Article Review

388

JOURNAL OF BLACK PSYCHOLOGY, Vol. 35 No. 3, August 2009 388-403 DOI: 10.1177/0095798409333621 © 2009 The Association of Black Psychologists

Reexamination of Young Children’s Racial Attitudes and Skin Tone Preferences

Phillip Jordan Maria Hernandez-Reif University of Alabama

This study continues the line of research on children’s racial preferences that dates to Kenneth and Mamie Clark’s classic research that revealed that Black children preferred White dolls and attributed more positive characteristics to White dolls than to Black dolls. In the current research, the authors examined Black and White preschool children’s preferences for cartoon characters of different skin tones and whether their preferences shifted after listening to a moral story depicting a Black child as a hero. Findings are discussed in the context of how method of presentation may affect young children’s responses and children’s preferences in relation to race awareness and self-identification.

Keywords: preschool children; skin tone; racial preferences; attitudes

More than 65 years ago, Kenneth and Mamie Clark conducted a landmark study using White and Black dolls to examine Black children’s view of skin tones (Clark & Clark, 1939, 1947). Specifically, the Clark and Clark paradigm, referred to as the “Doll Study,” measured young Black children’s preferences,

AUTHORS’ NOTE: We thank the preschool children who participated in the study, their parents, teachers, and the preschool directors. Additionally, we thank Dr. Nancy Campbell for editing comments on the manuscript, Dr. Jason Scofield for assistance with cartoon software, Jennifer Rice who sketched the cartoons, and Hylan Noble and Maggie Porter who assisted with data collection. This study was supported by a scholarship to Phillip Jordan from the Ronald McNair Post-Baccalaureate Achievement Program. Address correspondence to: Maria Hernandez-Reif, University of Alabama, Department of Human Development and Family Studies, 223 Child Development Research Center, Box 870160, Tuscaloosa, AL 35487-0160; e-mail: [email protected].

EDITOR’S NOTE: This article was accepted under the editorship of Shawn O. Utsey.

at TEMPLE UNIV on December 2, 2016jbp.sagepub.comDownloaded from

Jordan, Hernandez-Reif / RACIAL ATTITUDES OF YOUNG CHILDREN 389

racial awareness, and racial self-identification. The Clarks showed young Black children a Black doll and a White doll and asked the children to

give me the doll that: (1) you like to play with or the doll you like best, (2) is the nice doll, (3) looks bad, (4) is a nice color, (5) looks like a White child, (6) looks like a colored child, (7) looks like a Negro child, (8) looks like you.

Questions 1 through 4 were designed to disclose preferences; Questions 5 through 7 examined children’s knowledge of differences between races; and Question 8 probed children’s self-identification (Clark, & Clark, 1939, 1947). The results of the Clark and Clark (1947) study revealed that 67% of Black children preferred to play with White dolls, 59% chose the White doll as the nice doll, and 60% chose the White doll as having a nice color. Additionally, 59% chose the Black doll as being the one that “looks bad.” Interestingly, overall only 58% of Black children selected the Black doll as the one that “looks like you.”

Closer examination of the data suggested that self-identification with the Black doll was related to the Black children’s own skin tone, with the dis- tribution for choosing the Black doll as looking like them being only 20% for light Black children, 73% for medium skin tone Black children, and 81% for dark skin tone Black children. When measuring self-identification (Question 8) in relation to age, only 61% of Black 3-year-old children chose the Black doll as the one that looked like them. In contrast, 87% of 7-year-old Black children accurately chose the Black doll as the one that looked like them.

The overall results of the studies conducted by the Clarks illustrated that young Black children raised in the 1930s preferred White dolls and judged the White dolls as superior to duplicate dolls of Black skin color. Replication studies in the decades that followed revealed that White children identified with their skin tone more often than Black children (Goodman, 1952). In contrast, Black children were inclined to reject their own ethnic group and had greater preferences for White skin tone (Clark & Clark, 1947; Greenwald & Oppenheim, 1968; Lewis & Biber, 1951; Morland, 1962, 1966). Studies also revealed that Black children were persuaded by the majority norms, values, and pessimistic judgments about their racial or ethnic group (Clark, 1955; Stevenson & Stewart, 1958). The studies by Clark and Clark and oth- ers influenced the state of segregation in the United States and ultimately contributed to the Brown v. Board of Education (1954) decision to eradicate racial segregation in schools and other public facilities.

In a recent documentary, Davis (2006), a young high school filmmaker, examined young Black children’s self-perception and preferences for skin

at TEMPLE UNIV on December 2, 2016jbp.sagepub.comDownloaded from

390 JOURNAL OF BLACK PSYCHOLOGY / AUGUST 2009

tones. The documentary included videotaped clips of young Black children sitting in front of a table with identically clothed Black and White baby dolls. The children were asked questions modeled after those posed by Clark and Clark in their 1939 study, such as “Show me the doll ‘that you like best,’ ‘that is the nice doll,’ and ‘that looks bad.’ According to the filmmaker, 15 out of the 21 children in the documentary preferred the White doll. No statistics were provided on the children’s responses to the other two ques- tions, although a number of children were shown selecting the White doll as “the nice doll” and the Black doll as the one “that looks bad.”

The Davis (2006) documentary generated renewed interest in young children’s skin tone preferences because it suggested that several genera- tions later Black U.S. children continue to have negative views about their skin tone and show strong preferences for White over Black skin tones. Because racial discrimination is detrimental to both White and Black chil- dren’s self-concept, it is important to continue to study young children’s views on skin tones and develop paradigms that attenuate children’s nega- tive perception of Black skin tone. In a pre-post study, researchers exam- ined whether Black and White children’s skin tone preferences could be altered (Powell-Hopson & Hopson, 1992). Specifically, researchers modeled pro-Black positive behaviors when responding to the Clarks’ racial prefer- ence questions. They also read a story depicting a Black child in a positive manner to the children in the study. In the pretest, only 35% of the Black children and 25% of the White children showed a preference for the Black doll. However, following the researchers’ modeling and positive storytell- ing, approximately 67% of the Black and White children showed a prefer- ence for the Black doll when responding to the Clarks’ racial preference questions. This finding suggests that preschool children’s racial preferences may be altered with behavioral techniques and by presenting favorable stories of Black children.

The present study replicated, and extended, prior research on young Black and White children’s skin tone preferences using an experimental design. Extension on prior research included (1) modernizing the stimuli by using computer-generated cartoon characters as opposed to dolls; (2) using several skin tone variations, as opposed to only Black and White skin tones; and (3) assessing whether children’s preferences shifted from their baseline responses after listening to a story of a Black cartoon character saving a baby duck, as in the Powell-Hopson and Hopson (1992) study. In addition to asking children to choose between Black and White cartoons, they were also told that they could choose to say “neither” or “both” in response to the experimenter’s questions. The participants tested were preschool children between 3 and 5 years of age. The lower age range was chosen because as early as age 3 the knowledge of racial differences becomes conspicuous

at TEMPLE UNIV on December 2, 2016jbp.sagepub.comDownloaded from

Jordan, Hernandez-Reif / RACIAL ATTITUDES OF YOUNG CHILDREN 391

and children appear to differentiate skin color as well as other physical characteristics (Clark & Clark, 1947; Goodman, 1952; Porter, 1971; Stevenson & Stewart, 1958). If skin tone preferences and attitudes persist, young Black children were expected in the current study to prefer and report more posi- tive attitudes for the White over the Black cartoons, and overall, a greater number of Black children were expected to self-identify with the White cartoon than the Black cartoon. Additionally, White preschool children were expected to show preferences and correctly identify with those within their race, that is, self-identify with the White cartoons. We also hypothe- sized that after listening to a story in which a Black child saves a baby duck Black and White preschool children would attribute more positive charac- teristics to the Black cartoon and that Black children would identify more with those within their race. However, we expected White children to consistently identify with their own skin color.

METHOD

PARTICIPANTS

Parents with children between 3 and 5 years of age attending area pre- schools were invited to participate. Those who consented were asked to complete a background demographic questionnaire on their child, includ- ing their child’s age, gender, ethnic background, and skin tone (white, light brown, medium brown, or black). Additionally, parents were asked their occupation and education to determine their socioeconomic status based on the Hollingshead two-factor index (Hollingshead, 1975).

The sample consisted of 40 Black and White preschool children (M = 53.0 months; SD = 8.1) recruited from preschool programs and day cares in Tuscaloosa and Huntsville, Alabama. These two cities were chosen because they represent urban progressive cities in the state of Alabama and are where the authors of the study resided. The participants included 10 male and 10 female Black children (M = 54.0 months) and 9 male and 11 female White children (M = 52.0 months). The majority of the children (Black = 70%; White = 60%) were from middle socioeconomic families. The study had the university’s institutional review board approval as well as parental written consent and children’s verbal assent.

The children were randomly assigned to either a moral or a nonmoral story (control) group. Children assigned to the moral group were told a short story about a little Black child saving a baby duck (moral story group), and those assigned to the nonmoral story group were told a story about a child and a baby

at TEMPLE UNIV on December 2, 2016jbp.sagepub.comDownloaded from

392 JOURNAL OF BLACK PSYCHOLOGY / AUGUST 2009

duck without reference to the skin tone of the character child in the story or reference to saving the duck (control group). The purpose of the group assign- ment was to determine if the moral script would influence children’s responses to skin tone preferences and attitudes from pre- to posttest. The children were individually tested, and after completing the study, they received a sticker in appreciation of their participation.

Precautions were taken to control for experimenter bias and demand characteristics. First, prior to starting the recruitment and testing phase of the study, a session was held with the experimenters to discuss the importance of being mindful not to use inflection or gestures that might influence the children’s responses. The experimenters practiced posing each question in a nonjudgmental, monotone manner prior to the start of the study. In addition, each child was told at the beginning of the testing session and throughout the testing that there was no right or wrong answer and that they could respond to any question by saying “neither” or “both.”

MATERIALS

Cartoons were sketched by hand and redrawn using Macromedia Flash Player on a Dell Inspiron E 1705 laptop computer. For the first phase of the study, one unique cartoon body was drawn to compose four cartoons (see Figure 1). The type of clothing (pants and T-shirts), color of the cloth- ing, and hairstyle (short pigtails for female cartoons and short-cropped hair for male cartoons) drawn on the four cartoons were identical. The cartoons differed only in skin tone—white, light brown, medium brown, and black— and were presented in a random order across the screen.

For the second phase of the study, which consisted of a pre-post test, eight additional cartoons (four boys and four girls) were created and redrawn on Macromedia Flash Player. These drawings also depicted young cartoon children with vivid expressions, dressed in identical T-shirts and jeans (male) or skirts (female). The children were shown two cartoons of their gender (i.e., boys were presented with male cartoons and girls were presented with female cartoons) for the pretest, and then shown two new, but again same-gender, cartoons for the posttest. For each test, one cartoon was Black with dark hair and the other was White with light hair. The car- toons were identical in every aspect, except for their skin tone (one Black and one White in each pair) and dress, which changed from the pretest to the posttest. The cartoons were presented to the children on a 17-inch Dell laptop using Microsoft PowerPoint 2002. The side of the stimulus presenta- tion was counterbalanced on the first trial so that for half of the children the Black cartoon was on the right side and for the other half of the group the

at TEMPLE UNIV on December 2, 2016jbp.sagepub.comDownloaded from

Jordan, Hernandez-Reif / RACIAL ATTITUDES OF YOUNG CHILDREN 393

Black cartoon was on the left side of the screen. For the remaining trials, the side of presentation of the Black cartoon was randomly assigned to the right or the left side.

PROCEDURE

Following parental consent, the children were tested individually by a researcher. The children were first asked for permission by the examiners to ask them questions about cartoons on a laptop (the assent process). Only one child did not initially assent to participating because he was preparing to have a snack. However, the child assented later in the day to participating and was tested. The participants were tested in a research room or a desig- nated area within their preschool. The testing lasted approximately 10 to 12 minutes and consisted of (1) a skin tone variation preference question, (2) a

Figure 1: Samples of Cartoons Used for the Ranking Study

at TEMPLE UNIV on December 2, 2016jbp.sagepub.comDownloaded from

394 JOURNAL OF BLACK PSYCHOLOGY / AUGUST 2009

pretest, (3) an intervention component (telling of a moral story or a non- moral control story), and (4) a posttest.

Skin tone variation preference question. The examiner presented four identical cartoon characters that varied only in skin tones: white, light brown, brown, and black. The cartoons’ skin colors were randomly paired across the monitor. The examiner asked the child to point to or show the cartoon who the child would like as a “best friend.” This was taken as an indication of the child’s skin tone preference.

Pretest. Children were shown two cartoons of the same gender as the child participant. The cartoons were identical, with the exception that one cartoon was White and one Black. The cartoons were paired side by side, and the side on which the Black or White cartoon was presented was counterbalanced across the participants. The child was asked to point to or show the cartoon on the laptop monitor in relation to the following questions: Who (1) you would like as a “playmate or best friend,” (2) “looks nice,” (3) “looks bad,” (4) has a “nice skin color,” (5) “looks like a White child,” (6) “looks like a Black child,” and (7) “looks like you” (Clark & Clark, 1939, 1947). Immediately follow- ing each question, and before the children responded, they were told by the experimenter that they could also say “none” or “both.”

Intervention component. For the intervention phase, which followed the pretest, half of the children were told a positive moral story about a Black child who saves a baby duck, and the other half of the children were told a story about a child who sees a baby duck, without making reference to the child’s skin tone color or the moral character of the child. The purpose of the intervention was to determine if the moral script influenced children’s racial preferences and attitudes in a posttest.

Posttest. The posttest was identical to the pretest, except that different cartoons were shown, as already described. As in the pretest, the children were told that they could respond to Questions 1 through 7 choosing Black, White, “none” (i.e., neither Black nor White), and “both” Black and White.

RESULTS

The study is a randomized, mixed-model design, with one fixed factor (children’s race) and one random factor (story assignment). Nonparametric

at TEMPLE UNIV on December 2, 2016jbp.sagepub.comDownloaded from

Jordan, Hernandez-Reif / RACIAL ATTITUDES OF YOUNG CHILDREN 395

test statistics were used because of the categorical or nominal nature of the data, except for the analyses on some of the data on the background questionnaire that were continuous (ages) and required parametric statistical testing.

BACKGROUND QUESTIONNAIRE

The two groups of children (Black and White) were similar in age (M = 53 months), and comparable numbers of male and female children participated in both groups (all ps > .05). The mothers were about 34 years old and the fathers averaged 37.5 years of age. The two groups of children did not differ on parents’ age or socioeconomic status (all ps > .05).

SKIN TONE VARIATION PREFERENCE QUESTION

Chi-square analyses on the range of the four cartoons’ skin tones revealed no differences between Black and White children’s responses, χ2 = 2.60, p > .05, or differences between boys’ and girls’ responses, χ2 = 3.15, p > .05, on the skin tone of the cartoon they chose for their “best friend.” In addition, no single cartoon was preferred over any other cartoon, irrespec- tive of children’s race, χ2 = 2.40, p > .05.

MALE VERSUS FEMALE CHILDREN’S RESPONSES TO THE SEVEN QUESTIONS

Chi-square gender analyses on the responses to the seven pretest ques- tions on the two cartoons revealed only a trend for the question on which cartoon “looks Black,” with more boys (89%) than girls (67%) identifying the Black cartoon as the one that looks Black, χ2 = 2.98, p = .08.

PRETEST (BASELINE) RESPONSES

The pretest data for the seven questions were first analyzed by group assignment (moral story, control) and then separately analyzed by children’s race (Black, White) to examine baseline differences. The analyses revealed no significant pretest differences according to group assignment (all ps > .05). However, significant baseline differences were detected when the data were analyzed by race (Black vs. White children’s responses), including that (1) more Black (91%) than White children (65%) accurately identified the Black cartoon as the one that “looks Black,” χ2 = 3.58, p = .058; (2) more Black (75%) than White children (15%) selected the Black cartoon as the one

at TEMPLE UNIV on December 2, 2016jbp.sagepub.comDownloaded from

396 JOURNAL OF BLACK PSYCHOLOGY / AUGUST 2009

that “looks like you,” χ2 = 14.76, p < .01; and (3) more White (80%) than Black children (25%) selected the White cartoon as the one that “looks like you,” χ2 = 14.76, p < .01.

PRE-POST TEST ANALYSES

Nonparametric analyses of the Black versus the White children’s responses to the seven questions before versus after the storytelling (see Table 1) revealed the following: (1) for the “best friend” question, a trend for the Black children in the control group, with more Black children (56%) consistently selecting the White cartoon over the Black cartoon (44%), χ2 = 2.72, p = .099; (2) for the “looks nice” question, a trend for the White children assigned to the moral story condition, with more White children selecting the White cartoon (50%) than the Black cartoon (36%), χ2 = 8.02, p = .091; (3) for the “looks bad” question, significant differences pre-post for both the Black and White children assigned to the moral story group, with significantly fewer Black children selecting the Black cartoon after hearing the story (27% vs. 36%), χ2 = 10.63, p < .05. Surprisingly, more White children selected the Black cartoon (55%) than the White cartoon (27%) as the one that “looks bad” after hearing the story of the little Black boy who saved a duck, χ2 = 24.75, p < .01; (4) for the “nice skin color” question, significant differences in the pre-post responses of the White children assigned to the moral story group, with more White children selecting the White cartoon (82%) after versus before the story (62%), χ2 = 4.28, p < .05; (5) for the “looks White” question, significant differ- ences in the pre-post responses of the White children in the moral story group, with more White children selecting the White cartoon after the sto- rytelling (82%) than before the storytelling (64%), χ2 = 4.82, p < .05; (6) for the “looks like a Black child,” no significant differences were found; and (7) for the “looks like you” question, significant differences in the responses of the White and the Black children, with significantly fewer White children assigned to the moral story selecting the White cartoon (73%) after the story compared with before (82%) the story, χ2 =15.89, p < .01, and significantly fewer Black children in the control group selecting the Black cartoon after the story (56%) compared with before the story (67%), χ2 = 5.63, p < .05. White children in the control group also identified sig- nificantly more with the White cartoon (78%) than with the Black cartoon (22%), χ2 = 9.00, p < .01.

Secondary analyses conducted to examine preferences and attitudes for the Black children by their skin color (light brown skin tone vs. medium- dark brown skin tone) revealed differences only for the question of who

at TEMPLE UNIV on December 2, 2016jbp.sagepub.comDownloaded from

397

T A

B L

E 1

C h

il d

re n

’s P

re -P

os t

T es

t R

es p

on se

s b

y C

h il

d re

n ’s

R ac

e an

d M

or al

S to

ry A

ss ig

n m

en t

(M or

al S

to ry

V

er su

s C

on tr

ol G

ro u

p )

W h

en S

h ow

n a

B la

ck a

n d

a W

h it

e C

ar to

on P

ai re

d S

id e

b y

S id

e

B

la ck

C hi

ld re

n W

hi te

C hi

ld re

n

M

or al

S to

ry

C on

tr ol

M

or al

S to

ry

C on

tr ol

P

re te

st

P os

tt es

t To

ta l

P re

te st

P

os tt

es t

To ta

l P

re te

st

P os

tt es

t To

ta l

P re

te st

P

os t

To ta

l W

hi ch

C ar

to on

(%

) (%

) (%

) (%

) (%

) (%

) (%

) (%

) (%

) (%

) (%

) (%

)

Q 1.

W ou

ld b

e yo

ur b

es t f

ri en

d B

la ck

55

45

50

44

44

44

* 55

18

37

22

22

22

W hi

te

45

55

50

56

56

56

36

73

54

67

78

72 .5

N ei

th er

11

5.

5 B

ot h

9 9

9

Q

2. L

oo ks

n ic

e B

la ck

36

45

40

67

67

67

36

36

36

* 44

67

56

W hi

te

64

55

60

33

33

33

55

46

50

56

33

44 N

ei th

er

B

ot h

9 18

14

Q

3. L

oo ks

b ad

B

la ck

36

27

32

**

56

44

50

55

55

55 **

* 67

33

50

W hi

te

27

46

36

11

33

22

27

27

27

11

56

34 N

ei th

er

36

27

32

33

22

28

9 9

9 22

11

16

B ot

h

9

9 9

Q 4.

H as

n ic

e sk

in c

ol or

B la

ck

18

82

50

44

44

44

36

18

27 **

56

33

44

.5 W

hi te

82

18

50

56

56

56

64

82

73

44

56

50

(c on

ti nu

ed )

at TEMPLE UNIV on December 2, 2016jbp.sagepub.comDownloaded from

398

T A

B L

E 1

(c

on ti

n u

ed )

B

la ck

C hi

ld re

n W

hi te

C hi

ld re

n

M

or al

S to

ry

C on

tr ol

M

or al

S to

ry

C on

tr ol

P

re te

st

P os

tt es

t To

ta l

P re

te st

P

os tt

es t

To ta

l P

re te

st

P os

tt es

t To

ta l

P re

te st

P

os t

To ta

l W

hi ch

C ar

to on

(%

) (%

) (%

) (%

) (%

) (%

) (%

) (%

) (%

) (%

) (%

) (%

)

N ei

th er

11

5.

5 B

ot h

Q

5. L

oo ks

W hi

te

B la

ck

18

9

36

18

27

**

22

22

W hi

te

82

10 0

91

10 0

90

95

64

82

73

78

78

78 N

ei th

er

11

5

B ot

h

Q 6.

L oo

ks B

la ck

B

la ck

82

10

0 91

10

0 10

0 10

0 64

10

0 82

67

89

78

W hi

te

18

9

36

18

33

11

22 N

ei th

er

B

ot h

Q

7. L

oo ks

li ke

y ou

B

la ck

82

82

82

67

56

62

**

9 9

9* **

22

22

22

** *

W hi

te

18

18

18

33

44

38

82

73

77

78

78

78 N

ei th

er

9 18

14

B

ot h

*p <

.1 0.

* *p

< .0

5. *

** p

< .0

1.

at TEMPLE UNIV on December 2, 2016jbp.sagepub.comDownloaded from

Jordan, Hernandez-Reif / RACIAL ATTITUDES OF YOUNG CHILDREN 399

“looks like you.” Specifically, on the pretest, 65% of light brown skin tone children (vs. 100% of light brown skin tone children on the posttest) chose the Black cartoon as the one that “looks like you,” χ2 = 9.59, p = .002.

DISCUSSION

For more than half a century, American studies have revealed that when presented with Black and White dolls, young Black children attribute more positive qualities to White dolls and more negative attributes to Black dolls (Clark & Clark, 1939, 1947; Davis, 2006; Greenwald & Oppenheim, 1968; Powell-Hopson & Hopson, 1992). Studies have ruled out that socioeco- nomic status plays a role in children’s preferences and attitudes toward skin tones (Gopaul-McNicol, 1995; Landreth & Johnson, 1953; Morland, 1966; Porter, 1971). These findings have been interpreted as Black U.S. children having negative views about their skin tone and showing strong preferences for White over Black skin tones.

A critique of studies in Black identity (1936-1967), however, suggests that one should be cautious in interpreting previous research as evidence that Black children have feelings of shame or self-hatred because of flaws in the design of studies (Cross, 1991). In particular, concerns have been raised about the internal and external validity of previous research, mostly related to demand characteristics and experimenter bias (Banks, 1976; Rosenberg, 1989). The present study reexamined the issue of young chil- dren’s skin tone preferences by testing both Black and White preschool children and taking precautions to control for demand characteristics and experimenter bias. Controls for demand characteristics in the present study included informing children that there were no right or wrong answers; in addition, children were told that they could say “none” or “both” in response to any question. To control for experimenter bias, the experimenters were trained to deliberately use monotone speech and refrain from making gestures or comments that could influence the children’s responses. Although the current study had controls for potential design flaws identified in the litera- ture, the issue of whether Black children experience shame or feelings of self-hatred remains an issue that has not been resolved, particularly because of studies suggesting a relationship between adolescent skin color and self-esteem (Robinson & Ward, 1995).

Irrespective of the potential that previous study designs were flawed, one objective of the present study was to examine if skin tone preferences were still evident among preschool children. Based on the past 50 years of

at TEMPLE UNIV on December 2, 2016jbp.sagepub.comDownloaded from

400 JOURNAL OF BLACK PSYCHOLOGY / AUGUST 2009

research, and a recent documentary (Davis, 2006), preschool children were expected to continue to demonstrate favoring White skin tone. Another objective of the present study was to examine the effects of a brief vignette about a good deed performed by a Black character as a possible pre-post paradigm to use with young children to promote positive views about Black skin tones. At least one study found that reinforcing and modeling positive behaviors toward a Black doll and presenting positive color-word associa- tions (e.g., “the Black doll is beautiful just like you”) served as a positive catalyst for changing children’s preferences and attitudes toward skin tone (Powell-Hopson & Hopson, 1992).

In the current study, children were questioned about their skin tone pref- erences using cartoon stimuli that ranged in skin tones from white, light, medium, to black skin tone. In addition, children’s preferences and attitudes about skin tones were assessed using the conventional pairing of White and Black stimuli. Moreover, as already discussed, in the current study children also had the option of indicating a preference for “both” or “neither” skin tones, as opposed to the traditional forced-choice Clark doll technique that required that children choose between a Black doll and a White doll. In mod- ernizing the study, computerized cartoons, as opposed to dolls, were con- structed for the stimuli because the current generation of young children is more familiar with computer-generated images than dolls and may respond better to cartoons.

The most revealing finding of the current study was that children showed no skin tone preference when shown a variety of skin tone cartoons and asked who they would pick as their “best friend.” This discrepant find- ing might relate to the children having more choices because of the greater skin tone variations of the cartoon characters presented to them (i.e., white, light brown, medium brown, and black), as opposed to having to choose between a Black or a White doll as in previous studies (Clark & Clark, 1939; Davis, 2006; Greenwald & Oppenheim, 1968; Lewis & Biber, 1951; Morland, 1962, 1966) or it may relate to that measures were taken to con- trol for demand characteristics. The hypothesis that more options may affect children’s skin tone preferences was supported in the pre-post phase of the current study where children did demonstrate skin tone preferences, as previously reported in the literature. Specifically, the Black children in the control group showed a trend toward selecting the White cartoon, as opposed to the Black cartoon, as their “best friend” when shown a Black and a White cartoon character. Taken together, these findings are both encouraging and worrisome. They are encouraging because they suggest that children’s view about skin tone may be positively influenced by the number

at TEMPLE UNIV on December 2, 2016jbp.sagepub.comDownloaded from

Jordan, Hernandez-Reif / RACIAL ATTITUDES OF YOUNG CHILDREN 401

of options they have, with more options attenuating skin tone preferences. The findings are discouraging because they also suggest that even when measures are taken to control for demand characteristics, when presented with only two skin tone choices, young Black children appear to hold nega- tive views about Black skin tones in that they favored the White over the Black skin tone cartoon. A replication study with a larger sample size would be important to conduct to examine whether indeed providing more skin tone options leads to children being less likely to show skin tone prefer- ences and whether limiting options is associated with children being more likely to show skin tone preferences.

Another finding from the current study was that portraying Black chil- dren in a positive light altered Black, but not White, children’s view about Black skin tones. After hearing a story of a little Black boy who saved a baby duck, significantly fewer Black children selected the Black cartoon as the one that “looks bad.” In contrast, after hearing the same story, White children continued to select the Black cartoon as the one that “looks bad” and the White cartoon as the one that has the “nice skin color.” Somewhat puzzling, however, was the finding that whereas White children showed a preference for selecting the White cartoon as the one with the “nice skin color,” they also showed a trend toward selecting both the White and the Black cartoon as the one that “looks nice” after hearing the story of the lit- tle Black boy who saved the baby duck. Additional research is needed to examine young Black and White children’s threshold in shifting preferences for skin tones.

One reason for the discrepancies in the responses of the Black versus the White children might relate to their knowledge of racial differences. In the pretest, approximately 90% of Black children, compared with only 65% of White children, correctly identified the Black cartoon as the one that “looks Black.” Similarly, while more than 90% of Black children accurately identi- fied which cartoon “looks White,” only about 70% of White children cor- rectly selected the White cartoon in the pretest as looking White. Interestingly, the White children became better at selecting which cartoon was White and which was Black after the storytelling, suggesting that the story might have helped White children be more mindful of skin tones.

That additional research is needed in assessing and educating preschool children about skin tones was evident in that some Black and White chil- dren appeared to have difficulty self-identifying with their skin tones. For the control groups, more than one third of the Black children identified their skin tone as White, and more than one fifth of the White children identified themselves as looking Black. These high percentages may relate to preschool children being less aware of skin tones or perhaps not yet

at TEMPLE UNIV on December 2, 2016jbp.sagepub.comDownloaded from

402 JOURNAL OF BLACK PSYCHOLOGY / AUGUST 2009

knowing their colors. Nonetheless, that the majority of the White children self-identified with the White cartoon and that the majority of the Black children self-identified as looking more like the Black cartoon suggest that overall both Black and White preschool children are aware of their skin tone color. Interestingly, for the Black children, incorrect responding on the self-identification question might have related to the children’s own skin tone variation (i.e., light brown, brown, black, or white) in that about one third of the Black children with light skin did not initially identify with the Black cartoon. Self-identification and knowledge of race may be separate con- cepts that appear to be evident as early as 3 to 5 years of age and may relate to social cultural experiences. Future studies are needed that examine the issues of social cultural influences and children’s own skin tone on chil- dren’s racial awareness and self-identification. In the current study, the sample size did not permit examining these issues.

In summary, in the present day study of preschool children living in Alabama, knowledge of race did not negatively affect children’s skin tone preferences. Further studies are needed to explore which factors predict young children’s skin tone preferences and negative attitudes. Research should also assess biracial/multiracial children’s skin tone preferences and attitudes, as well as examine those of other races (e.g., Hispanics, Asians, and Native Americans). As the United States moves forward into the 21st century, research and programs educating young children about cultural and racial diversity are needed to foster healthy attitudes that will hopefully continue as children grow into adolescence and adulthood.

REFERENCES

Banks, W. C. (1976). White preference in blacks: A paradigm in search of a phenomenon. Psychological Bulletin, 83, 1179-1186.

Brown v. Board of Education of Topeka, 347 U.S. 483 (1954). Clark, K. B. (1955). Prejudice and your child. Boston: Beacon. Clark, K. B., & Clark, M. P. (1939). The development of consciousness of self and the emer-

gence of racial identification in Negro preschool children. Journal of Social Psychology, 10, 591-599.

Clark, K. B., & Clark, M. P. (1947). Racial identification and preference in Negro children. In T. M. Newcomb & E. L. Hartley (Eds.), Readings in social psychology. New York: Holt, Rinehart & Winston.

Cross, W. E. (1991). Shades of Black: Diversity in African-American identity. Philadelphia: Temple University Press.

Davis, K. (2007). A girl like me [Documentary film]. Retrieved October 2007 from http:// www.kiridavis.com/index.phpoption=com_content&task=view&id=17&Itemid=88888901

at TEMPLE UNIV on December 2, 2016jbp.sagepub.comDownloaded from

Jordan, Hernandez-Reif / RACIAL ATTITUDES OF YOUNG CHILDREN 403

Goodman, M. E. (1952). Race awareness in young children. Cambridge, MA: Addison- Wesley.

Gopaul-McNicol, S. A. (1995). A cross-cultural examination of racial identity and racial pref- erence of preschool children in the West Indies. Journal of Cross-Cultural Psychology, 26, 141-152.

Greenwald, H. J., & Oppenheim, D. B. (1968). Reported magnitude of self misidentification among Negro children—Artifact? Journal of Personality and Social Psychology, 8, 49-52.

Hollingshead, A. (1975). Four factor index of social status. Unpublished manuscript, Yale University, New Haven, CT.

Landreth, C., & Johnson, B. C. (1953). Young children’s responses to a picture and inset test designed to reveal reactions to persons of different skin color. Child Development, 24, 63-80.

Lewis, C., & Biber, B. (1951). Reactions of Negro children toward Negro and White teachers. Experimental Education, 20, 97-104.

Morland, J. (1962). Racial acceptance and preference of nursery school children in a southern city. Merrill-Palmer Quarterly, 8, 271-280.

Morland, J. K. (1966). A comparison of race awareness in northern and southern children. American Journal of Orthopsychiatry, 36, 22-31.

Porter, J. D. R. (1971). Black child, White child. Cambridge, MA: Harvard University Press. Powell-Hopson, D., & Hopson, D. (1992). Implications of doll color preferences among Black

preschool children and White preschool children. In A. Burlew, K. Hoard, W. Banks, H. McAdoo, & D. Azibo (Eds.), African American psychology: Theory, research, and practice (pp. 183-189). Newbury Park, CA: Sage.

Robinson, T. L., & Ward, J. V. (1995). African-American adolescents and skin color. Journal of Black Psychology, 21, 256-274.

Rosenberg, M. (1989). Old myths die hard: The case of black self-esteem. Revue Internationale de Psychologie Sociale, 2, 356-365.

Stevenson, H. W., & Stewart, E. C. A. (1958). A developmental study of race awareness in young children. Child Development, 29, 399-410.

at TEMPLE UNIV on December 2, 2016jbp.sagepub.comDownloaded from