Van Essen, 2004). Charles Eliot for example warned about the incapability of female teachers for educating adolescent boys (Blair, 1979, in Timmerman & Van Essen, 2004). Around 1900, the objections really started to have a clear common ground. Schools were better equipped for girls, and the media spoke of “the women peril” (Ayres, 1905, & Weiler, 1989, in Timmerman & Van Essen, 2004). However, keeping in mind the
1
Hoff Sommers, 2000, p.13. 2
4 recurring concern on the topic in over a century, remarkable few researchers have focussed their research on this.
Also, children spent a lot of their childhood at (pre-)school. According to Tavecchio (2008), 99% of the teachers in preschool and 85% of the teachers in primary school are female. This means that during the day, children spend the majority of their time with female adults. According to Delfos (2004), this feminization in the environments the children spent the majority of their time, is especially a problem in the school environment. She addresses four disadvantages of the feminization in primary education. The first disadvantage is that the child will be viewed from only one gender-perspective; both gender-perspectives are thought to complement each other. Secondly, the sex roles the child will see during the day are mostly female, showing how to be a female. Third, the activities that promote technical challenges and exploring the world will not be offered as much as with male teachers. Fourth, the typical boy way of expressing problems will be reacted to with resistance, and thus perceived by boys as problem and stress increasing (Delfos, 2004).
A second concern for the feminization of society is the increase of single-mother families. In the Netherlands, between 2000 and 2012 for example, the increase of single-parent households was 1.33%, and the increase of single-mother households was 1.29%, compared with an increase of 1.04% of mother-father households (CBS, 2012). As a result, the focus of research has shifted more and more to the possible consequences of growing up in a single-parent family without a gender appropriate role model.
In the family home, fathers begin playing an increasing role in the upbringing of their children (Tavecchio, 2008). This is also due to the fact that research shows that fathers are a unique factor in the upbringing of their children. The involvement of fathers in the upbringing of their children results in more adolescents having an appropriate male role model at home. Nevertheless, not all fathers play an increasing role in the upbringing of their children,
5 especially in single-mother families. If children grow up in single-mother families without an involved father-figure, or appropriate male role model, children will only have a female as a role model until adolescence, at school ánd at home.
One of the theories on which the feminization of society is based is the social learning theory. The social learning theory explains the acquisition of new behaviour through
observation of the behaviour of others, or through direct experience. Bandura (1971) states that almost all behaviours displayed by a person are learned. Role modelling plays an important role in this process, because seeing a person doing a behaviour also provides a mental image of the behaviour: It is much harder to form an image or to correctly interpret social cues and descriptions of behaviour than it is while watching a role model displaying the behaviour. This can also help to correct and to perfect behaviour (Bandura, 1971). Acquiring new behaviour is accelerated if an appropriate role model will be used. An important factor in this process is associational preferences for the role model: A person who associates itself with the role model will observe more thoroughly and acquire the behaviour quicker, compared to a person with a non-associated role model (Bandura, 1971). This process is similar with the role modelling of gender roles (Bussey & Bandura, 1999), in which teachers and parents play an active role.
If the social learning theory was to be linked with the theory of Hoff Sommers, it says that without proper male role models in the classroom, boys do not know how to behave themselves in an appropriate way. This would mean that female behaviour would be more promoted than male behaviour, and this will go against the nature of boys being boys. Also, girls would be favoured over boys and this would have consequences for their performances and attitudes towards school. However, this view of Sommers is not based on research. Veendrick, Tavecchio, and Doornendal (2004) concluded that the feminization of education in The Netherlands is only seen in primary school. More importantly for this notion is that
6 boys and girls did not appear to differ by more than one point on their grades at the end of primary school. Also, no evidence was found that a ‘girl standard’ was enforced upon the boys, which could have had a negative effect on their well-being and their performances and attitudes towards school. So, no differences were found in primary school. Researchers however did find differences among boys and girls in vocational school on their grades and performances (Veendrick, Tavecchio, & Doornendal, 2004). Except, this could not be
attributed to the effect of the feminization in primary school, as has already been argued here. Neither could this be the result of feminization in vocational school, because only a third of the vocational school teachers are female (Veendrick, Tavecchio, & Doornendal, 2004). The conclusion made by Sionders (2002) based on a study done in Flanders, indicates that no differences exist in the gender of the teacher for the grades of the students. In math education for example, having a female teacher did not result in higher grades for the female adolescents in having an appropriate role model, neither did it make a difference for the grades of the male adolescents. To conclude, except for some later unfounded studies, all studies show no results corresponding to the theories on feminization of education (Sionders, 2002;
Timmerman & Van Essen, 2004).
In line with the (unfounded) concerns of the feminization of education, the widely held belief is that the feminization of society: the single-parent household, and in particular the single-mother household, has a negative effect on the well-being and gender development of boys. The theory is that the gender of the parent does effect the well-being of the child, and that it needs both a mother and a father (Biblarz & Stacey, 2010). The results of the study of Biblarz and Stacey (2010) support this: a two-parent household, with compatible parents, seems to do better than a single-parent household. Whether the parents are male or female however, does not have an effect on this outcome, nor does it have an effect on the well-being of the children (Biblarz & Stacey, 2010). The results of the study of Bos, Goldberg, Van
7 Gelderen, and Gartrell (2012) support this. Their research on planned lesbian families shows that no differences were found between adolescents with or without a male role model on femininity and masculinity personality characteristics. However, as the researchers state in their discussion, their research was with a specific group, lesbian families. Further research is needed to be able to generalize the results.
The present study focusses on the question whether adolescents from single-mother families scored differently on feminine and masculine personality characteristics and psychosocial well-being than adolescents from traditional father-mother families. Also, we examined differences between family types in (possible) associations between feminine and masculine personality characteristics and psychological well-being.
Method Participants
The 1,518 adolescents who participated in this study reside in Amsterdam, the
Netherlands. They were recruited to participate in a larger study on adolescent psychological health, relationships, and school experiences.
A question about the kind of family structures the adolescents were living in was asked in the survey, and several different family types were reported by the adolescents. The current study is focussed on single-mother families and traditional father-mother families, therefore, only the data of the adolescents in single-mother families (n = 121) and traditional father-mother families (n = 1,216) were used. Data of adolescents who reported to live in a different kind of family were excluded (n = 181, e.g. foster family or step-parent family). As a result, the sample consisted of 1,337 participants, with an age range of 11 to 18 years old (M = 14.6, SD = 1.04), of which 51.2% (n = 685) was male.
The majority of the participants (94.2%) were born in the Netherlands, 1.4% was born in another Western country, and 4.1% was born in a non-Western country. The adolescents
8 were divided over three different schools with different education levels: VMBO
vocational school) (24.0%), HAVO (general secondary school) (27.1%) and VWO (pre-university school) (48.9%), and they were in their first- (27.2%), second- (28.0%), or third-year (44.5%).
Sampling procedures
Data of the adolescents were gathered during the school year of 2009-2010.
Adolescents were recruited with the help of the school principals of the secondary schools. The school principals of thirty-two secondary schools were invited to participate in the study and received a letter of the University of Amsterdam. In this letter the schools were informed about the procedure of the survey. Meaning that if they participated, the students were asked to fill in the questionnaire during class. Schools were also asked, when they were willing to participate, to deliver a letter to the parent(s) of the adolescents. The overall amount of the participants was underage, which meant that in accordance with the Dutch ethical guidelines, parental consent was needed for the adolescents’ participation. The parents received a letter that described the purpose and nature of the study, and asked the parents to sign and return an enclosed form if they did not want their child to participate (passive consent). The parents of thirty-eight students did not allow their children to participate in the study. Reasons for refusal were not required to include, therefore, the reasons are unknown.
No significant differences were found in the demographic characteristics between the adolescents in the single-mother and traditional father-mother families on age, gender, and class (see Table 1). However, significant differences were found on ethnicity and the education level. Adolescents from single-mother families had more often a non-Western ethnicity than adolescents from traditional father-mother families. The pre-vocational
education level (VMBO) is overrepresented and the pre-university education level (VWO) is underrepresented in single-mother families.
9
Measures
The data were gathered by means of adolescent reports on demographic
characteristics, masculinity and femininity personality characteristics, psychosocial well-being (behavioural and emotional problems, self-esteem, and social interaction anxiety), and substance use.
Masculinity and femininity personality characteristics. Masculine and feminine
personality characteristics were measured with a short version of the Bem Sex Role Inventory (BSRI, 1974). This is a 15-item self-report questionnaire which intends to independently assess the adolescent’s self-report on the possession of socially desirable, stereotypical feminine and masculine personality characteristics (Bem, 1974). The masculinity subscale comprises of eight items (e.g. “superior”), and the femininity subscale comprises of 7 items (e.g. “gentle”). For each item the adolescent was asked to indicate whether a characteristic would apply to their personality (1 = almost never and 7 = always). The Cronbach’s alpha’s for the subscales in the present study were .79 (masculinity) and .85 (femininity).
Psychosocial well-being of the adolescent. To measure the psychosocial well-being
in this study we focussed on several aspects of psychosocial well-being, namely behavioural and emotional problems, self-esteem, and social interaction anxiety.
Behavioural and emotional problems. To assess the behavioural and emotional
problems in our participants we used a short version of the Youth Self Report (YSR), which involved a 77-item self-report questionnaire. The adolescents answered the questions regarding their present condition and their condition during the previous six months (Achenbach, 2001).
The YSR comprises of several syndrome-scales, and the following scales were used in the present study: Anxious/Depressed (16 items; e.g. ‘I feel lonely’), Withdrawn/Depressed (7 items; e.g. ‘I refuse to talk’), Somatic Complaints (9 items; e.g. ‘I feel tired’), Rule-Breaking
10 Behaviour (11 items; e.g. ‘I use alcohol or drugs’), Aggressive Behaviour (19 items; e.g. ‘I am mean to others’), and Conduct Problems (15 items; e.g. ‘I disobey at school and at home’). For all 77 items, the adolescents could choose one of three answers categories (0 =
strongly disagree and 2 = strongly agree or often). The Cronbach’s alpha’s for the
syndrome-scales in the present study were .88 (Anxious/Depressed), .69 (Withdrawn/Depressed), .81 (Somatic Complaints), .69 (Rule-Breaking Behaviour), .82 (Aggressive Behaviour), and .76 (Conduct Problems).
Self-esteem. Adolescents’ self-esteem was measured with the Rosenberg Self-Esteem
Scale (RSES) (Rosenberg, 1965). This instrument consists of 10 statements (e.g. “On the whole, I am satisfied with myself”), with answer categories ranging from 1 (Strongly
disagree) up to 4 (Strongly agree). The Cronbach’s alpha in the present study was .53, which
is low, but sufficient.
Social anxiety. Social anxiety was measured with a short version of the Social
Interaction Anxiety Scale (SIAS) (Mattick & Clarke 1998). This 10-item self-report
questionnaire includes statements such as “I have difficulty making eye contact with others”, or “I feel tense if I am alone with just one other person”. For each item the adolescents were asked how often they were encountered with certain thoughts, feeling and behaviour,
described in the statements (1 = not at all characteristic or true of me and 5 = extremely
characteristic or true of me). The Cronbach’s alpha in the present study was .68.
Substance use. Substance use of the adolescents was measured with three questions on
the use of tobacco, alcohol, and hashish. The questionnaire consisted of general questions, or questions about the last four weeks or last six months. The questions were derived from a large survey carried out by the Trimbos Institute on substance among Dutch adolescents (Monshouwer, Smit, De Graaf, Van Os, & Vollebergh, 2005). For tobacco, the question was asked: “How many cigarettes, on average, did you smoke the last 4 weeks?”. The answer
11 categories ranged from 0 “I have never smoked” to 7 “More than 20 cigarettes per day”. For alcohol the question was “When you drink alcohol, how much do you usually drink?” The answer categories for this question were: 0 “I do not drink alcohol” to 5 “Over more than six drinks”. For hashish, the question was asked “How many times did you smoke hashish in the last 6 months?” The answer categories ranged from 0 “Zero times” to 10 “Over more than 24 times”.
Analyses
For the first aim, three 2 (family type: 1 = traditional father-mother family, 2 = single-mother family) by 2 (gender: 1 = boy, 2 = female) between-subjects multivariate analysis of
covariance’s (MANCOVA) were carried out. The first MANCOVA was performed on two dependent variables: femininity and masculinity. The second MANCOVA was performed on eight dependent variables of behavioural and emotional problems, reflecting the
psychological well-being of the adolescents: Anxious/depressed, withdrawn/depressed, somatic complaints, rule-breaking behaviour, aggressive behaviour, and conduct problems. Self-esteem, social anxiety, and the three variables of substance use (cigarette, alcohol, and hashish use) were the dependent variables in the third MANCOVA.
For all MANCOVA’s, education level and ethnicity were used as covariates: As described above, family types differed on these two demographic characteristics. To be sure that possible differences between adolescents in both family types on the dependent variables are not confounded by ethnicity and education, we entered these variables as covariate in the multivariate analyses.
To assess whether gender role traits (femininity and masculinity) are related to psychological well-being (behavioural and emotional problems, self-esteem, social anxiety and substance use) and whether these associations are different for adolescents in single-mother families and traditional father-single-mother families, series of hierarchical regression
12 analyses were conducted (for each psychosocial variable one). In each analyses gender and family type were entered in Step 1. In Step 2 masculinity and femininity were added, and in Step 3 the interaction between masculinity and family type, and the interaction between femininity and family type.
Results
Adolescents’ Feminine and Masculine Personality Characteristics
We started with examining the possibility of differences between adolescents in single-mother and traditional father-mother families on feminine and masculine personality characteristics. Table 2 presents the means and standard deviations for the BSRI feminine and masculine personality characteristics subscale scores separately for adolescents in single-mother and traditional father-single-mother families, and separately by gender. The two (family structure) by two (adolescents’ gender) MANCOVA on feminine and masculine personality characteristics showed a significant multivariate main effect for gender (Wilkes λ = .95, F (2, 1187) = 31.79, p < .001, η2= .05); however, the main effect for family type was not significant (Wilkes λ = 1.00, F (2, 1187) = 1.00, p = .368, η2= .00), nor the family type x gender
interaction (Wilkes λ = 1.00, F (2, 1187) = .25, p = .779, η2= .00).
Univariate F tests indicated that the multivariate main effect of gender was localized to feminine personality characteristics (p < .001), with a higher mean score for adolescent girls.
Adolescents’ Behavioural and Emotional Problems
In this section, we examined whether family structure has an influence on the adolescents’ behavioural and emotional problems. For the analysis, we used the several subscales of the YSR of the adolescents (see Table 2) as dependent variables in a two (family structure) by two (adolescents’ gender) MANCOVA The main effect of gender (Wilkes λ =
13 .94, F (6, 1197) = 12.10, p < .001, η2= .06) was significant. The main effect of family
structure however was not significant (Wilkes λ = .99, F (6, 1197) = 1.26, p = .271, η2= .01), nor was the multivariate interaction between family structure and gender (Wilkes λ = 1.00, F (6, 1197) = .19, p = .981, η2= .00).
Univariate F tests indicated that the multivariate main effect of gender was localized to anxious/depressed (p < .001), withdrawn/depressed (p < .001), and somatic problems (p < .001) with a higher mean score on these subscales for adolescent girls. In contrast, adolescent boys showed higher scores on the subscales measuring rule-breaking behavior (p < .05) and conduct problems (p < .001).
Adolescents’ Self-Esteem, Social Interaction Anxiety and Substance Use
The influence of family structure on the adolescents’ self-esteem, social interaction anxiety and substance use was examined in this section. A two (family structure) by two (adolescents’ gender) MANCOVA was conducted with the adolescents’ self-esteem, social interaction anxiety, and cigarette, alcohol, and hashish use as dependent variables. The main effect of gender (Wilkes λ = .99, F (5, 1089) = 1.51, p = .183, η2= .01) and family structure were not significant (Wilkes λ = 1.00, F (5, 1089) = .79, p = .556, η2= .00), nor was the multivariate interaction between family structure and gender (Wilkes λ = 1.00, F (5, 1089) = .98, p = .428, η2= .00).
Associations Between Feminine and Masculine Personality Characteristics and Behavioural and Emotional Problems
In this section, the relationship between feminine and masculine personality
characteristics and behavioural and emotional problems is examined in association with two different family types and the adolescent’s gender. Table 3 presents the results of the
regression analyses, with gender and family type entered in Step 1. The BSRI feminine and masculine traits subscale scores were entered in Step 2, and BSRI feminine traits subscale
14 scores × family type and BSRI masculine traits subscale scores × family type were entered in Step 3. Separate analyses were conducted with anxious behaviour, withdrawn behaviour, somatic problems, rule-breaking behaviour, aggressive behaviour, and conduct problems as dependent variables.
On two of the six dependent variables, Step 1 in the hierarchical regression analysis was significant (Anxious Behaviour (R2 = .04) and Somatic Problems (R2 = .04)). On four of
the six variables, Step 2 was significant, namely Withdrawn Behaviour (ΔR2 = .01),
Rule-breaking Behaviour (ΔR2 = .05), Aggressive Behaviour (ΔR2 = .06), and Conduct Problems
(ΔR2 = .08). Step 3 added nothing for these four variables.
Gender was significant in Step 1 for Anxious behaviour (β = .20) and Rule-breaking Behaviour (β = -.14), and in Step 3 (and also in Step 2) for Withdrawn Behaviour (β = .09), Somatic Problems (β = .20), and Conduct Problems (β = -.09). Family type was only for Conduct Problems in Step 1 significant (β = .06).
BSRI feminine traits was significant in Step 3 for Rule-breaking Behaviour (β = -.28), Aggressive Behaviour (β = -.29), and Conduct Problems (β = -.36). BSRI masculine traits was significant in Step 3 (and also in Step 2) for Withdrawn Behaviour (β = -.12), Rule-breaking Behaviour (β = .18), Aggressive Behaviour (β = .24), and Conduct Problems (β = .20). No significant interaction-effects were in Step 3.
Associations Between Feminine and Masculine Personality Characteristics and Self-Esteem, Social Interaction Anxiety and Substance Use
The relationship between feminine and masculine personality characteristics and self-esteem, social interaction anxiety, and substance use is examined in association with two different family types and the adolescents’ gender. Table 3 presents the results of the regression analyses, with gender and family type entered in Step 1. The BSRI feminine and masculine traits subscale scores were entered in Step 2, and BSRI feminine traits subscale
15 scores × family type and BSRI masculine traits subscale scores × family type were entered in Step 3. Separate analyses were conducted with self-esteem, social interaction anxiety,
cigarette use, alcohol use and hashish use as dependent variables.
Step 1 in the hierarchical regression analysis was on two of the five variables significant (Alcohol Use (F = 4.90, R2 = .01) and Hashish Use (F = 9.40, R2 = .02)). Gender and family
type are also significant for these two variables: More male adolescents than female
adolescents smoked hashish, and more adolescents in single-mother families smoked hashish compared to adolescents in traditional father-mother families. The addition of the BSRI feminine traits is in Step 2 for all the variables significant, as is BSRI masculine traits, except for Self-Esteem. Family type is also for Hashish Use significant (β = .08).
The interaction effects of BSRI feminine traits and family type is only for Hashish Use significant in Step 3 (β = -.10), and accounts for 4 % of the variance. For the directionality of the effect simple slopes analyses were performed, using the method described by Jose (2013). The simple slopes analyses showed that the negative effects of the interaction of femininity and family type on hashish use were not significant for the traditional father-mother families (β = -.22, p = .682), nor for the single-mother families (β = -.58, p = .367). However, as can be seen the Figure 1, the difference in hashish use between the adolescents in single-mother and traditional father-mother families is larger. So, they might not be significant for a single family type, but compared with each other, the difference is significant.
Discussion
The aim of this study was to determine whether differences exist between adolescents in traditional father-mother families and adolescents in single-mother families on gender role traits and psychosocial well-being. Also, whether there are differences between family types in (possible) associations with gender role traits and psychosocial well-being.
16 The results showed that no differences exist between the two family types on gender role traits. Associations between some of the psychosocial well-being variables and gender role traits were found. Interaction between family type and gender role traits was only significant for hashish use.
These findings are similar to the findings of Bos et al. (2012). Their study showed that there are no differences in gender role traits and psychological adjustment between
adolescents of lesbian parents and traditional father-mother families.
The study of Timmerman and Van Essen (2004) gives a different view of the
feminization of education. At the end of primary education, mostly taught by female teachers, children only show slight differences on their scores on math (higher score for boys) and language (higher score for girls). The root of the problem of boys underachieving in school seems to lie in their time spent on vocational schools. However, the majority of teachers in Dutch vocational education is male (Timmerman & Van Essen, 2004; Veendrick, Tavecchio, & Doornedal, 2004). Apparently, the problem lies elsewhere. Feminization of the family home is the next stop. The increasing number of single-parent families, mostly single-mother families, has also been the subject of speculation for the ‘boys problem’. The results of this study however, contradict this . Whether children grow up in single-mother of traditional father-mother families does not seem to have an effect on their well-being or their masculinity and femininity. Of course, arguments can be made that it might possibly the combination of a single-mother family home and female primary education teacher which lead to problems in adolescents. This could only be true if these two factors alone are an indicator, but they are not. The results do not even leave room for any speculation: no results of the studies named above indicate feminization of society.
As stated above, the research done for this paper does not corroborate the theory of feminization of the society. Why were no differences found between the adolescents in the
17 two families? A first explanation could be that The Netherlands are categorized as a very feminine country (also compared to other countries in Europe), whereas The United States, where Hoff Sommers lives and the recent discussion of the ‘boys problem’ started, is categorized as a very masculine country (Hofstede, 1995). The Netherlands are even
categorized as the most feminine countries in the world (De Waij, 2009). A feminine country has a society in which emotional gender roles overlap: both men and women are expected to be gentle and humble and focused on the quality of living (Hofstede, 2002). The Netherlands being a feminine country could mean that adolescents identify themselves more with both male and female characteristics and therefor do not really differentiate between gender role models.
Another reason that could have contributed for not finding significant results is that the demographic data of the parents of the adolescents were not included in the analyses. Research done in the US showed that adolescents from single-parent families with a low social-economic status (SES), display lower a well-being (McLanahan, 1997) Third, no data exists on whether the adolescents from the single-mother families have contact with their father. If these children still see their father, they still have a male role mode. Whereas now we assumed that they do not have contact with their father and that their mother is their only role model at home.
A limitation of this study was that no comparisons could be made for adolescents in single-father families. Data on only 22 adolescents from single-father families were collected, which was not enough for external validation. Therefore, their data were not included in the study. A second limitation of this study is that no information was gathered on whether the adolescents of the single-mother families were still in contact with their father. The
assumption made in this study that the adolescents did not have contact with their father and that their mother was their only role model at home could be wrong. Another limitation is that
18 the way the data was retrieved was not multi-informant: only the adolescents were asked to fill in the questionnaire. The parents and teachers were not questioned. The role of the parents is reported from the adolescents’ perspective and the parents’ perspective could have given us more insight. Additional information could have suggested that a parent might be gender non-conform.
Another limitation of this study is that substance use and its’ use in frequency and quantity is known to be underreported by its’ users, which means that the data is most likely inaccurate. Also, the age range of the adolescents in this study is 11 to 18 years old. For some of the adolescents it is not allowed by law to smoke and drink, which means that their report on substance use is not based on their use, but on the law. And lastly, the adolescents in this study all resided in Amsterdam. Their attitudes could be different from adolescents in other cities in the Netherlands: this is not a representative sample of the Dutch population.
A strength of this study is that it made a comparison between data on adolescents from single-mother families and on adolescents from traditional father-mother families. In the Netherlands, no other study has been conducted with data on adolescents from single-mother families on feminization of the family home.
The significant finding of the interaction between family type and gender role traits on hashish use could be explained by the city the adolescents resided in. All the adolescents resided in Amsterdam at the time of the data collection. Different outcomes could have been found if data were collected in other large cities in The Netherlands, or small cities and villages.
The theoretical importance of this study is its findings that no significant differences were found for the interaction between family type and gender role traits, except for hashish use. This means that the focus which is currently on the unfairly divided gender roles and the
19 harm it has on boys, can be turned to other reasons why boys will not do as well as girls in high school.
For future research it could be interesting to compare traditional father-mother families with single-mother families, single-father families, and same-sex families. It might also be best to collect data from different cities and villages in the Netherlands, for a
representative sample of the Dutch population.
References
Achenbach, T. M. (1991). Manual for the Youth Self-Report and profile. Burlington: University of Vermont, Department of Psychiatry.
Achenbach, T. M. (2001). Youth Self Report for ages 11-18 (YSR). ASEBA, Burlington: University of Vermont. Retrieved from http://www.aseba.org/forms/ysr.pdf
Bandura, A. (1971). Social learning theory. General Learning Press, 1-46. Retrieved from http://www.jku.at/org/content/e54521/e54528/e54529/e178059/
Bandura_SocialLearningTheory_ger.pdf
Beaty, L. A. (1995). Effects of paternal absence on male adolescents’ peer relations and self- image. Adolescence, 30, 873-880. Retrieved from http://search.proquest.com/docview/ 1295890678?accountid=14615
Bem, S. L. (1974). The measurement of psychological androgyny. Journal of Consulting and
Clinical Psychology, 42, 155-162. Retrieved from http://lib.uva.nl/primo_library/
libweb/action/dlSearch.do?institution=UVA&onCampus=false&query=any,contains, medline4823550&indx=1&dym=true&lang=eng&group=GUEST&loc=adaptor, primo_central_multiple_fe&vid=UVA
20
Social Forces, 75, 1319-1339. Retrieved from http://www.heinonline.org/HOL/
Page?page=1319&handle=hein.journals%2Fjosf75&collection=journals Biblarz, T. J., & Raftery, A. E. (1999). Family structure, educational attainment, and
socioeconomic success: Rethinking the “Pathology of Matriarchy”. American
Journal of Sociology, 105, 321-365. doi:10.1086/210314
Biblarz, T. J., & Stacey, J. (2010). How does the gender of the parents matter? Journal of
Marriage and Family, 72, 3-22. doi:10.1111/j.1741-3737.2009.00678.x
Bos, H. M. W., Goldberg, N., Van Gelderen, L., & Gartrell, N. (2012). Adolescents of the U.S. National Longitudinal Lesbian Family Study: Male role models, gender role traits, and psychological adjustment. Gender & Society,26, 603-638.
doi: 10.1177/0891243212445465
Bramlett, M. D., & Blumberg, S. J. (2007). Family structure and children’s physical and mental health. Health Affairs, 26, 549-558. doi: 10.1377/hlthaff.26.2.549
Bussy, K., & Bandura, A. (1999). Social cognitive theory of gender development and
differentiation. Psychological Review, 106, 676-713. Retrieved from http://lib.uva.nl/ primo_library/libweb/action/dlSearch.do?institution=UVA&onCampus=false& query=any,contains,apa_articlesrev_106_4_676&indx=1&dym=true&lang=eng &group=GUEST&loc=adaptor,primo_central_multiple_fe&vid=UVA
Calhoun Davis, E., & Friel, L. V. (2001). Adolescent sexuality: Disentangling the effects of family structure and family context. Journal of Marriage and Family, 63, 669-681. Retrieved from http://www.jstor.org/stable/3654641?origin=JSTOR-pdf
Carver, P. R., Yunger, J. L., & Perry, D. G. (2003). Gender identity and adjustment in middle childhood. Sex Roles, 49, 95-109. doi:0360-0025/03/0800-0095/0
Centraal Bureau voor de Statistiek (CBS) (2012). Huishoudens; grootte, samenstelling, positive in het huishouden, 1 januari [Households; extent, structure, position within
21 the household, January 1st]. Retrieved from http://statline.cbs.nl/StatWeb/publication/ ?VW=T&DM=SLNL&PA=37312&D1=a&D2=a,!1-4,!6-7&HD=130522-
1040&HDR=G1&STB=T.
Collier, K. L., Bos, H. M. W., Merry, M. S., & Sandfort, T. G. M. (2013). Gender, ethnicity, religiosity, and same-sex sexual attraction and the acceptance of same-sex sexuality and gender non-conformity. Sex Roles, 68, 724-737. doi:10.1007/s11199-012-0135-5 Crozier, P. W. (2001). Forcing boys to be boys: The persecution of gender non-conforming
youth: The war against boys: How misguided feminism is harming our young men. By Christima Hoff Sommers. BC Third World LJ, 21, 123-315. Retrieved from http://www.heinonline.org/HOL/Page?page=123&handle=hein.journals%2Fbctw21 &collection=journals
De Waij, D. (2009). Zo dichtbij, maar toch zo ver weg [So close, yet so far away]. Retrieved from: http://proactivity.nl/Afbeeldingen/Leiderschapsstijlen%20NL%20BE.pdf Delfos, M. (2004). Een jongen is geen meisje [A boy is not a girl]. De Wereld van het Jonge
Kind, 31, 183-186. Retrieved from http://www.mdelfos.nl/2004-wjk-verschil.pdf
Demuth, S., & Brown, S. L. (2004). Family structure, family processes, and adolescent delinquency: The significance of parental absence versus parental gender. Journal of
Research in Crime and Delinquency, 41, 58-81. doi:10.1177/0022427803256236
Downey, D. B., Ainsworth-Darnell, J. W., & Dufur, M. J. (1998). Sex of parent and
children’s well-being in single-parent households. Journal of Marriage and Family,
60, 878-893. Retrieved from http://www.jstor.org/stable/353631?origin=JSTOR-pdf
Hawkins, D. N., Amato, P. R., & King, V. (2006). Parent-adolescent influence of parent gender and residence. Journal of Marriage and Family, 68(1), 125-136. Retrieved from http://www.jstor.org/stable/3600361?origin=JSTOR-pdf
22 Hilton, J. M., & Devall, E. L. (1998). Comparison of parenting and children’s behaviour in
single-mother, single-father, and intact families. Journal of Divorce and Remarriage,
29(3-4), 23-54. doi:10.1300/J087v29n03_02
Hilton, J. M., Desrochers, S., & Devall, E. L. (2001). Comparison of role demand,
relationships, and child functioning in single-mother, single-father, and intact families.
Journal of Divorce and Remarriage, 35(1-2), 29-56. doi:10.1300/J087v35n01_02
Hoff Sommers, C. (2000). The war against boys: How misguided feminism is harming our
young men. United States, New York: Simon & Schuster Paperbacks
Hofstede, G. (1995). Allemaal andersdenken; omgaan met cultuurverschillen [All dissident thinkers; dealing with cultural differences]. Amsterdam: Uitgeverij Contact.
Hofstede, G. (2002). Culturele diversiteit in de Nederlandse samenleving [Cultural diversity in the Dutch society]. Culturele Diversiteit, 28(5), 9-18. Retrieved from
http://wodc.nl/images/jv0205-volledige-tekst_tcm44-57424.PDF#page=10 Jose, P. E. (2013). ModGraph-I: A programme to compute cell means for the graphical
display of moderational analyses: The internet version, Version 3.0. Victoria
University of Wellington, Wellington, New Zealand. Retrieved from http://pavlov.psyc.vuw.ac.nz/paul-jose/modgraph/
MacCallum, F., & Golombok, S. (2004). Children raised in fatherless families from infancy: A follow-up of children of lesbian and single heterosexual mothers at early
adolescence. Journal of Child Psychology and Psychiatry, 45, 1407-1419. doi: 10.1111/j.1469-7610.2004.00324.x
McLanahan, S. (1997). Parent absence or poverty: Which matters more?. In G. Duncan & J. Brooks-Gunn (Eds.), Consequences of growing up poor (pp. 35–48). New York : Russell Sage Foundation.
23 use: Does onset shift to younger ages? Findings from 1988 to 2003 from the Dutch National School Survey on Substance Use. Addiction, 100, 963-970. doi:10.1111/ j.1360-0443.2005.01088.x
Pfiffner, L. J., McBurnett, K., & Rathouz, P. J. (2001). Father absence and familial antisocial characteristics. Journal of Abnormal Child Psychology, 29, 357-367. Retrieved from http://link.springer.com/content/pdf/10.1023%2FA%3A1010421301435.pdf
Powell, B., & Downey, D. B. (1997). Living in single-parent households: An investigation of the same-sex hypothesis. American Sociological Review, 62, 521-539. Retrieved from http://www.jstor.org/stable/2657424?origin=JSTOR-pdf
Rosenberg, M. (1965). The measurement of self-esteem. In M. Rosenberg (Ed.), Society and
the adolescent self-image (pp. 16-36). Princeton, New Jersey: Princeton University
Press.
Silverstein, L. B., & Auerbach, C. F. (1999). Deconstructing the essential father. American
Psychologist, 54, 397-407. Retrieved from http://lib.uva.nl/primo_library/libweb/
action/dlSearch.do?institution=UVA&onCampus=false&query=any,contains, ericEJ590650&indx=1&dym=true&lang=eng&group=GUEST&loc=adaptor, primo_central_multiple_fe&vid=UVA
Sionders, J. (2002). De gevolgen van de feminisering van het leerkrachtenberoep in het
secundair onderwijs: een empirische analyse [The effects of the feminization of the
teaching profession in secondary education: an empirical analysis]. (Dissertation) Brussel: Vakgroep Sociologie, Onderzoeksgroep TOR, Vrije Universiteit Brussel. Tavecchio, L. (2008). Seksespecifieke opvoeder-kind relaties in de kinderopvang [Gender-
specific parent-child relationships in pre-school]. Retrieved from http://www.eco3.nl/ eCache/DEF/1/11/248.html
24 Timmerman, G., & Van Essen, M. (2004). De mythe van het ‘vrouwengevaar’. Een
historiserende inventarisatie van (inter)nationaal onderzoek naar de relatie tussen feminisering en ‘jongensproblemen’ in het onderwijs [The myth of the ‘women peril’. A historical review of (inter)national studies into the relationship between
feminization and boys’ problems]. Pedagogiek, 24, 57-71. Retrieved from http:// www.pedagogiek-online.nl/index.php/pedagogiek/article/viewFile/223/222
Twenge, J. M. (1997). Changes in masculine and feminine traits over time: A meta-analysis.
Sex Roles, 36, 305-325. Retrieved from http://link.springer.com/content/pdf/
10.1007%2FBF02766650.pdf
Veendrick, L., Tavecchio, L., & Doornenbal, J. (2004). Jongens als probleem [Boys as a problem]. Pedagogiek, 24(1), 12-22. Retrieved from http://www.pedagogiek- online.nl/index.php/pedagogiek/article/viewFile/220/219
Yeung, W. J., Sandberg, J. F., Davis-Keane, P. E., & Hofferth, S. L. (2001). Children’s time with fathers in intact families. Journal of Marriage and Family, 63(1), 136-154. Retrieved from http://www.jstor.org/stable/3599964?origin=JSTOR-pdf
25
Appendix A
Table 1
Descriptive Statistics for Adolescent Demographic Characteristics by Family Type
Measure Family Type χ2 df p-value
Single-Mother Family Traditional Father-Mother Family Age 5.18 7 .64 M 14.72 (112) 14.54 (1133) SD 1.04 1.04 Gender, %(n) 0.15 1 .70 Boys 52.9 (64) 51.1 (621) Girls 47.1 (57) 48.9 (595) Ethnicity, %(n) 5.77 1 .02 Dutch / other Western country 91.7 (111) 96.3 (1168) Non-Western country 8.3 (10) 3.7 (45) Education level, %(n) 12.89 2 .00 Pre-vocational 36.7 (40) 22.8 (252) General secondary 28.4 (31) 26.9 (298) Pre-university 34.9 (38) 50.3 (556) Class, %(n) 0.07 2 .97 1st year 28.1 (34) 27.1 (329) 2nd year 27.3 (33) 28.2 (342) 3rd year 44.6 (54) 44.6 (541)
Table 2
Descriptive Statistics for Adolescent Feminine and Masculine traits and Psychosocial Well-Being by Family Type and Gender
Single Mother Family
Traditional Father Mother Family
Girls Boys F Value
Family Type
F Value Gender BSRI Feminine and
Masculine Traits Femininity 0.65 40.71*** M (SD) 4.71 (1.21) 4.86 (1.05) 5.22 (0.90) 4.51 (1.09) Masculinity 0 .27 0.30 M (SD) 4.51 (1.07) 4.53 (1.01) 4.51 (1.00) 4.55 (1.03) Psychosocial well-being Anxious behaviour 0.60 20.92*** M (SD) 0.28 (0.34) 0.31 (0.32) 0.37 (0.36) 0.25 (0.26) Withdrawn behaviour 0.04 7.95** M (SD) 0.37 (0.33) 0.37 (0.33) 0.41 (0.34) 0.33 (0.32) Somatic problems 0.00 23.67*** M (SD) 0.33 (0.36) 0.33 (0.36) 0.41 (0.38) 0.26 (0.31) Rule-breaking behaviour 1.40 4.09* M (SD) 0.35 (0.29) 0.31 (0.24) 0.29 (0.23) 0.35 (0.26) Aggressive behaviour 1.48 0.02 M (SD) 0.41 (0.30) 0.38 (0.25) 0.38 (0.24) 0.39 (0.27) Conduct problems 3.33 12.32*** M (SD) 0.31 (0.26) 0.26 (0.22) 0.22 (0.19) 0.31 (0.25) Self-esteem 0.33 2.16 M (SD) 2.84 (0.40) 2.82 (0.40) 2.84 (0.40) 2.81 (0.41) Social Anxiety 0.47 0.00 M (SD) 1.88 (0.54) 1.92 (0.57) 1.94 (0.55) 1.90 (0.59) Cigarette Use 0.00 1.15 M (SD) 0.46 (1.22) 0.41 (1.09) 0.37 (0.99) 0.45 (1.20) Alcohol Use 0.61 0.00 M (SD) 1.00 (1.51) 0.81 (1.34) 0.76 (1.26) 0.90 (1.44) Hashish Use 1.72 4.14* M (SD) 0.42 (1.57) 0.21 (1.10) 0.11 (0.79) 0.34 (1.42) * ρ < .05, ** ρ < .01, *** ρ < .001
27 Table 3
Hierarchical Regression Analyses with Behavioural and Emotional Problems as Dependent Variables
Anxious Behaviour Withdrawn Behaviour Somatic Problems Rule-breaking Behaviour Aggressive Behaviour Conduct Problems B SE β B SE β B SE β B SE β B SE β B SE β Step 1 Gender .12 .02 .20*** .07 .02 .11*** .15 .02 .21*** -.07 .01 -.14*** -.02 .01 -.03 -.09 .01 -.21*** Family Type -.02 .03 -.02 -.00 .03 -.00 -.01 .04 -.01 .04 .02 .04 .04 .02 .04 .05 .02 .06* R2 .04 .01 .04 .02 .00 .05 F 26.01*** 7.72*** 29.34*** 13.66*** 1.77 32.91*** Step 2 Gender .11 .02 .17 .06 .02 .09** .14 .02 .21** -.02 .01 -.05 .03 .02 .06* -.04 .01 -.09** Family Type -.02 .03 -.02 -.01 .03 -.01 -.01 .04 -.01 .03 .02 .03 .03 .02 .03 .04 .02 .05 BSRI feminine traits .02 .01 .07 .01 .01 .05 .00 .01 .01 -.06 .01 -.28*** -.07 .01 -.29*** -.07 .01 -.36*** BSRI masculine traits -.03 .01 -.06 -.04 .01 -.12*** .01 .01 .03 .04 .01 .18*** .06 .01 .24*** .04 .01 .20***
R2 .04 .02 .04 .07 .06 .13 F 14.20*** 7.25*** 15.00*** 24.01*** 21.38*** 47.47*** ΔR2 .00 .01 .00 .05 .06 .08 ΔF 2.33 6.72** .67 33.69*** 40.88*** 59.10*** Step 3 Gender .12 .02 .18 .06 .02 .09** .14 .02 .20*** -.02 .01 -.04 .03 .02 .06* -.04 .01 -.09** Family Type -.01 .03 -.01 .01 .03 .01 -.01 .04 -.01 .03 .02 .04 .03 .02 .03 .04 .02 .05 BSRI feminine traits .02 .01 .06 .01 .01 .04 .00 .01 .01 -.06 .01 -.28*** -.07 .01 -.29*** -.07 .01 -.36*** BSRI masculine traits -.03 .01 -.06 -.04 .01 -.12** .01 .01 .03 .04 .01 .19*** .06 .01 .24*** .04 .01 .20*** BSRI feminine traits ×
family type
.06 .03 .07 .07 .03 .07* .00 .03 .00 .01 .02 .01 .01 .02 .01 -.01 .02 -.01 BSRI masculine traits ×
family type -.05 .03 -.05 .02 .03 .01 .05 .03 .04 -.03 .03 -.04 -.01 .02 -.01 -.00 .02 -.00 R2 .05 .03 .05 .07 .06 .13 F 10.32 6.30*** 10.49*** 16.29 14.26*** 31.62*** ΔR2 .00 .01 .00 .00 .00 .00 ΔF 2.50 4.33 1.45 .85 .08 .06 * ρ < .05, ** ρ < .01, *** ρ < .001
28 Table 3 continued
Hierarchical Regression Analyses with Psychosocial Well-being and Substance Use as Dependent Variables
Psychosocial Well-being Substance Use
Self-Esteem Social Anxiety Cigarettes Use Alcohol Use Hashish Use
B SE β B SE β B SE β B SE β B SE β Step 1 Gender .04 .02 .05 .06 .03 .05 -.09 .06 -.04 -.17 .07 -.06* -.21 .07 -.09** Family Type .01 .04 .01 -.06 .06 -.03 .14 .11 .04 .27 .13 .06* .35 .12 .08** R2 .00 .00 .00 .01 .02 F 1.57 2.25 1.97 4.90** 9.40*** Step 2 Gender .00 .03 .00 .02 .04 .02 -.01 .07 -.01 -.07 .08 -.03 -.09 .07 -.04 Family Type .01 .04 .01 -.05 .06 -.02 .12 .11 .03 .24 .13 .05 .34 .12 .08** BSRI feminine traits .05 .01 .14*** .05 .02 .01** -.11 .04 -.11** -.14 .04 -.12** -.17 .04 -.15*** BSRI masculine traits .00 .01 .00 -.06 .02 -.10** .11 .04 .11** .23 .04 .18*** .09 .04 .08*
R2 .02 .01 .01 .03 .03 F 6.54*** 3.73** 4.02** 9.90*** 9.40*** ΔR2 .02 .01 .01 .02 .01 ΔF 11.48*** 5.19** 6.06** 14.79*** 9.28*** Step 3 Gender .00 .03 .00 .20 .04 .02 -.01 .07 -.00 -.07 .08 -.03 -.10 .07 -.04 Family Type .01 .04 .01 -.05 .06 -.02 .13 .11 .03 .23 .13 .05 .32 .12 .08** BSRI feminine traits .05 .01 .14*** .06 .02 .10** -.12 .04 -.11** -.14 .04 -.11** -.16 .04 -.14*** BSRI masculine traits .00 .01 .01 -.06 .02 -.11** .11 .04 -.11** .23 .04 .18*** .08 .04 .07* BSRI feminine traits ×
family type
.05 .04 .04 .00 .06 .00 .03 .10 .01 -.13 .12 -.03 -.34 .11 -.10** BSRI masculine traits ×
family type .06 .04 .05 .10 .06 .05 -.15 .11 -.04 -.00 .13 -.00 .06 .12 .02 R2 .03 .02 .01 .03 .04 F 5.58 3.13** 3.00** 6.83*** 8.06 ΔR2 .01 .00 .00 .00 .01 ΔF 3.61* 1.92 .96 .69 5.25** * ρ < .05, ** ρ < .01, *** ρ < .001
29
Figure 1. Interaction effect of femininity and family type (with single-mother families in purple and traditional father-mother families in blue) on
hashish use of adolescents. -0,2 -0,1 0 0,1 0,2 0,3 0,4 0,5 0,6 0,7
low med high
H ash ish U se Femininity Family Type Single-Mot Traditiona
