The Project at a Glance
This project began when Lone Star Policy Institute Director Michael Wood asked Tower Scholar Destiny Rose Murphy, "Is it easier to beat your wife in Fort Worth than it is in Dallas?" The answer, as it turns out, is complicated. Due to the federalist nature of government, to begin to answer this question the researcher must first decide what level of our government to focus research on. The data and research in this website focus on the office of the District Attorney (including Criminal District Attorneys, as well as misdemeanor data in counties that elect County Attorneys). The DA's office was chosen as the focus of this work because it has been identified by recent research as one of the most important actors in local criminal justice jurisdictions (Pfaff).
Having chosen the DA's office as its focus, this research sought to gather data on the prosecution and sentencing trends from each DA's Office in Texas, and to see if that data revealed any high level trends across the state. Although the original intent of this project was to discover those high-level trends, the most interesting part of the project quickly became the data-gathering stage, in which it was discovered that most Texas DA's offices are recording very little data, and only making a tiny fraction of that recorded-data public. The most important findings from this project have therefore centered on DA transparency in Texas.
Below you can find detailed information on the finding and gathering of the data used in this project, as well as the reasoning behind the digital presentation of this data. On the Findings page you can read about the high-level criminal justice trends this research uncovered, as well as a detailed description of findings on transparency, and suggestions for future research. Finally, on the About page you can learn about the institutions and people who made this happen: the Lone Star Policy Institute, the John G. Tower Center for Political Studies, and Tower Scholar Destiny Rose Murphy.
Finding, Aggregating, and Evaluating Data
Alternatively: How the Sausage got Made
The majority of the data used in this project was pulled from the Texas Office of Court Administration (OCA) website, which offers some county-wide reports of misdemeanor and felony activity. This data is reported to the OCA by the County Clerk for each county, and is the best public source of information about county criminal filings; in fact, when asked for data about misdemeanors and felonies many DA offices either pointed to this website as the source or used this website to pull information about their own counties, which they then reported to the researcher. However, during an interview the DA of Fort Bend County (John F. Healey, Jr. at the time) explained that he believes the data to be inaccurate, not only due to reporting mistakes, but also due to the simple nature of the reports, which do not include measures like diversion programs. As such, the reliability of this data cannot be confirmed.
The two OCA forms containing most of the data used for this research are the Misdemeanor Activity Detail and the Felony Activity Detail. At the time of this research the function to export the information in these forms to Excel was broken, as was the function to export the information to Word. So, researcher Destiny Rose built a Ruby program to download the 3,556 misdemeanor and felony reports for individual counties for the years 2011 - 2017 (the years with the most reliable data), and then used Tabula to turn those PDF's into CSV's. Finally, Destiny Rose made another Ruby program to aggregate information from the 3,556 CSV's into one central spreadsheet. The number of each kind of misdemeanor and felony filed for each county for each year was recorded, as was sentencing data (the number of each kind of sentence) for all crimes.
A similar method was used to obtain yearly county population estimates from the Texas State Library and Archives Commission, which was recorded for each year for each county. Neither the state government nor the state Democratic or Republican parties maintain historical lists of the name and party of each DA in office each year, so, as a proxy, Destiny Rose used the Texas Office of the Secretary of State's (SOS) historical election data website to gather information about DA election winners. Unfortunately, the SOS's database does not provide information about DA's who may have been appointed, nor does it appear to have consistent data about all counties. As such, the names and party affiliations of DA's from this database may not be completely accurate.
With the above information gathered into a single spreadsheet, the statistical analysis software SPSS 24 was used to analyze the data. Simple linear and multivariate regressions were conducted via the software, which are detailed in the Findings page.
Displaying Data: Why a Website?
The choice to display this research via a website was made for several reasons. First, the information detailed in this website is public, and so it should be actually accessible to the public. District Attorneys are elected officials, which means the workings of the DA's office should not be a secret to the office's own constituency. Unfortunately, much of the data displayed in this website required a great deal of technical skill to retrieve. It is estimated that, without coding knowledge, creating the final spreadsheet of data used to generate this website would have taken approximately 200 hours of manual data entry. Even worse than the immense time such a project would take, aggregating that data by hand would make this project even more open to the risk of error, which, compounded with the already somewhat unreliable nature of the data itself, would likely make the data useless. Because of the systematized way the data in this website was retrieved, however, a great deal of human error has been avoided. Hopefully, this website will provide Texans with a more accessible way to learn about the criminal justice system of their home county and will highlight the need for real transparency in District Attorneys offices.
Finally, the data in this website may be incomplete, but it is also a starting place for future research. Whether you are a hobbyist or a scholar, you are welcome to use and build off of this data as much as you can. Below, you will find a link to the final CSV spreadsheet used to generate the graphs and stats in this website. I would love to help further this kind of research, so please feel free to steal the data below for your own work, so long as you reach out to me at firstname.lastname@example.org to talk about research collaboration and any questions you may have. Remember to cite this website if you use it for scholarly work, and consider posting your findings online (or send them to me and I will post them here for you). Research about public institutions like DA's offices belong in the public forum, not just written in legalese and published in an academic journal, so be sure to share your work as much as you can.County-Per-Year Data CSV
Pfaff, John F. Locked in: the True Causes of Mass Incarceration--and How to Achieve Real Reform. Basic Books, 2017.