The Flow of Race and Education

A chord diagram visualizes these relationships by drawing quadratic Bézier curves between arcs. The source and target arcs represents two mirrored subsets of the total population, such as the number of people with black hair that prefer blonde hair, and the number of people with blonde hair that prefer black hair. This chord diagram shows how each race listed distributes amongst the various levels of educational attainment. There are some very obvious patterns that emerge when each group is specified (which can be done by mousing over it). The sums of each racial group is 100, but the sums of the educational categories are larger than 100 because it is accounting for each race separately and the data is in percentages, but it is a useful visualization nonetheless. The data is pulled from the 2018 census.

By mousing over each subsection of each race, or the specific connecting line of interest you can view the percent that “achieves” that educational attainment, or the racial distribution of that particular type of education.

Put more simply, this is just a good way to visualize the following table that shows the proportion of races fulfilling each degree. Green represents a race having greater educational attainment than the average in that category, while yellow represents a lower educational attainment, and white represents an equivalent educational attainment. Blue is used to highlight All - the variable to which each racial category is being compared.

Geography

This section shows some of the basic geography of the US in terms of education and race. The illustration of which come in county level data and state level data. The clear pattern that emerges is how densely populated hispanic and black communities are, while white populations are extremely widespread.

Education Quality and Race

To really determine an association a direct link between macroeconomic factors can be created by looking at state level education quality data. The provision of affordable, and quality education to families is one of the inherent ideals behind “equality of opportunity” and is necessary for educational and income mobility between generations. Should a state be failing to provide that to certain races, and if it is comprised of a higher proportion of a minority race then the affects upon the entirety is multiplied greatly by the geographic compactness of non-white races across the US. The following regression analysis and graph separately show very similar ideas with similar mathematical bases - that as racial proportion of black and hispanic students increases in a state, education quality is generally decreasing, producing a disproportional impact on minorities. If Asian’s are weighted separately we see that they actually enjoy a slight improvement in public school educational quality as their proportional makeup of the state increases, although if they are weighted against the proportional increase in educational quality for white students as in the regression table, the difference is slightly negative. The graph perfectly illustrates the educational opportunity gap in the US, although white people make up a majority of the US, when breaking down to state level educational data - which is where most educational provision and funding come from - there is a disproportionate lack of opportunity being provided to minorities, especially black and hispanic populations.

Looking at this in a table with a single regression comparing the weighted association of different races with educational quality scores we see that after accounting for the proportional makeup of a state, and race, there are decreases in eduational quality for all races relative to the educational quality score of white communities weighted for proportional makeup. The effect is however much larger for black and hispanic educational quality than for asian educational equality given the barely negative coefficient for asians, which indicates that if you are asian you can expect educational quality to decrease 0.06 points relative to white educational quality after weighting for proportions. For black and hispanic educational quality the expected decline is 0.338, and 0.369 points relative to white educational quality.

Conclusions

There are clear racial inequities in educational attainment, table 1 indicates that such inequities result in all races other than white or asian, the most common final education achievement is a high school degree. Intergenerational income mobility has been a well studied subject in economics, but educational mobility might offer a better lens to analyze this subject as the majority of income mobility comes from level of educational attainment. Given that this investigation found distinct levels of geographical disparity comparison of educational attainment by race should be considered with longitudinal data, as well as with structural breaks for counties in which minorities are actually a majority to determine the extent of educational opportunity disparities. Since education achieves its ultimate value fairly early in the life cycle, and is more directly influenced by parents, it serves as a measure of early life opportunities available to the child. Therefore measures of educational mobility should be developed, and used to compare different cohorts of children to assess levels of “equality of opportunity”. Student-teacher ratios do not appear to have large quantitative associations with race in each state after accounting for proportional makeup. Geography is nonetheless a strong determinant of educational quality, as evidenced by table 1 and figure 5, and educational opportunity is certainly strongly disproportionate by state, and therefore also by race.