Goyette, Kimberly A., Farrie, Danielle, & Freely, Joshua
This School’s Gone Downhill: Racial Change and Perceived School Quality among Whites
How does perceptions of quality change when racial composition of schools change?
Journal Name or Institutional Affiliation:
Vol. 59, No. 2, Pp. 155-176
Results suggest that white residents may initially be threatened by racial change and judge declining school quality according to the racial change itself. As a consequence, white families may flee these integrating schools and neighborhoods, further contributing to school and neighborhood segregation.
Results show that a growing proportion of blacks influences white neighborhood residents’ perceptions that school quality is declining over and above the current status of schools and changes in that status over the past four to five years, but only for schools that were predominantly white at the beginning of the four- to five-year period, in 1999–2000.
Those near schools with greater proportions in poverty, with high incidences of violence, and lower test scores (indicating high distress on the school distress index) are more likely to say school quality is decreasing than are those who live among less troubled schools.
White residents near schools that saw the proportion of blacks increase up to seven percentage points are significantly more likely to report that school quality declined as black representation increased.
In schools where black populations are decreasing, larger decreases are actually associated with a greater likelihood of saying that the schools have declined. In fact, among those who saw declines in black representation (represented by the spline for change of –8.42 percent to 0 percent), there is a statistically significant negative slope.
The smaller the decrease (the less negative the change) in the percentage of black students, then, the less likely neighborhood residents are to say that school quality has declined in the past five years. The larger the decrease (the more negative the change), the more likely residents are to say that school quality has decreased for the neighborhood residents that are represented by this spline.
Journal Article Empirical Research
Race, Racial Composition, School Quality, Segregation
Secondary Survey Data
Method of Analysis:
Heads of households in the Philadelphia Metropolitan region
Unit of Analysis:
- The main source of data for this study is the Philadelphia Area Survey (PAS).
-The PAS is a random telephone survey of heads of households in the Philadelphia Metropolitan region administered in 2003 and 2004, with 2,000 respondents interviewed in total.
The survey has a 36 percent response rate.
Households with children under 18 and those in the city of Philadelphia are slightly overrepresented, which is corrected with normalized sample weights. Applying sample weights decreased the overall sample size to 1,901.
Then the researchers excluded respondents who did not answer the school quality question (155) and those with missing address information (184). This resulted in a sample size of 1,563. Because the researchers are interested in whites’ perceptions of racial change, they further limited the sample to only white respondents, resulting in an overall weighted N of 1,150.
Dependent Variable: perceived school quality. This variable is derived from the survey question “Within the past five years, do you think the quality of public schools in your community has increased, decreased, or stayed the same?” The categories of this variable are collapsed into a dichotomous variable where values of 1 represent respondents who perceived that the quality of their local public schools had decreased and 0 represents those who said otherwise.
Independent variables are split into three categories; Individual characteristics, Neighborhood characteristics, and Local school characteristics.
Individual characteristics include age, gender, if they have children under 18, education status (dichotomous do you have a BA or higher), income (dichotomous if income is $40K or more), if they were a homeowner, and if they lived in their house for 5 or more years.
Neighborhood characteristics is the medium housing value.
Local school characteristics- average percent white (1999-2000 school year), percent receiving free or reduced lunch, standardized reading score, violent incidents per 1000 students, percent failing reading exam. -Change in percent black, change in percent other or minority.
This study is framed on Contact theory.