School-to-Prison Pipeline: School Funding

As you may know, education spending in the United States is chronically low, totaling six percent of the total federal funding per year.


chart (2)


According to the U.S. Department of Education, the primary source of funding in the U.S. is intended to be from states and localities (approximately 83 cents per dollar, as of the 2004-2005 school year). The issue with this is the high rate of income inequality in the United States, particularly by location (rural, suburban, or urban). As such, those who live in poorer areas are able to contribute less to students’ education, and school funding, as well as educational quality, suffer. Lack of funding in schools and school districts often coincides with low rates of educational attainment and high rates of school closures (leading to overcrowding and high dropout rates). These issues correlate with higher rates of imprisonment, lending to the “school-to-prison pipeline” that academics and activists have become aware of.

Since funding of schools is intended to be mostly by locality, it is valuable to assess how poverty rates in different localities might indicate a link between school resource allocation and the school-to-prison pipeline.

Without accounting for locality, the average elementary and secondary education funding by local sources is approximately 44% (as of 2011).




When taking locality and poverty rates into account using the same data set, there is a clear trend in the allocation of local funding versus total funding (which includes federal, state, and local funding).




In areas of low poverty, all localities are able to contribute more to schooling, High poverty rates in all localities produce the opposite effect, as one might expect. Interestingly, suburbs and cities with high poverty rates are able to contribute roughly the same percentage to funding, while rural localities are able to contribute a lower percentage to school funding.

Schools that are funded less by the community receive less funding overall and also enjoy less community engagement with the schools and student education. Student educational attainment is lessened by these factors and research shows higher rates of arrest and incarceration among those with lower educational attainment.

These findings suggest that school resource allocation does not consist of a single number across the board, or a certain number of cents out of a dollar. One must take into account locality and poverty rate in determining school funding and understanding how the method of funding schools in the United States privileges certain students over others, creating educational opportunities for some and all but ensuring criminal records for others.

Additionally, funding for prisoners is increasing at a much higher rate than funding for students — nearly three times as quickly. Coupled with the chronic rise in America’s prison population, funding priority at federal and state levels seems to be given to prisons rather than schools. This results in resource-strapped schools, particularly in areas that are already limited in funding (i.e. poor locations, cities, and rural areas). Through this funding structure, students continue to be shuttled through the school-to-prison pipeline from schools to prisons, perpetuating the funding cycle and the pipeline.




The issues of the school-to-prison pipeline in relation to school funding is two-pronged then; the methods through which the majority of school funding is received work against high-need, high-risk areas that already have limited community resources, and the allocation of funding to rapidly growing prisons instead of schools. A re-examination of funding priorities and methods may go a long way in alleviating one aspect of the school-to-prison pipeline and resource inequality for students from kindergarten onward.

Enter your e-mail address to receive notifications of new blog posts.
You can leave the list at any time. Removal instructions are included in each message.

Powered by WPNewsman


Please like & share:

Historical Data Requires Historical Finesse


Utilizing contemporary tools to analyze historical data provides a unique way to approach historical research, but can prove to be an arduous process as modern tools may not be compatible with historical data. This summer, I have been working with Professor Sarah Purcell to create maps for her book on spectacle funerals of key figures during the U.S. Civil War and Reconstruction. Most commonly, famous bodily remains traveled from city to city on railroads, in some cases on a special funeral train, though they also traveled on rivers and in one case, across the Atlantic Ocean. Nearly every historical figure discussed in the book has an accompanying map which charts their extended funeral processional route. Using GIS technology, we are able to juxtapose census and election data with the geographic routes in highly analytical maps.

In order to layer election data onto the map for Col. Elmer Ellsworth (died 1861), I gathered county-level election data from the Interuniversity Consortium for Political and Social Research (ICPSR) and county-level census data from the National Historical Geographic Information System (NHGIS). I then needed to combine the ICPSR election data and the NHGIS census data in a joined spreadsheet before importing the data into ArcGIS software to link the data to its county location.

At first, I thought we could link the data using something called a “FIPS code.” In an effort to standardize big data and allow for easy joining of tables by location, the Federal Information Processing Standard assigned each county in the United States during the 1990s with a unique five-digit code, more commonly known as a FIPS code. The first two digits are the state FIPS code and the last three are the county code within the state. For example, the FIPS code for Poweshiek County, Iowa is 19157. This code is assigned to the current borders of Poweshiek County. Yet the data I was analyzing is from 1860. Poweshiek County in 2015 represents a different land area than Poweshiek County in 1860. Thus, joining ICPSR and NHGIS data from the 19th century could not be completed using FIPS codes without introducing historical inaccuracy in the maps.

In order to join two tables of data in any computer program, there must be a common column between them. From ICPSR, I had a table of county-level election data from 1860 and from NHGIS, I had a table of county-level 1860 census data. If I were to join data tables of current counties, the FIPS code would serve as my common column. However, instead of using FIPS codes to join the data, I created a common column using the name of the county and state. Creating a unique name for each county assures that I correctly joined the historic county data to the historic county borders. Poweshiek County’s unique identifier would be: “PoweshiekIowa.” I quickly discovered that joining data by this concatenated column was not without error. I went through each county individually to discover discrepancies, many of which resulted from spelling inconsistencies between the two databases.

After cleaning the data, the tables joined neatly. Using GIS, I then linked the combined election and census dataset to the geographic borders of the counties on the electronic map. I color coded the map by political party. The darker shade of each color show where the political party won the majority of votes in the county (greater than 50%), while the lighter shade of the color shows where the party won a plurality of the votes in the county. As you can see from the map’s multiple colors, unlike modern American politics, the 1860 presidential election involved more than two prominent political parties including Republicans, Northern and Southern Democrats, and the Constitutional Union Party. The political divide between North and South is clearly apparent along the Mason-Dixon Line between Pennsylvania and Maryland foreshadowing the sectional conflict of the American Civil War nearly six months after the election.

Mapping historical data is certainly a different process than mapping current data and can prove to be more time-consuming and complex. Though current tools (like FIPS codes) can help standardize mapping techniques, they may not be applicable in historical data settings and current tools may need to be discarded or updated. Historical FIPS codes, anyone?

Enter your e-mail address to receive notifications of new blog posts.
You can leave the list at any time. Removal instructions are included in each message.

Powered by WPNewsman

Please like & share: