Color and racial attitudes in white, black and biracial children

Social Behavior and Personality, 1998 by Neto, Felix, Paiva, Lizalia

Previous studies of children in United States, Western Europe and Asia have demonstrated a bias favoring the color white relative to the color black, and a bias favoring light-skinned figures relative to dark-skinned figures. In this study of eight-year old children, procedures used in previous studies were administered to biracial children of mixed black and white parentage and to black and white children of monoracial parentage. Both types of bias were found among the three racial categories, providing additional evidence that the pro-white and pro-light-skinned biases are pancultural tendencies. The biases were not different by gender, but they were significantly different by race. Mean color attitudes of white children were significantly different from biracial children, such that the white children displayed a stronger pro-white/anti-black bias than the biracial children. Mean racial attitudes of white children were significantly different from black and biracial children, with white children showing more pro light-skinned bias than black and biracial children. It appears that the biracial category should be considered in research on color and racial attitudes.

The existence of a pancultural tendency to associate positive evaluation with the color white and negative evaluation with the color black has been documented among both adults (Adams & Osgood, 1973; Williams & Morland, 1976; Neto, 1998) and children (Best, Field, & Williams, 1976; Best, Naylor, & Williams, 1975; Dent, 1976; Iwawaki, Sonoo, Williams, & Best, 1978; Neto & Williams, 1997).

Adams and Osgood (1973) found that young adult subjects rated the color name white more positively than black in all of the 23 language-culture groups studied in Europe, Asia and the Americas. Williams and Morland (1976) obtained the same finding in the seven American, European, and Oriental groups which they studied.

The childhood development of the evaluative meanings of white and black has been studied in the United States, in several European countries and in Japan, employing a picture-story technique known as the Color Meaning Test II (CMT II; Williams, Boswell, & Best, 1975). Studies in the United States, reviewed by Williams and Morland (1976), have demonstrated pro-white/anti-black bias among both Euro-American and Afro-American preschoolers, with the bias being weaker in the latter group. Pro-white/anti-black (W /B-) bias has also been demonstrated among 5- and 6-year-old children in England and Scotland (Dent,1976), in France and Italy (Best et al., 1975), in Germany (Best et al., 1976), in Japan (Iwawaki et al., 1978) and in Portugal (Neto & Williams, 1997). The pancultural nature of this pro-white/anti-black bias has led to speculation that this tendency may be attributable, at least in part, to the diurnal nature of Homo Sapiens which is related to the nature of the human retina (poor night vision) and which results in a learned preference for light over darkness (Williams & Morland, 1976).

Several investigators have studied the possible relationship of W /B- bias to the tendency of many young children to evaluate light-skinned persons more positively than dark-skinned persons, as revealed by the Preschool Racial Attitude Measure II (PRAM II; Williams, Best, Boswell, Mattson, & Graves, 1975). Pro-light-skinned/anti-dark-skinned bias (LS /DS-) is usually found among Euroand Afro-American preschool children (Williams and Morland, 1976), where it is often given a racial interpretation, i.e., it is assumed to reflect the children's learning experiences in a multiracial society where light-skinned persons generally enjoy a more privileged position and prejudice against dark-skinned persons is often encountered. An exclusive reliance on such an interpretation has been challenged by the demonstration of LS /DS- bias among preschool children in France and Italy (Best et al., 1975), in Germany (Best et al., 1976), in Japan (Iwawaki et al., 1978) and in Portugal (Neto & Williams, 1997). Since young children in these countries have little contact with dark-skinned persons or with the concept of race as it relates to skin color, the LS /DS- bias which they display must be attributed to other sources, possibly to a general pro-light/anti-dark bias which is related to the W /B- bias which they also display. Support for a functional relationship between LS /DS- bias and W /B- bias is seen in the positive correlations found between the two bias scores (CMT II and PRAM II) among children in all countries studied, (Williams & Morland, 1976).

In a recent study both types of bias were found among Portuguese children providing additional evidence that the pro-white and pro-light-skinned biases are pancultural tendencies (Neto & Williams, 1997). The biases were not different by gender, but they were higher among eight-year-old than among five-yearold children.

This paper reports findings of a study in which translated versions of CMT II and PRAM II were administered to eight-year-old children living in Portugal. It was expected that the findings in Portugal would provide additional evidence of the pancultural tendencies toward W /B- and LS /DS- biases observed in previous studies. Beyond this, the present study explored findings for biracial children, examining these relative to the findings for white and black children of monoracial parentage.