Counterstereotyping can change children’s thinking about boys’ and girls’ toy preferences
Introduction
Children’s environments are replete with messages that girls and boys like fundamentally different kinds of things. Storybooks (Berry and Wilkins, 2017, Gooden and Gooden, 2001), toy marketing (Blakemore and Centers, 2005, Murnen et al., 2016, Reich et al., 2018), television (Davis, 2003, Kahlenberg and Hein, 2010, Leaper et al., 2002), and movies (Arnold et al., 2015, Streiff and Dundes, 2017) portray stereotypical representations of boys’ and girls’ interests. For example, the girls’ section of the Disney Store’s website features pink and purple colors, jewelry, and cosmetics, whereas the boys’ section features red, black, and brown colors, building toys, weapons, and vehicles (Auster & Mansbach, 2012).
Children are also exposed to stereotypes in their own homes. Beginning in infancy, parents often fill boys’ and girls’ rooms with gender-stereotypical items: Boys’ rooms typically display sports equipment, vehicles, and the color blue, whereas girls’ rooms typically display jewelry, dolls, and the color pink (MacPhee and Prendergast, 2019, Pomerleau et al., 1990, Witt, 1997). Parents also tend to buy their children gender-stereotypical toys (Weisgram & Bruun, 2018) and reject their children’s requests for counterstereotypical gifts (Etaugh and Liss, 1992, Robinson and Morris, 1986). Furthermore, parents—even those with egalitarian motives—often steer their boys toward stereotypically masculine activities and away from stereotypically feminine ones (Freeman, 2007, Halpern and Perry-Jenkins, 2016, Kane, 2006).
Beginning early in life, children are aware of, and behave in accordance with, stereotypical messages in their environments (for complete reviews, see Martin and Ruble, 2004, Ruble et al., 2007). For example, young children associate stereotypically feminine items such as dolls and tea sets with girls but not with boys, and they associate stereotypically masculine items such as trucks and tools with boys but not with girls (e.g., Cherney and Dempsey, 2010, Freeman, 2007, Todd et al., 2018). In addition to being knowledgeable about gender stereotypes regarding toy preferences, young children act in alignment with widely held stereotypes. For example, when asked to choose toys for others, 3- to 5-year-olds avoid choosing trucks for girls (favoring dolls instead) and avoid choosing dolls for boys (favoring trucks instead) (Cowan and Hoffman, 1986, Eisenberg et al., 1982). Children of this age also make similar decisions when choosing toys for themselves (Fabes et al., 2003, Halim et al., 2013, Zosuls et al., 2009). Furthermore, young children will exclude children whose gender does not match the gender commonly associated with the play activity at hand (Theimer, Killen, & Stangor, 2001).
Despite ample research documenting children’s gender stereotyping, less is known about effective strategies for addressing early-emerging stereotypical beliefs. Yet, reducing children’s stereotyping is important because such biases can constrain children’s interest in, and access to, advantageous experiences. For example, block play, which is stereotypically associated with boys (Miller, 1987, Sherman, 1967), provides opportunities to hone spatial skills (Levine, Ratliff, Huttenlocher, & Cannon, 2012). Similarly, doll play, which is stereotypically associated with girls (Miller, Lurye, Zosuls, & Ruble, 2009), allows children to practice their socioemotional skills (Li & Wong, 2016). Because peers exclude children from activities based on gender stereotypes (Theimer et al., 2001), and because children become less interested in toys they think are “for” another gender (Martin, Eisenbud, & Rose, 1995), stereotyping may cause children to miss opportunities to develop important skills. Given its problematic nature, it is critical to develop methods to reduce stereotyping early in development.
One frequently tested approach for reducing gender stereotyping is counterstereotyping (Lenton et al., 2009, Pruden and Abad, 2013). Counterstereotyping interventions highlight specific examples of stereotype-inconsistent information. For example, a researcher hoping to alter the belief that only women are nurses might expose participants to vignettes about male nurses. Although counterstereotyping is commonly used to combat adults’ stereotypes (Lenton et al., 2009), there have been just a few informative interventions targeting children’s gender stereotypes (e.g., for occupations, see Coyle and Liben, 2016, Scott and Feldman-Summers, 1979, Sherman and Zurbriggen, 2014, Steinke et al., 2007, Weeks and Porter, 1983; for musical instruments, see Pickering & Repacholi, 2001). Further, the results of such studies have been mixed (for a review, see Durkin, 1985): Some return positive effects (e.g., Pickering and Repacholi, 2001, Scott and Feldman-Summers, 1979, Sherman and Zurbriggen, 2014), and others do not (e.g., Coyle and Liben, 2016, Steinke et al., 2007, Weeks and Porter, 1983).
The failure of some counterstereotyping interventions aligns with the idea that it is generally easier to build up biases than to reduce them (Sherif, 1954). The challenge of reducing stereotyping is likely amplified when stereotypes accurately reflect the statistics in a child’s environment (Jussim et al., 2009, Swim, 1994). In the case of toy stereotypes, for example, boys are less likely than girls to play with dolls and girls are less likely than boys to play with vehicles (Servin, Bohlin, & Berlin, 1999).
Even among apparently successful interventions, mechanisms and sources of change can be difficult to discern. For instance, in a recent study focused on changing young children’s toy stereotypes, children assigned to a counterstereotyping condition showed less stereotyping compared with those assigned to a stereotype-consistent condition (Spinner, Cameron, & Calogero, 2018). However, the study did not contain a stereotype-neutral condition or pretest; thus, it is difficult to determine whether the counterstereotyping intervention changed children’s baseline thinking (for similar studies, see Ashton, 1983, Green et al., 2004, Pike and Jennings, 2005).
In the current research, we sought to improve on prior interventions focused on children’s gender stereotyping by (a) choosing intervention strategies rooted in theories of stereotype acquisition and (b) carefully deconstructing our intervention approach over multiple conditions and studies in order to shed light on mechanisms underlying the observed intervention effects. Across four studies, we investigated whether counterstereotypical information changed children’s gender stereotypes about toy preferences. We focused in particular on two highly stereotyped toys: dolls (associated with girls and not with boys) and trucks (associated with boys and not with girls) (Blakemore & Centers, 2005).
The current experiments targeted 5- and 6-year-old children (but also included 4- and 7-year-old children whose parents returned signed consent forms). This age range provides a particularly stringent test of the intervention’s effectiveness because this is a time during development when children’s gender stereotypes about toys are particularly rigid (Martin and Ruble, 2004, Martin and Ruble, 2010, Weisgram et al., 2014). We did not have specific hypotheses based on participant age or gender; however, analyses focused on participant age and gender are available on OSF (https://osf.io/jdmqz/?view_only=a582d398d2264789aba318c9b1aee788).
We reasoned that to be successful, an intervention must be convincing enough to confront both existing messages in children’s environments and children’s own observations of what other children like. To this end, Experiment 1 participants received counterstereotypical messages with phrasing that has been theorized to encourage generalization of attributes within a category (i.e., language that bolsters stereotype acquisition); each counterstereotypical message contained gender category labels (“girls” and “boys”; see Arthur et al., 2008, Baron et al., 2014, Bigler and Liben, 2007, Diesendruck and HaLevi, 2006, Patterson and Bigler, 2006, Rhodes and Gelman, 2008, Roberts et al., 2017, Waxman, 2010) and employed generic phrasing (e.g., “girls like trucks”; see Bian and Cimpian, 2017, Cimpian and Markman, 2008, Cimpian and Markman, 2011, Gelman, 2003, Rhodes et al., 2012). An adult actor (described as a “teacher”) delivered the intervention messages because children in our participants’ age range trust information provided by adults and teachers (Corriveau and Harris, 2009, Jaswal et al., 2010) and children commonly hear generic statements about gender categories from adults (Bigler and Liben, 2007, Endendijk et al., 2014, Gelman et al., 2004).
Section snippets
Experiment 1
Experiment 1 compared children’s gender counterstereotypical beliefs about dolls and trucks after an intervention or control manipulation. Statements in both conditions contained gender group labels and generic phrasing. Participants in the intervention condition heard counterstereotypical statements about boys’ and girls’ toy preferences (e.g., “girls like trucks”), and participants in the control condition heard stereotype-irrelevant content (e.g., “girls like apples”). The control condition
Experiment 2
Experiment 2 implemented a 2 (generic phrasing or demonstrative plural phrasing) × 2 (gender label or no gender label) between-participants design. As in Experiment 1, the experimenter was unaware of condition assignment and a teacher provided counterstereotypical information. We hypothesized that interventions featuring either generic language or gender labels would be more effective at increasing participants’ counterstereotyping from pretest to posttest than an intervention containing
Experiment 3
Experiment 3 employed an intervention that communicated counterstereotypical information via demonstrative singular phrasing without gender group labels (e.g., “this kid likes trucks”). The experimenter was unaware of condition assignment, the content and number of conditions (one condition), and hypotheses.
Participants
We tested 36 children (18 girls; Mage = 5.45 years, range = 4.05–6.85). Most children were White (72.22%) and had at least one parent with a college degree (58.33%).
Materials, procedure, and design
Participants saw a photograph of an unfamiliar target child in front of three toys: a baby doll in yellow clothing, a yellow and red dump truck, and a yellow maraca. Half of the male (and half of the female) participants selected a toy for a female target, and half of the male (and half of the female) participants selected a toy for
Experiment 4b
Participants in Experiment 4b were randomly assigned to one of three intervention conditions. One condition used the generic–gender label teaching videos from Experiments 1 and 2. The other conditions featured either four (four exemplars condition) or one (single exemplar condition) individual counterstereotypical exemplars per teaching video. The experimenter was unaware of condition assignment.
We predicted that participants in all conditions would counterstereotype more at posttest than at
Summary of effects
The current research provides consistent support for the efficacy of counterstereotyping as a means to change children’s thinking about boys’ and girls’ toy preferences. From pretest to posttest in every intervention condition (see Table 3 for a summary), children became more likely to indicate that boys liked dolls and that girls liked trucks. The only condition where this change was not observed was the control condition of Experiment 1 in which participants learned stereotype-irrelevant
Acknowledgments
This research was supported by a Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD) grant (R01 HD070890) to K.S. and National Science Foundation (NSF) Graduate Research Fellowships to R.A.K. and M.P.R. This study was also supported in part by a core grant to the Waisman Center from the NICHD (U54 HD090256). We thank Bailey Immel for assistance with data collection. Thanks also to Patricia G. Devine, Katherine D. Kinzler, and Tory Ash for helpful comments on
References (91)
Self-perception theory
- et al.
Preschool children’s use of cues to generic meaning
Cognition
(2008) Psychological essentialism in children
Trends in Cognitive Sciences
(2004)- et al.
Cognitive models of stereotype change: III. Subtyping and the perceived typicality of disconfirming group members
Journal of Experimental Social Psychology
(1992) Acquiring generic knowledge
Trends in Cognitive Sciences
(2000)- et al.
Group presence, category labels, and generic statements influence children to treat descriptive group regularities as prescriptive
Journal of Experimental Child Psychology
(2017) - et al.
Early preschool environments and gender: Effects of gender pedagogy in Sweden
Journal of Experimental Child Psychology
(2017) - et al.
Pint-sized public relations: The development of reputation management
Trends in Cognitive Sciences
(2018) ANCOVA versus change from baseline: More power in randomized studies, more bias in nonrandomized studies
Journal of Clinical Epidemiology
(2006)- et al.
A second look at the impact of nontraditional vocational role models and curriculum on the vocational role preferences of kindergarten children
Journal of Vocational Behavior
(1983)