This study is based on interviews and observations
of children, aged 8 to 10, and their
families. The data were collected over time
in three research phases. Phase one involved
observations in two third-grade classrooms
in a public school in the Midwestern community
of "Lawrenceville."4 After conducting
observations for two months, I grouped
the families into social class (and race) categories
based on information provided by
educators. I then chose every third name,
and sent a letter to the child's home asking
the mother and father to participate in separate
interviews. Over 90 percent of parents
agreed, for a total of 32 children (16 white
and 16 African American). A black graduate
student and I interviewed all mothers and
most fathers (or guardians) of the children.
Each interview lasted 90 to 120