Another amazing bgsu blog

Archive for the ‘Research Methods’ Category

Research Performance Progress Report for July-December 2013

without comments


The purpose of the research project is to promote police integrity by gaining a better understanding of police crime and agency responses to officer arrests. The study provides a wealth of data on a phenomena that relates directly to police integrity—data that police executives have not previously had access to because this information did not exist in any useable format.

In the previous reporting period we substantially completed tasks 1-4 100% complete, task 5 was 95% complete, task 6 was 75% complete, task 7 was 86.5% complete, task 10 was 82% complete, task 11 was ongoing, and task 12 was 15% complete. Tasks 8, 9, 13, 14, 15, and 16 had not yet been undertaken. On January 28, 2014, we again revised the project timeline. As of the close of business on December 31, 2013, tasks 1-5 are 100% complete, task 6 is 95% complete, task 7 is 97% complete, task 8 is 100% complete, task 9 is 20% complete, task 10 is 95% complete, task 11 is ongoing, task 12 is 20% complete, tasks 13 and 14 are 100% complete, task 15 is 85% complete, and task 16 has not yet been undertaken.

As of the close of business on December 31, 2013, a total of 8,880 police crime arrest cases involving 7,396 individual officers have been logged in our integrated relational, object-oriented, and digital imaging database. The arrested officers were employed by 2,816 nonfederal law enforcement agencies in 1,396 counties and independent cities in all 50 states and the District of Columbia. Of these, 6,799 cases involve 5,580 individual officers arrested for one or more crimes during the time period of January 1, 2005, through December 31, 2011. The officers arrested in years 2005-2011 were employed by 2,339 nonfederal law enforcement agencies in 1,212 counties and independent cities in all 50 states and the District of Columbia. Included in the database are cases from 2012 and 2013 (which are not included in this study), including 1,111 arrest cases for officers arrested during year 2012, and 970 arrest cases for officers arrested during year 2013. Prior to July 1, 2013, there were 8,154 police crime arrest cases (involving 6,853 individual sworn law enforcement officers) that had previously been logged in our database. Thus, during this reporting period we added 726 new police crime arrest cases and 543 more sworn officers to our database. The new cases include 141 additional cases involving 28 individual officers who were arrested during the years 2005-2011. In sum, during this reporting period, we expanded our database of known police crime arrest cases by 8.2% overall (an increase of 2.1% for the study years 2005-2011 arrest cases).

The project utilizes OnBase, an enterprise-level content management (ECM) database system. We continue to make enhancements and deploy customized application integrations to support the research project. The integrated project database includes a digital imaging ECM database with a full-text search optical character recognition (OCR) search engine, a relational and object-oriented database, and an electronic coding instrument. The project database also includes a video file object-oriented database that is integrated within OnBase. The integrated relational and digital imaging database includes electronic case log-in procedures and allows us to structure and search data in different ways for content analyses. The video database is used to triangulate data obtained from television news sources as a supplement to our other database libraries.

Coding of content is completed using a computerized coding instrument that utilizes a customized version of the IBM/SPSS Data Collection / Data Entry Author and Interviewer software application. The coding instrument system pulls information from the relational database into the on-screen coding instrument for each case to be coded, thus reducing coder duplication of effort and the potential for coding errors. We continue to make enhancements to our project database in OnBase. During this reporting period we designed and deployed an e-coding sheet that creates an electronic facsimile of the project’s paper coding sheet for each case that has been coded with the computer-based coding instrument. The e-coding sheets are imported and indexed in OnBase in Microsoft Word (doc.) files so that coded data can be retrieved and viewed without viewing a raw spreadsheet data file. Throughout this project, we have found that training (and routine periodic retraining) of graduate student research assistants and the process of coding the content of case file records is more time-consuming and slower than we anticipated when we developed the project timeline. Implementation of the electronic coding instrument has allowed us to streamline data coding processes, reduce potential coding and data entry errors, and ensure a high degree of intercoder reliability.

The digital imaging database includes 242,998 pages of digital images (consisting of 188,883 scanned pages and 54,115 imported pages). These digital image files consist of 19,526 TIFF case document files, 20,019 TIFF coding sheet document files, 2,479 e-coding sheet MS Word files, and 5,418 PDF PACER document files. Of these, 54,115 pages were added since July 1, 2013 (consisting of 2,663 case document files, 4,510 coding sheet document files, 283 PACER PDF document files, and 2,479 e-coding sheet MS Word files). The PACER files consist of 70,153 pages of federal court docket sheets, pleadings, court orders, sentencing records, and other docket entries. The e-coding sheets consist of 52,059 pages in MS Word document files that were created electronically from an SPSS data file for all of the cases that were coded using the Interviewer electronic coding instrument (thus making it possible to examine coded data for each case in OnBase without having to resort to the original data file spreadsheet). Also included in the database are 1,267 news media video files.

In sum, our enhanced content management system database currently includes more than 244,265 pages and object-oriented files of news articles, court records, news videos, and coding sheets that document the criminal arrests of more than 7,396 police officers since the beginning of 2005. The arrested officers were employed by nonfederal state, local, special, and tribal law enforcement agencies in all 50 states and the District of Columbia.


There are three major goals of the project. The first goal of this research is to determine the nature and extent of police crime in the United States. The second goal is to determine what factors influence how a police organization responds to arrests of officers. The third and final goal of the research is to foster police integrity by exploring whether police crime and officer arrests correlate with other forms of police misconduct.


The project’s principal investigator, Dr. Philip Stinson, delivered a one-hour presentation on the project’s major findings at a mini-conference on police misconduct that was sponsored by the American Psychological Association’s Division on Psychologists in Public Service, Section on Police and Public Safety, in July 2013 as part of the annual meeting of the American Psychological Association in Honolulu, HI.


Results have been disseminated to communities of interest through the writing and publication of refereed journal articles, magazine articles, and research briefs. We also maintain a project website and blog, and produce a monthly audio podcast that is available on iTunes.

We proactively disseminate our research results to communities of interest. To that end, we have engaged in outreach activities to reach members of communities who are not usually aware of these research activities for the purpose of enhancing public understanding and increasing interest in learning and careers in criminological research. We have published three short research-in-brief articles reporting our research findings in Police Chief, a practitioner-oriented magazine published by the International Association of Chiefs of Police. We also make PDF copies of one-sheet research briefs explaining our research studies and findings on our web-site and post timely project-related entries on our blog. Additionally, we distribute an audio podcast on iTunes where we discuss the research project, studies, and related findings.

The web analytic reports for the Police Integrity Lost Project Blog indicate that for the time period July 1, 2013, through December 31, 2013, there were 341 unique visitors to the blog. Traffic to the blog was primarily from the United States (88.569%) where visitors to the blog were located in 119 cities in 37 states. There were also visitors to the blog from 15 foreign countries: India, Philippines, United Kingdom, Canada, Germany, Australia, Indonesia, Hong Kong, Japan, Kenya, Cambodia, Leotho, Malaysia, Thailand, and Vietnam. Traffic to the blog included organic search traffic (47.5%), direct traffic (39.5%), and referrals (4.9%).

The Police Integrity Lost Podcast is available on iTunes domestically as well as internationally, and interested persons can stream, download, and/or subscribe to the podcast audio files directly from the iTunes client application. Web analytics for the iTunes podcast episodes indicate that for this reporting period there were 1,333 mp3 file hits, indicating that there were 1,333 instances when someone either streamed (listened to) all or part of one of our podcast episodes and/or downloaded mp3 audio files. Of those hits, the largest number of requests directed to our pod server to stream and/or download the mp3 audio files during this reporting period was 141 hits for the “Police Crime Research Findings” podcast episode (Police Integrity Lost Podcast Episode 13) and 133 hits for the “Police Crime Sexual Misconduct” podcast episode (Police Integrity Lost Podcast Episode 11).


As of the close of business on December 31, 2013, tasks 1-5 are 100% complete, task 6 is 95% complete, task 7 is 97% complete, task 8 is 100% complete, task 9 is 20% complete, task 10 is 95% complete, task 11 is ongoing, task 12 is 20% complete, tasks 13 and 14 are 100% complete, task 15 is 85% complete, and task 16 has not yet been undertaken. Currently, one research paper (a study of police sexual misconduct arrests) is under review for publication consideration at a refereed journal. We are also currently writing a second paper that focuses on the victims of police sexual misconduct, which will be submitted to a refereed journal for publication consideration within the next few months. We will be making a research presentation on our study of victims of police sexual misconduct in February 2014 at the annual conference of the Academy of Criminal Justice Sciences in Philadelphia, Pennsylvania. We will also continue to produce our monthly podcast episodes for iTunes and disseminate research briefs on the various studies as completed. The project website and blog will also be updated with new postings periodically. The primary focus, however, during the next reporting period is continuing with statistical data analyses, writing, and submitting the final technical report and completing other related items to complete the final requirements of the current NIJ-funded project.


The project has produced four peer-reviewed journal articles, three research brief trade-magazine articles, six one-sheet research briefs, 17 mp3 audio podcast episodes, and 13 research presentations.


The project’s principal investigator, Dr. Philip Stinson, was recently recognized by the Metropolitan Police Department in Washington, D.C., in recognition for his assistance during the past two years in responding to research-related inquiries from the police department.


A six-month no-cost extension of the project was approved by NIJ on August 14, 2013, because it has taken longer for us to complete the content analysis than originally expected. The new project end date is June 30, 2014. Currently the draft final technical report is due to be submitted no later than March 31, 2014. Pursuant to a newly revised timeline, we now anticipate that the draft final technical report will be submitted on or before April 30, 2014.

Written by Phil Stinson

January 31st, 2014 at 4:29 pm

Using Google News for Data Collection: Podcast Available on iTunes

without comments

The March 2013 episode of the Police Integrity Lost podcast is Using Google News for Data Collection: Police Crime Research Methods Part 2. In this podcast episode, Steve Brewer (Penn State Shenango) and John Liederbach (Bowling Green State University) question Phil Stinson (Bowling Green State University) on his use of the Google News search engine and Google Alerts as part of his research methodology to collect data for research studies on police crime in the United States.

Written by Phil Stinson

March 9th, 2013 at 2:01 pm

Podcast Episode on Research Methods: Classification Tree Analysis

without comments

The February 2013 episode of the Police Integrity Lost podcast features Phil Stinson (Bowling Green State University) and Steve Brewer (Penn State – Shenango) discussing their use of classification tree analysis (also called “decision tree” analysis) techniques as part of their predictive analytic statistical procedures in the study of police crime. Classification tree analysis is used as a statistical technique to uncover the causal pathways between independent predictors and an outcome (dependent) variable of interest in a regression model.

The Police Integrity Lost podcast is available exclusively on iTunes.

Written by Phil Stinson

February 26th, 2013 at 1:27 pm

Skip to toolbar