Toy Preference Test

1. Abstract

The Toy Preference Test, developed by Brown University researcher Lenore A. DeLucia in 1963, is an early psychometric assessment designed to evaluate sex-role identification and gender-typed play interests in pediatric populations ranging from kindergarten through the fourth grade (approximate chronological ages 5 to 10 years). Conceived as an objective, nonverbal alternative to projective drawing tests and culturally ambiguous pictorial techniques such as the It Scale for Children (ITSC), the instrument utilizes photographic depictions of familiar play objects carefully pre-scaled along an empirical masculinity-femininity continuum. The test protocol is divided into two equivalent parallel forms, designated Set A and Set B, each consisting of 24 paired photographic stimuli (48 items per full administration across both sets) representing contrasting masculine and feminine activity markers.

Psychometrically, stimulus pairs were originally standardized using adult normative ratings across a 9-point Likert-type scale (ranging from 1 = strongly masculine to 9 = strongly feminine), ensuring high stimulus discriminability. Administration employs a forced-choice matching paradigm where children indicate their personal preference within each paired presentation. Scoring assigns 1 point to sex-congruent (sex-appropriate) selections and 0 points to sex-incongruent (opposite-sex) selections, yielding an overall composite score ranging from 0 to 24 points per set (or 48 points across combined sets), where higher scores indicate stronger stereotypic sex-role preference and conformity. Empirical investigations report test-retest reliability coefficients ranging from Spearman’s rho = .67 to .72 when administered under conditions accounting for examiner sex-bias effects. While the instrument lacks modern multidimensional confirmatory factor analytic validation and reflects mid-twentieth-century essentialist conceptualizations of biological sex and gender binaries, it remains a foundational historical milestone in the psychometrics of child development, early gender social learning, and pediatric behavioral measurement.

2. Keywords

Toy Preference Test, Lenore A. DeLucia, sex-role identification, gender typing, childhood development, toy preference methodology, play behavior assessment, psychometrics, test-retest reliability, forced-choice format, sex-typed socialization, early childhood measurement.

3. Authors

The Toy Preference Test was conceptualized, operationalized, and psychometrically standardized by Lenore A. DeLucia, Ph.D.

• Institutional Affiliation: Department of Psychology, Brown University, Providence, Rhode Island, United States.
• Historical Context: Dr. DeLucia conducted this research during the early 1960s, a transformative period in American developmental psychology characterized by increasing empirical scrutiny of psychoanalytic tenets regarding identification and the ascendancy of behavioral and social learning paradigms.
• Contact and Archival Status: Given that the original standardizations and research reports were published in 1963, primary archival materials, photographic plates, and technical manuals are retained within university research libraries and historical developmental archives. Correspondence regarding historical administration protocols is managed via university repository holdings and psychological measurement archives.

4. Purpose

The primary purpose of the Toy Preference Test is the quantitative assessment of sex-role preferences and sex-role identification in young children between the ages of 5 and 10 years. During the mid-twentieth century, clinical and developmental researchers faced severe methodological bottlenecks when attempting to measure the internalization of culturally defined gender roles. Prior clinical assessments relied disproportionately upon projective techniques—such as the Draw-A-Person test, the Machover figure-drawing diagnostic, and the Thematic Apperception Test—which suffered from notoriously low inter-rater reliability, unstandardized stimulus dimensions, and subjective interpretive rubrics. Furthermore, the dominant forced-choice measure of that era, the It Scale for Children developed by Daniel G. Brown (1956), employed an ambiguous child-figure drawing named “It” that was systematically confounded by child respondents perceiving “It” predominantly as male, thereby biasing female respondents toward masculine-coded response patterns.

DeLucia designed the Toy Preference Test to resolve these psychometric limitations by eliminating projective ambiguity and personal figure identification artifacts. The instrument isolates manifest preference for concrete cultural artifacts (toys) that serve as ecological markers of socialization. The test operationalizes two specific clinical and research goals:

• Direct Measurement of Manifest Preference: Providing an objective, ecologically valid index of whether a child selectively approaches or avoids toys culturally designated for their assigned sex, free from the confounding influence of an ambiguous intermediary cartoon figure.
• Assessment of Differential Developmental Trajectories: Permitting longitudinal and cross-sectional comparisons across educational levels (kindergarten, first grade, second grade, third grade, and fourth grade) to map how institutionalized schooling, peer reinforcement, and cognitive developmental maturation influence sex-typed behavioral choices over time.

In applied and clinical settings, the test was originally utilized to investigate deviations from normative gender-typed development, to evaluate the influence of parental absence (particularly father absence during early childhood) on male sex-typing, and to measure the social-cognitive impact of diverse familial structures. In basic developmental research, it offered an experimental dependent variable for laboratory investigations exploring reinforcement paradigms, modeling, and observational learning mechanisms.

5. Psychological Construct

The theoretical construct measured by the Toy Preference Test is sex-role preference, viewed historically as an operational surrogate for underlying sex-role identification. Psychologists of the mid-twentieth century, such as Jerome Kagan (1964) and Walter Mischel (1966), differentiated between three interrelated but distinct components of gender-related development:

1. Sex-Role Identity: The individual’s internal, covert knowledge and self-categorization of their biological sex and psychological alignment with cultural prototypes.
2. Sex-Role Adoption: The overt, publicly observable pattern of behavior, demeanor, and interpersonal conduct that signifies masculinity or femininity to external observers.
3. Sex-Role Preference: The preferential selection of, attraction to, or desire for specific activities, occupational goals, peer associations, and play objects culturally associated with a particular sex role.

The Toy Preference Test specifically targets sex-role preference. The construct assumes that cultural artifacts such as toys are not merely recreational instruments; rather, they serve as powerful socialization vehicles embedded with prescriptive norms, occupational affordances, and cognitive schemas. A child who consistently selects a masculine-typed toy (e.g., a miniature tool kit, construction equipment, or motorized vehicles) is choosing an object associated with mechanical manipulation, spatial orientation, independent mastery, and direct agency. Conversely, selection of a feminine-typed toy (e.g., domestic miniature appliances, baby dolls, tea sets, or beauty accessories) reflects an orientation toward nurturance, domestic roles, dyadic caretaking, and interpersonal aesthetics.

The construct is dimensionalized in the instrument along a bipolar continuum ranging from high sex-appropriate preference to low sex-appropriate (or cross-sex) preference. In DeLucia’s model:

• Conformity / Masculine Identification (for Boys): Operationalized as the systematic selection of masculine-valenced objects when paired directly against equivalent feminine-valenced objects. High scores signify internalization of male sex-role standards.
• Conformity / Feminine Identification (for Girls): Operationalized as the systematic selection of feminine-valenced objects when paired against masculine alternatives. High scores reflect internalization of traditional female sex-role expectations.
• Cross-Sex Role Preference: Operationalized as a marked tendency for a child to select objects culturally standardized for the opposite sex. Within the 1963 empirical paradigm, cross-sex selections in boys were viewed as psychometrically significant indicators of non-conforming sex-typing, whereas cross-sex selections in young girls (often exhibiting broader behavioral latitude, colloquially termed “tomboyism”) were recognized as developmentally normative due to asymmetrical cultural reinforcement schedules.

Importantly, the construct relies entirely on cultural consensus definitions of gender-appropriate behavior established through adult normative scaling, treating sex-role alignment as a direct reflection of differential reinforcement and cultural assimilation.

6. Theoretical Framework

The formulation of the Toy Preference Test was situated at the crossroads of three major theoretical movements that defined twentieth-century developmental psychology: psychoanalytic identification theory, social learning theory, and cognitive-developmental theory.

Psychoanalytic Formulations of Identification

Historically, the study of sex-role acquisition originated within Freudian psychoanalysis, which posited that identification occurs during the phallic stage of psychosexual development (ages 3 to 6). According to classic Freudian drive theory, the resolution of the Oedipus complex in boys and the Electra complex in girls motivates the child to repress eroticized parental strivings and identify defensively with the same-sex parent (“identification with the aggressor” or anaclitic identification). Through this internal psychological introjection, the child adopts the behavioral dispositions, taboos, and material preferences of the same-sex adult. DeLucia sought an empirical, objective metric that could operationalize the end-state of this putative identification process without requiring the speculative clinical inferences inherent in psychoanalytic interviews.

Social Learning Theory and Reinforcement Architecture

The prevailing paradigm during the development of the test in the early 1960s was social learning theory, championed by Albert Bandura and Richard Walters. Social learning theorists argued that sex-typing does not emerge from mysterious internal drives, but rather through overt environmental contingencies: differential reinforcement, punishment, and vicarious observational learning. Parents, educators, and peers actively reward children for selecting toys that conform to normative expectations (e.g., praise for a boy playing with blocks; reprimand or discouragement for playing with dolls). Concurrently, children observe same-sex models navigating real-world environments, internalizing the perceived utility and valence of sex-typed artifacts. The Toy Preference Test operates within this conceptual architecture by presenting pairs of stimuli that embody these socialized reinforcement histories.

Cognitive-Developmental Formulations

Shortly after DeLucia’s publication, Lawrence Kohlberg (1966) published his landmark cognitive-developmental theory of sex-role concepts. Kohlberg posited that children are active cognitive agents who structure their physical and social reality through self-categorization. Once a child achieves basic gender identity (categorizing oneself as a boy or girl) and progresses toward gender constancy (understanding that sex is biologically invariant across time and situational transformations), they actively seek out objects, activities, and behaviors consistent with that cognitive label. In Kohlberg’s paradigm, “I am a boy, therefore I want to do boy things; playing with trucks is a boy thing; therefore, I value trucks.” The Toy Preference Test captures the behavioral manifestations of this cognitive schema matching, tracing the developmental trajectory as children consolidate their gender-typed schemas across the elementary school years.

7. Validity

The psychometric evaluation of the Toy Preference Test’s validity reveals important empirical strengths alongside distinct methodological limitations typical of mid-century psychological test construction.

Content and Construct Validity

The instrument possesses strong face and content validity derived from its meticulous empirical item-selection procedure. DeLucia did not arbitrarily assign masculine or feminine labels to toys. Instead, the author assembled an initial corpus of 52 widely familiar childhood toys and subjected them to empirical scaling by independent adult panels. Judges evaluated each play object on a 9-point continuum (1 = strongly masculine; 9 = strongly feminine). Only items demonstrating extreme mean values and narrow standard deviations—indicating high inter-judge consensus—were selected for the definitive stimulus pool. This rigorous item-derivation procedure ensured that the stimulus pairs represented genuine, culturally robust exemplars of sex-typed artifacts.

In terms of construct validity, empirical administration across kindergarten through fourth-grade cohorts revealed distinct developmental and sex-differentiated trends consistent with theoretical predictions:

• Male Stereotyping Trajectory: Male participants consistently achieved significantly higher sex-appropriate preference scores than female participants across virtually all age bands. This finding aligned directly with developmental literature demonstrating that boys experience more rigid, punitive socialization pressures against engaging in opposite-sex play behaviors.
• Age Trends: Scores across successive grade levels demonstrated developmental consolidation. Older boys (third and fourth graders) exhibited near-ceiling conformity in rejecting feminine-valenced toys, validating the construct as an index of progressive cultural learning.

Discriminant and Methodological Validity (Examiner Effects)

A major focus of DeLucia’s psychometric investigation was the evaluation of demand characteristics and examiner-induced bias. Recognizing that children frequently modify their play declarations to conform to adult social expectations, the author systematically manipulated the sex of the experimenter. Construct validity analyses indicated that children’s observed toy preferences were significantly modulated by whether the administrator was male or female, revealing that the test effectively captured situational self-presentation and sex-typed social monitoring rather than merely reflexive motor choices.

Validity Limitations

No formal modern criterion-related validity studies (such as correlating Toy Preference Test scores with longitudinal behavioral observation of unstructured playground activities or teacher-rated gender conformity inventories) were conducted within contemporary psychometric frameworks. Additionally, the original validation studies did not compute modern structural equation models or multigroup invariance metrics to verify whether the construct of toy preference operated identically across racial, socio-economic, and regional sub-populations.

8. Reliability

The stability of scores over time represents one of the most critical psychometric dimensions evaluated in the original standardization of the Toy Preference Test.

Test-Retest Reliability

DeLucia evaluated temporal stability by re-administering the instrument across longitudinal test intervals among elementary school children. Given the developmental volatility of early childhood preferences, establishing stability was essential for validating the instrument as an assessment of stable personality/role identification rather than transient mood states:

• Opposite-Sex Experimenter Stability: When the test protocol was administered by an opposite-sex experimenter during both baseline and retest sessions, the rank-order test-retest reliability coefficient ranged from Spearman’s rho = .67 to .72. In developmental psychometrics, coefficients within this range are considered moderate-to-strong, indicating that children maintain consistent relative standing in their sex-role preference rankings over time despite natural fluctuations in situational interest.
• Examiner-Shift Attenuation: When the sex of the experimenter was alternated between test and retest sessions (e.g., initial administration by a male experimenter followed by retest by a female experimenter), reliability coefficients exhibited significant attenuation. This attenuation empirically demonstrated that examiner sex acts as an active environmental cue that triggers differential social desirability responses, particularly among young boys sensitive to masculine role enforcement.

Alternate-Form and Internal Consistency

The construction of two parallel sets—Set A and Set B, each consisting of 24 paired items—allowed for the evaluation of parallel-forms equivalence. Both sets were balanced based on adult normative stimulus ratings. While modern psychometric evaluations routinely calculate Cronbach’s alpha, McDonald’s omega, and split-half reliability, these metrics were not systematically reported in the original 1963 monograph. However, the high degree of stimulus matching across Set A and Set B provided structural assurance of cross-form equivalence, minimizing measurement error due to specific toy familiarity.

9. Factor Analysis

At the time of the instrument’s inception in 1963, modern computer-assisted factor analysis—specifically Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA)—was not routinely applied to forced-choice, paired-comparison developmental tests. Consequently, no empirical factor structure, matrix of item loadings, or structural equation model was reported in the original literature.

Latent Dimensionality and Contemporary Structural Interpretations

From a theoretical and psychometric perspective, the Toy Preference Test was constructed under an implicit unidimensional assumption: that sex-role identification existed as a single, bipolar continuum extending from extreme masculinity at one terminus to extreme femininity at the other. Under this classical bipolar model, selecting a masculine toy automatically implied the rejection of femininity, and vice versa. Each of the 24 forced-choice pairings represented an operational indicator of this singular underlying latent trait ($ heta$).

Subsequent psychometric scholarship in gender role measurement—most notably catalyzed by Sandra Bem’s (1974) work on androgyny and the development of the Bem Sex-Role Inventory (BSRI)—critiqued the unexamined bipolarity of mid-century instruments like DeLucia’s. Had contemporary factor analytic procedures been applied to the Toy Preference Test, empirical data would likely have revealed multidimensionality rather than strict unidimensionality. Potential latent factors underlying the stimulus pool include:

• Active/Rough-and-Tumble Mastery: Factor loadings clustered around heavy machinery, military toys, construction equipment, and outdoor sporting implements.
• Nurturance and Domestic Affiliation: Factor loadings driven by baby dolls, infant caretaking accessories, kitchen appliances, and grooming tools.
• Aesthetic/Fine-Motor Orientation: Factor loadings reflecting arts, crafts, and decorative play.

Furthermore, because the forced-choice paired-comparison format imposes ipsative constraints on the resulting data (preference for one pole mechanically depresses the score of the alternative pole), standard Pearson correlation matrices among items are structurally distorted. Modern psychometric evaluation of paired-comparison instruments requires item response theory (IRT) models designed for forced-choice data, such as the Bradley-Terry-Luce model or the Thurstonian IRT model, to estimate latent parameters without the statistical artifacts inherent in classical ipsative scoring.

10. Instrument / Measurement Tool

The structural characteristics, administration protocol, and scoring mechanics of the Toy Preference Test are detailed below:

• Test Classification: Objective, nonverbal, forced-choice pictorial preference test.
• Target Population: Children in early to middle childhood (kindergarten through fourth grade; chronological ages 5 to 10 years).
• Stimulus Material: High-contrast black-and-white photographic plates displaying 52 familiar, culturally standardized children’s play objects.
• Form Structure: Divided into two balanced, parallel forms:
  • Set A: Comprising 24 distinct pairs of toy photographs.
  • Set B: Comprising 24 alternate, parallel pairs of toy photographs.
• Stimulus Pairing Logic: Each individual item presentation consists of exactly two juxtaposed photographs:
  • One photographic stimulus representing an empirically rated masculine-typed toy (e.g., dump truck, hammer, train, soldier).
  • One photographic stimulus representing an empirically rated feminine-typed toy (e.g., doll carriage, sewing kit, tea set, nurse kit).
• Item Selection Standardization: Prior to test assembly, toys were rated by adult judges on a 9-point scale ranging from 1 (strongly masculine) to 9 (strongly feminine). Only items occupying extreme ends of the distribution with minimal variance were paired.
• Administration Procedure: Individual administration conducted in a quiet, distraction-free experimental room. The examiner presents the photographic pairs sequentially in a fixed, standardized order. For each pair, the examiner instructs the child: “Look at these two toys. Point to the one that you would like to play with the best.”
• Response Modality: Nonverbal gestural response (pointing) or verbal indication, minimizing expressive language demands.
• Scoring Algorithm:
  • A choice matching the child’s own biological/assigned sex (e.g., a boy selecting the masculine-typed toy, or a girl selecting the feminine-typed toy) is scored as 1 point (sex-appropriate selection).
  • A choice selecting the opposite sex-typed toy is scored as 0 points (sex-inappropriate / cross-sex selection).
• Total Score Range:
  • Single Set Administration (Set A or Set B): Scores range from 0 to 24 points.
  • Combined Administration (Sets A and B): Scores range from 0 to 48 points.
• Score Interpretation:
  • 20 – 24 (or 40 – 48): High sex-role conformity / strong traditional sex-typed preference.
  • 12 – 19 (or 24 – 39): Moderate sex-role preference with intermediate cross-sex play interests.
  • 0 – 11 (or 0 – 23): Pronounced cross-sex preference / non-conforming sex-role preference.

11. Permissions & Fee and Test Year

• Publication Year: 1963.
• Original Author: Lenore A. DeLucia, Ph.D. (Brown University).
• Original Publication Outlet: Published within academic peer-reviewed developmental literature in the United States (DeLucia, 1963).
• Copyright and Commercial Status: The Toy Preference Test is not marketed as a commercial clinical kit by commercial publishers (such as Pearson Clinical, Western Psychological Services, or PAR). The technical parameters, stimulus classifications, and standardization procedures were published in the public academic literature for scholarly and experimental research.
• Permissions and Research Access: Researchers wishing to replicate or adapt the visual stimuli for non-commercial historical or psychological investigations may consult the original primary literature. Reproduction of the specific archival photographic plates or original stimulus cards must respect relevant academic publication copyrights and fair-use guidelines. Contact with original institutional archives (Brown University Department of Psychology) may be required for historical source access.
• Fee: Historically, there is no standardized licensing fee for academic research use; however, acquisition of archival microfiche or university repository documents may incur institutional document retrieval fees.

12. References

• Bandura, A. (1969). Social-learning theory of identificatory processes. In D. A. Goslin (Ed.), Handbook of socialization theory and research (pp. 213–262). Rand McNally.
• Bem, S. L. (1974). The measurement of psychological androgyny. Journal of Consulting and Clinical Psychology, 42(2), 155–162. https://doi.org/10.1037/h0036215
• Brown, D. G. (1956). Sex-role preference in young children. Psychological Monographs: General and Applied, 70(14), 1–19. https://doi.org/10.1037/h0093726
• Brown, D. G. (1957). Masculinity-femininity development in children. Journal of Consulting Psychology, 21(3), 197–202. https://doi.org/10.1037/h0047391
• DeLucia, L. A. (1963). The Toy Preference Test: A measure of sex-role identification. Child Development, 34(1), 107–117. https://doi.org/10.2307/1126834
• Kagan, J. (1964). Acquisition and significance of sex-typing and sex-role identity. In M. L. Hoffman & L. W. Hoffman (Eds.), Review of child development research (Vol. 1, pp. 137–167). Russell Sage Foundation.
• Kohlberg, L. (1966). A cognitive-developmental analysis of children’s sex-role concepts and attitudes. In E. E. Maccoby (Ed.), The development of sex differences (pp. 82–173). Stanford University Press.
• Maccoby, E. E., & Jacklin, C. N. (1974). The psychology of sex differences. Stanford University Press.
• Mischel, W. (1966). A social-learning view of sex differences in behavior. In E. E. Maccoby (Ed.), The development of sex differences (pp. 56–81). Stanford University Press.
• Ruble, D. N., Martin, C. L., & Berenbaum, S. A. (2006). Gender development. In W. Damon & R. M. Lerner (Series Eds.) & N. Eisenberg (Vol. Ed.), Handbook of child psychology: Vol. 3. Social, emotional, and personality development (6th ed., pp. 858–966). John Wiley & Sons.

13. Items of the Scale