Social Pretend Play in Korean- and Anglo-American Preschoolers

Child Development, June 1997, Volume 68, Number 3, Pages 544-556

Jo Ann M. Farver and Yoolim Lee Shin

Ninety-two preschoolers (46 Anglo- and 46 Korean-American) were observed during free play activities and videotaped in an experimental toy play setting. Cultural differences were examined in the frequency of social pretend play, communicative strategies, and pretend play themes. Anglo-American children engaged in more pretend play during free play activities than Korean-American children. In the experimental setting, there were no cultural differences in the frequency of pretend play; however, there were significant differences in children's communicative strategies and in their play themes. Korean-American children's play included everyday activity and family role themes, whereas Anglo-American children enacted danger in the environment and fantastic themes. Anglo-American children described their own actions, rejected their partners' suggestions, and used directives, whereas Korean-American children described their partners' actions and used tag questions, semantic ties, statements of agreement, and polite requests. The findings suggest that play is a common activity for most children. However, the thematic content and the communicative strategies used to structure and maintain pretend play are influenced by culture.

INTRODUCTION

This study compared Korean- and Anglo-American preschoolers' play behavior to understand how culture influences the frequency and expression of pretend play. A major task in developmental theory is to explain how structure in culture-specific settings shapes children's development. Super and Harkness (1986) propose that the characteristics of children's physical and social environments set the course for developmental incomes. According to this theory regularities within settings, customs, and cultural belief systems organize children's developmental experiences and provide the information from which they construct the rules of their culture. Current research on children's play in the United States and other societies supports this notion by stressing the importance of culture-specific practices in structuring and organizing the environment in which children's social interaction and play activities take place (Farver, 1993; Farver & Howes, 1993; Gaskins, 1996; Haight & Miller, 1993; Rogoff, Goncu, Mistry, & Mosier, 1991; Roopnarine, Johnson, & Hooper, 1994).

Factors that have been shown to influence children's play behavior across cultures are the physical and social contexts (i.e., settings, props, time, and personnel available for play), and adult beliefs about play. For example, in a study examining Korean- and Anglo-American preschoolers' play, Farver, Kim, and Lee (1995) found that Anglo-American children engaged in more frequent social pretend play than Korean-American children. These differences were attributed to the way teachers organized the preschool environments and structured children's play opportunities based on their beliefs about play, their

developmental goals for young children, and their emphasis on the development of culturally valued skills. In the Korean-American setting, where play was not considered beneficial for early childhood education, teachers organized their classrooms and curriculum to encourage the development of academic skills, and children's social interaction and play were limited to brief periods of outdoor activity. In contrast, consistent with their beliefs about the positive value of play and their educational goals for young children, Anglo-American teachers provided an environment that facilitated self-expression, social skills, and cognitive development, and children had many opportunities for social interaction and play.

The present study extends this prior work by examining Korean- and Anglo-American children's play in two situations: during free play activities in their preschools and in a quasi-experimental condition where children were provided with imaginative toys in a playroom setting. The first objective was to determine if the cultural differences we found in the frequency of children's social pretend play in the first study were related to facets of the culturally mediated contexts provided in the two preschool settings. A second objective was to understand how culture influences children's expression of pretend play by examining the communicative strategies they use to organize and maintain play with a partner, and the thematic content of their play episodes.

Prior research conducted with Anglo-American children has indicated that by age 4, most children

Farver and Shin 545

have mastered the basic elements of social pretend play (Howes, Unger, & Matheson, 1991). They are able to enact social roles, superimpose story lines or scripts on their activities (Forys & McCune-Nicolich, 1984; Garvey & Brendt, 1977; Rubin, Fein, & Vanden-berg, 1983; Schwartzman, 1978), and use metacom-munication to establish and maintain play sequences (Giffin, 1984).

In her work, Garvey (1984, 1990) has noted that much of children's talk is directed toward creating and maintaining the social pretend experience. During play children must attract their partner's interest, establish a common frame of reference by naming or inventing objects and assigning roles, and then col-laboratively construct a joint line of action and discourse. Consequently, children must monitor the interaction and consistently acknowledge and expand on their partners' contributions.

Analyses of play dialogues suggest that children rely on particular communicative strategies to make their actions and ideas intelligible and shareable with a partner (Corsaro, 1986; Farver, 1992a, 1992b; Goncu & Kessel, 1984; Nelson & Seidman, 1984). These include suggestions for activities to take place, role assignments (Fein & Rivkin, 1986), descriptions of action to inform the partner about the meaning of the play sequences (Giffin, 1984), semantic ties to expand on the partner's contribution (Goncu & Kessel, 1984, 1988), and tag questions to elicit acknowledgment or agreement (Corsaro, 1986).

Whereas research illustrates that much of Anglo-American children's social pretend play involves self-explanation, negotiation, and the development of shared realities (Farver, 1992a, 1992b; Giffin, 1984; Goncu, 1989; Goncu & Kessel, 1984, 1988), it is unclear how children from other cultural backgrounds coordinate social pretend play. According to Garvey (1990) and others (Schiefflin & Ochs, 1986), all communication, as well as that used by children in pretend play, is governed by the pragmatics of language, the social conventions of a particular culture, and cultural norms about self-expression. For example, young children often adapt their speech to fit social contexts. They use "motherese" to address a younger child (Shatz & Gelman, 1973; Tomasello & Manle, 1985) and are more formal in their choice of words and syntax when they role-play characters such as fathers, mothers, or teachers (Owens, 1990).

The implicit rules for using language effectively and appropriately have been shown to differ in the Korean and American cultures (Kim, 1991; Lee & Lee, 1987). In Korean culture social status is reflected in language use. Based on Confucianism, Korean individuals are assigned a hierarchical social position re-

flecting their age, role, and gender that is acknowledged through the use of honorifics (Chu, 1978; Kim, 1991). Thus a person of lower status uses a different form of the Korean language when addressing someone of higher status.

There are also differences in Korean and American styles of communication. In Korean culture, where harmonious relationships are valued, communication is rarely direct or confrontational. Instead, one talks "around" an issue and relies on the other's sensitivity to understand the point of the conversation (Chu, 1978; Kim, 1991; Lee & Lee, 1987). Korean childrearing practice also discourages children from expressing their own opinions or asserting themselves especially when their ideas differ from those of family or friends (Kim & Choi, 1994; Lee & Lee, 1987). Korean children are taught to control the display of emotion, to value group harmony, and to minimize conflict in social interaction (Kim, 1991). In contrast, in American culture where parents encourage the development of self-reliance, self-expression, and independent action (Hoffman, 1988; LeVine, 1980; Rohner & Pettengill, 1985; Whiting & Edwards, 1988), children's style of communication is direct, but their social interaction is also often conflictual.

Research has also shown that Korean and American cultures have different views about the self, others, and interdependence of the two. In an empirical study examining the psychological connotations of "me" among Korean and American respondents, Maday and Szalay (1976) found that Americans focused on the individuated self, or the "I," whereas, Koreans emphasized the relatedness and interdependence among individuals, or the "we." The Korean interdependent view of the self is generally expressed in a collectivist orientation. Chong, which is the affective bond uniting Korean families and social groups, is associated with being attentive to others' needs, attempting to "read" their minds, and maintaining harmonious shared experiences (Choi, Kim, & Choi, 1993). In contrast, the American independent view of the self is generally expressed in an individualistic orientation where attending to the self, individuality, self-assertion, and "saying" what is on one's mind is encouraged (Markus & Kitayama, 1991). Cultural variations in social conventions, communication styles, and the construal of the self can be expected to influence the nature of individual experience, social behavior, and self-expression as well as how children engage in and communicate social pretend play.

Previous research on pretend play themes suggests that imaginative scenarios depict children's knowledge of real-life events (Bretherton, 1984), their

546 Child Development

concerns (Fein & Rivkin, 1986), and their attempts to organize and make sense of their experiences (Bruner, 1990; Paley, 1986). Garvey (1990) and others (Black, 1989; Corsaro, 1985) have identified play themes that are linked to familiar real-life experiences (e.g., cooking, shopping, sleeping, birthday parties, family relationships, and so on) and fantastic themes which draw on childhood stories (e.g., Cinderella, Black Beauty, monsters, ghosts, and so on).

In accounting for variations in play themes, Black (1989) suggests that familiar everyday themes may be common among young children because they are easily shared and facilitate long bouts of pretend play by reducing the likelihood of arguments about rules and procedures. Differences in children's play themes may also be related to their social and personal goals during peer interaction. Several theorists have suggested that children who are less concerned with maintaining harmonious collaborative play and more concerned with pursuing their own individual interests may choose to enact more fantastic imaginative play themes than familiar themes (Connolly, Doyle, & Reznick, 1988; Garvey, 1990; Martini, 1994).

Cross-cultural studies demonstrate that children enact themes that portray universal familiar activities (e.g., buying, cooking, and eating food; and family relationships), as well as culture-specific themes that reflect activities that are important to a particular society. For example, in the Marquesas Islands, Martini (1994) observed children pretending to paddle canoes and to hunt and fish. In India, Roopnarine, Hos-sain, Gill, and Brophy (1994) observed children enacting traditional celebrations and folk tales. In the United States, Farver (1992a) observed children pretending to cut down trees in the mountains and to eat lunch with Santa Claus because the data were collected during the American Christmas holidays.

Based on the results of our prior study on Korean-and Anglo-American children's play behavior, and on the cultural differences noted in the research literature on the socialization of Korean and American children, we expected that there would be cultural differences in the frequency and expression of children's pretend play. First, because the Korean- and Anglo-American preschool settings in the current study were very similar to those where we conducted our first study, we expected that the culturally defined social environments would shape children's pretend play during free play activities and in the experimental play condition. Accordingly, we hypothesized that Korean-American children would engage in less social pretend play than Anglo-American children in both play conditions.

Second, based on the culturally divergent views of

the self (interdependent versus independent), and in the social conventions of communication, we expected that there would be cultural differences in children's use of communicative strategies and in the thematic content of their pretend play sequences. We predicted that Korean-American children would be sensitive to relative social status and would avoid using strategies that required them to direct another child's behavior, to set and enforce rules, or to decide roles or scripts. Instead, we hypothesized that Korean-American children would be attentive to their partner's ideas and would use strategies to facilitate cooperative and harmonious dyadic play, such as describing their partner's action (rather than their own action); statements of agreement and tag questions to elicit their partner's acknowledgment or approval-polite requests for information, access to the toys or to play a certain character; and semantic ties to build on their partner's contributions. In contrast, we expected Anglo-American children to be relatively self-assertive in their use of communicative strategies by describing their own actions (rather than their partner's); directing their partner's play behavior; rejecting their partner's suggestions for play activities, themes, and roles; and using calls for attention to gain their partner's immediate attention.

Finally, we hypothesized that Korean-American children would enact familiar play themes that require little explanation and negotiation and in turn minimize potential social conflict. We predicted that Anglo-American children would enact fantastic play themes that require more negotiation, explanation, and peer direction than familiar play themes.

To address our hypotheses children were observed during free play activities in their preschools and were videotaped with a familiar same-sex partner in a quasi-experimental toy play session. Videotaped play sequences were examined for communicative strategies and thematic content. Examples of Korean- and Anglo-American children's discourse during these play sequences appear below. Two Anglo-American children enacted the following:

A: Dragon horse! My dragon is fighting. Hey--Look out! (High voice speaking for the figure. A throws dragon on the castle roof)

B: (Picks up dragon)

A: No! I want him. He's fighting. Don't go in there. (Tries to take dragon from B's hand while closing the castle drawbridge)

B: I killed the dragon, sir. (In low voice speaking for the knight figure)

A: No you didn't. He's alive. See? ("Walks" the dragon over to the castle)

Farver and Shin 547

B: I'm in a fire! (In high voice speaking for the figure; holds the knight figure)

A: No, he's not in the fire, he isn't!

B: Yes, he is! (Regular speaking voice) And you're the naughty dragon. Naughty dragon! And you are sick. They are all sick and they all go to jail. The castle is closed. Halleluiah! (Low voice speaking for the figure)

A: Nobody worries, cause I'm okay. (Puts the dragon on a 'bed" in the castle and picks up a horse)

B: I want to ride on that with you.

A: No, you ride your own horse.

Two Korean-American children enacted the following:

A: He is a king. There are two kings. (Picks up the king and knight figures and shows them to B)

B: What? (Looks at figures)

A: There are two kings.

B: Ah, right.

A: He died. Let's make him alive. Let's put him here like this. (Puts king on a "bed") Then he will be alive. Then put him like that. He is not alive yet.

B: What? (Looks at figure as it rolls off the bed)

A: He is not alive yet.

B: She must be dead, too. (Holds queen figure, then hands to A)

A: Put her here too. (Puts queen on another bed next to the king) Ah alive! He was alive!

B: Then what about her? (Pointing to the queen)

A: She isn't alive yet.

B: Why?

A: Because she was just dead.

B: Oh yes, she is dead.

METHOD

Participants

The participants were 92 (46 Anglo- and 46 Korean-American) preschool children (half girls) ranging in age from 44 to 63 months (Korean-American, M = 48.50, SD = 3.56; Anglo-American, M = 48.10, SD = 4.56). All children attended full-day preschool programs for at least 5 months prior to the data collection. The Anglo-American children were enrolled in a university-based preschool located in a suburban community in the western United States. They were all from intact, white, middle-class families. The Korean-American children attended two all-Korean pre-schools located in a Koreatown section of a west coast

city. They were all from intact, middle-class Korean immigrant families.

The Preschool Settings

The Anglo-American preschool had an adult-child ratio of 1:7 with 12-18 children per class. Teachers were all white and middle class and held college degrees in early childhood education. The play-based curriculum included daily group-time activities, self-select learning centers, and three 30 min outdoor free play periods. The preschool curriculum was based on the National Association for the Education of the Young Child Developmentally Appropriate Practice Guidelines. The two Korean preschools were similar in adult-child ratio of 1:7, average group size (12 children), and philosophy. The directors and teachers were all middle-class, Korean-born immigrants. The preschools had bilingual programs emphasizing pre-academic skills. Although the daily schedules included indoor free play and three 30 min outdoor free play periods, large group and teacher-directed instruction dominated the curriculum.

Procedure and Measures

Observations of Play Behavior

To ensure that the observations were representative of children's play behavior, children in both groups were observed individually, in random order, on four separate occasions during free play activities in their preschools over a 4 month period. Observations began when a child started to interact with a peer and continued for 5 min whether or not the interaction was maintained.

The target child's highest level of play during each 60 sec interval was noted using a scale developed by Howes (1988). Solitary play was coded when the child was playing alone. Parallel play was coded when the child and partner(s) engaged in the same or similar activity but did not make eye contact or engage in social interaction. Simple social play was coded when the child and partner(s) engaged in the same or similar activity and directed social bids to each other (i.e., smiled, offered or received an object). Complementary and reciprocal play was coded when the child and partners) engaged in social play with a turn-taking structure and role reversal. Social pretend play was coded when the child and partner(s) engaged in fantasy play (i.e., children acted or used objects in an "as if" manner, engaged in scripted pretend play, or enacted complementary pretend roles such as mother and baby).

548 Child Development

Four types of predominant affect were coded and were used in determining children's behaviorally identified friends for the experimental procedure. Neutral affect was coded when the target child had a unresponsive face and looked emotionally absent. Positive affect was coded when the target child smiled or laughed. Shared positive affect was coded when the target child and partner(s) both smiled or laughed. Negative affect was coded when the target child frowned, displayed anger, or showed distress. The name of the target child's play partner(s) was noted for each 60 sec interval.

The Experimental Play Condition

Each child was paired with a self-chosen same-sex play partner, and the dyads were videotaped in a familiar playroom at their preschools for 20 min while playing with a Fisher-Price Castle and its accompanying accessories (a king, queen, prince, princess, knight, two horses and a carriage, a pink dragon, beds, and chairs).

The dyads were formed by allowing the target child to select a play partner from his or her behavior-ally identified friends. Behaviorally identified friends were determined from the free play observations. These friends were defined as peers who maintained proximity (by being within 3 ft of each other) and expressed shared positive affect while engaged in social play in at least 30% of the combined observations of the two children (Howes, 1988). A 2 (sex) X 2 (culture) analysis of variance test for the number of behaviorally identified friends was not significant. Boys and girls in both groups distributed play among three to four partners.

The videotapes were transcribed verbatim. All statements made by the children as well as any vocalizations such as paralinguistic cues or behaviors that served a communicative function were recorded.

Social play complexity. The transcriptions were keyed to the videotaped footage and were divided into 20 60-sec intervals. Each 60 sec interval was coded for the target child's highest level of play complexity using the Howes (1988) scale described above.

Communicative strategies. All 60 sec intervals containing social pretend play were isolated and were coded for 10 communicative strategies. Strategies were scored for the number of times they occurred during each 60 sec interval. The first seven communicative strategies defined below were identified by Corsaro (1986) and the following three were derived from the videotapes using the grounded theory method (Glaser & Strauss, 1967). Using this inductive approach, the communicative strategies that did not

fit Corsaro's scheme were noted, categorized, and coded based on the frequency of their occurrence (i.e., statements of agreement, reject play, and polite requests). The purpose of this approach was to uncover culturally specific patterns in children's use of communicative strategies as they occurred in the play context.

Descriptions of action were declarative statements accompanying ongoing activity which described what the target child or partner was doing. For example, a child says, "I'm going into the forest" as she moves the horse to the back of the castle (describe own action). The play partner responds, "Oh no, you're going into the forest where the dragon lives" (describe partner's action). Semantic ties were statements that added new elements to the partner's previous contribution. For example, child A: "It's raining. Gonna rain." Child B: "Rain! Its gonna be a rainstorm." Tag questions were verbal devices normally placed at the end of a statement to elicit a response or an acknowledgment. These included, "We're playing prince and princess, right?" "I'll put the chair here, Ok?" "This is the bad dragon, isn't he?" Directives were declaratives used to control the partner's action, for example, "No, that guy won't fit on the horse"; "Don't do that!" "You have to put the queen on her chair." Directives could also occur within scripts or role play. For example, the "queen" says to the "dragon," "You go into the dungeon, you bad dragon!" Calls for attention were utterances used to gain the partner's immediate attention. These included the partner's name, "Hey!" or "Look!" Paralinguistic cues were changes in intonation and pitch to mark fantasy and the animation of objects, for example, the "vroom vroom" sounds of car engines, the "rooooarrrr" of a dragon, and the use of high or low voices to mark role enactment. Reject play was coded when a suggestion for play was refused or ignored or when a child refused to share the toys or prohibited the partner from taking part in the play. For example, child A: "We're playing prince and princess, and you can be the prince." Child B: "No, I'm not playing that! I'm the mother!"; or "You can't have the horse, 'cause I'm having it." Statements of agreement included the acceptance or approval of the partner's suggestions for play themes or roles, for example, child A: "Let's put these guys in the dungeon cause they're bad." Child B: "OK, let's put them in." Polite requests included questions that recognized the right of the partner to refuse the request, for example, "Could I please be the king?" "Can we call the police to kill the dragon?" or "Should we be sisters or princesses?"

Thematic content. Four pretend play themes were coded for each 60 sec interval that contained social

Farver and Shin 549

pretend play using a scale adopted from Rosenberg (1985) and Farver (1992a, 1993). A theme was defined as "a plan or logical temporal order of events, that was identifiable either by the participants' explicit comments on the plan or by an obvious relation between actions" (Dunn & Dale, 1984, p. 137). Family relations included setting up a family structure, identifying the figures as family members, or enacting family roles. Everyday activities included sleeping, cooking and eating, dressing, fixing things, driving cars, making telephone calls, studying, going places, getting married, and having birthday parties. Danger in environment included aggressive behaviors (killing, shooting, and injuring characters, or crashing and running over with horses), and dangerous situations (disasters, storms, accidents, getting lost, or becoming ill). Fantastic themes included extraordinary actions performed by fantasy characters.

Reliability

The Korean-American children were observed by the bilingual second author and a bilingual graduate student assistant. The Anglo-American children were observed by two graduate student assistants. Prior to the data collection in both settings, the first author trained the second author and all research assistants to a criterion of 90% agreement.

To establish reliability for the Anglo-American sample, the first author and the assistants simultaneously recorded 10 5-min observations. Observations were compared and scored. This procedure was repeated four times during the data collection period to minimize interobserver drift. To establish reliability for the Korean-American observations, the second author and assistant used the same procedure described above for the Anglo-American sample. Cohen's kappas interobserver reliabilities on the individual variables ranged from .88 to .94 {median - .90) for the Korean-American sample, and from .85 to .92 (median = .89) for the Anglo-American sample.

The Korean-American videotapes were transcribed by the second author and a bilingual bicul-tural assistant, and the Anglo-American videotapes were transcribed by four psychology undergraduate students. The Anglo- and Korean-American videotapes were coded by the second author and a bilingual bicultural graduate student assistant who was blind to the study's objectives. The second author trained her assistant by using eight randomly chosen videotaped play sessions (four from each culture, two from each sex dyad). To establish reliability, eight additional videotapes were randomly selected (four from each culture) and were coded indepen-

dently by the second author and the assistant. Cohen's kappas for agreement on segmenting the tapes into 60 sec intervals, for coding the level of play complexity, the four play themes, and the number of times the 10 communicative strategies occurred during the play episodes, ranged from .87 to .92. Similar reliability checks performed midway and at the end of the coding ranged from .87 to .97. Because the data were coded by the second author, careful attention was given to training and establishing reliability with her assistants.

Acculturation Questionnaire

Korean parents completed an acculturation measure (Cuellar, Harris, & Jasso, 1980) translated into the Korean language which consisted of questions about family demography, language preference, and cultural identification rated on a 5 point Likert scale. This measure differentiates five distinct types of Korean-Americans from (1) very Korean to (5) very Anglicized. Acculturation ratings were computed for each family using the guidelines provided with the measure.

RESULTS

Korean Sample Characteristics

Mean scores were computed for the Korean families' responses to the acculturation scale by summing their answers and dividing by the total. The 46 families' acculturation scores ranged from 1.03 (1 = very Korean) to 1.86 (2 = 1.5 generation Korean, i.e., the term designating the first assimilated generation) (median = 1.38). All 46 children were born in the United States, and their parents and grandparents were all born in Korea. The parents had immigrated to the United States 6-11 years ago (M = 9.6 years) and were employed as professionals in the Korean community or were self-employed. Ninety percent of the families spoke only Korean at home and expressed their preference for speaking Korean rather than English outside the home. About 80% identified themselves as "Korean" and the other 20% considered themselves as "1.5 generation." Eighty-five percent associated almost exclusively with other Korean-Americans. Families watched Korean and American television and movies about equally.

Comparison of Social Pretend Play by Culture and Play Condition

To ensure that children's play partners were the same in both play conditions, intervals during free

550 Child Development

Table 1 Proportions of Social Pretend Play by Culture and Play Condition

			Culture
Pretend Play Condition	Korean-American	Anglo-American	F Culture	F Setting	F Culture X Setting
Pretend Play Condition		% SD	F Culture	F Setting	F Culture X Setting	% SD
Free play Toy play	.28 (.05) .45 (.15)	.36 (.09) .49 (.17)	12.31***	52.90***	1.26

***p < .001.

play in which children were observed to interact with their self-chosen behaviorally identified friend were selected for analysis. Proportions were calculated for the frequency of social pretend play in the free play and experimental conditions. To avoid violating the assumptions of multivariate analysis of variance (MANOVA) tests with proportional variables, an arc-sine transformation was conducted. The transformed variables were used in the subsequent analyses.

To examine the hypothesis that there would be cultural and play setting differences in children's social pretend play, the proportions of social pretend play were compared using a 2 (culture) X 2 (sex) repeated-measures MANOVA test with play condition as the repeated factor. Because there were no significant results for sex it was dropped from further analysis.

As shown in Table 1, there were significant findings for culture and play condition. The Anglo-American children engaged in a higher proportion of social pretend play than the Korean-American children, F(l, 90) = 12.31, p < .001. There was a higher proportion of social pretend play in the toy play condition than in the free play condition, F(l, 90) = 52.90, p < .000. There was no significant interaction of culture and condition.

Comparison of Communicative Strategies in the Toy Play Condition by Culture

To examine the hypothesis that there would be cultural differences in the strategies children used to construct and maintain social pretend play, frequencies of the 10 strategies were compared for culture and sex using a MANOVA test. Because there were no significant findings for sex it was dropped from further analysis.

There were significant multivariate F results for the effects of culture and strategies, F(10, 81) = 19.29,

p < .000. Univariate F tests displayed in Table 2 show that Korean-American children described their partner's action, F(l, 90) = 13.58, p < .01, used semantic ties, F(l, 90) = 9.23, p < .01, tag questions, F(l, 90) = 9.79, p < .01, statements of agreement, F(l, 90) = 84.58, p < .000, and polite requests, F(l, 90) = 120.69, p < .000, more frequently than did Anglo-American children. Anglo-American children described their own actions, F(l, 90) = 9.73, p < .01, used directives, F(l, 90) = 33.08, p < .000, and rejected their partners' play, F(l, 90) = 7.23, p < .05, more frequently than did Korean-American children. There were no significant findings for calls for attention or paralinguis-tic cues.

Frequency of Social Pretend Play Themes in the Toy Play Condition by Culture

To examine the hypothesis that there would be cultural differences in children's pretend play themes, the frequencies of the four play themes (family roles, everyday and social activities, danger in environment, and fantastic themes) were compared by culture using a MANOVA test. There were significant multivariate F results for the effects of culture and play themes, F(4, 87) = 7.79, p < .001.

Univariate F tests, displayed in Table 3, show that family role, F(l, 90) = 5.31, p < .05, and everyday activity themes, F(l, 90) = 26.41, p < .001, were more common among Korean-American children than among Anglo-American children. Fantastic F(l, 90) = 8.30, p < .01, and danger in the environment themes, F(l, 90) = 5.69, p < .05, were more common among Anglo-American children than among Korean-American children.

DISCUSSION

The Korean-Americans studied here provided a unique opportunity to examine the influence of a

Farver and Shin 551

Table 2 Frequency of Communicative Strategies in Social Pretend Play in the Experimental Toy Play Condition by Culture

			Culture
	Korean-		Anglo-
	American		American
Strategies	M	SD	M	SD	F
Describe own action	2.91	(2.26)	4.82	(3.48)	9.73"
Describe partner's action	9.32	(4.79)	4.36	(7.75)	13.58"*
Semantic ties	9.52	(8.83)	4.56	(6.65)	9.23**
Tag questions	2.28	(3.88)	.41	(1.14)	9.79**
Directives	1.13	(1.43)	5.28	(4.67)	33.08*"
Statement of agreement	6.30	(2.96)	1.45	(1.99)	84.58***
Polite requests	13.80	(6.45)	2.56	(2.55)	120.69"*
Call for attention	1.98	(2.23)	2.02	(2.09)	.09
Reject play	2.71	(3.29)	4.23	(3.77)	7.23*
Paralinguistic cues	5.59	(6.21)	5.95	(5.58)	.08

* p < .05; ** p < .01; *** p <

.001.

non-Western culture on children's expression of social pretend play. The sample can be considered fairly representative of Korean culture primarily because the families had low scores on the acculturation measure, which indicated that they were not extensively assimilated into American culture. The parents were all reared in Korea and were recent immigrants to the United States. They reported associating primarily with other Koreans within their immediate community and resided within a Koreatown inner-city neighborhood that is densely populated with other relatively nonacculturated Korean immigrants. Furthermore, recent research suggests that Korean immigrants continue to be strongly influenced by their traditional culture long after moving to the United States, which often results in intergenera-tional conflict within families (Kahng, 1990; Kim, 1995; Lee, 1989; Lee & Lee, 1990; Min, 1988; Yu, Phillips, & Yang, 1981).

The results suggest that there were two distinct culturally-defined social environments which in part determined children's opportunities for social interaction and pretend play. Therefore, the finding that Korean-American children engaged in less social pretend play during free play activities than Anglo-American children is not surprising and confirms the results from our previous study. This finding can be explained by the similarity in the preschool environments. In both studies, the Korean-American pre-schools had few, if any materials available for pretend play, and existing toys were limited to gross motor play (i.e., bikes, climbing structures, scooters, and so on). Also, because culturally valued school readiness tasks were emphasized and teacher directed activities were the most common, children had few opportunities to interact socially with their peers. In contrast, in the Anglo-American preschools in the previous and current study, teachers encouraged

Table 3 Frequency of Social Pretend Play Themes in the Experimental Toy Play Condition by Culture

Culture

Korean-American

Anglo-American

Themes

Family roles	.48	(.86)	.15	(.42)	5.31*
Everyday activities	3.20	(3.49)	.48	(.84)	26.41***
Danger in the environment	1.48	(2.36)	3.07	(3.84)	5.69*
Fantastic themes	.28	(.62)	.96	(1.46)	8.30**

*p < .05; **p < .01; ***p < .001.

552 Child Development

children's play activities and social interaction by providing materials and curriculums where they were free to wander between learning centers, to interact with their peers, and to play with available toys.

The results did not confirm our second hypothesis that there would be cultural differences in children's social pretend play in the experimental toy play condition. In the experimental condition, where children were provided with imaginative toys in a quiet playroom setting, there were no significant cultural differences in children's social pretend play. In both cultures, nearly half of the toy play intervals involved social pretend level play. This finding supports the notion that the frequency of children's pretend play is dependent on facets of particular sociocultural contexts.

The findings supported our third hypothesis that there would be cultural differences in children's use of communicative strategies in their social pretend play. Korean-American children were nonconfronta-tional and nonassertive in their dyadic play, and they used communicative strategies that minimized social conflict. They less frequently directed or rejected their partner's play and more frequently described their partner's play actions (rather than their own), and used semantic ties, tag questions, statements of agreement, and polite requests. The following is a typical example of the Korean-American children's communicative strategies:

A: Let's play here. He is a king, isn't he? (Picks up the king figure and shows it to B)

B: Yes. Let's put him here. (Puts king figure on the "bed")

A: He is the bad guy, isn't he? (Picks up the "knight" figure, shows B)

B: Yes, but the good guy caught and killed him, right?

A: Right. The bad guy will die here. (Puts the knight in the castle dungeon/jail)

B: Yes, let's go to sleep now.

A: The good guy caught him, right?

B: Yes, the good guy caught him, so he died.

A: So the jail is broken open.

B: Right. And everybody died . . . the sun died, too.

In contrast, the Anglo-American children described their own actions, frequently directed their partners' play activities, and rejected their partners' ideas or play behaviors. The following is a typical example of Anglo-American children's communicative strategies:

A: Where's the king?

B: I need the dragon. Where's the dragon?

A: I'm having that. (A takes the dragon from B) And I am your king! Do not obey the bad king! (High voice) And here was the queen. Miss Lunchmea, tell them what you want to say. (Low voice) We shall never hurt the people. (High voice speaking for the queen figure)

B: I want this one. (Taking the dragon from A) Help! Help! Help! (High voice) The dragon is getting her! (Low voice)

A: I'll save you! (High voice; tries to take the dragon from B)

B: No! No! Help! Help! (High voice) I want to be the dragon now. (Regular voice)

A: No! The dragon's the queen now. You have this horse now. (Hands В the horse)

B: I'm the horse hero now! and I'll save every one.

A: No, I'm the hero of you. I'm the hero of you 'cause I'm so cute.

B: Well, I'm not playing then. (B gets up to leave the play)

These results suggest that children's use of communicative strategies were consistent with their culture's norms about social behavior and self-expression. Korean-American children, whose culture emphasizes a relational mode characterized by group interdependence, sensitivity to others, and a collec-tivist orientation (Kim & Choi, 1994), used communicative strategies that reflected these cultural values. On the other hand, Anglo-American children, whose culture emphasizes an aggregate mode characterized by independence, a preoccupation with the self and its expression, and a individualistic orientation (Kim & Choi, 1994), were direct in their communication and were not hesitant to speak their minds and make their ideas known. Moreover, the findings that Anglo-American children frequently rejected their partner's play activities suggests that their play was more conflictual than the Korean-American children's play.

The findings supported our fourth prediction that there would be cultural differences in children's pretend play themes. Korean-American children enacted more familiar everyday activities and family role themes, whereas Anglo-American children enacted more dangerous and fantastic themes. Based on previous studies that have suggested that fantastic thematic play is a more mature form of pretense which requires advanced ideational skills (Rubin et al., 1983; Saltz, Dixon, & Johnson, 1977), the findings could in-

Farver and Shin 553

dicate that the Anglo-American children were more skilled in enacting complex social pretend play themes than were the Korean-American children.

Although there are few studies specifically examining Korean-American children's play, research with Anglo-American and other non-Western populations has shown that the form and content of children's pretend play is influenced by the abilities and cultural values of the children's playmates, whether they are parents, siblings, or same-age peers (Farver & Howes, 1993; Farver & Wimbarti, 1995; Gaskins, 1996; Goncu et al., 1991; Haight & Miller, 1993). Accordingly, play can be expected to vary with a child's particular experience. For example, in settings where pretend play is encouraged and believed to be an important factor in development, children who have had a variety of playmates and have acquired a broad repertoire of play styles and experiences may have a distinct advantage over children who have had few such opportunities (Garvey, 1990, p. 132).

Whereas the thematic differences in children's play could be explained by differences in their experiences, it is also possible that Korean- and Anglo-American children may have had different social goals in their play interactions. For example, consistent with the Korean culture's emphasis on harmonious interpersonal relationships, the Korean-American children may have minimized social conflict by enacting familiar realistic themes that required little negotiation and were easily shared with their play partners. On the other hand, Anglo-American children, whose culture promotes the notion of equality and encourages self-reliant social behavior and independent thinking, may have been more interested in pursuing their own interests rather than promoting shared realities.

Research in the Marquesas Islands supports this notion. Martini (1994) reports that Marquesan children are sensitive to status differences and are intolerant of dominant children who try to "boss" other children or make rules for play. Instead, children maintain group solidarity and avoid social conflict and the need for play negotiation by engaging in familiar games that are played in the same way from one time to the next, and in which everyone performs the same actions at the same time.

Parental attitudes may have influenced Korean-American children's choice of play theme. It has been reported that Korean-American parents discourage children from expressing childhood fears and anxieties (Strom, Park, & Daniels, 1986) and encourage the early development of emotional control (Chu,

1978). Consequently, the Korean-American children could have been hesitant about expressing dangerous and frightening experiences in their play.

Similarly, exposure to television cannot be ruled out as a possible influence on Anglo-American children's choice of fantastic or dangerous play themes. Although Korean-American parents reported that they watch English- and Korean-language television about equally, it is possible that the Anglo-American children were exposed to more superhero cartoons and similar television programs than were the Korean-American children.

The finding that children in both cultures used paralinguistic cues to animate objects or to signal pretend role enactments indicates that children adapted their speech to fit the social context. However, there were some cultural variations. For example, we noticed that the Korean-American children's paralinguistic devices often involved greater formality, or the honorific form of the Korean language, when pretending to be high status characters, whereas the Anglo-American children varied the pitch of their voice when they spoke out of the pretend frame, enacted roles, and provided speech or character dialogue. Hence, children's play seemed to reflect the social conventions and rules for behavior that are learned from the adults in their culture and practiced in play with their peers.

The relative absence of sex differences in children's play behavior is somewhat surprising given the fact that the Korean-American children came from fairly traditional families which tend to emphasize sex-stereotyped behavior, and the American research literature on play contains many findings for sex differences. Although we have no clear explanation for these findings, we suggest that the absence of sex differences in the play episodes may be related to the teachers' educational goals and how the classrooms were organized. For example, in the Anglo-American settings one of the program goals was to provide an anti-bias curriculum in which teachers make conscious efforts to reduce sex-role stereotypes, and in the Korean-American setting, children were expected to display "proper" school behavior by being quiet, cooperative, and group-oriented. Thus both preschools could have been in effect, "diminishing" sex-typed behavior, albeit in different ways and for different ends. The absence of sex differences in children's use of communicative strategies is not surprising, as contrasting findings have been reported in the research literature and there is little agreement on this topic (Corsaro, 1986; Farver, 1993; Goncu & Kessel, 1984; Sachs, Goldman, & Chaille, 1984).

554 Child Development

These results suggest that social pretend play is a common facet of early childhood. However its frequency, expression, and the contexts in which it occurs are influenced by culture. Although one could argue that the differences found in children's play in the current study were not cultural but rather a result of situational constraints, we suggest that the school settings were extensions of the broader cultures and that the children bring different preferences, prior understandings, and skills to the common experimental situation. We base this argument on the notion that all settings, situations, or activities are culturally bound in that they are always determined by the meanings inherent in the culture of the participants (Rogoff, 1982). In this sense, "neither children nor their families experience the environment directly, but only as mediated by the local categories, beliefs, and norms that reflect their cultural traditions and guide the design of the man-made environments in which children are raised" (LeVine, 1991, p. 59).

These findings concur with Tobin, Wu, and Davidson's (1989) research, which illustrated how facets of Japanese, Chinese, and American preschool settings are linked to cultural contexts and, particularly, to historical values and contemporary societal needs. In the Japanese setting, where a sense of group identity and a concern for others were emphasized, Japanese teachers tended to downplay individual comparison and establish cohesion by involving children in large group activities. In the Chinese preschool, where the development of citizenship, discipline, and task perseverance were considered important, children were involved in very similar activities under the direction of the teacher.

Overall, the current study suggests that sociocul-tural variations in the contexts in which children develop and interact are associated with children's social behavior, their self-expression, and how they experience and coordinate play with a partner. Nevertheless, in interpreting these results it should be noted that observations and videotaped play sequences were only 20 min each. Also, the sample size was small and limited to middle-class children and their families. Children's play behavior, parental beliefs, and cultural values may be different among other socioeconomic groups (McLoyd, 1982; Tudge, Lee, & Putman, 1995) and among families who have different experiences with the immigration process and varying levels of acculturation.

Also, it is not the intention here to "culturally stereotype" either group. Scores on the acculturation measure and informal observations made by the second author, who is familiar with her own Korean culture, show that the Korean-American families and

the preschool staff were very traditional in their lifestyles, values, and behavior. Thus, the contrasts between the Korean- and Anglo-American children found in the current study may appear quite dramatic to a reader who is relatively unfamiliar with Korean culture. Also, there is much variation within any cultural group, and the Korean-American participants in the current study may be viewed as being more similar to Koreans in Korea than to Korean-American immigrants in the United States. Part of this distinction may be due to the fact that most, if not all of the Korean-American families studied here remained relatively isolated from the rest of the city in a Koreatown neighborhood. In any case, the participants discussed here represent only two examples of Korean- and Anglo-American groups. Therefore, the results should be replicated with other samples.

ACKNOWLEDGMENTS

The authors are grateful to the children who participated in this study, their parents, and the preschool staff.

ADDRESSES AND AFFILIATIONS

Corresponding author: J. M. Farver, Department of Psychology, SGM 501, University of Southern California, Los Angeles, CA 90089-1061; e-mail: Farver@-rcf.usc.edu. Yoolim Lee Shin is at the Samsung Welfare Foundation, Early Childhood and Development Center, Seoul.

REFERENCES

Black, B. (1989). Interactive pretense: Social and symbolic skills in preschool groups. Merrill-Palmer Quarterly, 35, 379-397.

Bretherton, I. (Ed.). (1984). Symbolic play. New York: Academic Press.

Bruner, J. (1990). Culture and human development: A new look. Human Development, 33, 344-355.

Choi, S. C, Kim, U., & Choi, S. H. (1993). Korean culture and collective representation. In U. Kim & J. W. Berry (Eds.), Indigenous psychologies: Experience and research in cultural context. Newbury Park, CA: Sage.

Chu, H. (1978). The Korean learner in an American school. In Teaching for cross-cultural understanding. Arlington, VA: Arlington Public Schools.

Connolly, J. A., Doyle, А. В., & Reznick, E. (1988). Social pretend play and social interaction in preschoolers. Journal of Applied Developmental Psychology, 9, 301-311.

Corsaro, W. A. (1985). Friendship and peer culture in the early years. Norwood, NJ: Ablex.

Corsaro, W. A. (1986). Discourse processes within peer cul-

Farver and Shin 555

ture: From a constructivist to an interpretative approach to childhood socialization. In P. Adler (Ed.), Play: Anthropological perspectives. New York: JAI Press.

Cuellar, I., Harris, L. C, & Jasso, R. (1980). An acculturation scale for normal Mexican-American and clinical populations. Hispanic Journal of Behavioral Sciences, 20,199-217.

Dunn, J., & Dale, N. (1984). I a daddy: Two-year olds' collaboration in joint pretend with a sibling and with a mother. In I. Bretherton (Ed.), Symbolic play. New York: Academic Press.

Farver, J. M. (1992a). An analysis of young American and Mexican children's play dialogues. In C. Howes, O. Un-ger, & C. Matheson (Eds.), The collaborative construction of pretend. New York: SUNY Press.

Farver, J. M. (1992b). Communicating shared meaning in social pretend play. Early Childhood Research Quarterly, 7, 1493-1503.

Farver, J. M. (1993). Cultural differences in scaffolding pretend play: A comparison of American and Mexican mother-child and sibling-child dyads. In K. MacDonald (Ed.), Parent child play: Descriptions and implications. New York: SUNY Press.

Farver, J. M., & Howes, С (1993). Cultural differences in American and Mexican mother-child pretend play. Merrill-Palmer Quarterly, 39, 344-358.

Farver, J. M., Kim, Y. K., & Lee, Y. (1995). Cultural differences in Korean- and Anglo-American preschoolers' social interaction and play behaviors. Child Development, 66, 1088-1099.

Farver, J. M., & Wimbarti, S. (1995). Indonesian toddlers' social play with their mothers and older siblings. Child Development, 66, 501-516.

Fein, G., & Rivkin, M. (Eds.). (1986). The young child at play. Washington, DC: National Association for the Education of the Young Child.

Forys, S., & McCune-Nicolich, L. (1984). Shared pretend: Sociodramatic play at three years of age. In I. Bretherton (Ed.), Symbolic play. New York: Academic Press.

Garvey, С (1984). Children's talk. Oxford: Fontana.

Garvey, С (1990). Play. Cambridge, MA: Harvard University Press.

Garvey, C, & Brendt, R. (1977). The organization of pretend play. Paper presented at the meeting of the American Psychological Association, Chicago.

Gaskins, S. (1996). How Mayan parental theories come into play. In S. Harkness & C. Super (Eds.), Parents' cultural belief systems. New York: Guilford.

Giffin, H. (1984). The coordination of meaning in the creation of shared make-believe reality. In I. Bretherton (Ed.), Symbolic play (pp. 73-100). New York: Academic Press.

Glaser, В., & Strauss, A. (1967). The discovery of grounded theory. New York: Aldine.

Goncu, A. (1989). Models and features of pretense. Developmental Review, 4, 341-344.

Goncu, A., & Kessel, F. (1984). Children's play: A contextual-functional perspective. In F. S. Kessel & A. Goncu (Eds.), Analyzing children's play dialogues (New Directions for Child Development, Vol. 25). San Francisco: Jossey-Bass.

Goncu, A., & Kessel, F. (1988). Preschoolers' collaborative construction in planning and maintaining imaginative play. International Journal of Behavioral Development, 11, 327-344.

Goncu, A., Mistry, J., & Mosier, С (1991). Cultural variations in toddlers' play. In L. Beizer & P. Miller (Chairs), Cultural dimensions of pretend play in infancy and early childhood. Symposium conducted at the meeting of the Society for Research in Child Development, Seattle.

Haight, W., & Miller, P. (1993). Pretending at home. New York: SUNY Press.

Hoffman, L. (1988). Cross-cultural differences in childrear-ing goals. In R. Levine, P. Miller, & M. Maxwell (Eds.), Parenting behaviors in diverse societies: New directions for child development. San Francisco: Freeman.

Howes, C. (1988). Peer interaction of young children. Monographs for the Society of Research in Child Development, 53(1, Serial No. 217).

Howes, C, Unger, O., & Matheson, С (Eds.). (1991). The collaborative construction of pretend. New York: SUNY Press.

Kahng, T. J. (1990). Identity into unity: Some speculation on patterns of assimilation and education. Korean Observer, 21,1-11.

Kim, J. U. (1991). The Koreans: Their mind and behavior. Seoul: Kyobo Centre Co.

Kim, S. J. (1995). The effects of parenting style, cultural conflict, and peer relations on academic achievement and psychological adjustment among Korean immigrant adolescents. Unpublished dissertation, University of Southern California, Los Angeles.

Kim, U., & Choi, S. H. (1994). Individualism, collectivism, and child development: A Korean perspective. In P. Greenfield & R. Cocking (Eds.), Cross-cultural roots of minority child development. Hillsdale, NJ: Erlbaum.

Lee, D. C, & Lee, E. H. (1990). Korean immigrant families in America: Role and value conflict. Korean Observer, 22, 31-42.

Lee, E. H., & Lee, K. W. (1987). The Korean mother's socialization process for children (in Korean). Seoul: Ehwa Women's University.

Lee, S. B. (1989). Childrearing practices of Korean mothers in the United States in relation to acculturation: A study of ethnic persistence. Dissertation Abstracts International, 50, 11 A.

LeVine, R. (1980). Anthropology and child development: New directions for child development. San Francisco: Jossey-Bass.

LeVine, R. (1991). Cultural environments in child development. In W. Damon (Ed.), Child development today and tomorrow. New York: Aldine.

Maday, B. C, & Szalay, L. B. (1976). Psychological correlates of family socialization in the United States and Korea. In T. Williams (Ed.), Psychological anthropology. The Hague: Mouton.

Markus, H. R., & Kitayama, S. (1991). Culture and the self: Implications for cognition, emotion, and motivation. Psychological Review, 98, 224-253.

Martini, M. (1994). Peer interactions in Polynesia: A view

556 Child Development

from the Marquesas. In J. Roopnarine, J. Johnson, & F. Hooper (Eds.), Children's play in diverse cultures. New York: SUNY Press.

McLoyd, V. (1982). Social class differences in sociodramatic play: A critical review. Developmental Review, 2, 1-30.

Min, P. G. (1988). The Korean family. In С. Н. Mindel & R. W. Habenstein (Eds.), Ethnic families in America: Patterns and variations New York: Elsevier.

Nelson, K., & Seidman, S. (1984). Playing with scripts. In I. Bretherton (Ed.), Symbolic play. New York: Academic Press.

Owens, R. E. (1990). Development of communication, language, and speech. In G. Shames & E. Wigg (Eds.), Human communication disorders. Columbus, OH: Merrill.

Paley, V. (1986). Boys and girls: Super heroes in the doll corner. Chicago: University of Chicago Press.

Rogoff, B. (1982). Integrating context and cognitive development. In M. E. Lamb & A. L. Brown (Eds.), Advances in developmental psychology (vol. 2). Hillsdale, NJ: Erlbaum.

Rogoff, В., Goncu, A., Mistry, J., & Mosier, С (1991). Cultural variations in role relations in socialization of toddlers. In M. Bornstein (Ed.), Cultural approaches to parenting. Hillsdale, NJ: Erlbaum.

Rohner, R. P., & Pettengill, S. M. (1985). Perceived parental acceptance / rejection and parental control among Korean adolescents. Child Development, 56, 524-528.

Roopnarine, J., Hossain, Z., Gill, P., & Brophy, H. (1994). Play in the East Indian context. In J. Roopnarine, J. Johnson, & F. Hooper, (Eds.), Children's play in diverse cultures. New York: SUNY Press.

Roopnarine, J., Johnson, J., & Hooper, F. (1994). Children's play in diverse cultures. New York: SUNY Press.

Rosenberg, D. M. (1985). Fantasy play and socio-emotional development. Paper presented at the Society for Research in Child Development, Toronto.

Rubin, K. H., Fein, G. G., & Vandenberg, B. (1983). Play. In E. M. Hetherington (Ed.), P. H. Mussen (Series Ed.), Handbook of child psychology: Vol. 4. Socialization, personality, and social development. New York: Wiley.

Sachs, J., Goldman, J., & Chaille, С (1984). Planning in pre-

tend play: Using language to coordinate narrative development. In A. Pelligrini & T. Yawkey (Eds.), The development of oral and written language in social context. Norwalk, NJ: Ablex.

Saltz, E., Dixon, D., & Johnson, J. (1977). Training disadvan-taged preschoolers on various fantasy activities: Effects on cognitive functioning and impulse control. Child Development, 43, 367-380.

Schiefflin, В., & Ochs, E. (1986). Language socialization across cultures. Cambridge, MA: Cambridge University Press.

Schwartzman, H. (1978). Transformations: The anthropology of children's play. New York: Plenum Press.

Shatz, M., & Gelman, R. (1973). The development of communication skills: Modification in the speech of young children as a function of the listener. Monographs of the Society for Research in Child Development, 38(Serial No. 152).

Strom, R., Park, S. H., & Daniels, S. (1986). Adjustment of Korean immigrant families. Educational and Psychological Adjustment Research, 6, 312-327.

Super, C, & Harkness, S. (1986). The developmental niche: A conceptualization at the interface of child and culture. International Journal of Behavioral Development, 9, 1-25.

Tobin, J., Wu, D., & Davidson, D. (1989). Preschool in three cultures: Japan, China, and the United States. New Haven, CT: Yale University Press.

Tomasello, M., & Manle, S. (1985). Pragmatics of sibling speech to one-year-olds. Child Development, 56, 911-917.

Tudge, J., Lee, S., & Putman, S. (1995). Young children's play in socio-cultural context: Examples from South Korea and North America. In A. Goncu & A. Nicolopoulou (Chairs), The pretend play of cultures: Cultures of pretend play. Symposium conducted at the biennial meeting of the Society for Research in Child Development, Indianapolis.

Whiting, В., & Edwards, С. Р. (1988). Children of different worlds: The formation of social behavior. Cambridge, MA: Harvard University Press.

Yu, E. E., Phillips, E., & Yang, E. (Eds.). (1981). Koreans in Los Angeles. Los Angeles: Koryo Research Institute.

Оставить комментарий © Copyright Farver М. and Yoolim Lee Shin (kazgugnk@yahoo.com) Обновлено: 08/09/2011. 142k. Статистика. Статья: США		Ваша оценка:

Farver М. and Yoolim Lee Shin: другие произведения.

Social Pretend Play in Korean- and Anglo-American Preschoolers