Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Early childhood gender socialization : subtitle implications of sex-typed toys and play on adulthood.. 2005

You don't seem to have a PDF reader installed, try download the pdf

Item Metadata

Download

Media
ubc_2005-0305.pdf [ 3.8MB ]
Metadata
JSON: 1.0091959.json
JSON-LD: 1.0091959+ld.json
RDF/XML (Pretty): 1.0091959.xml
RDF/JSON: 1.0091959+rdf.json
Turtle: 1.0091959+rdf-turtle.txt
N-Triples: 1.0091959+rdf-ntriples.txt
Citation
1.0091959.ris

Full Text

E A R L Y C H I L D H O O D G E N D E R S O C I A L I Z A T I O N : I M P L I C A T I O N S O F S E X - T Y P E D T O Y S A N D P L A Y O N A D U L T H O O D C A R E E R O U T C O M E S B y K A T H E R I N E R H O D E S B . A . P sycho logy U n i v e r s i t y o f B r i t i s h C o l u m b i a , 2002 M A S T E R O F A R T S In T H E F A C U L T Y O F G R A D U A T E S T U D I E S ( F A M I L Y S T U D I E S ) U N I V E R S I T Y O F B R I T I S H C O L U M B I A M a r c h 2005 © Kather ine Rhodes, 2005 11 Abstract Sex-typed childhood toys are used as indicators o f parent-child gender socialization. Sex-typed toys indicate gender roles and expectations parents expose their young children to. The present study's purpose is to test whether early chi ldhood gender socialization is related to adulthood academic, career and family expectations. Gender role ideologies and gender schemas are hypothesized to mediate these relationships. T w o hundred and seventy-seven university students volunteered to complete surveys. The surveys measured the frequency o f play in feminine and masculine toys and games, neutral toys, and wi th each parent. The dependent variables measure the number o f women enrolled in respondents' declared university majors (specializations), as we l l as their expected commitment to future occupation, parenting, marital, and home care roles. Expected role commitments are measured using the L i fe Ro l e Salience Scale (Amatea et a l . , 1986). Gender role ideologies are measured by using the short version o f the Attitudes Towards W o m e n Scale (Spence, Helmreich, & Stapp, 1973), and gender schemas are measured by using Bern ' s (1974) short version o f the Bern Sex Role Inventory. The results o f the study found that exposure to sex-typed toys i n early chi ldhood is related to expected commitment levels to future occupational, parenting, marital, and home care roles, as we l l as to enrollment in female dominated university majors. Gender role ideologies and gender schemas mediate more o f the relationships tested in the male sample than in the female sample. iii Table o f Contents Abs t rac t » L i s t o f Tables v L i s t o f F igures v i Introduct ion 2 Research Quest ions 3 Theore t ica l Issues o f Gender Soc ia l i za t ion 3 L e a r n i n g T h e o r y 4 Gender Schema Theory 5 Theories i n Gene ra l 6 Gender Research to Date 9 T o y s and Parental Influences 11 The Importance o f Sex-typed T o y s 12 Parental Influences 14 Career Expectat ions and Un ive r s i t y Special izat ions 18 R e p l i c a t i o n 2 2 Theore t ica l M o d e l 26 Hypotheses 26 M e t h o d s 26 Dependent Var iab les 26 Independent Var iab les 27 C o n t r o l Va r i ab l e s 29 Procedure 31 iv Data Analysis 32 Results 33 Discussion $g Limitations 65 References 70 Appendix A -75 Appendix B gg V List of Tables 34 Table 1 J Table 2 3 5 Table 3 3 8 39 Table 4... J Table 5 4 0 44 Table 6 * Table 7 4 6 47 Table 8 H / 49 Table 9 • Table 10. ••• 5 0 Table 11 5 2 Table 12 8 6 Table 13 8 7 Table 14 8 8 5 89 vi List of Figures Figure 1 24 Figure 2 53 Figure 3 54 Figure 4 55 2 Introduction Despite fluctuations i n interest, gender research has been a popular area for psychologists and sociologists, particularly in the 1980's (Ruble & Mar t in , 1998). A l l k n o w n societies have established gender roles through sex-typed labor divisions (Eagly, W o o d , & Diekman, 2000), wh ich makes this area universal and interesting to social scientists. The process o f learning and acquiring these sets o f social rules originates from the primary social unit. A s parents are one o f the most influential agents o f social information for young children (Maccoby , 1992; Schwartz & Markham, 1985), parents become a key social izing agent to gender development. The earliest example begins at birth, when sons and daughters are subjected to differential sex socialization. Parents perceive their children differently according to the sex o f the chi ld (Fagot & Leinbach, 1987). Gender and sex-typed behaviors become observable in children as young as 18 months ( O ' B r i e n & Huston, 1985; Caldera, Huston, & O ' B r i e n , 1989; Fagot, 1974). The family context provides children wi th the earliest sex role socialization (Fagot & Leinbach, 1987), and is thus a popular focal point o f gender research. Toys are used by parents and children to learn about gender rules and expectations. Toys have been used i n many studies to indicate how children learn gender (e.g., Giul iano, Popp, & Knight , 2000; O ' B r i e n & Huston, 1985; Eisenberg, W o l c h i c k , Hernandez, & Pasternack, 1985; Langlo is & D o w n s , 1980; Roopnarine, 1986; Fisher-Thompson, 1993). M u c h research has been implemented to assess the process o f how gender socialization occurs through the encouragement o f sex-typed toy play, but some questions still remain. These questions ask what the importance o f this vast body o f research is , and what the outcomes o f variations in this process mean for individuals. One significant effect o f childhood gender socialization is that women who played with masculine toys as children are more l ike ly to be athletic. Giul iano, et al . (2000) conducted a study that analyzed what toys and play groups 3 female athletes and non-athletes were exposed to in childhood. They found that, compared to female non-athletes, female athletes were more l ikely to have played with masculine toys, and to have had support and encouragement from peers and siblings. The findings o f this study describe h o w childhood gender socialization can influence adult outcomes. The Giul iano et al . (2000) study outcomes indicate the importance o f gender socialization in chi ldhood through the use o f toys is a topic needing further research. Research Questions The purpose o f this study is to examine what the possible adulthood outcomes o f chi ldhood gender socialization are. The questions for this study are as follows: D o childhood sex-typed toys affect career or academic choices? Does the sex o f play groups i n chi ldhood predict these choices in adulthood? Does parent-child play predict academic and career choices? Is this process mediated b y gender schemas or gender ideologies? These research questions w i l l be tested in this study in order to assess the possible outcomes o f parent-child gender socialization. The proposed research w i l l examine h o w social indicators o f parent-child gender socialization can be l inked to gender specific outcomes in adulthood. Toys , play groups, and parental play i n early childhood w i l l be the indicators o f gender socialization. Universi ty majors and career expectations w i l l be the adulthood outcomes. The fo l lowing sections w i l l discuss the theoretical issues behind this research, the past research, and the research methods. Theoretical Issues o f Gender Socialization M a n y theories have been applied and developed to guide gender socialization research. However , many o f these theories apply only to certain areas o f gender development, as no theory thus far can account for the entire developmental process (Fagot & Leinbach , 1989). Theories from where the research questions are rooted w i l l be discussed in the next section. 4 These theories focus on parental gender socialization and its impact on later life outcomes, such as career choices. Learning Theory Learning theory can be used to explain h o w individuals learn gender. The assumptions o f learning theory postulate that we learn gender appropriate, or sex-typed, behaviors and roles through punishments and rewards (Ruble & M a r t i n , 1998; Bern, 1983, 1981). Rewards and punishments in this paradigm must be conceptualized in the social context; such responses to sex-typed behaviors, whether congruent or incongruent wi th the individual ' s sex, can range from facial expressions, to involvement o f play, to verbal cues (Roopnarine, 1986; Langlois & D o w n s , 1980). Chi ldren imitate sex-typed behaviors that parents, or other authority figures, model (Ruble & Mar t in , 1998). A s an example, i f a female chi ld chose to play wi th a do l l , or any other feminine-typed toy, this behavior wou ld be met with positive reactions. A n y other congruent sex-typed play wou ld also receive positive reactions. However , should this same female chi ld play wi th a masculine toy, parents and others w o u l d frown, state their disapproval, or engage in another form o f social punishment. Al though this theory identifies the sources o f gender information, social learning theory implies that children are relatively inactive participants in the socialization process (Bern, 1983). Parents are seen as the information providers and modelers, while children are simply absorbing gender information. M a n y studies that apply this inactive assumption to observe the process o f parental socialization on young children forget that gender socialization is a process, not a sequence. Fo r example, Fagot and Hagan (1991) observed parent-child toy selections and coded video recordings o f children's selections, fol lowed by parents' reactions (rewards and punishments). This sequence o f events assumes that the process ends when a parent reacts. This sequence omits the possibili ty that children may be observing the parents' reactions while 5 making the toy selection in the form o f social referencing. In assuming that children are not engaging in social referencing, children appear to be unmotivated to seek gender information. Al though social learning theory is used in many infancy and early chi ldhood gender socialization studies (e.g., Fagot & Hagan, 1991), it fails to explain why gender information is so readily processed, why gender is the first and largest categorical system that children learn, or i f factors affect gender development beyond early chi ldhood and into adulthood. This theory does not account for the variation in children's gender development, and arguably not within multi-sibling families. Despite these issues, social learning theory has been an important paradigm to assess the gender socialization o f infants and young children. Gender Schema Theory Gender schemas are the networks that contain gender and gender-related information (Ruble & Mar t in , 1998; Bern, 1981). Gender schemas help to seek out information related to gender and organize new information to be related to gender (Bern, 1981). Sex typing is created from individual gender-schematic processing capabilities o f organizing information into gender categories (Bern, 1983; Roopnarine & Mount , 1987). The focus on individual abilities al lows for variations in gender development. This can explain why siblings can have varying levels o f masculinity and femininity, even though they grew up in the same family environment. Gender schema theory assumes that appropriate sex-typed norms come from the surrounding social environment, wh ich acknowledges the learning processes involved in gender development (Bern, 1983). Parents create the social environment for their infants and young children, by providing gender information to them. Chi ldren learn to use their developing gender schemas to understand and organize information. Societal gender expectations give children sex-typed information they w i l l in turn relate to their o w n sense o f gender, wh ich is also related to their self concept (Bem^ 1983). Chi ldren w i l l learn information 6 relevant to their o w n gender prior to information o f the opposite gender. Gi r l s possess schemas wi th greater female knowledge, and boys w i l l ho ld more masculine knowledge (Levy , 1999). The cognitive and learning elements o f gender schema theory combine the concepts o f learning and cognitive development theories to create a more comprehensive theory that applies to more areas o f gender development. This theory explains how and why children learn from the gender related information provided by their family. A s gender schema theory is a theory o f process (Bern, 1983), it can be critiqued for its lack o f accounting for the content o f schemas. However , the lack o f content description explains differences in gender development within families and across social definitions o f gender. The accumulation o f gender-related information can be explained, but the gender information cannot be. Gender schema theory also fails to explain when, or i f the process o f gender schema development is completed. The flexibility o f gender schemas is also not explained. Gender schema theory does not explain i f gender schemas change over time, or i f there is a point at which they are completed. This theory does not say whether gender schemas receive information throughout the life course that continues to shape them. Theories in General Learning theory and gender schema theory are only two perspectives on gender. The concept o f gender is a complex topic for social research, where many perspectives and theories have tried to study its many facets. Conne l l (1995) describes the complexity o f studying masculinity (and gender) i n social research and discusses the rationale that masculinity exists only where femininity exists. The idea that each gender must exist in order for the concept to be real adds to the complexity o f the concept, as not a l l perspectives v i ew masculinity and femininity as being polar opposites. Fo r example, Eagly (1987) argues that the maintenance o f sex differences depends on gender roles, that are occupied strongly by one o f the sexes. The 7 gender roles serve as a guideline for what type o f employment, social role, and status either sex should hold. Gender roles are not necessarily v iewed as being opposite from one another, but masculine and feminine roles must exist together in order for gender roles to be a va l id concept. Conne l l (1995) describes four main areas o f thought for where most o f the gender perspectives can be categorized into. Essentialism focuses on one aspect o f the definition o f masculinity and defines male life in terms o f that definition only. These perspectives are too vague to applied to the process o f gender socialization. Posi t ivism relies on the factual findings o f social research to define masculinity and femininity in order to describe h o w men and women are. This perspective relies on scales and statistics that al low for the concept o f gender to be quantified. Gender schema theory falls into this category, as gender is assumed to be relative i n terms o f the quantity o f masculine and feminine traits the individual holds. Learning theory can also be quantified by measuring gender role ideologies, but also relies on normative definitions o f gender roles. Normative perspectives interpret gender difference i n terms o f h o w each gender should behave (Connel l , 1995). The assumptions o f learning theory fall into this category, where children learn what gender roles are appropriate for themselves as w e l l as for others. The fourth perspective, k n o w n as the semiotic perspective, defines one gender in terms o f not being the other gender; the symbolic differences between the genders are compared in contrast to the other. Connel l ' s (1995) categorization o f gender perspectives demonstrates that there are many theories surrounding gender, however, these perspectives emphasize different aspects o f the concept. There are many theories that are applied to gender research, however they do not explain the process o f acquiring gender in early childhood, and h o w this process affects one 's gender i n adulthood. Tajfel and Turner 's (1978) social identity theory posits that individuals 8 within a social group w i l l act and behave as other members o f that group do. Individuals are assessed based on how they identify with a social category. F r o m this perspective, being a man w o u l d cause men to act and behave i n similar fashions that other men do (Ruble & Mar t in , 1998). This theory explains the learning o f gender in broad and general terms, but takes on a macro-level explanation o f how gender exists, instead o f explaining the specific parent- chi ld gender socialization process. This theory also does not explain h o w differences o f gender exist within a group. F o r example, in a group o f adult women, why do some describe themselves as being more ' feminine' than other members o f the group? Another contextual theory that does not explain the process o f acquiring gender information is the situation-based context perspective. This perspective views the context o f the situation as an important contributor to the development o f gender. West and Zimmerman (1987) explain that gender is something that is "done" and expressed through actions, such as behavior, speech, and thought. D o i n g gender is a theory that can only be applied to those who already have an individual sense o f their gender, and cannot be used to explain the process o f gender learning or processing. This perspective is not applicable to early chi ldhood socialization where children do not have a sense o f what gender is. Some theories, such as evolutionary schools o f thought, base their understanding o f masculinity and femininity on the reproductive restraints that men and women hold (Ruble & Mar t in , 1998). These theories argue that gender roles originate from these reproductive differences. These theories ignore the social factors that influence the development o f gender through socialization, and do not take into consideration the evolution o f gender roles themselves. Gender is a complex and vast concept that is not only debatable i n definition, but also in the use o f the concept in research. There are many more theories and perspectives on gender 9 which can be applied to specific areas o f the development o f gender. Learning theory and gender schema theory are two competing theories in the paradigm o f early chi ldhood gender socialization, and are appropriately applied to the research questions in this study. Past research has used these theories i n this area o f study, and w i l l thus be used i n this present study. G i v e n the complicated nature o f gender, and the many ways o f defining and measuring the concept, it is clear that a l l facets w i l l not be able to be included in this study. The idea that gender is a socially constructed concept is an assumption that holds for many o f the social theories discussed, and w i l l be an assumption in this study. This means that any biological influences w i l l not be tested. Gender Research to Date Al though there have been some changes in the distinctions between sex and gender, for this discussion, gender w i l l be defined as the amount o f masculine and feminine characteristics, including preferences, behaviors, and aspirations. Mascul ine characteristics are the attributes that social norms deem to be appropriate for males and feminine characteristics are the attributes deemed appropriate for females. Sex w i l l be defined as the biological classification that is used to distinguish males from females, characterized by sex organs. Sex-typed behaviors are actions congruent wi th either masculine or feminine characteristics. F o r example, putting on makeup is considered to be a feminine sex-typed behavior, and fixing a car is considered to be a masculine sex-typed behavior. Gender has been distinguished from "sex" i n that it develops through social and cultural processes, and is thus an acquired description, rather than biological ly detemiined (West & Zimmerman, 1987). However , from infancy, biological sex determines how parents, and others, w i l l react to us as children and adults (Fagot & Leinbach, 1987; Pomerleau, Bo lduc , Ma lcu i t & Cossette, 1990). Fo r example, parental personality expectations for daughters are 10 different from those held for their sons (Maccoby & Jackl in , 1974). Research continues to support the idea that parents assume their children are predisposed to developing sex-typed characteristics. Parents in turn respond to the sex o f infants in sex-typed manners according to the infant's sex (Pomerleau et a l . , 1990). Parents and others must hold certain gender related assumptions in order to provide appropriate gender information to children. A s parents usually provide the most care for their infants, they become the largest source o f social information to their children. Chi ldren learn gender roles from the sex-typed behaviors that parents model (Schwartz & M a r k h a m , 1985). Parents model sex-typed behaviors through implicit and explicit demonstrations. Expl ic i t modeling occurs in such ways as when parents vocalize what is appropriate for girls and boys. Such explicit gender information reflects the gender expectations that parents hold and practice. F o r example, traditional parents engage in orating stories wi th achievement themes tied to masculinity; these themes are reflective o f parents' gender expectations for masculine roles (Fiese & Ski l lman, 2000). Implicit modeling occurs when parents do not directly communicate ideas o f gender. F o r example, parents can display their gender ideologies wi th daily routines such as driving. W h e n the father drives the car and the mother sits in the passenger seat, gender i n this context is modeled implici t ly. These sex-typed roles that are learned from parental models are assumed by researchers in tins area to be used and displayed i n social contexts outside the family context (Lindsey & M i z e , 2001). Chi ldren ' s bedrooms are an important gender r ich environment parents create. Homes have been found to have children's bedrooms containing sex-typed colors, clothing (Pomerleau et al . , 1990), and toys ( O ' B r i e n & Huston, 1985). B y surrounding children's bedrooms with information regarding their gender, girls develop interests in feminine items, and boys in masculine items. Parents assume that their daughters are predisposed to appropriate sex-typed 11 behaviors. This assumption becomes a logical explanation for the parents to account for their daughters' interests in the color pink, rather than to assume the responsibility for creating the feminized environment o f a pink bedroom fi l led wi th pink bears and Barbie dolls. However , it cannot be assumed that infant bedrooms are the only gender environments that parents create for their children. Toys and Parental Influences Research in the area o f parent-child gender socialization has relied heavily on toys as a measure o f both parental gender roles and expectations, and o f children's gender development. Toys occupy a large amount o f time in children's lives and are important to gender socialization. Parents can interact wi th their children in the form o f play, which can also involve the use o f toys. This interaction provides children wi th messages regarding which behaviors are appropriate for their gender. Chi ldren are discouraged to engage in cross gender play and with cross gender toys; parents encourage their children to play with toys that are congruent wi th their sex (Fisher-Thompson, 1990), and to play in congruent sex-typed play styles (Lindsey & M i z e , 2001). Parents provide social information that w i l l guide their children's toy selections Important Ages The majority o f research in the parent-child gender socialization in the play context looks at children as young as 12 months old . In a longitudinal study, Fagot and Hagan (1991) found 18 months was the age at wh ich parents used the most rewards and punishments, compared to other age groups in their longitudinal study. They argue that parents feel a need to educate their children at 18 months o f age regarding the appropriate social norms o f their society. B y 3 V2 years o f age, children can make sex appropriate play decisions and toy selections in peer interaction settings where parents are not influencing the children (Fagot & 12 Leinbach, 1987). Subsequently, Fagot and Hagan (1991) found that parents displayed fewer reactions to the appropriateness of sex-typed play for their 5 year-old children, compared to when their children were 18 months old. Fagot and Huston (1991) argue that the decrease of attention to sex-typed play guidance through rewards and punishments is no longer required at this age. Children should have learned the basic gender rules and be able to maintain them. The Importance of Sex-Typed Toys Given the research on toy selections and play styles presented, the question remains: why are toy and gender related research findings important? In order to answer this question, sex-typed toys must properly be defined and understood. Masculine toys are what a given set of social norms rules as being appropriate for boys to play with, and vice versa for girls' toys. Guns, army toys, weapons, football uniforms and cards, airplanes, vehicles (cars, trucks, trains), sports balls, weight lifting gear, and tools are examples of toys that past research has found to be masculine toys; jewelry boxes, dolls and soft dolls, sewing kits, kitchen gadgets, hair dryers and telephones are feminine toys (Fisher-Thompson, 1990; Idle et al., 1993). Upon closer analysis of the common sex-typed toys in this list, it becomes apparent that masculine toys promote spatial movement. Sparfkin, Serbin, Denier and Connor (1983) summarized research findings that provided strong evidence that children who play more often with masculine toys outperform on various spatial tests children who play most often with feminine toys. Trucks (mobile toys) allow children to utilize their play space to its full extent and have been found to elicit the greatest amount of movement (Liss, 1983), just as sports equipment promotes physical movement of the entire body. Feminine toys promote domestic and imaginative play; dolls are the only toys that elicit nurturing play (Liss, 1983). The majority of feminine toys do not take children's imaginations beyond the domestic paradigm. While feminine toys do not encourage 13 spatial abilities, tbey teach girls about traditional female roles. A s children are learning about what toys are appropriate for them to play wi th , their gender schemas are becoming more concrete. Parents w i l l control wh ich toys children have access to, and these toys help to shape and become a part o f children's gender schemas. Chi ldren w i l l develop a sense o f femininity or masculinity from their toy exposure, and the toys w i l l become recognized as congruent with their o w n gender schemas. B o t h learning and gender schema theories can argue what role toys have in the development o f gender in childhood. However , it is not clear i f gender schemas or ideologies o f these theories mediate adulthood outcomes from these early gender experiences. Mascul ine toys in gir ls ' lives have been found by Giul iano et al . 's (2000) study to be related to adult outcomes o f athletic abilities. W h e n female athletes and non-athletes were asked to give retrospective accounts o f their play styles and toys from their chi ldhood, the female athletes were found to have had more access to masculine toys. P lay ing wi th masculine toys was related to personal athletic confidence, orientations towards winning, and competitiveness. The encouragement these female athletes were receiving in their chi ldhood indicates that encouragement in sports participation (sports usually being masculine) affected their athletic careers. Giul iano et al. (2000) also found that women who played wi th masculine toys and participated in male play groups were more l ikely to become varsity athletes. The results o f this study provide support for the argument that masculine toys and play groups affect specific developmental outcomes. The implications o f the sex o f play groups are important considerations when assessing later life choices. The findings that relate masculine toy play in early childhood athletic careers compared to non-athletes make this study an important element to the proposed research. The strong relationships that Giul iano et al . (2000) found wi th masculine toys is evidence that toys and play groups have a significant effect i n the process o f gender socialization. The findings o f this study are crucial when considering the 14 research on gender socialization with sex-typed toys. Although Giul iano et al . (2000) were measuring concepts related to athleticism, the relations between chi ldhood toys and play groups to adulthood suggest that toys may affect other adulthood outcomes. Fo r example, career expectations and post-secondary fields o f study could be mediated b y chi ldhood gender socialization. It is also unclear as to whether gender role ideologies or gender schemas are mediating factors in this process. The past research on the parent-child gender socialization process through the use o f toys has failed to look beyond the childhood stage and to assess the implications and outcomes o f this socialization. Parental Influences Rewards and Punishments. Fac ia l expressions that parents display upon the presentation o f toys have been found to be a mechanism that parents use to socialize gender in their young children. Twelve month o ld boys receive more positive reactions to masculine play, and negative assertive behaviors, while mothers give more positive reactions to twelve month o ld girls who played wi th feminine toys than d id fathers (Fagot & Hagan, 1991). Caldera, Huston and O ' B r i e n (1989) found that when parent-child dyads were presented with toys i n a laboratory setting, both parents and children exhibited excited facial expressions i f the toy was consistent wi th their own sex. Toys that were inconsistent wi th the sex o f the parent were received with less excitement, especially for fathers. These same parents responded verbally to feminine toys wi th more teaching and praise, while masculine toys elicited more animated sounds and negative comments. Caldera et a l . ' s (1989) and Fagot and Hagan ' s (1991) studies demonstrate h o w toys are used as tools o f gender socialization by parents. However , they do not imply that parents are consciously aware o f the information provided in their reactions to toys. Parental Toy Influences. Parents are most l ike ly not aware o f h o w they are reacting to 15 their children and their toys. Fagot's (1974) study had three of their twelve family sample show displeasure and upset when asked about their own sex differences in reaction to girls' and boys' play. One would assume such attitudes would predict a less traditional pattern of gender socialization from these parents, however, Fagot (1974) found that the 12 to 18 month old children of these parents did not behave and play differently than the other children of the same sex in the study. Although these parents were distraught at the idea of differential treatment of boys in girls, they were not aware that they were engaging in it. The evidence of parent-child toy play may appear to imply that the child plays a very inactive part in the toy selection process. The learning process requires those around the infant to provide information, sometimes in the form of a toy. Children have been found to accept toys from their parents almost all the time (Idle et al., 1993). The lack of opposition to parental toy selection reiterates the parental socialization influences; children are highly accepting of what parents offer them. Even for children who do not accept a toy selected by their parents, children are still likely to consider the toy, and are extremely unlikely to reject the toy (Idle et al., 1993). This occurrence illustrates the impact that parents have on sex-typed toy preferences, as children are highly likely to accept the toys their parents provide them with. Parents are the primary toy purchasers for their children (Fisher-Thompson, 1993). However, external influences on which toys children prefer must be considered. For example, children receive toys as a gifts from relatives and family friends. Toys as gifts were just as likely to be sex-typed as non-sex-typed when people outside of the immediate family were providing them. Most of the toys that are received as gifts from parents are usually sex-typed, especially if they are requested by the child (Fisher-Thompson, 1993). However, in the event of a cross gender toy being received, the parent is still able to control the gender socialization process. As mentioned previously, parents' reactions to toys guide the toy selections of the 16 chi ld. Fagot and Hagan (1991) found that parents w o u l d react after a toy was selected by the chi ld ; this reaction acts as a reference to wh ich toys are thought to be appropriate. A cross gender toy that is given to a chi ld wou ld be met with the same negative reactions as d id a l l other cross gender toy presentation studies found. Gender of the Parent. Mothers and fathers have different gender expectations for their children. The sex o f the parent has implications on the gender socialization o f their children. Fagot (1974) found that fathers generally identify more activities as being sex appropriate than mothers do. In play situations, fathers spend equal amounts o f time playing with masculine and neutral toys, whereas mothers spend more time wi th neutral toys (Caldera et al . , 1989). W h e n ranking toys along a desirability scale, fathers and men were found to give the highest (and extreme) ratings to masculine toys, and the lowest ratings to fennnine toys (Idle et a l . , 1993; Fisher Thompson, 1990). Idle et a l . , (1993) found that mothers had similar ratings as fathers, while Fisher-Thompson (1990) found that female subjects considered more toys as neutral, rather than sex-typed. Preferences toward certain toys should also indicate what types o f toys parents w o u l d want to use in play with their children. Fathers also engage in more physical play wi th their sons, whi le mothers do not show any differences i n physical play wi th sons or daughters (Lindsey & M i z e , 2001). These sex differences parents exhibit demonstrate how fathers and mothers assert their gender by displaying, or doing, sex-typed behaviors. Fagot and Hagan 's (1991) male subjects gave the least amount o f positive reactions to cross sex play to their 18-month o ld sons, while mothers made no distinction. Parents overall gave more positive responses to sons who were engaging in masculine-typed play. W h e n selecting toys as gifts, males are more l ikely to purchase sex- typed toys than non-sex-typed toys, especially when buying for children other than their own (Fisher-Thompson, 1993). These findings suggest that fathers provide and display more 17 stereotypical masculine gender information to sons. Fathers hold stronger gender stereotypes than mothers, and thus, give more sex-typed messages to their sons. In contrast, daughters appear to be receiving more neutral than sex-typed messages, compared to the sex-typed information that sons receive overall. Men appear to display their gender through interacting with children in sex-typed manners (e.g., toys purchasing, play styles), especially with their sons, indicating that men receive more social rewards for their gender appropriate behavior Context of Gender and Sex-typed Toys. The context of social interaction is also important to the socialization process. Parents can define the context in terms of gender appropriateness by choosing when to interact with their children (Lindsey & Mize, 2001). Research in the area of parent-child toy selection has found that associative play and activities are valid measures of positive social interaction (e.g., Fagot & Hagan, 1991). As an example, when a father asks a son to help fix the family car, he is choosing the context of interaction outside of the house to teach his son a new skill. If the father does not interact with his son on the same level inside the house doing domestic work, the father has defined auto mechanics as a masculine context. Lindsey and Mize (2001) found that parents were more likely to choose to engage in pretense play (a form of play that involves using objects to represent other objects) with daughters, while choosing to engage in physical play with their sons more often than any other form of play. Parents in these contexts were teaching their cliildren how certain types of play styles are more appropriate for girls or for boys. Toys in these play situations are also used by parents to define sex-typed play styles, as well as to indicate which toys are meant for boys, and which are meant for girls. Children learn a large amount about gender from their toys (Schwartz & Markham, 1985). Idle, Wood, and Desmarais (1993) found that fathers in play situations with their children would choose to spend more time with toys that were masculine typed, while mothers chose neutral toys. These findings exemplify how parents create sex- 18 typed contexts for learning. However , both the sex-typed contextual and environmental influences that parents create for their children may be very detailed and meticulously researched, but the questions pertaining to the effects and outcomes o f varying gender socialization must be addressed. Questions or studies pertaining to the importance o f gender information i n early childhood must be addressed in order to validate the importance o f these studies. Career Expectations and University Specializations Career expectations are also affected by gender. Gender roles and gender role expectations appear to influence career paths that are taken. For example, women who place high salience on personal and family life, compared to those women who d id not, have been found to be less l ike ly to choose the sciences as university specializations (Ware & L e e , 1988). However , mathematically oriented university majors (which are male dominated) for men are positively related to personal and future familial obligations (Ware & L e e , 1988). This gender difference suggests that perhaps male dominated domains o f study are helpful i n pursuing full time careers that require, or are thought to require, a higher priority over family obligations. B o t h male and female college students report equal desire for having children at some point in their lives (Schroeder, B l o o d , & M a l u s o , 1993). However , male and female careers offer different paths; female career paths usually entail considering time off for rearing and raising children. A de-emphasis on familial responsibilities wou ld deter traditional females from pursuing such fields o f study. Jackson, Gardner, and Sull ivan (1992) found that on average women, compared to men, expect to take more time off from their careers in order to focus on chi ld care. These women reported they expected an average o f 3 years away from their paid work, while men only predicted 1 year away from their jobs. Fami ly related gender 19 differences also exist within the careers themselves. Women have reported that accommodating jobs to family life and development opportunities are more important than men perceive to be (Jackson et al., 1992). While both men and women want to have children, women tend to face more career interruptions to accommodate these responsibilities. Men who pursue masculine careers are more interested in extrinsic job features, such as money, little supervision, leadership and prestige; women are concerned with intrinsic features such as creativity, working with people, and steady progression (Lyson, 1984). Women have reported that they want both a career and a traditional family life (Schroeder et al.,1993), which is more attainable through intrinsic careers. Since intrinsic careers are not as focused on independence and leadership, they are easier to leave and to return to. The features of intrinsic careers allow women to accommodate their familial goals. Despite the desire to have a career that accommodates child rearing, undergraduate women assume that they will endure more role frustration when they become parents, compared to the levels of role frustration men predicted their future wives would face (Schroeder et al., 1993). Although women are preparing for maternal roles by choosing careers that accommodate this desire, they are still aware that this could conflict with their career goals. Family Influences. The influence of the gender role expectations of female social and familial responsibilities may be linked to the gender socialization process. According to Lewko et al. (1993), family support and encouragement predicted daughters enrolling in science specializations, while internal factors, such as motivation, predicted male participation in the sciences. This study indicates that careers in scientific domains are expected to be filled by males, who rely on their own motivation to seek these careers. Women must be socialized through family encouragement and away from traditional female roles to choose masculine 20 fields. Fathers appear to have a large influence in this process. Ti l leczek and L e w k o (2001) found that females whose fathers were employed i n the sciences were 3.5 times more l ike ly than other females to pursue science careers themselves. Maternal science occupations d id not predict daughters' science specializations. However , having a mother who was employed predicts more egalitarian attitudes towards parenting roles (Schroeder et al . , 1993). These findings reiterate the amount o f influence that fathers have over the gender socialization and career choices o f their children. Al though these fathers were employed in traditionally masculine fields where the mothers were not, family support, as indicated by L e w k o et al . (1993), influences female participation in male dominated careers. In the past, families were less l ikely to pay for their daughters to go to post-secondary institutions, and then in the late 1980's, were less l ikely to provide computer training for their daughters (Eccles , 1987). In these instances, parents were using their gender ideologies to guide their behaviors i n relation to their daughter's education careers, and steer them away from non-traditional careers. In doing this, traditional roles, such as motherhood, are emphasized. Application of Theory. B o t h gender schemas and gender ideologies are important factors in sex-typed career decisions. E c c l e s ' (1987) literature review describes h o w specific elements o f these theories account for many variables in career decisions for men and women. For example, the perception o f available fields o f study and employment are influenced by individual gender schemas. Whi l e some career options may not be considered due to being unaware o f their existence, many options simply do not fit into individuals ' gender schemas, and are therefore not considered when a decision is made. W h e n a career or academic option matches their gender schema, individuals w i l l consider these options. Conversely, gender roles affect career choices by influencing perceptions o f careers. Eccles (1987) reported that gender role socialization has a negative effect on women ' s confidence i n their abilities compared to 21 men's confidence. This confidence deficit leads women to feel the need to work harder in order to attain similar career goals that men have. Perceptions o f success are important variables, as they predict levels o f performance. Gender role socialization was also reported by Eccles (1987) to impose different personal priorities on career decisions. W o m e n are more l ike ly to prioritize children over careers than men. Some individuals choose career fields that are not considered to be gender appropriate. F o r example, when a woman decides to enter into mechanical engineering, or when a man decides to enroll in nursing in university, these sex-typed choices are incongruent wi th the individuals ' behaviors. L y s o n (1984) found that men and women who pursue fields mat are incongruent with their sex hold work value orientations that fall between those o f traditional men and women. However , the men and women i n the fields that are incongruent wi th their sexes were more similar in attitudes to traditional men and women, compared to each other. In other words, women in non-traditional fields have similar values as traditional women. W o m e n i n non-traditional fields do not have similar attitudes to men in non-traditional fields. M e n and women in non-traditional areas o f study and work still ho ld more traditional values. In this instance, gender schemas might mediate the career and field decisions that are being made. The non-traditional careers might be congruent with their gender schemas, but their gender role ideologies could be more traditional. I f an individual 's gender schema is more similar to a career that is not congruent with their gender, the work value orientations are accounted for by their traditional gender role ideologies. F r o m these examples, it is unclear whether gender schemas or gender role ideologies mediate career decisions. In relation to the toy research, the questions that remain ask whether toys i n early chi ldhood are true predictors o f later life career choices, such as university specializations. Learning theory w o u l d hypothesize that children learn career and academic gender roles from 22 their parents. F o r example, when a daughter's academic performance is equal in math and Engl i sh , parents w i l l still assume that her academic performance i n Engl i sh exceeds her performance in math (Eccles , 1987). Parents socialize their children through the assumptions o f their own gender role ideologies. These gender roles should predict wh ich university majors students register in . In contrast, gender schema theory w o u l d argue that university and career choices are reflections o f individual 's masculinity or femininity. Regardless o f which theoretical perspective mediates career decisions, the gendered environments that parents provide in early childhood need further examination. The toys and play styles that parents controlled in childhood are important elements o f the process o f gender formation. Career choices are l ike ly to be predicted by the sex-typed nature o f childhood toys and play environments. The path that leads to career decisions is as follows: parents socialize gender in their children in early chi ldhood through toys and play; children learn gender ideologies and create their individual gender schemas; in adulthood, career and educational choices are reflections o f gender ideologies or gender schemas that were learned and created i n childhood. Repl icat ion The majority o f the cited research on parental gender socialization wi th the use o f toys and play is over fifteen years old. The research designs and findings o f these study are still important contributions to this field o f research. However , as gender ideologies towards female and family roles change over time, these studies are in need o f replication and re-evaluation. The proposed research combines the findings o f the parent-child toy studies wi th the E c c l e s ' (1987) review findings. This study is guided by the findings on gender role and gender schema influences on career and adulthood outcomes. Theoretical M o d e l Figure 1 is an illustration o f the models that represent the relationships between early 23 childhood gender influences and the type of university major that has been declared. The outcome variable in model one is defined as the percentage of female students enrolled in the declared major. The more females that are enrolled in the declared fields indicate feminine academic fields, and lower female enrollment indicates higher masculine fields of study. The previously discussed research has indicated that parents create gender environments for their children through toys, play groups, and reactions to their children's behaviors. These components of children's gendered environments serve as indicators of parental gender and gender role expectations for their children. All together, these are considered as the independent variables in both models. However, with the abundance of research done in this field, the question remains, does this phenomenon have any influence on children beyond childhood? And specifically, do the gender messages learned in childhood influence later life behaviors? As Giuliano et al. (2000) found, predictors of athleticism and confidence included sex of play groups and toys. However, no research so far has found whether or not academic and career choices are influenced or predicted by early childhood gender socialization. For these reasons, the outcome variables in model tow are career expectations and university specializations. The models in Figure 1 illustrate the hypothesis that there is a connection or relationship between parental gender socialization through play media, and later life decisions, such as career expectations and academic fields of study. For example, for those children that do play with masculine toys more often, are they more likely to choose masculine university majors?. Would they also be more likely to assume a masculine role in work and family life? And do femmine toys predict the opposite? The models in Figure 1.0 illustrate the direction of these proposed relationships. Model 1 predicts the path that links early childhood gender socialization to declared specializations. This model is testing whether the theories can explain the relationships to the 24 Figure I. Models 1 and 2 of Hypothesized Relationships. Gender Ro le Ideologies Parent-child Gender Socialization •> Univers i ty Specializations Gender Schemas Gender Ro le Ideologies Parent-Child Gender Socialization Occupational Ro l e Commitment arental Ro le Commitment Mar i t a l Ro l e Commitment H o m e Care Ro le Commitment Gender Schemas 25 action o f declaring and pursuing academic fields that may be sex-typed. M o d e l 2 is predicting the commitment to future family and work roles that may be related to early gender socialization. The path o f this relationship may be described or predicted better by various theoretical frameworks. Whi l e learning theory and gender schema theory are both able to describe the gender developed in childhood, neither theory has attempted to explain how this process mediates adulthood outcomes. Bo th models illustrate how parent-child gender socialization can be mediated by either theoretical paradigm. I f the childhood serves as a time where information is sorted into the mind through schematic processing, then gender schema theory w o u l d be best suited to describe the mediated effects o f gender socialization into adulthood. Gender categories wou ld influence later life decisions by basing career, family, and academic interests on what is considered congruent wi th individuals ' gender schemas. However , i f learning theory's assumptions that children learn through observations, models, rewards and punishments, then the learned gender roles wou ld mediate the relationship between gender socialization and adulthood outcomes. The individual wou ld thus rely on their gender role ideologies in order to make the decisions that are congruent with it. B o t h o f these theoretical assumptions must be tested in order to understand this process. Thus, the question o f which theoretical paradigm can explain the effects o f childhood gender socialization is asked i n this study. Whi l e other variables are not able to be measured in this study, this model is testing the relationship between parent-child gender socialization and academic and career outcomes. For example, media influences are not measurable within the parameters o f this retrospective research study. However , despite the lack o f measures available to consider media and other influences on gender socialization, the strength o f the model 's assumptions w i l l indicate the 26 amount o f influence parents have. Hypotheses The fol lowing hypotheses have been derived from models one and two: H I : Sex-typed childhood toys predict sex-typed university specializations. H 2 : Parent-child play predicts sex-typed university majors. H 3 . The sex o f play groups predicts sex-typed university majors H 4 : Sex-typed childhood toys predict sex-typed career choices and expectations. H 5 : Parent-child play predicts sex-typed career choices and expectations. H 6 : The sex o f play groups predicts sex-typed career choices and expectations. H 7 : Gender role ideologies mediate the relationships tested in hypotheses 1 through 6. H 8 : Gender schemas mediate the relationships tested in hypotheses 1 through 6. Methods Dependent Variables. University Major/Specialization. Univers i ty majors or specializations are the areas in which students have registered as their primary field o f study. Students at the university are required to have declared their specialization at the end o f their second year o f study. Specializations indicate which area o f interest the student w i l l be focusing on for the larger part o f their academic careers. The surveys had an open ended response item where respondents were asked to indicate their declared specialization. The university 's Planning and Institutional Research ( P A I R ) office provided a list o f a l l specializations with their respective breakdown by sex o f students who are enrolled in the program. The percentage o f female enrollment in the declared specialization o f the respondents was used for analysis. Career Expectations. Career expectations are the assumptions that respondents hold about their future employment and earnings beyond the university setting. The measures for 27 career expectations determine whether individuals place salience to occupational and familial responsibilities. The L i f e Ro le Salience Scale ( L R S S ) (Amatea, Cross , Cla rk & B o b b y , 1986) consists o f items that measure role expectations for w o r k and family, based on the levels o f importance and commitment to each o f the roles (see Appendix B to v i ew survey).These items are applicable to both sexes. A l l o f Amatea et al . 's (1986) four life role constructs in the L R S S are used for the purposes o f this study. These constructs are occupational, parental, marital and homecare roles. On ly the role commitment dimension o f these constructs w i l l be included. The role reward value dimension o f this scale does not apply to this research. A n example o f h o w the scale measures career expectations is as follows: higher levels o f commitment to the occupational role indicate masculine gender ideologies. In contrast, higher parental role commitment to parenthood indicates more feminine traits. These four dimensions o f the constructs are measured by using a five-point Likert-type attitude scale on levels o f agreement. Respondents were asked to indicate the extent to which the statements about these constructs best describes their attitudes out o f five forced choice options. Examples o f these questions are: ' I f I choose not to have children, I w i l l regret i t ' , and, T expect to devote whatever time and energy it takes to move up in my job/career f ie ld ' . The L R S S ' s validity has been tested across student and married couple populations (McCutcheon , 1998). Independent Variables Sex-Typed Childhood Toys and Play. Sex-typed childhood toys are toys that are categorized to being appropriate for one sex. Subjects were be asked to recal l wh ich toys they owned and played wi th during their chi ldhood years. Al though events in infancy cannot be recalled from memory o f experience, the subjects should have some knowledge o f the toys they were exposed to. F o r instance, photographs o f the chi ld wi th the toy, or i f the toy has been kept as a keepsake o f early childhood, should a l low the respondent to recal l such toys. 28 Chi ldhood toys are measured in the survey by giving respondents the opportunity to indicate toys that they played wi th in childhood. The toys that appear in the survey in the forced choice list include popular sex-typed toys that wou ld be congruent with the era o f when respondents were young children. The sex-typed toys are toys that past research has found to appropriate for one sex, and not the other. The toys in the survey are based in Fisher-Thompson's (1990) extensive inventory o f adult sex-typed toy categorizations. On ly toys from Fisher-Thompson's (1990) inventory that correspond with those used in Giul iano et al . 's (2000) survey research were used and adapted to attain a universal interpretation o f what sex-typed toys should be classified for this research. Those toys that correspond from the Fisher-Thompson (1990) assessment with the Giul iano et al . (2000) research have maintained nearly the same sex-typed categories twenty years prior to this research (e.g., O ' B r i e n & Huston, 1985). A l l items measuring toys and play w i l l have an option labeled "other" where the respondents can write a toy or a form o f play not on the list. The mean scores for the toys and game play variables are used in the analysis. The bivariate correlation between the feminine toy play and feminine game play variables is .80, p = 0, and the reliability analysis alpha is .89. Feminine toy play average is combined with feminine games average to create one variable, feminine toys and game play. The bivariate correlation between masculine toy play and masculine game play is .67, p = 0, and the reliability analysis alpha is .81. Mascul ine toy play average was combined wi th masculine games average to create one variable, masculine toy and game play. Neutral toy play is calculated by taking the mean score for the frequency o f neutral toy play. Parent-child play. Parent chi ld play is defined as the amount o f time and gender based play that subjects recall playing with in early childhood. Participants were asked i n the survey to indicate wh ich parent they remember playing wi th more. Respondents answered on a 5 point 29 scale whether they played each with their mothers and fathers never (1) to very often (5). These items measure which parent was more readily available for gender information in early childhood. As the research indicates, fathers and mothers provide different levels of sex-typed information in play sessions. Sex of Play Groups. The sex of play groups is defined as which sex of playmates children had in early childhood. The sex of the play groups will help to measure another aspect of what other types of gender information children were exposed to. If the child engaged to play with children of the opposite sex, then this could create a different gendered environment than children who played only with same-sex peers. The cross sex-play could allow access to toys that are not congruent with their sex. The survey asked what play groups respondents played in when they were younger. The survey asked respondents to recall the frequency of play with each gender around the time when they were in preschool and first grade, when children are around 6 Vi years of age, and far more likely to have opposite sexed friends (Maccoby, 1990). However, these two measures do not have enough variation when gender is controlled for. Respondents report playing with only same-sexed peers in their early childhood. The lack of variance in measures indicates that they cannot be defined as variables, and were therefore dropped from the analysis. Hyppotheses 3 and 5 could not be tested. Control Variables Parents' Occupations. Parents occupations are defined as what type of careers parents have. The survey asked respondents which occupations both their mothers and fathers hold, and what type of category the occupations fall into (e.g., trades, health professions). Responses are coded into a new variable based on employment and stay at home professions. These control variables are tested to indicate whether having employed or stay at home parents has an effect on the outcome variables. This variable is included to ascertain whether fathers' and/or 30 mothers' occupations (as seen in Ti l leczek & L e w k o ' s (2001) study) can predict academic paths, or career expectations. Parent's Level of Education. Parents' education levels are controlled for in a l l regressions. Respondents are asked to indicate the highest level o f education both their mothers and fathers have obtained. The responses range from some elementary school to post- graduate degrees, and are coded hierarchically. Mediating Variables. Gender Role Ideologies. Gender ideologies are the belief systems o f h o w men and women should behave and interact wi th the social wor ld . Gender ideologies are learned through imitation o f gender roles, wh ich are learned in early childhood. These ideologies are products o f learning that children engage in from infancy. Learning theory assumes that children learn gender roles and behavioral social expectations from parental interactions in infancy. Learning theory should be measured with a scale that reflects ideologies and beliefs o f roles, wh ich reflect their learning environments as children. The br ief 15 item version Attitudes Towards W o m e n Scale ( A W S ) (Spence, Helmreich, & Stapp, 1973) was included i n the survey to measure gender and gender role ideologies. The A W S is the most frequently used scale o f gender role attitudes ( M c H u g h & Frieze, 1997) and has more than satisfactory psychometric properties i n measuring gender-role ideologies (Spence & Hahn, 1997). The shorter version was used as it has been found to have reliability scores in the mid -.80s and higher, as w e l l as having a unifactorial structure (Spence & Hahn, 1997; Spence et a l . , 1973). In the fifteen item scale, respondents are asked to respond to statements about women ' s roles. The forced choice responses are asked on a four point scale the degree to wh ich they agree to the statements. Scores are calculated to give an overall gender ideology score. Gender Schemas. Gender schema theory's assumption that gender is learned through a 31 categorizing schema system w i l l be tested. U s i n g the short version o f Bern 's (1974) Bern Sex Ro le Inventory ( B S R I ) , respondents gender schemas in adulthood were measured. The short version o f the B S R I measures different components o f the gender schema and was tested i f the gender schema was a better predictor o f gender decisions regarding sex-typed academic specializations and career toys. The original 60-item version was not used, as it is considered by many as being less psychometric than the short version (Hoffman & Borders , 2001). The items from the B S R I are scored on a 7 point scale. Respondents use this scale to indicate how items describe them. Hoffman and Borders (2001) did a twenty-five year review o f the B S R I and found that the short version has higher internal consistency than the original version, and demonstrates reliability and validity. The standardized scores are used in the analysis. Procedure Permission to enter into undergraduate courses was obtained prior to recruiting participants from university classes. Students were approached during class time with a br ief presentation outlining the research without giving away the hypotheses and assumptions behind the research. The surveys were given in addressed envelopes with prepaid postage. I f the respondents did not want to mai l the surveys, the researcher came to the fol lowing 2 scheduled classes. Classes in various departments were surveyed in order to obtain variation in declared specializations in the sample. Data Analysis The principle statistical tests are O L S regression. A s most o f the survey consists o f scale variables, regression is needed to interpret the varying levels o f the independent variables to the levels o f dependent variables. Mul t ip le regression is used to test the relationships in the models. Ba ron and K e n n y ' s (1986) explanation o f h o w to measure mediating effects is used to measure the gender ideology and gender schema effects on the outcome variables. B y 32 definition, variations in independent variables must account for variations in the mediator, which in turn accounts variations in the outcome variables, or criterion (Baron & Kenny, 1986; Holmbeck, 1997). From the models' descriptions, variations in socialization account for variations in gender ideologies and gender schemas. These mediators may account for variations in career and university decisions. Multiple regression is used in four conditions to test each independent variable. The first analysis assesses the significance of the relationship between the independent variables and the mediating variables. The significance of the path between the independent variables and the outcome variables are assessed next. Then both the independent variables and the mediators are used in the third equation to predict the outcome variables, where Baron and Kenny (1986) recommend simultaneous entry as opposed to hierarchical regression. This is recommended so that the mediator is controlled for when assessing the effect of the independent variable on the outcome variables, and the independent variable is controlled for when assessing the effect of the mediators on the outcome variables. When the mediator is controlled for, the relationship between the independent variables and the outcome variables should be smaller than the relationship when it is not controlled for (in the second equation) (Holmbeck, 1997). Relationships that do not attain statistical significance will be reported. Due to the small sample size, p levels up to .10 will be reported. The focus will be on the relationships between the variables that are found within the sample. As the target sample consists of respondents that have volunteered to participate in the study, the sample is a non-probability sample. The respondents are not randomly selected, and are not representative of a population. The nature of non-probability samples limits inference to a larger population, for both statistically significant and non-significant relationships. Relationships that do not reach significance cannot be inferred beyond to other populations. 33 For all o f the relationships tested, parental occupation and level o f education are controlled. T o enhance clarity, the control variables are not reported in the tables. The models are tested for their overall relationships with a l l five independent and mediating variables. W h e n a dummy variable for gender is used in the regressions, many o f the relationships do not hold for the overall sample, indicating that the gender o f the respondent changes the relationships. The models are thus also tested for female only and male only respondents. A s the sample was not randomly selected from the general population, regressions that y ie ld relationships that are not significant w i l l be discussed in this section. The issue o f the significance levels w i l l be discussed in the limitations sections. Results Sample Description Three hundred and sixty-eight surveys were handed out to nine senior level courses, and one lower level course. O f the 368 surveys, 279 were returned and 277 were useable. The overall response rate is 75.8%. The courses that were surveyed were from varying departments and subjects (Table 1). The sample is comprised o f 175 women and 102 men. The average age o f the respondents is 21.9 years. The sample's ethnic origin is mainly A s i a n and Nor th Amer ican (see Table 1). The respondents have highly educated parents: the majority o f mothers and fathers had post-secondary training. The majority o f the respondents (79.1%) have declared their majors, while 20 .2% have not. The respondents that had not declared their majors indicated which majors they w i l l be declaring. Referring to Table 1.1, over hal f o f the sample indicated they have declared specializations that have between 50 and 75 percent females registered in their majors. 34 Table 1. Sample Description. Independent Variable Total Percent Class Surveyed Fami ly Studies 106 38 .3% C i v i l Engineering 53 19 .1% Mathematics 20 9.0% Sociology 19 7.2% Elec t r ica l Engineering 14 . 5 . 1 % Computer Science 14 5 .1% History 14 5 .1% Philosophy 12 4 . 3 % Ethnic Origin A s i a n 111 4 0 . 1 % Nor th-Amer ican 90 32 .5% European 26 9.4% M i x e d Background 22 7.9% South-East A s i a n 20 7.2% M i d d l e Eastern 6 2 .2% Afr ican 2 0.7% Mothe r ' s Highest L e v e l o f Educat ion Post-Graduate Degree 19 6.9% Universi ty Degree 115 41 .5% Some Post-Secondary 62 22.4% Highschool D i p l o m a 49 17.7% Some or no Highschool 27 9.8% Father 's Highest L e v e l o f Educat ion Post-Graduate Degree 53 19 .1% Universi ty Degree 112 40 .4% Some Post-Secondary 47 17.0% Highschool D i p l o m a 32 11.6% Some or no Highschool 31 11.2% Note. n=277. 35 Table 2. Percentages of Females Enrolled in Declared Majors/Specializations. Independent Variable Total Percent W o m e n 175 63 .2% Percent female enrolment In Specialization - 64 .6% M e n 102 36 .8% Percent female enrolment In Specialization - 43 .2% Declared Specialization 219 7 9 . 1 % Specialization Enrollment 2 5 % or less female 40 14.6% 25-50% female 41 14.9% 50-75% female 145 52 .3% 75-100% female . 48 17 .1% Note. n = 277. HP. Sex-typed Childhood Toys and Play Predict Sex-typed University Specializations. Overall. The beta for overall regression between the feminine toys and play variable and the percent female enrollment in specializations is .41 (p = 0) (Table 3). P laying with feminine toys and games explains 17.2% o f the variance in the relationship. There is a moderate, negative relationship between the masculine toys and play variable and the percent female enrollment variable (fi = -.27, p = 0) for the overall sample. Mascul ine toys and game play explains 7.8% o f the variance i n the specialization variable. Neutral toy play was also posit ively related to the percent o f females i n the declared major for the overall sample. These relationships indicate that overall , there is a positive relationship between playing with feminine toys and in feminine games and choosing female dominated university specializations. P lay ing more frequently wi th feminine toys and games, and with neutral toys, is related to declaring a specialization with a higher female to male ratio. Higher frequency o f 36 play wi th masculine toys is related to being enrolled in specializations wi th lower female enrollment. Sex-typed childhood toys predicts enrolling i n sex-typed university majors, corifinning hypothesis 1 for the overall sample. Women. W h e n the direct relationship for the women in the sample is tested, the correlation is very small and not significant between feminine toy and game play and traditionally female specializations (Table 4). A m o n g female respondents, an extremely weak direct negative relationship between masculine toy and game play and the number o f females enrolled in declared majors. The relationship between neutral toy play and female enrollment in university specializations is positive, but weak and not significant. These results indicate that playing with any sex-typed toys in chi ldhood for women does not predict enrolling i n sex- typed university majors in adulthood. Hypothesis 1 does not hold for the women in this sample. Men. The beta for the relationship for the men in the sample between feminine toys and play and traditionally female majors is .20, and is close to significance (p =.06) (Table 5). P laying with feminine toys and games explains 7% o f the variance in declaring a female dominated major for the men in the sample. A higher reported frequency o f playing with feminine toys and games in childhood is related to enrolling in university specializations that are more traditionally female. Mascul ine toy and game play and neutral toys are not significantly related to enrolling in sex-typed majors. Hypothesis 1 only holds true for men for feminine-typed play. H2: Parent-child Play Predicts Sex-typed University Majors. Overall. The strength o f the overall relationship between the frequency o f playing with mother and enrolling in female dominated specializations is very weak and not significant (Table 3). There is no relationship for the overall model between playing wi th fathers and enrolling in female dominated university specializations. Hypothesis 2 does not hold true for 37 the overall model, as playing with mothers and fathers does not predict sex-typed university majors. Women. Fo r women, playing more frequently with their mothers is negatively and weakly related to enrolling in female dominated specializations (Table 4). This relationship is not significant. P lay ing with fathers is not related to sex-typed specialization. There is no correlation between the frequency o f playing with fathers and the sex-typed nature o f the university majors for women. Hypothesis 2 is not supported for the women in the sample. Men. The frequency o f play wi th mothers and wi th fathers in chi ldhood are both weakly related sex-typed specializations (Table 5). The relationship for playing wi th fathers is stronger, and explains 4 .7% o f the variance in declaring female dominated university majors. However , neither relationship is significant. Despite the significance level , playing with both mothers and fathers more often in chi ldhood is weakly related to enrolling in more feminine- typed majors for men, which provides evidence that hypothesis 2 holds for men. H4: Sex-typed Childhood Toys and Play Predict Sex-typed Career Choices and Expectations. Overall. In the overall model , playing with feminine toys and games has a weak negative relationship to the occupational role commitment, a weak positive relationship to both parenting and marital role commitments, and a moderate positive relationship to home care commitment. The relationship to the home care role is the only significant relationship. However , feminine play explains only 1.9% o f the variance in home care commitment levels. P laying with masculine toys and games has only one significant relationship to any o f the role commitment variables. Mascul ine toy and game play in childhood is positively related to occupational role commitment. Neutral toy play is negatively related to occupational role commitment. Neutral play is has a weak, positive relationship to the home care role commitment, though not significant (.11, p =.09). P lay ing more frequently wi th neutral toys in Table 3 Regressions for the Overall Sample Between Childhood Socialization Variables and Adulthood Outcome Variables Independent Variables Female Major Occupation Role Parenting Role B 13 B 13 B B Marital Role 13 Home Care Role B 13 Toy Variables Feminine Toys and Games R2 Masculine Toys and Games R2 Neutral Toy Play R2 Parent Play Variables Playing with Mother R2 Playing with Father R2 8.89** .17 -7.47** .08 5.08** .04 .85 ,01 -.47 .01 .41 -.27 .20 .04 -.02 -.07 .02 .14** .03 -.12* .03 -.05 .01 -.08* .02 .10 .16 -.15 -.08 -.12 .05 .03 -.02 .03 .08 .04 .04 .03 .12** .06 .06 -.02 .10 .07 .17 .04 .01 -.05 .01 .02 .01 .07* .03 .05 .02 .07 -.07 .04 .14 .09 .08* .02 -.09 .01 .09f .02 .12* .03 .11* .03 .12 .10 .11 .16 .15 Note. All regressions controlled for parent education and employment variables; tp̂ O.10; *p<0.050; **p̂ 0.01, n - 111. Table 4 Regressions for Women's Relationships Between Childhood Socialization Variables and Adulthood Outcome Variables Independent Variables Female Major Parenting Role B 13 B B B B Marital Role B Home Care Role B B Toy Variables Feminine Toys and Games R2 Masculine Toys and Games R2 Neutral Toys and Games R2 Parent Play Variables Playing with Mother R2 Playing with Father R2 1.02 .01 -.51 .01 1.93 .01 -1.62 .01 -.35 .01 .04 -.02 .09 -.10 -.02 .06 .02 .08 .02 -.01 .02 .02 .02 -.11* .04 .06 .06 .01 .04 -.17 .12 .04 .10 .03 .14 t .05 .07 .04 .16 ** .07 .11 .08 .15 .10 .22 .15 * .03 .00 .00 .05 .01 .07 .02 .03 .01 .17 .00 .07 .13 .05 .05 .01 -.05 .01 .05 .01 .04 .01 .09f .02 .05 .04 .06 .06 .13 Note. All regressions controlled for parent education and employment variables; fp<0.10; *p̂ 0.050; **p̂ 0.01, n - 175. CO CO Table 5 Regressions for Men's Relationships Between Childhood Socialization Variables and Adulthood Outcome Variables Independent Female Major Occupation Role Parenting Role Marital Role Home Care Role Variables ; B 13 B B B B B B B B Toy Variables Feminine Toys and Games 10.16f .20 -.13 -.08 -.23 -.13 -.06 -.05 .22 .13 R2 .07 .03 .05 .03 .03 Masculine Toys and Games 5.40 .15 .18 .15 -.07 -.06 -.21* -.24 -.04 -.03 R2 .05 .04 .04 .08 .01 Neutral Toys and Games 2.12 .09 .21* -.26 -.02 -.02 .00 -.01 .09 .11 R2 .04 .08 .03 .03 .02 Parent Play Variables Playing with Mother 2.48 .11 -.19** -.27 -.01 -.01 .08 .16 .23** .30 R2 .04 .09 .03 .05 .10 Playing with Father 2.70 .13 -.08 -.11 .08 .12 .09 .17 .16* .23 R2 .05 .03 .05 .05 .06 Note. All regressions controlled for parent education and employment variables; fp<0.10; *p̂ 0.050; **p̂ 0.01, n— 105. 41 childhood is correlated with having lower occupational role comrnitrnent, and is weakly related to having higher comrnitrnent to home care roles. Similarly, playing more often with feminine toys and games in childhood is moderately related to being less committed to future occupational roles, and moderately related to having higher commitment to caring for future homes. Women. Women's scores on the feminine toys and game play variable are significantly and positively related to the marital role variable; all other relationships were not significant. There are no significant relationships between playing with masculine toys and games in childhood and any of the LRSS role dimensions. These relationships are also very weak. Playing with neutral toys in childhood is positively related to parental role commitment, though this relationship did not reach significance (fi =.15, p =.051; R2 = .03). For the women in the sample, playing more frequently feminine with toys and games is correlated to higher commitment to future marital roles. Playing with more frequently with neutral toys is related to women's higher commitment to future parenting roles. Men. Men's feminine toy and game play scores are not significantly related to the LRSS dimensions; however, there is a moderate negative correlation with the parental role (R^.05), and a positive correlation with the home care role (R*=.03). Playing with masculine toys and games in childhood is negatively related to marital role commitment, and explains 8.2 % of the variation in the marital commitment. This is the only significant relationship for masculine toys and games on the LRSS dimensions for the men; however, there was a positive relationship to occupational role commitment (6=15, p =.14). Childhood neutral toy play produces a negative relationship to occupational role comrnitment. The relationships to the other role commitment variables are not significant. Playing in feminine play is related to lower parenting role commitment, and higher home care commitment for men. Masculine toy 42 and game play for men is related to lower commitment to future marital relationships. Neutral toy play predicts higher occupational role commitment for men. H5: Parent-child Play Predicts Sex-typed Career Choices and Expectations. Overall. P laying wi th mothers in childhood is positively related to marital role commitment and home care role commitment. P laying with fathers was related to three o f the o f the four L R S S role dimensions. P laying more often with fathers in childhood was related to lower occupational role commitment, and higher for both parenting and home care role commitment in adulthood. Women. The female respondents have two moderate relationships between playing with their mothers and being committed to future parenting and marital roles. P lay ing with fathers for women is negatively related to occupational role commitment, and positively to parental role commitment. P lay ing wi th fathers explains 4.2 % o f the variance o f occupational role commitment scores, and 7.2% o f the variance in the parenting role commitment scores. There is a moderate positive relationship to the home care role, though it is not significant (13 =.13, p =.09). Higher frequency o f play wi th mothers in chi ldhood is correlated with holding higher commitment to being a parent and mamtaining a marital relationship for the women in the sample. W o m e n who played more often wi th fathers i n childhood are more l ikely to be more committed to being a parent, and are less l ikely to be committed to future employment. Men. For the male respondents, playing wi th mothers in childhood is negatively related to occupational role commitment, and explains 8.8% o f the variance in the role. There is a moderate positive relationship to the marital role commitment, but it was not significant (13= 16; p =.12). P laying wi th both mothers and fathers in childhood was positively related to the home care role commitment ( R 2 = .10; R 2 = .06). The positive relationships to the martial and parenting roles are moderate in size, though not significant. There was a weak negative 43 relationship to the occupational role. Less commitment to future occupational roles is related to playing more frequently with both parents in childhood, especially with mothers. Playing more frequently wi th both parents in childhood is also related to being more committed to maintaining a marital relationship and maintaining future homes i n adulthood. P lay ing with fathers is moderately related to being committed to being a parent. HI:- Gender Role Ideologies Mediate the Relationships tested in Hypotheses l,2,4,and 5. Overall There is no relationship between playing with fathers and gender role ideologies, wh ich discounts any mediating relationships wi th this independent variable for the overall sample (Tables 6 & 7). W h e n the percent female enrollment is tested as a dependent variable, the regression analysis indicates that gender ideologies have a mediating effect on the fol lowing independent variables: feminine and masculine toy and game play, neutral toy play, and the frequency o f play wi th mothers, though this relationships is weak and not significant. Compared to the direct relationship, playing with masculine toys and games in chi ldhood explains more o f the variance (12.4%) in enrolling in female dominated majors when gender ideologies are tested as a mediator. Gender role ideologies mediate only one other relationship, wh ich is between feminine toy and game play and the occupational role. The direct relationship is not significant, but the relationship decreases close to zero in the mediating regression. The perception o f what gender appropriate roles should be mediates relationships for the overall sample between sex-typed chi ldhood toy play and declaring traditionally female university majors. This perception also explains part o f the relationship between childhood feminine play and being committed to future employment. Women. Fo r the women in the sample, gender role ideologies cannot be tested as a mediator, as there is no relationship between neutral toys and gender role ideologies. The weak positive correlation between playing with feminine toys in childhood is not significant, but 44 Table 6 Regressions for Overall Sample's Mediating Regression Between Neutral Toy Play and Adulthood Outcome Variables : Independent Variables Female Major Occupation Role Parenting Role B 13 ' B 13 B 13 Feminine Toys and Games 8.03** .37 -.08f -.02 .07 .09 Gender Role Ideologies 7.95* .14 .06 .04 -.17 -.09 R2 .19 .02 .04 Feminine Toys and Games 8.50** .39 -.05 -.08 .01 .02 Gender Schemas .14 .08 .00 -.09 .01** .18 R2 .18 .03 .06 Masculine Toys and Games 12.50** .22 .14 ** .16 -.02 -.02 Gender Role Ideologies .02 .00 .04 .02 -.13 -.07 R2 .12 .03 .03 Masculine Toys and Games -.62** -.23 .12 * .14 .03 .03 Gender Schemas .21* .12 .00 -.08 .01** .19 R2 .09 .04 .06 Neutral Toy Play 3.50* .15 -.13 * -.16 . io t .12 Gender Role Ideologies 12.22** .22 .06 .04 -.17 -.09 R2 .09 :03 .04 Neutral Toy Play 4.50** .18 -.11 * -.13 .06 .07 Gender Schemas .25* .14 .00 -.09 .01** .17 R2 .06 .04 .06 Note. All regressions controlled for parent education and employment variables; |p<0.10; *p<0.050; **p<0.01, n — 277. Table 6 Continued Regressions for Overall Sample's Mediating Regression Between Childhood Socialization Variables and Adulthood Outcome Variables 45 Independent Marital Role Home Care Role Variables B B B B Feminine Toys and Games .04 .08 .11* .15 Gender Role Ideologies -.06 -.04 -.20f -.11 R2 .01 .03 Feminine Toys and Games .01 .03 .07 .09 Gender Schemas .01* .15 .oit .11 R2 .03 .03 Masculine Toys and Games -.05 -.07 -.iot -.11 Gender Role Ideologies -.04 .00 -.14 -.08 R2 .01 .02 Masculine Toys and Games -.02 -.03 -.06 -.07 Gender Schemas .01* .15 .oit .12 R2 .03 .03 Neutral Toy Play .03 .05 .11* .13 Gender Role Ideologies -.04 -.03 -.17 -.09 R2 .01 .02 Neutral Toy Play .01 .02 .07 .08 Gender Schemas .01* .15 .01* .12 R2 .03 .03 Note. All regressions controlled for parent education and employment variables; fp<0.10; *p<0.050; **p<0.01, n — 211. Table 7 Regressions for Overall Sample's Mediating Relationships Between Parent Play and Adulthood Outcome Variables Independent Female Major Occupation Role Parenting Role Marital Role Home Care Role Variables B 13 B 13 B 13 B 13 B 13 Playing with Mother .01 .00 -.06 -.08 .06 .08 .08* .14 .12** .17 Gender Role Ideologies 14.12** .25 .03 .02 -.15 -.08 -.06 -.04 -.17 -.09 R2 .06 .01 .04 .03 .04 Playing with Mother .51 .02 -.05 -.07 .04 .05 .07* .12 .01* .14 Gender Schemas .30** .17 -.01t -.11 .01** .18 .01* .15 .02* .13 R2 .04 .03 .06 .05 .05 Playing with Father -.67 .03 -.08* -.12 .12** .17 .05 .09 .11* .16 Gender Role Ideologies 14.26** .25 .01 .01 -.13 -.07 -.03 -.02 -.13 -.07 R2 .07 .02 .06 .02 .03 Playing with Father -.13 -.01 -.09* -.13 .19 .05| .10 .11** .17 Gender Schemas .30** .17 -.01* -.12 .01** .20 .01** .17 .01* .15 R2 .03 .04 .10 .04 .05 Note. All regressions controlled for parent education and employment variables; "fp<0.10; *p<0.050; **p̂ 0.01, n = 277. CD Table 8 47 Women's Regressions for Mediating Relationships Between Childhood Socialization and Adulthood Outcome Variables Independent Variables Female Major Occupation Role Parenting Role B B B 6 B B Feminine Toys and Games .75 .03 .04 .04 .13 .11 Gender Role Ideologies 4.93 .09 .31* .16 -.12 -.06 R2 .01 .04 .04 Feminine Toys and Games .39 .01 .08 .07 .11 .09 Gender Schemas .20| .14 .00 -.08 .01 .11 R2 .02 .03 .05 Masculine Toys and Games -1.01 -.03 .04 .04 .12 .09 Gender Role Ideologies 5.41 .10 •31t .16 -.13 -.06 R2 .02 .04 .03 Masculine Toys and Games .02 .00 .07 .06 .12 .09 Gender Schemas .20t .14 .00 -.06 .011 .13 R2 .02 .02 .05 Neutral Toys and Games 1.76 .08 -.02 -.02 .15* .16 Gender Role Ideologies 4.73 .09 .32* .17 -.12 -.06 R2 .02 .04 .05 Neutral Toys and Games 1.64 .07 -.01 .00 .13| .14 Gender Schemas .19t .13 .00 -.07 .01 .11 R2 .03 .02 .06 Note. All regressions controlled for parent education and employment variables; fp^O. 10; *p<0.050; **p<0.01, n — 175. 48 Table 8 Continued Women's Regressions for Mediating Relationships Between Childhood Socialization and Adulthood Outcome Variables . Independent Marital Role Home Care Role Variables B 13 B 13 Feminine Toys and Games .16* .18 .06 .06 Gender Role Ideologies -.20 -.12 -.26 -.13 R2 .04 .02 Feminine Toys and Games .14* .15 .04 .04 Gender Schemas .00 .08 .00 .06 R2 .04 .01 Masculine Toys and Games .02 .02 -.03 -.02 Gender Role Ideologies -.17 -.10 -.24 -.12 R2 .01 .02 Masculine Toys and Games .01 .01 -.05 -.04 Gender Schemas .00 .10 .00 .06 R2 .01 .01 Neutral Toys and Games .06 .08 .06 .07 Gender Role Ideologies -.18 -.11 -.26 -.13 R2 .02 .02 Neutral Toys and Games .04 .06 .05 .05 Gender Schemas .00 .09 .00 .06 R2 .02 .01 Note. All regressions controlled for parent education and employment variables; fP^O-10; *p̂ 0.050; **p̂ 0.01, n — 175. Table 9 Women's Regressions for Mediating Relationships Between Parent Play and Adulthood Outcome Variables Independent Variables Female Major fl Occupation Role Parenting Role Marital Role Home Care Role B B 13 B 13 B 13 B 13 WOMEN Playing with Mother Gender Role Ideologies R2 Playing with Mother Gender Schemas R2 -1.7 5.4 .02 -1.75 •21t .03 .10 .10 .10 .14 .02 .32* .04 .03 .00 .02 .03 .16 .05 -.08 .07 -.11 .04 .01 .01 .05 .11 -.05 .10 .12 .07f -.18 .03 .07 .00 .03 .13 .12 .10 .04 -.26 .02 .04 .00 .01 .07 -.13 .06 .07 Playing with Father Gender Role Ideologies R2 Playing with Father Gender Schemas R2 -.27 5.04 .01 -.10 .20t .02 -.02 .10 -.01 .14 -.10* .31* .07 -.12* ,01 .05 -.16 .16 -.18 -.09 .15** -.07 .07 .01* .09 .22 -.03 .23 .15 .02 -.16 .02 .03 .54 .02 .04 -.10 .06 .12 .09f -.23 .03 .10t .00 .03 .13 -.12 .14 .09 Note. All regressions controlled for parent education and employment variables; tP^O.10; *p<0.050; **p̂ 0.01, n= 175. -t>. CD Table 10 50 Men's Regressions for Mediating Relationships Between Childhood Socialization and Adulthood Outcome Variables Independent Female Major Occupation Role Parenting Role Variables B B B B B B Mediating: Feminine Toys and Games 8.77| .17 -.10 -.06 -.20 -.11 Gender Role Ideologies 9.12 .17 -.21 -.12 -.19 -.10 R2 .10 .04 .06 Feminine Toys and Games 10.08J .20 -.09 -.06 -.33 -.18 Gender Schemas .02 .01 -.01 -.12 .02* .30 R2 .07 .04 .13 Mediating: Masculine Toys and Games 5.39 .15 .18 .15 -.07 -.06 Gender Role Ideologies 10.66| .20 -.23 -.13 -.22 -.12 R2 .09 .06 .05 Mediating: Masculine Toys and Games 5.90 .16 .15 .13 .00 .00 Gender Schemas .13 .08 -.01 -.11 .02* .26 R2 .06 .05 .10 Mediating: Neutral Toys and Games .88 .04 -.19* -.24 .01 .01 Gender Role Ideologies 10. iot .19 -.12 -.07 -.23 -.12 R2 .07 .09 .05 Neutral Toys and Games 2.00 .08 -.19 * -.24 -.05 -.07 Gender Schemas .05 .03 -.01 -.09 .02** .28 R2 .04 .09 .11 Note. All regressions controlled for parent education and employment variables; tp^O. 10; *p̂ 0.050; **p̂ 0.01, n - 102. Table 10 Continued Men's Regressions for Mediating Relationships Between Childhood Socialization and Adulthood Outcome Variables Independent Marital Role Home Care Role Variables B B B B Feminine Toys and Games -.08 -.07 ' .25 .14 Gender Role Ideologies .15 -.12 -.15 -.09 R2 .04 .03 - Feminine Toys and Games -.13 -.10 .16 .09 Gender Schemas .01** .29 .oit .19 R2 .11 .06 Masculine Toys and Games -.21* -.24 -.04 -.03 Gender Role Ideologies .14 .11 -.11 -.06 R2 .09 .02 Masculine Toys and Games -.17f -.20 .01 .01 Gender Schemas .01* .23 .01* .21 R2 .13 .05 Neutral Toys and Games -.02 -.04 .12 .14 Gender Role Ideologies .15 .12 -.18 -.10 R2 .04 .03 Neutral Toys and Games -.03 -.06 .06 .08 Gender Schemas .01** .28 .oit .20 R2 .10 .06 Note. All regressions controlled for parent education and employment variables; fp^O. 10; *p̂ 0.050; **p<0.01, n— 102. Table 11 Men's Regress ions for Mediating Relationships Between Parent Play and Adulthood Outcome Variables Independent . Female Major Occupation Role Parenting Role Marital Role Home Care Role Variables B 13 B 6 B 13 B (3 B 13 Mediating: Playing with Mother 1.51 .07 -.18 -.25 .02 .03 .08 .14 .25** .33 Gender Role Ideologies 9.73f .18 -.12 -.07 -.24 -.13 .09 .07 -.26 -.15 R2 .07 .09 .05 .05 .11 Playing with Mother 2.40 .11 -.06 -.09 .11 .16 .08 .16 .18* .26 Gender Schemas .05 .03 -.19 -.11 -.30 -.16 .09 .07 -.23 -.13 R2 .04 .04 .07 .06 .07 -.19 Mediating: Playing with Father 1.79 .09 -.06 -.09 .11 .16 .08 .16 .18* .26 Gender Role Ideologies 9.48 .18 -.19 -.11 -.30 -.16 .09 .07 -.23 -.06 R2 .08 .04 .07 .06 .07 Mediating: Playing with Father 2.8 .13 -.07 -.11 .08 .11 .08f .17 .16 * .23 Gender Schemas .08 .05 -.07 -.13 .02 ** .26 .01** .27 .01 * .21 R2 .05 .05 .12 .11 .10 Note. All regressions controlled for parent education and employment variables; fp<0.10; *p<0.050; **p<0.01, n= 102. N> Figure 2. Women's Relationships. Neutral Toy Play (. 154^ Occupational Role Commitment Parental Role Commitment Marital Role Commitment Home Care Role Commitment Figure 3. Men's Relationships: Model 1. Feminine Toys & Play (19? Masculine Toys & Play U471 * / &/ V Play with Mother * Play with Father tp^O.10; «p s 0 .050; "psO.Ol; ^mediatedrelationships O l Figure 4. Men's Relationships: Model 2, +P<0.10; *p,0.050; **p.O.OI; ^cdiatcd^atioasbips 56 strong enough to test mediating relationships (Tables 8 & 9). Gender role ideologies have a mediating effect only on the relationship between feminine toy and game play and the percent female enrollment. The direct relationship is too small (and not significant) to conclude a mediating relationship. Gender role ideologies do not mediate any o f the relationships between sex-typed toy play in chi ldhood and declaring traditionally female or male majors in university. The direct relationships between masculine toy and game play and occupational and home care commitments are weak and not significant, but are mediated by gender role ideologies as wel l . Men. F o r the male respondents, there was no relationship between masculine toy play and gender ideologies, thus it cannot be defined as a mediator. Though the relationship from feminine toy play to gender ideologies is not significant, the size o f the relationship gives justification to test it as a mediator (Tables 10 & 11). W h e n the regressions that test the percent female enrollment in majors as the dependent variable are conducted, the independent variables that are mediated by gender role ideologies are feminine toy and game play, neutral toy play, and the frequency o f play wi th both parents. However , the feminine toy play relationship is the closest to significance (6 = .200, p =.06). The impact o f playing with feminine toys and games and childhood on parental role commitment is partially mediated by gender role ideologies, though the direct relationship is not significant. The meditating model explains 5.9 % o f the variation i n the parenting role. Gender role ideologies explain part o f the relationship between neutral toy play and occupational commitment, but no other relationships to the other L R S S roles. The impact o f the frequency o f playing with parents on being committed to future employment is mediated by gender role ideologies, though only the frequency o f play wi th mothers' direct relationship is significant. There is also a mediating (but not significant) effect between frequency o f play wi th mothers and the marital role commitment. Be ing committed to future occupational roles 57 and being committed to mamtaining future marital roles is related to time spent playing with mothers and fathers in childhood, as well as by individuals' gender role ideologies. Playing more with feminine toys for men in childhood is related to lower commitment to parenting roles, and is mediated by gender role ideologies. H8: Gender Schemas Mediate the Relationships tested in Hypotheses l,2,4,and 5. Overall. In the overall sample, gender schemas mediate the relationships between feminine toys and game play and the following variables: the percentage of females enrolled in declared specializations, the occupation role, and the home care role. The weak relationships to the parenting and marital roles are also mediated by gender schemas, but these relationships are not significant. Although none of these relationships decrease to zero when the mediator is added into the regression equations, mediating effects are still found. Gender schemas mediate the relationships the overall sample has between engaging in feminine play in childhood, and the extent to which their chosen majors are traditional, as well as their commitment to their future occupational and home care roles. Gender schemas also mediate the relationships between childhood masculine toy and game play and neutral toy play on the percent of females enrolled in declared majors. Masculine toy play and gender schemas explain 8.6% of the variance in declaring female dominated majors, which is less than the gender ideologies explain. The relationship between masculine play and occupational role commitment is also mediated by gender schemas: the direct relationships to the parenting and marital roles are weak and not significant, but gender schemas still act as mediators. Gender schemas mediate all relationships between neutral toy play and the four role dimensions, though only the direct relationship to the occupational role was significant. None of these are perfect mediating relationships, as the direct relationships do not decrease to zero. 58 Neither the frequency of play with mothers nor, the frequency of play with fathers produce relationships strong enough to define gender schemas as a mediator. Though gender schemas do produce mediating effects on some of the tested relationships with the parent play measures, by definition, a mediator must be caused by the independent variable. These weak causal relationships are too small to definitively argue that gender schemas are mediators. Women. The women's frequency of play with mothers is not related to gender schemas, and thus gender schemas do not mediate any relationships with the mother play variable. Gender schemas have a mediating effect on the three sex-typed toy play independent variables when regressions are run to the percent female enrollment in university majors (Figure 2). However, it must be noted that none of the direct relationships are significant, and are weak. This indicates that gender schemas, or individual self-perceptions of masculinity and femininity, explain a small part of the relationships between playing with sex-typed toys in childhood and whether respondents chose to enroll in traditionally female or male university majors. The positive relationship between playing with feminine toys and games in childhood and marital role commitment is mediated by gender schemas. This effect is small; feminine play explains 4.4% of the variance in the marital role. Although the parenting role commitment variable does not have a significant relationship with feminine toys and games, gender schemas have a small mediating effect on the correlation. Masculine toys and games for the women in the sample did not create any significant relationships to the LRSS dimensions, though there is a mediating effect on the occupational role. Neutral toy play has a direct relationship to the parenting role commitment, but not significant ( 6 = .14; p =.051) and is slightly mediated by gender schemas. Gender schemas do not mediate any relationships between playing with fathers and the LRSS roles. 59 Men. For the men in the sample, only the toy play variables are related to gender schemas, though masculine play and neutral toy play are not significant (B= -.19; p =.06, = 17: p =.10). Despite the significance levels, the strength of the relationships and the small size of the male sample are considerations for testing gender schemas as mediators. Out of the three toy play variables, only the relationship between neutral toy play and percent female enrollment in declared specializations is mediated by gender schemas; this mediating effect is not very strong. This indicates that self-perceptions of masculine and feminine traits have a small impact on the relationship to how traditionally female the declared university specialization. Gender schemas mediate the relationship between feminine toys and game play on the home care role, though the direct relationship is not significant. The relationships between all three toy variables to the occupational role commitment variable are also mediated by gender schemas, however, only the direct relationship from neutral toy play is significant. The effect of masculine toys and game play on the marital role for men is mediated by gender schemas, as the direct relationship decreases. The relationship to occupational role commitment is not significant (.15, p =.14), but is mediated by gender schemas. Gender schemas explain part of the relationship between feminine childhood play and declaring traditionally female major, as well as the relationship between playing in masculine play and the extent to which men are committed to future marital roles. Neutral toy play effects on occupational role commitment is mediated by gender schemas Discussion The purpose of this study is not to find direct relationships between sex-typed toys and play and adulthood career choices, but rather to find out if early childhood gender socialization predict specific adulthood outcomes. Sex-typed toys are used as indicators of the gender 60 socialization subjects were exposed to i n early childhood. These indicators are tested in relation to adulthood career and university expectations and outcomes to find whether early chi ldhood gender socialization matters in terms o f outcomes. The questions are centered around finding out i f chi ldhood variables are important factors that carry through to adulthood. The focus o f this study was to understand the gender socialization process from infancy into adulthood. The parameters o f this research only a l low for certain variables to be tested, and it must be made clear that there are many intervening variables that affect gender development from chi ldhood into adulthood. Feminine socialization on young children does matter for some adults and their career outcomes. M e n are particularly sensitive to this type o f socialization, as it can predict their chances o f enrolling in traditionally female university specializations. B e i n g exposed to feminine socialization in early chi ldhood may be an indicator that parents are more wi l l ing to expose these boys to other types o f fenrinine socialization such as feminine academic options, wh ich opens the doors for the men to be comfortable to pursue these academic fields. For men, however, masculine socialization also has the same effect on their academic outcomes. This relationship can be explained by the fact that boys receive more sex-typed messages in their childhood, and masculine play may be a universal part o f growing up as a boy. However , there is strong evidence that masculine socialization for men and feminine socialization for women predicts marital relationship expectations and commitment levels. Be ing exposed to gender appropriate social izing agents, and neutral toys for women, teaches children about the relationship roles they should expect in adulthood. W o m e n ' s traditional roles include mamtaining the emotional climate in marriages, which mirrors the findings o f this study. These early social izing agents could be reflections o f the gender roles that were instil led in the respondents i n childhood. Specific toys such as dolls and dollhouses teach women to 61 expect to be involved in maintaining their marriages, while the lack o f these toys in men's development may inhibit the sense o f obligation to take responsibility for relationship maintenance. Research has shown that neutral toy play is encouraged more often in girls than in boys (e.g., Caldera et a l . , 1989). N o t only are girls encouraged to play more frequently wi th neutral toys, and mothers play more frequently with neutral toys with their children (Idle et a l . , 1993). Mothers ' neutral toy play serves as a model o f feminine-typed behavior. In the case o f being committed to future marital roles, neutral toy play is associated wi th higher role commitment for women. These results fo l low the findings from previous research, and emphasize the role o f neutral toys in feminine socialization. P laying more frequently with neutral toys for men is associated wi th being less committed to future employment and careers. The ' feminine' nature o f the neutral toys suggests that more exposure to these toys for boys teaches feminine roles, or fewer masculine roles. Commitment to careers is traditionally a masculine role, and increased exposure to neutral play is related to a decrease in the desire to maintain a career for men. These findings then pose more questions about the definition o f neutral toys. I f neutral toys are encouraged more for girls, and less for boys, and considering the findings regarding occupational commitment for men, then neutral toys may be another description for feminine toys. Neutral toys may be a term that describes feminine toys that do not hold the same degree o f feminine qualities as feminine toys, but are categorically recognized as being more appropriate for girls than for boys. The implications o f the findings regarding neutral toy play add to the complexity o f defining and testing gender in social research. The feminization o f neutral toy play also suggests that masculinity and femininity are not absolute categories along a dichotomous dimension, but can be conceptualized in varying degrees. I f neutral toys are milder forms o f feminine toys, then the question o f what is defined as a milder form o f 62 masculine toys also arises. P laying wi th parents also raises some questions about h o w gender roles are being taught and learned. T o date, it is k n o w n that fathers w i l l choose sex-typed toys in play situations more often that mothers, who prefer neutral toys (Idle et al . , 1993); these play interactions with toys teach children about gender roles (Schwartz & Marham, 1985). Fo r the men in the study, the more they played with mothers, the more committed they were to their future careers as w e l l as their marital roles. P laying with their fathers was slightly related to lower occupational commitment, and higher marital commitment. It can be hypothesized that for boys who play more often with their mothers, less importance on maintaining a career and more emphasis on marital roles may be taught. F o r the fathers who spent more time playing with their sons, they could have been stepping out o f their role as the family 'breadwinner ' to emphasize the importance o f having other roles, such as a stronger involvement in parenting. F o r girls, more time spent wi th fathers is associated with lower occupational commitment. This association might stem from gender-appropriate play that fathers engage in , wh ich emphasizes gender roles and wou ld educate girls on their future roles as women. Playing more with fathers in childhood also predicts stronger desires for parenting roles i n women. The more time spent wi th fathers i n childhood is related to higher parenting expectations. Lindsey and M i z e (2001) found the context o f the parent-child interactions is also important factors in parent-child gender socialization. The context o f the play situations respondents had wi th their parents, especially with their fathers, may an important variable that can explain the findings from this study. Playing with both parents increases the desire for men to maintain their future homes; however, the home care role dimension can be interpreted as either a traditionally female or a traditionally male task. The nature o f these questions does not specify what type o f home care 63 is desired to maintain. The context o f the upkeep o f the home is debatable, as the traditionally feminine context resides within the home (laundry, cleaning, cooking) and the traditionally masculine context is centered around the exterior o f the home ( lawn care, garbage, painting, roofing). The desire for commitment to home care roles may be l inked to the contextual research, where Lindsey and M i z e (2001) found that parents choose to interact wi th their children in sex-typed contexts. Both mothers and fathers may have contributed to their children's commitment levels to home care by defining certain home care tasks as being sex- typed. For example, mothers may emphasize the necessity for a clean house, and fathers may emphasize the need for regular lawn care. Gender ideologies and gender schemas are important variables that were considered in the execution o f this study. The extent to wh ich one defines their own masculinity and femininity as w e l l as one's interpretations o f what appropriate gender roles are, both account for adulthood career choices and expectations. Just as sex-typed toys cannot be considered as perfect predictors o f adulthood gender-related decisions, gender ideologies and schemas cannot be considered as perfect mediators. Neither schemas nor ideologies are found to mediate a l l o f the tested relationships. Gender schemas and gender role ideologies are also cannot be defined as perfect mediators. For example, gender schemas mediate both relationships between women who played more wi th feminine toys and for men who played more wi th masculine toys to their expected levels o f marital commitment. Personal reports o f masculine and feminine traits account for some respondents' marital commitment levels. Gender ideologies explain part o f the relationship between men playing more with feminine toys in chi ldhood and their increased l ikel ihood o f enrolling in female dominated majors in university. The fact that the mediating effects for both gender ideologies and schemas are not consistent over adulthood outcomes or for both genders reiterates the complexity o f the concept o f gender and is not easily quantified 64 or measured. Gender development and socialization are difficult to describe i n finite terms o f development processes. The mediating relationships exemplify the fact that gender socialization is dependent upon many different variables and factors, as w e l l as context. T o understand the process o f how gender schemas and ideologies affect adulthood decisions in relation to early childhood socialization, the findings o f the tested mediated relationships must be understood. Thus far, it has been found that gender ideologies do not mediate women 's relationships between any childhood sex-typed toy and game play and any o f the tested adulthood outcomes, nor any o f the relationships between playing with parents and adulthood outcomes. M e n ' s gender ideologies have some mediating effects between the frequencies o f chi ldhood feminine toy play to both enrolling in female dominated majors, as w e l l as to their commitment to being parents. The relationship between neutral toy play and enrolling in female dominated majors is also mediated by men's gender ideologies. The relationships between playing with both parents to enrolling in female dominated majors and to being less committed to careers and more committed to marital roles are mediated by gender ideologies for men. M e n ' s ideas o f what appropriate gender roles are can explain only some o f these relationships, in contrast wi th the lack o f influence ideologies have for women ' s adulthood outcomes. Fo r men, gender role ideologies are important factors in gender-related adulthood career decisions, and not for women. Gender schemas do have a mediating effect on the women ' s correlation between playing more with feminine and neutral toys and being more committed to marital roles. M e n are more influenced by their gender schemas than women are when describing their career and family expectations. M e n ' s relationships between feminine play and home care role commitment, neutral play and enrolling in female majors, and masculine play and occupational and marital commitments are a l l influenced by their gender schemas. M e n ' s chi ldhood gender 65 socialization and adulthood outcomes are affected by their sense of masculinity and femininity, and in more contexts than for women. In general, men's academic paths and career and family expectations are dependent upon both their gender ideologies and their gender schemas. In comparison, gender schemas affect only some of women's adulthood outcomes, while their gender role ideologies are not influential variables in those relationships. To conclude, it is apparent from this study that early childhood gender socialization matters where adulthood academic and career paths are concerned. The gender of the individual, their time spent playing with sex-typed toys and their parents, as well as the specific outcomes are important variables in the socialization process. Personal perceptions of masculinity and femininity as well as gender ideologies explain some of these relationships, and more so for men than for women. It is also evident that the path from early childhood gender socialization to adulthood gender-based decisions is not direct. The complexity of the path of gender socialization is related to the complexity of the concept of gender itself, where the concept has yet to be clearly defined and understood. Limitations A select number of relationships that did not reach statistical significance were reported in the results section. Relationships that had a p level between .05 and .10 "were reported. These relationships were reported and considered to be important to the models tested for several reasons. The participants were not randomly selected from their population, as the respondents voluntarily participated in the study. The sample in the study is not a representative sample of the population as all are university students, and the majority of them are in their third and fourth years of study. This non-probability sample creates limitations to inference of findings to a population beyond the sample in the study. The results of this study are confined to the selected the sample. These results may occur only within the tested sample. The sample size of 66 each gender category is a l ikely contributor to the fact that some relationships d id not reach significance. The male sample was smaller than the female sample. The concepts o f masculinity and fernininity can also raise some concerns. The concepts o f masculinity and feniininity within this study are based on previous literature that defines gender in their own words. For example, to Bern (1981, 1983) the concept o f being 'mascul ine ' is relevant to a coefficient that is derived from the number o f feminine and masculine traits the individual reports owning. B y quantifying their traits, the B S R I can give a raw score to an individual 's perceptions o f their o w n sense o f masculinity and fennninity, and place their score on a gender continuum to describe their gender. Gender ideologies are simply measures o f h o w the individual assumes social roles are appropriate for both sexes. The measure o f a feminine career path within this study is the number o f female students in a declared university specialization. Whi l e these examples define facets o f the concept o f gender, each can also be considered as being separate from the other. The discrepancy in the continuity o f the definitions o f gender can be a limitation o f the research, as they may not be defining the exact same concept. Bo th theoretical perspectives approach the concept o f gender as being a categorical concept, and not fluid. This limits the findings as this approach implies that gender can be sorted into discrete categories, further implying that there is no overlap between the categories. A p p l y i n g a f luid approach to the concept o f gender may have yielded different results and definitions o f sex-typed behaviors. This is an issue for al l gender research, but also defines the variations that are seen within gender itself. This study may be missing any new definitions o f gender currently held by the participants o f the study, as w e l l as the cultural definitions held within the community o f the participants. Certain variables that contribute to the gender socialization development from infancy to adulthood cannot be measured within the parameters o f this study. Effects o f the media, such 67 as television, video games, internet and magazines are factors that cannot be measured in this proposed research. Despite research that indicates the importance o f the family 's influence on childhood socialization (e.g., M a c c o b y , 1992), the importance o f increasing external media influences cannot be captured in this model. Some unique variables in individual families are other limitations. Whi le this study can ask about the number and sex o f siblings, the effects o f these factors cannot be measured. Hav ing older sisters as opposed to being the oldest brother or coming from a cultural system that has different expectations for gender appropriate behavior could mediate the influences o f family and mainstream cultures. Personality and other individual variations are also other possible factors in the model . The retrospective nature o f the study also creates some concern over the reliability o f the accounts o f toy and play styles. Respondents w i l l be asked about details o f their lives from over 15 years ago. These detailed events act as cued recall , wh ich helps the retrieval o f memories that have been stored for periods o f time. Engaging i n recall facilitates more recall wi th similar information (Schwarz, Hippler , & Noelle-Neumann,1994). The questions in the survey are designed to act as cued recall , instead o f open-ended responses, and the questions regarding toys and play i n childhood are similar, and should thus make the memory retrieval o f childhood play experiences more readily available. The reliability o f the Giul iano et al. (2000) study indicates that this method is acceptable. Gi lmar t in (1987) also conducted a retrospective study where male respondents were asked i f they played with masculine and feminine toys when they were 5 and 12 years o f age. The range o f ages o f the respondents ranged from 19 to 50 years o f age. These respondents were asked to recall toys they had as long as 45 years prior to the study. These studies are only two examples o f h o w research uses recall to get an account o f chi ldhood experiences. Overa l l , normative chi ldhood experiences are reliably recalled (Maughan & 68 Rutter, 1997). Adults appear to be reliable in their accounts of their episodic memories, which includes childhood memories. However, normative memory decay and distortions are variables that cannot be measured in this study. Forgetting events and not remembering events accurately are common issues in retrospective recall. For example, forgetting about certain toys one had, and distortions such as highly emotional events are factors that cannot be controlled for. Maughan and Rutter (1997) point out that questionnaires have been found to generate more accurate and detailed responses than interviews in some cases. While some recalled information may be forgotten or distorted, toys and friendships that had significant meaning in childhood are memorable experiences. Gender schemas may also affect the retrospective nature of this study, and could affect the validity of recalling childhood experiences. Gender schemas act as information processors that organize information into gender categories, and could thus create false recollections of the frequency of play with toys, parents, and games. For example, highly sex-typed individuals may readily recall childhood play that congruent with their gender more than cross-gender play. The accuracy of recalling gender-related information may be subject to individuals' salience of their own gender schemas. Neither gender ideologies nor gender schemas can be considered as perfect mediators between early parent-child gender socialization and adulthood career and academic outcomes. There are many other external and mediating factors that were outside of the parameters of this study that could affect adulthood career choices and expectations. Gender schemas appear to have more of a mediating effect on the relationships between childhood gender socialization and adulthood academic and career paths and expectations. The sense of holding masculine and feminine traits can explain more relationships between sex-typed childhood gender socialization and holding traditional and non-traditional expectations about future career and 69 family roles, compared to gender role ideologies. Ideologies about how the genders should behave are not as strong a variable in the gender socialization process. Despite the degree of the impact of gender ideologies and gender schemas as mediators, it is important to note that both affect certain aspect of the gender socialization process. 70 References Amatea,, E . S . , Cross , E . G . , Clark , J . E . & B o b b y , C . L . (1986). Assess ing the work and family expectations o f career oriented men and women: The L i f e R o l e Salience Scales. Journal Marriage and the Family, 48, 831-831. Baron , R . M . & Kenny , D . A . (1986). The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51,1173-1182. Bern, S .L . (1981). Gender schema theory: A cognitive account o f sex typing. Psychological Review, 88, 354-364. Bern, S .L . (1983). Gender schema theory and its implications for chi ld development: Ra i s ing gender-aschematic children in a gender-schematic society. Journal of Women in Culture and Society, 8, 599-616. Bridgeman, B . & Wendler , C . (1991). Gender differences in predictors o f college mathematics performance and i n college mathematics course grades. Journal of Educational Psychology, 83,275-284. Caldera, Y . M . , Huston, A . C . & O ' B r i e n , M . (1989). Socia l interactions and play patterns o f parents and toddlers with feminine, masculine, and neutral toys. Child Development, 60, 70-76. Connel l , R . W . (1995). Masculinities. Cambridge, U K : Pol i ty Press. Connor , J . M . , Serbin, L . A . , & Shackman, M . (1977). Sex differences i n children's response to training on visual-spatial test. Developmental Psychology, 13, 293-294. Connor , J . M . , Schackman, M . , & Serbin, L .A . (1978) . Sex-related differences in response to practice on a visual-spatial test and generalization to a related test. Child Development, 71 49, 24-29. Eagly , A . L . , W o o d , W . , & Diekman, A . B . (2000). Socia l role theory o f sex differences and similarities: A current appraisal. In T. Ecc les & H . M . Trautner (Eds.), The Developmental Social Psychology of Gender (pp.123-174). M a h w a h , N J : Lawrence Er lbaum Associates. Ecc les , J . S. (1987). Gender roles ans women 's achievement-related decisions. Psychology of Women Quarterly, 11, 135-172. Eisenberg, N . , W o l c h i k , S .A . , Hernandez, R . & Pasternack, J . F . (1985). Parental socialization o f young children's play: A short-term longitudinal study. Child Development, 56, 1509-1513. Fagot, B . I. (1974). Sex differences in toddlers' behavior and parental reaction. Developmental Psychology, 10, 554-558. Fagot, B . I . & Hagan, R . (1991). Observations o f parent reactions to sex-stereotyped behaviors: A g e and sex effects. Child Development, 62,617-628. Fagot, B . I . & Leinbach, M . D . (1989). The young chi ld ' s gender schema: Environmental input, internal organizations. Child Development, 60), 663-672. Fagot, B . I.. & Leinbach, M . D . (1987) Social izat ion o f sex roles wi t l i in the family. In D . B . Carter (Ed.) , Current Conceptions of Sex Roles and Sex Typing (pp.89-100). N e w Y o r k , N Y : Praeger Publishers. Fiese, B . H . & Ski l lman, G . (2000). Gender differences in family stories: Modera t ing influence o f parent gender role and chi ld gender. Sex Roles, 43, 267-283. Fisher-Thompson, D . (1993). Adul t toy purchases for children: Factors affecting sex-typed toy selection. Journal of Applied Developmental Psychology, 14, 385-406. Fisher-Thompson, D . (1990). Adul t sex-typing o f children's toys. Sex Roles, 23, 291-303. 72 Gilmart in, B . G . (1987). Shyness and Love: Causes, Consequences and Treatment. N e w Y o r k , N Y : Universi ty Press o f Amer ica . Giul iano, T . A . , Popp, K . E . & Knight , J . L . (2000). Footballs versus Barbie dolls: Ch i ldhood play activities as predictors o f sport participation. Sex Roles, 42,159-181. Hoffman, R . M . & Borders , L . D . (2001). Twenty-five years after the Bern Sex Ro le Inventory: A reassessment and new issues regarding classification variability. Measurement & Evaluation in Counseling & Development, 34, 39-50. Holmbeck , G . N . (1997). Toward terminological, conceptual, and statistical clarity in the study o f moderators and mediators: Examples form the chi ld-cl in ical and pediatric psychology literatures. Journal of Consulting and Clinical Psychology, 65, 599-610. Idle, T. , W o o d , E . & Desmarais, S. (1993). Gender role socialization in toy play situations: Mothers and fathers wi th their sons and daughters. Sex Roles, 28,679-691. Jackson, L . A . , Gardner, P . D . & Sul l ivan, L . A . (1992). Expla in ing gender differences i n self- pay expectations: Social comparison standards o f fair pay. Journal of Applied Psychology, 77, 651 -663. Langlois , J . H . & Downs , A . C . (1980). Mothers , fathers, and peers as socialization agents o f sex-typed play behaviors in young children. Child Development, 51, 1217-1247. L e w k o , J . H . , H e i n , C , Garg , R . , & Tesson, G . (1993). Transitions o f adolescents in science career pathways. In: P . A n i s e f & P . A x e l r o d (Eds), Transitions: Schooling and Employment in Canada. Toronto: Thompson Educational Publishing. Lindsey, E . W . & M i z e , J . (2001). Contextual differences i n parent-child play: Implications for children's gender role development. Sex Roles, 46,155-175. L i s s , M . B . (1983). Learning gender-related skills through play. In M . B . L i s s (Ed.) , Social and Cognitive Skills: Sex Roles and Children's Play. San Fransisco, C A : Academic Press. 73 L e v y , G . D . (1999). Gender-typed and non-gender-typed category awareness in toddlers. Sex Roles, 41, 851-873. L y s o n , T . A . (1984). Sex differences in the choice o f a male or female career line: A n analysis o f background characteristics and work values. Work and Occupations, 11,131-146. M a c c o b y , E.E. (1992) . The role o f parents in the socialization o f children: A n historical overview. Developmental Psychology, 28,1006-1016. M a c c o b y , E . E . & Jackl in C . N . (1974). The Psychology of Sex Differences. Stanford, C A : Stanford Universi ty Press. Maughan, B . & Rutter, M . (1997). Retrospective reporting o f childhood adversity: Issues in assessing long-term recall . Journal of Personality Disorders, 11,19-33. McCutcheneon , L . E . (1998). L i fe Ro le Salience Scales: Addi t iona l evidence for construct validity. Psychological Reports, 83, 1307-1314. M c H u g h , M . C . & Hanson Frieze, I. (1997). The measurement o f gender role attitudes: A review and commentary. Psychology of Women Quarterly, 21, 1-16. O ' B r i e n , M . & Huston, A . C . (1985). Development o f sex-typed play behavior in toddlers. Developmental Psychology, 21, 866-871. Pomerleau, A . , Bo lduc , D . , Malcu i t , G . & Cossette, L . (1990). P ink or blue: Environmental gender stereotypes in the first years o f life. Sex Roles, 22, 356-367. Roopnarine, J . L . (1986). Mothers ' and fathers' behaviors toward the toy play o f their infant sons and daughters. Sex Roles, 14, 59-68. Roopnarine, J . L . & Mounts , N . S . (1987). Current theoretical issues in sex roles and sex typing. In D . B . Carter (Ed.) , Current Conceptions of Sex Roles and Sex Typing (pp.7-32). N e w Y o r k , N Y : Praeger Publishers. Ruble , D . N . & Mar t in , C L . (1998). Gender development. Handbook of Child Psychology: 74 Socialization, Personality, and Social Development ( V o l . 5, pp. 993-1016). N e w Y o r k : W i l e y . Schroeder, K . A . & Ma luso , D . (1993). Gender differences and similarities between male and female undergraduate students regarding expectations for career and family roles. College Student Journal, 27, 237-249. Schwartz, L . A . & M a r k h a m , W . T . (1985). Sex differences i n children's toy advertisements. Sex Roles, 12, 157-170. Signorella, M . L . , Jamison, W . , & Krupa , M.H. (1989) . Predicting spatial performance from gender stereotyping in activity preferences and in self concept. Developmental Psychology, 25, 89-95. Sparfkin, C , Serbin, L . A . , Denie rC. , & Connor , J . M . (1983). Sex-differentiated play: Cognit ive consequences and early childhood interventions. In. M . A . L i s s (Ed.) , Social and Cognitive Skills: Sex Roles and Children's Play. San Fransisco, C A : Academic Press. Spence, J .T. & Hahn, E . D . (1997). The Attitudes Toward W o m e n Scale and attitude change in college students. Psychology of Women Quarterly, 21, 17-32. Spence, J . , T. , Helmreich, R . & Stapp, J . (1973) A short version o f the Attitudes T o w a r d W o m e n Scale ( A W S ) . Bulletin of Psychonomic Society, 2, 219-220. Ti l leczek, K . C . & L e w k o , J . H . (2001). Factors influencing the pursuit o f health and science careers for Canadian adolescents in transition from school to work. Journal of Youth Studies, 4 ,415-428. Ware , N . C . & Lee , V . E . ( 1988). Sex differences in choice o f college science majors. American Educational Research Journal, 25, 593-614. West , C , & Zimmerman, D . H . (1987). D o i n g gender. Gender & Society, 1,125-151. 76 Parti : Background information 1. What is your gender? Male Female 2. What category best describes your background? Check ( X) ONE: North American South American Central American African Western European Eastern European Middle Eastern Asian South East Asian Australian Other: 3. How old are you? Years 4. What is your parents' Highest Level of Education? Please put a check one box for EACH parent: MOTHER'S highest level of education: Some Highschool Highschool Diploma Some Post Secondary University/College Degree Post-Graduate Degree(MD.,PhD.,Law...) FATHER'S highest level of education: Some Highschool Highschool Diploma Some Post Secondary University/College Degree Post-Graduate Degree(MD.,PhD..) 5. What type of occupations do your parents currently hold? If they are retired, what was the last position they held? MOTHER'S Occupation Business/Finance/Administration Trades/Transport/Equipment Operations Science Related (Chemist, Social...) Management Occupation Sales and Service Health Occupation Stay at Home Other FATHER'S Occupation Business/Finance/Administration Trades/Transport/Equipment Operations Science Related (Chemist, Social...) Management Occupation Sales and Service Health Occupation Stay at Home Other 77 What is the Title of your mother's occupation? (Ex: Nurse, doctor ) What is the Title of your father's occupation? 6. What academic year are you currently in? First Second Third Fourth Other 7.a) Have you declared your major (specialization)? Yes No 7.b)What major (specialization) have you declared? (Ex.: Psychology, Marketing, Nursing...). If you have not declared a major, please indicate what major you will be declaring: 8 . Which Degree Program are you pursuing? | | Bachelor of Science [^Bachelor of Applied Sciences [^Bachelorof Sci.(Natural Res.Cns.) Bachelor of Arts 1 Bachelor of Commerce | | Bachelor of Education (Elementary) | [ Bachelor of Education (Middle) EZ1 Bachelor of Education (Secondary) I I Bachelor of Environmental Design I | Bachelor of Fine Arts [ | Bachelor of Human Kinetics | | Bachelor of Laws • Other: Check one: 1 I Bachelor of Medical Lab Sciences | | Bachelor of Midwifery | | Bachelor of Music | | Bachelor of Science (Agroecology) I Bachelor of Science (Global Rpsourpps) O Bachelor of Science (FNH) [ [ Bachelor of Science (Forestry) | | Bachelor of Science (Occup. Thrpy.) [ | Bachelor of Science (Phys. Thrpy.) | [ Bachelor of Science (Nursing) | | Bachelor of Science Pharmacy | | Bachelor of Social Work 78 Part 2: Personality Measures 9. Directions: For the following words listed below, rate the extent to which you think the trait (word) describers you. Write the number in the box next to each word to indicate your response. Use the scale provided: E X A M P L E : Emotional 1 2 3 4 5 6 7 Aggressive 1 2 3 4 5 6 7 1 2 3 4 5 6 7 Almost Usually Sometimes,but Occasionally Often Usually Almost never not Infrequently true true true always true true true Defend My Own Beliefs 1 2 3 4 5 6 7 Affectionate 1 2 3 4 5 6 7 Conscientious 1 2 3 4 5 6 7 Independent 1 2 3 4 5 6 7 Sympathetic 1 2 3 4 5 6 7 Moody 1 2 3 4 5 6 7 Assertive 1 2 3 4 5 6 7 Sensitive to needs of others 1 2 3 4 5 6 7 Reliable 1 2 3 4 5 6 7 Strong Personality 1 2 3 4 5 6 7 Understanding 1 2 3 4 5 6 7 Jealous 1 2 3 4 5 6 7 Forceful 1 2 3 4 5 6 7 Compassionate 1 2 3 4 5 6 7 Truthful 1 2 3 4 5 6 7 Have leadership abilities 1 2 3 4 5 6 7 79 1 2 3 4 5 6 7 Almost Usually Sometimes,but Occasionally Often Usually Almost never not Infrequently true true true always true true true Eager to soothe hurt feelings 1 2 3 4 5 6 7 Secretive 1 2 3 4 5 6 7 Willing to take risks 1 2 3 4 5 6 7 Warm 1 2 3 4 5 6 7 Adaptable 1 2 3 4 5 6 7 Dominant 1 2 3 4 5 6 7 Tender 1 2 3 4 5 6 7 Conceited 1 2 3 4 5 6 7 Willing to take a stand 1 2 3 4 5 6 7 Love children 1 2 3 4 5 6 7 Tactful 1 2 3 4 5 6 7 Aggressive 1 2 3 4 5 6 7 Gentle 1 2 3 4 5 6 7 Conventional 1 2 3 4 5 6 7 Please go to the next page! 80 10. Indicate the extent to which you agree with the following statements. Please C I R C L E the number that corresponds to you response according to the following scale. Disagree Somewhat Neither Disagree Somewhat Agree Disagree Nor Agree Agree 1 2 3 4 5 I want to work, but do not want to have a demanding career. 1 2 3 4 5 I expect to make many sacrifices as necessary in order to advance my career. 1 2 3 4 5 I value being involved in a career and expect to devote the time and efforts needed to develop it. 1 2 3 4 5 I expect to devote a significant amount of my time to building my future career and developing the necessary skills to advance it. 1 2 3 4 5 I expect to devote whatever time and energy it takes to move up in my future career field. 1 2 3 4 5 It is important to me to have some time for myself and my own development rather than have children and be responsible for their care. 1 2 3 4 5 I expect to devote my time to the rearing of my children. 1 2 3 4 5 I expect to be very involved in the day-to-day matters of rearing children of my own. 1 2 3 4 5 Becoming involved in the day-to-day care details of rearing children involves costs in other areas of my life which I am unwilling to make. 1 2 3 4 5 I do not expect to be very involved in child rearing. 1 2 3 4 5 I expect to commit whatever time is necessary to making my marriage partner feel loved. 1 2 3 4 5 Devoting a significant amount of my time to being with or doing things with a marriage partner is not something I expect to do. 1 2 3 4 5 I expect to put a lot of time and effort into building and maintaining a marital relationship. 1 2 3 4 5 Really involving myself in a marriage relationship involves costs in other areas which I am not willing to accept. 1 2 3 4 5 81 Disagree Somewhat Neither Disagree Somewhat Agree Disagree Nor Agree Agree 1 2 3 4 5 I expect to work hard to build a good marriage relationship even if it means limiting my opportunities to pursue other personal goals. 1 2 3 4 5 I expect to leave most of the day-to-day details of running a home to someone else. 1 2 3 4 5 I expect to devote the necessary time and attention to having a neat and attractive home. 1 2 3 4 5 I expect to be very much involved in caring for a home and making it attractive. 1 2 3 4 5 I expect to assume the responsibility for seeing that my home is well kept and well run. 1 2 3 4 5 Devoting a significant amount of my time to managing and caring for a home is not something I care to do. 1 2 3 4 5 11. The statements listed below describe attitudes toward the roles of women in society which different people have. Please indicate the extent to which Y O U agree with each statement. Circle the number that corresponds with the following scale: Strongly Disagree Disagree Mildly Agree Mildly Strongly Agree 1 2 3 4 Swearing and obscenity are more repulsive in the speech of a women, rather than of a man. 1 2 3 4 Under modern economic conditions with women being active outside the home, men should share in the household tasks such as washing the dishes and doing laundry. 1 2 3 4 It is insulting to have the "obey" clause remain in the marriage service. 1 2 3 4 A woman should be free as a man to propose marriage. 1 2 3 4 Women should worry less about their rights and more about becoming good wives and mothers. 1 2 3 4 Strongly Disagree 1 Disagree Mildly 2 Agree Mildly 3 8 2 Strongly Agree 4 Women should assume their rightful place in business and all professions along with men. 1 2 3 4 Women should not expect to go to exactly the same places or to have quite the same freedom of action as a man. 1 2 3 4 It is ridiculous for a woman to drive a train and for man to knit a sweater. 1 2 3 4 The intellectual leadership of a community should be largely in the hands of men. 1 2 3 4 Women should be given equal opportunity with men for apprenticeship in the various trades. 1 2 3 4 Women earning as much as their dates should bear equally the expense when they go out together. 1 2 3 4 Sons in a family should be given more encouragement to go to college than daughters. 1 2 3 4 In general, the father should have greater authority than the mother in the bringing up of children. 1 2 3 4 Economic and social freedom is worth far more to a woman than acceptance of the ideal femininity which has been set up by men. 1 2 3 4 There are many jobs in which men should be given preference over women in being hired or promoted. 1 2 3 4 Please go to the next page! 83 12. How often did you play with the following toys when you were younger? Using the scale below, please circle the number that corresponds to your response in the box next to each toy. 1 2 3 4 5 Never Rarely Occasionally Often Very Often Toy Guns 1 2 3 4 5 Barbies 1 2 3 4 5 Dishes/Tea Sets 1 2 3 4 5 Army Toys & Soldiers 1 2 3 4 5 Weight Lifting Toys 1 2 3 4 5 Sewing Kits 1 2 3 4 5 Painting Sets 1 2 3 4 5 Baby Dolls 1 2 3 4 5 Doll House 1 2 3 4 5 Play Doh & Modelling Clay 1 2 3 4 5 Tools & Carpentry Sets 1 2 3 4 5 Balls(footballs, soccer balls, baseballs..) 1 2 3 4 5 Jewellery/ Jewellery Boxes 1 2 3 4 5 Cars/Trucks/Airplanes 1 2 3 4 5 84 13. When you were younger, how often did you play the following games? Using the scale below, please circle the number that corresponds to your response in the box next to each game. 1 2 3 4 5 Never Rarely Occasionally Often Very Often House 1 2 3 4 5 Team Sports (soccer, baseball...) 1 2 3 4 5 "Teacher" 1 2 3 4 5 Video Games 1 2 3 4 5 "War" 1 2 3 4 5 Dress-up 1 2 3 4 5 Make-Believe or Pretend 1 . 2 3 4 5 Jump Rope 1 2 3 4 5 Hopscotch 1 2 3 4 5 14. Think back to when you were a young child, when you were in the ages between preschool and first grade. How often did you PLAY WITH: Never 1 Rarely 2 Occasionally 3 Often 4 Very Often 5 B O Y S 1 2 3 4 5 GIRLS 1 2 3 4 5 M O T H E R 1 2 3 4 5 F A T H E R 1 2 3 4 5 85 Thank you for your participation! Katherine Rhodes will come to your next class to pick up this survey. If for some reason you didn't bring this survey to that time, please either place this survey in the self addressed and postage paid envelope provided and put it in the mail,or, email Katherine Rhodes, and she will arrange to pick up this completed survey. Appendix B 86 Table 12 Mean Response Scores Variable W O M E N M E N Feminine Toy Play* 3.28 1.31 Feminine Game Play* 3.66 2.01 Masculine Toy Play* 2.20 3.42 Masculine Game Play* 2.53 3.73 Neutral Toy Play* 3.91 3.33 Frequency of Play with Mother * 3.36 3.12 Frequency of Play with Father* 2.84 3.10 Occupational Role Commitment* 3.67 3.87 Parental Role Commitment * 3.76 3.65 Marital Role Commitment* 4.33 4.30 Home Care Role Commitment* 3.71 3.57 Gender Ideology (ATW)** 3.48 3.26 Gender Schema (BSRl)t 3.49 -1.92 Note: *five point scale; **four point scale; f standardized t score; n = 277 87 Table 13 Regressions Between Childhood Socialization Variables and Mediators Independent Gender Ideologies Gender Schemas Variable B B B Overall Feminine Toys and Games .11** .28 3.15** .25 K2 .14 .07 Masculine Toys and Games -.06| -.11 -3.83** -.24 R2 .08 .06 Neutral Toy Play .10** .22 2.57** .18 R2 .11 .03 Playing with Mother .06** .16 1.11 .09 R2 .09 .01 Playing with Father .01 .04 -.94 -.08 R2 .07 .01 Note. All regressions controlled for parent education and employment variables; tP^O.10; *ps0.050; **p<0.0\,n — 277. 88 Table 14 Women's and Men's Regressions Between Childhood Socialization Variables and Mediating Variables Independent Variable Gender Ideologies Gender Schemas B 6 B B WOMEN Feminine Toys and Games R2 Masculine Toys and Games R2 Neutral Toy Play R2 Playing with Mother R2 Playing with Father R2 MEN Feminine Toys and Games R2 Masculine Toys and Games R2 Neutral Toy Play R2 Playing with Mother R2 Playing with Father R2 .06 .09 .11* .11 .04 .08 .02 .08 ,01 .08 .15 .09 .00 .06 .12** .13 .10* .12 .10* .12 .17 .08 .06 -.05 .15 .00 .26 .24 .25 3.09* .04 -2.60 .03 1.57 .02 .65 .01 -1.20 .02 5.52f .05 -4.02t .05 2.40 .04 1.20 .02 .16 .01 .16 .12 .10 .06 -.10 .19 -.19 .17 .01 Note. All regressions controlled for parent education and employment variables; fp̂ O.10; *p<0.050; **p<0.01, Women: n = 175; Men: n = 102. Table 15 Correlations Between all Independent and Dependent Variables Variables 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 1. Feminine Toys and Games 2. Masculine Toys and Games -.50 3. Neutral Toy Play .43 -.06 4. Play with Mother .19 .01 .26 5. Play with Father -.03 .19 .18 .52 6. Percent Female Enrollment .41 -.28 .20 .04 -.02 7. Occupational Role Comm. -.10 .16 -.13 -.09 -.12 -.07 8. Parenting Role Comm. .07 -.03 .12 .08 .18 .12 -.09 9. Marital Role Comm. .07 -.07 .05 .13 .08 .08 -.01 .41 10. Home Care Role Comm. .11 -.10 .10 .16 .15 .06 -.05 .26 .27 11. Gender Role Ideologies .27 -.11 .24 .16 .06 .23 .02 -.03 .01 -.07 12. Gender Schemas .25 -.25 .17 .08 -.07 .17 -.11 .18 .17 .14 .18 oo CO

Cite

Citation Scheme:

    

Usage Statistics

Country Views Downloads
Canada 49 10
United States 47 57
United Kingdom 33 0
Pakistan 5 0
Hong Kong 5 0
Philippines 4 0
Japan 3 0
China 2 8
France 1 0
Jamaica 1 0
Greece 1 1
India 1 0
Croatia 1 0
City Views Downloads
Unknown 79 19
Washington 18 0
Canterbury 9 0
Central District 5 0
Toronto 4 0
Lowell 4 0
Whiteland 3 1
Tokyo 3 0
Snohomish 2 0
London 2 0
Beckenham 2 0
Bellevue 2 1
Harpenden 2 0

{[{ mDataHeader[type] }]} {[{ month[type] }]} {[{ tData[type] }]}

Share

Share to:

Comment

Related Items