Early preschool environments and gender: Effects of gender pedagogy in Sweden

Highlights

• We compared children in a gender-neutral preschool to those attending other preschools in Sweden.

• Children at the gender-neutral school scored lower on a gender stereotyping measure.

• Children at the gender-neutral school were more willing to play with unfamiliar other-gender children.

• Children at the gender-neutral school were not less likely to notice another person’s gender.

• Differences in pedagogy are associated with how children think and feel about people based on their gender.

Abstract

To test how early social environments affect children’s consideration of gender, 3- to 6-year-old children (N = 80) enrolled in gender-neutral or typical preschool programs in the central district of a large Swedish city completed measures designed to assess their gender-based social preferences, stereotypes, and automatic encoding. Compared with children in typical preschools, a greater proportion of children in the gender-neutral school were interested in playing with unfamiliar other-gender children. In addition, children attending the gender-neutral preschool scored lower on a gender stereotyping measure than children attending typical preschools. Children at the gender-neutral school, however, were not less likely to automatically encode others’ gender. The findings suggest that gender-neutral pedagogy has moderate effects on how children think and feel about people of different genders but might not affect children’s tendency to spontaneously notice gender.

Introduction

Children are attuned to gender from an early age: infants in the United States can discriminate between male and female faces in looking time studies (Quinn et al., 2011) and look longer at faces that match the gender of their primary caregiver (Quinn, Yahr, Kuhn, Slater, & Pascalis, 2002). In addition, most U.S. children can label their own and others’ gender by the time they are 2 years old (Weinraub et al., 1984, Zosuls et al., 2009). Young children not only perceive gender but also use gender to guide their social preferences and inferences about others (Dunham et al., 2016, Martin and Halverson, 1981). Preschool-age children in the United States show gender in-group favoritism (Maccoby and Jacklin, 1987, Martin et al., 1999, Renno and Shutts, 2015, Shutts et al., 2013) and hold gender-based beliefs about others’ attributes (e.g., that boys like trucks and girls like dolls; Bauer and Coyne, 1997, Kuhn et al., 1978, Martin, 1989). Although much of the research on children’s consideration of gender has been conducted in the United States, studies reveal that young children in other cultures also categorize people by gender, prefer members of their own gender, and hold gender stereotypes (e.g., Brazil: de Guzman, Carlo, Ontai, Koller, & Knight, 2004; China: Knobloch, Callison, Chen, Fritzsche, & Zillmann, 2005; South Africa: Albert & Porter, 1986, and Muthukrishna & Sokoya, 2008; South Wales: Yee & Brown, 1994).

Although the prominence of gender in young children’s minds is well established (see Ruble, Martin, & Berenbaum, 2006, for a thorough review), the reason for the category’s preeminence is less clear (Martin, Ruble, & Szkrybalo, 2002). Some researchers have posited that gender is prominent because humans have an evolved specialized system dedicated to classifying and reasoning about others based on their gender (e.g., Cosmides, Tooby, & Kurzban, 2003; see Shutts, 2013, for a discussion). An alternative proposition in the field is that children focus on gender because their social environments highlight the importance of the category. For example, both gender schema theory (Bem, 1981, Bem, 1983) and developmental intergroup theory (Arthur et al., 2008, Bigler and Liben, 2007) emphasize that children receive significant input—from parents, teachers, and media—about gender categories and roles. Indeed, studies show that, at least in the United States, adults regularly use nouns and pronouns to mark gender when talking to children (Gelman, Taylor, & Nguyen, 2004; see Waxman, 2010, for a discussion). Moreover, teachers sometimes use gender to organize their classrooms (e.g., asking children to alternate by gender in seating; Bigler & Liben, 2007). Such practices may contribute to, or fully account for, children’s early and robust reliance on gender as a social category.

It is difficult to determine the role of social experience in guiding children’s reliance on gender categories because many young children spend significant periods of time in social environments where gender is emphasized—including preschools. Nevertheless, there are a small number of preschools that are committed to the practice of “gender-neutral” classroom environments in which teachers typically refrain from using gendered language and actively work to counteract gender stereotypes. Studying children who experience gender-neutral pedagogy (vs. more typical instruction) provides an unusual opportunity to examine the role of teachers’ behaviors in guiding children’s consideration of gender distinctions. The current research examined whether and how such schooling is associated with children’s reliance on gender information across a range of measures.

In addition to shedding light on the theoretical proposition that adults’ behavior plays a key role in children’s consideration of gender, research on gender-neutral pedagogy makes a practical contribution. Many have noted the negative effects of social exclusion, gender stereotyping, and unequal treatment on young children’s development (e.g., Andrews et al., 2016, Bian et al., 2017, Halpern et al., 2011), and gender-neutral pedagogy seeks to address such social problems through deemphasizing gender distinctions. Furthermore, gender-neutral pedagogy is of great international interest. For example, the first 20 results from a recent (February 2017) Google internet search for “Swedish gender neutral preschools” included relevant articles from The New York Times, The Washington Post, Newsweek, The Guardian, The Daily Mail, The Independent, and the British Broadcasting Corporation (BBC) (all published since 2010). Yet we are not aware of any research comparing children who attend gender-neutral preschools with those who attend more typical schools. Such research is needed because it bears on whether gender-neutral practices (which are effortful and require extensive teacher training) can affect children’s perceptions, feelings, and thoughts about gender.

Two studies have examined how highlighting gender categories in classrooms affects children’s gender attitudes and stereotyping. In one experiment (Bigler, 1995), 6- to 11-year-olds in the United States were randomly assigned to classrooms that differed in the emphasis teachers placed on gender over the course of 4 weeks. Teachers in the experimental classrooms frequently used gender noun labels to refer to students (e.g., “All the girls should be sitting down,” “Jack, be a good helper for the boys”) and designed classroom materials and activities that highlighted gender (e.g., a bulletin board with girls’ art on one side and boys’ art on the other, desks placed such that all the boys sat on one side of the room and all the girls sat on the other). Teachers in control classrooms treated their class as a unit and referred to students by their individual names only. Following the intervention, children in the experimental classrooms showed higher levels of gender stereotyping but showed similar levels of gender in-group favoritism to children in control classrooms. As Bigler (1995) noted, however, the gender in-group favoritism measures showed little variance, and most children were at ceiling. Thus, these measures might not have been sensitive enough to capture the effects of the intervention on children’s social preferences.

A second, more recent experiment in the United States (Hilliard & Liben, 2010) also manipulated the salience of gender categories for children—this time in a preschool setting. Over the course of 2 weeks, 3- to 5-year-old participants in a “high-salience” condition were exposed to classroom conditions similar to those implemented in Bigler’s (1995) experimental classrooms. Participants in “low-salience” classrooms experienced their usual classroom conditions, which were guided by school-level policies discouraging the use of gendered language or gendered classroom organization. From pretest to posttest, children in the high-salience condition showed increased gender stereotyping, less positive ratings of other-gender peers, and reduced play with other-gender peers, whereas children in the low-salience condition showed none of these effects. These findings, together with the gender stereotyping effects reported by Bigler (1995), are consistent with developmental intergroup theory (Bigler & Liben, 2007), which posits that when adults highlight a social distinction (e.g., through labeling and function use), it causes children to treat the highlighted distinction as a psychologically meaningful one.

Teachers in the experimental classrooms of Bigler, 1995, Hilliard and Liben, 2010 research were asked to go to great lengths to make gender salient. Furthermore, children in the experimental conditions experienced a dramatic and sudden change of classroom practices. Although the condition differences in both studies show that extreme gender-highlighting practices intensify children’s gender attitudes and stereotypes, the research cannot tell us whether subtler long-term differences in pedagogy—of the sort that distinguish practices in typical classrooms from practices in gender-neutral programs—affect how young children view gender. The answer to this question not only addresses the potency of gender socialization effects but also contributes to our understanding of whether promoting gender-neutral preschool environments is a meaningful practice during early childhood.

The current research focused on 3- to 6-year-old children in Sweden, a society with relatively egalitarian gender attitudes. Across the past 5 years (2011–2016), the World Economic Forum, 2015, World Economic Forum, 2016 gender gap index, for example, has consistently rated Sweden as the fourth most gender-equal society in the world based on the areas of economics, politics, education, and health. For comparison, the United States has ranged from 20th to 45th, with a median rank of 23rd. Participants came from two kinds of preschool settings, with some participants attending a preschool with several specific school policies and practices aimed at actively creating a gender-neutral environment and other participants attending more typical Swedish preschools. Throughout this article, we refer to the former schooling environment as “gender neutral” (GN) and the latter schooling environment as “typical” for ease of exposition. It is important to note, however, that all Swedish schools are required by law to aim for gender equality in classrooms; thus, the difference in gender neutrality between the GN and typical settings in the current research was a matter of degree rather than absolute. For reviews of the almost entirely qualitative and sociological literature discussing issues of gender in Swedish schools, see Edström, 2009, Heikkilä, 2017, Wahlström, 2003, Wernersson, 2009.

The government-mandated national curriculum for preschools in Sweden asserts that classroom practices affect preschoolers’ understanding of gender and prescribes that “girls and boys in preschools should have the same opportunities to test and develop abilities and interests outside the limitations of stereotyped gender roles” and that “preschools should counteract traditional gender roles and gender patterns” (Skolverket, 2011, p. 5). In practice, however, there is a great deal of variation in the implementation of government guidelines across preschools (Eidevald, 2009); even preschool teachers who intend to treat boys and girls in the same way are not always successful because treating boys and girls differently can be an unconscious habit (Odenbring, 2010, Wahlström, 2003). This is one reason why a small number of Swedish preschools have taken a further step of undergoing certification offered by the Swedish Federation for Lesbian, Gay, Bisexual, and Transgender Rights (RFSL). To achieve RFSL certification, all personnel undergo comprehensive training (certification takes 6–8 months) in treating all individuals according to their individual requirements regardless of gender or sexual orientation.

In the current study, the GN preschool, but not the typical preschools, pursued and attained RFSL certification, which therefore can be taken as an independent confirmation of differences in gender pedagogy between the schools. The specific policies of the GN school included avoiding gendered language as much as possible (e.g., by using the recently adopted Swedish gender-neutral pronoun and not saying “boy” or “girl”), modifying stories and songs to counteract rather than reinforce traditional gender roles and family structures, and avoiding some behaviors traditionally directed at one gender (e.g., commenting on the attractiveness of girls’ clothes). The Appendix presents information about the self-reported gender practices (and beliefs) of teachers at the GN and typical preschools in the current research.

We tested for potential effects of gender-neutral pedagogy by presenting children at the GN and typical schools with a series of measures. Participants first completed a task designed to assess their automatic encoding of other people’s gender. Participants then completed tasks meant to probe their gender attitudes and gender stereotyping. All tasks were chosen because of past research demonstrating their utility in measuring children’s emerging consideration of gender during the preschool years (e.g., LaFreniere et al., 1984, Leinbach et al., 1997, Shutts et al., 2013, Weisman et al., 2015), thereby allowing us to explore possible decreased consideration of gender among children in the GN school. At the end of the session, we also tested children’s ability to identify the gender of target stimuli from the tasks. The purpose of the identification task was to validate the stimuli and assess whether there were any differences between GN and typical school children’s ability to identify other people’s gender (which could be a confound in the study). Because our primary comparison of interest was across groups (i.e., typical vs. GN schooling), we chose to use a fixed task order so as not to increase within-group variance, which would reduce our power to detect between-group effects. The order of tasks—except for the position of the identification task—was arbitrary. The identification task was presented at the end of the session so that our use of gender labels would not cause any children to consider gender on the other tasks when they otherwise might not have done so.

Consistent with the tenets of developmental intergroup theory (Bigler & Liben, 2007), we hypothesized that children attending the GN school—where teachers deemphasize the importance of gender—would view gender as a less psychologically meaningful distinction than children in typical schools. Accordingly, we predicted that children at the GN school would be less likely to automatically take note of an individual’s gender (encoding task), less likely to use gender when deciding whether to play with another individual (social partner preference tasks), and less likely to use gender when reasoning about another person’s attributes (stereotyping task). Put another way, because adults in their environment do not emphasize that gender is a relevant dimension to consider, we predicted that children at the GN school would be less likely to consider other people’s gender. Yet another reason to posit that children at the GN school would be less likely to use gender to reason about another person’s attributes (stereotyping task) is that teachers at the GN school report working to actively counter gender stereotypes (see the Appendix); thus, children at the GN school likely have less access to information about gender stereotypes (which should result in lower scores on the stereotyping task).