Developing Models to Predict Giving Behavior of Non-Profit Donors; Analysis of the Use of Data Mining in Non-Profit Organizations
Ponos, Sofia, School of Engineering and Applied Science, University of Virginia
Scherer, William, University of Virginia
Elliott, Travis, University of Virginia
Data mining has greatly expanded into many industries over the past 20 years. This has changed how people view information collection and analysis. Deriving value from information has become crucial to organizations of every size. The technical portion of this thesis discusses using donor data from a client, the Children’s Inn at the National Institute of Health (NIH), to develop models that can predict future donor behavior. The science technology and society (STS) portion of this thesis will explore further how data mining has affected organizations using the Social Construction of Technology (SCOT) framework. It will also discuss the ethics of data mining through different ethical frameworks. Doing a technical project with sensitive data brings up ethical questions that relate to various stakeholders. Building donor behavior models not only requires technical expertise but also an awareness of the social factors that affect human behavior.
The Children’s Inn (or The Inn) at NIH is a non-profit organization that provides free room and board to patients and families participating in clinical research studies at the NIH. To be able to provide these services The Inn must solicit donations through direct mail appeals, online appeals, and events. In order to solicit donations most effectively this technical report discusses how models can assist in increasing donations from current donors and finding profitable new donors. These models will allow The Inn to take action in prioritizing high-value donors and acquiring new profitable donors. The Inn will be able to continue using these frameworks to invest time and resources more strategically into their fundraising efforts to maximize total donations while allowing for more time and resources for their mission of serving the families staying at the Inn.
The rapid increase in the availability of data and the use of data analysis has brought up many ethical and societal questions. The STS report will specifically look into how data mining has affected the non-profit space, what unique stakeholders are present and what ethical questions may arise. Data mining used in non-profits is often referred to as donor mining. The technique of analyzing donor behavior allows organizations to increase the dollar amount of donations and reach new donors. This developing analytical and technological approach affects all stakeholders involved including donors, organizations, data analysts and those who receive charitable donations. Donor mining affects all these groups differently and can have both positive and negative effects depending on how it is used. Donor mining cannot be done effectively without considering the social affects that the technology will have. It is integral for engineers to consider all the stakeholders that will be affected by their data analysis.
BS (Bachelor of Science)
nonprofit analytics, data-driven nonprofits, data mining, donor mining, predictive modeling, Markov, RFM analysis
School of Engineering and Applied Science Bachelor of Science in Engineering Systems and Environment
Technical Advisor: William T. Scherer
STS Advisor: Sean T. Elliott
Technical Team Members: Clare M. Hammonds, Josh Eiland, Shawn M. Weigand
All rights reserved (no additional license for public reuse)