In the article below, in the METHOD section what tools were used to collect the data?

That's a Boy's Toy: Gender-Typed Knowledge in Toddlers as a Function of Mother's Marital Status

ABSTRACT

A child who is highly gender schematic readily uses gender when processing new information. In the current study, we examined whether and how family structure predicts a child's level of gender-typed knowledge (as assessed by a gender-stereotype sorting task) once the category of gender is in place (as assessed by a gender-labeling task). It was predicted that children from more "traditional" family structures (married mothers) would have more gender-typed knowledge compared to children from less traditional families (unmarried mothers). Moreover, we explored if this relationship would be related to, at least in part, the greater frequency of androgynous behaviors (i.e., both masculine and feminine household activities) an unmarried mother performs. Twenty-eight children (age 2 to 3) were tested at local childcare centers. The mother of each child reported her marital status as well as how often she engaged in stereotypically masculine and feminine behaviors. As expected, mothers' marital status was associated with children's level of gender-typed knowledge, such that children with unmarried mothers had less gendertyped knowledge, in part due to the unmarried mother's greater frequency of androgynous behaviors. Implications for children's acquisition of gender-related stereotypes and the possible benefit of having mothers model both masculine and feminine behaviors are discussed.

METHODS

Participants. Forty-two children were recruited for the study; however, 14 children were excluded for the following reasons: did not correctly respond to at least 70% of the gender labeling items indicating they were either not able to label genders or were unable to perform the task (n = 7; 3 of which had at least an 87% bias for one response, and 1 who failed to respond to this section at all), mother did not return survey (n = 5), and technical difficulties (n = 2). This left us with a total of 28 children (n = 11 boys) who ranged in age from 2.16 to 3.99 years (M = 3.24 years, SD = 0.55 years). The children's mothers also participated. The children (and mothers) were recruited through 10 childcare centers in a medium-sized Midwestern town. The sample included children from both privately owned and public facilities (e.g., Head Start), large and small childcare centers, economically advantaged and disadvantaged areas of town, and centers that are both religious and secular. The average age of the mother was 31.00 years (SD = 6.28 years). Eighteen of the mothers were married whereas 10 were not (n = 1 divorced, 1 widowed, 8 never married); male and female children were approximately equal in each of these groups. Of the mothers, 64% worked full-time (67% of this group were married), 12% worked part-time (67% of this group were married), 16% were students (50% of this group were married), and 8% did not work outside of the home (50% of this group were married).

Materials. Items described in the materials for the Gender Labeling Test, Gender-Typed Knowledge Test, and the Androgynous Behavior Assessment were selected from an item pool based on pilot testing with undergraduates (N = 25) to determine Hupp et al. 393 the gender stereotype of the item (for similar pretesting procedures, see Cherney, 2005).

--Gender Labeling Test. This child measure was modeled after the gender labeling measure used by Leinbach and Fagot (1986) and was designed to examine if a child could appropriately label genders. Labeling signals the point at which children are able to make conscious use of the gender category and demonstrate organization of their understanding of gender (Fagot & Leinbach, 1995), and it is a basic component of gender knowledge. This task also gave children the chance to become familiar with the testing situation. This test consisted of 4 pictures of boys and 4 pictures of girls. Children were asked whether the person was a boy or a girl. Scores indicated the percentage correct for each child.

--Gender-Typed Knowledge Test. This child measure was modeled after Fagot et al. (1992) and consisted of 16 gender-stereotyped items (8 male, 8 female; e.g., football, butterfly). First, children were presented with two pictures, one that showed a man and boy together and one that showed a woman and girl together. These pictures remained the same throughout the test. Next, the children were shown a picture of a gender-stereotyped item. Finally, they were asked to identify who the item belonged to by pointing to either the pair of males or females. Given the content and the forced-choice nature of the design, this task assesses children's gender-typed knowledge about others (Signorella et al., 1993). Scores indicate the percentage of items that each child responded to in a stereotypical way; a high score represents a high level of gender-typed knowledge.

--Androgynous Behavior Assessment. In order to assess the household behaviors of the mothers, the mothers completed a 24-item survey that consisted of a list of behaviors that were stereotypically masculine (8 items; e.g., mowing the grass), feminine (8 items; e.g., doing the laundry), or neutral (8 items; e.g., driving children to school). The mothers were asked to indicate "how often you are the member of the household who is responsible for the following activities by circling the corresponding number" on a 5-item Likert-type scale ranging from 1 (never) to 5 (always). If no member of their household performed a particular activity, they were instructed to circle "Not Applicable." Separate frequency scores for masculine and feminine behaviors were calculated. Androgynous behaviors were indexed by the relatively high frequency of both masculine and feminine behaviors in relation to the other mothers within this sample.

--Family structure. We created a family structure measure based on demographic features. Mothers indicated marital status, employment status, and number and gender of other individuals in the home (e.g., mother's partner, sisters, brothers, uncles, aunts). 394 The Journal of Genetic Psychology

Procedure. After mothers gave consent, a female experimenter tested the children individually at their childcare centers. The children's measures were computerized, and the children were not given feedback for either measure. Each mother was mailed the androgyny and family structure survey and was asked to mail the survey back. During the child testing session, the children were first administered the Gender Labeling Test. Pictures were randomly presented to the participants on a computer screen, and the experimenter pressed a button according to the child's response. Next, the children were administered the Gender-typed Knowledge Test. Children were told that "Molly and Timmy need help picking up their room." They were shown a picture of a messy room. The children were told that they would see a series of toys, and they were to point to whether that toy belonged to Molly or Timmy. They were shown pictures of a woman-girl pair and a man-boy pair for each object. Children were then randomly presented with a picture of a gender-stereotyped item and asked to make their selection, which the experimenter recorded. Gender consistent sorting was indicative of a high level of gender-typed knowledge. At the end of the study, children were shown a picture of a clean room and congratulated.