Earning More & Taking Home Less: The Economic Disincentive of the Benefits Cliff
Advisors: Angie Bos (Political Science, Experiential Learning), Cameron Maneese
Team: Alec Monnie, Stacey Sinae Park, Margaret Simmering
The team was tasked by the Fund for Our Economic Future, United Way of Wayne and Holmes County, and the Wayne County Department of Job and Family Services to find solutions to the benefits cliff in Wayne County. The benefits cliff is a phenomenon in which an individual who receives public benefits receives an income increase which causes them to lose a disproportionate amount of public benefits. The team collaborated with local nonprofits, businesses, and governmental entities to create a booklet that detailed solutions in regards to the costs of child care, food, wellness, and transportation. This booklet was then distributed to over 1,000 businesses by the Wooster and Orrville Chambers of Commerce.
Affordable Housing/Homelessness
Advisors: Nate Addington (CDI), Brooke Krause (Economics)
Team: Camille Carr, Halen Gifford, Eduardo Vetancourt
The goals of the AMRE Affordable Housing and Homelessness project were to research and recommend ways to make periods of homelessness rare, brief and non-recurring in Wayne County, and to identify the best affordable housing practices for the community to implement. Studies have shown that when communities experience affordable housing shortages, the community will also likely experience an increase in the homeless population.
Workforce Expansion through Supportive Employment
Advisors: Brooke Krause (Economics), Cameron Maneese
Team: Michelle Bryant, Maggie Dougherty, NoaH Plotkin
The project consisted of researching ways to expand the workforce in Wayne County through inclusive hiring and supportive employment methods. In doing so, the team developed an HR toolkit which includes key findings and policy recommendations. The team partnered with two local companies which are currently in the process of implementing the team model through a pilot trial.
Ohio Light Opera (OLO) — Market Research, Economic Impact & Marketing Analysis
Advisors: Jim Burnell (Economics), Grit Herzmann (Psychology)
Team: Emily Farmer, Connor Hawkins, Jean Roberts (Ashesi University)
The 2019 Ohio Light Opera AMRE project marks the first part of a two-year project to complete market research and develop a strategic plan for the company’s future. Students completed an analysis on previous audience statistics and created several surveys to further data collection about the company’s patron base and local business impact. The team also made recommendations based on performing arts company social media research and business outreach. The recommendations and work that this year’s team completed will be used to create a sustainable plan for increasing the Ohio Light Opera’s patron base and adjusting the company’s repertoire for the future of light opera in Ohio.
PRC-Saltillo — Russian Language in Speech-Generating Devices
Advisors: Tatiana Filimonova & Michael Furman (German & Russian Studies), Nathan Sommer (Computer Science)
Team: Harry Dunham, Gillian Gregory, David Sasu (Ashesi University), Erin Tupman
PRC Saltillo tasked the team to identify the morphological rules of the Russian language, then use these rules to create software in Python, which when given an input word form, would produce the proper output forms. As the client’s target language was Croatian, a secondary goal of the project was to examine the similarities between Russian and Croatian, identify rules if there was time, and determine what parts of the Russian code could be repurposed for the future Croatian code. As such, this project consists of both programming and Slavic linguistics components.
Wooster Community Hospital — Data-Driven Analysis of Health Care Outcomes
Advisors: Drew Pasteur (Mathematics), Jen Bowen (Mathematics)
Team: Alex Hwang, Harriet Sudduth, Dan Zuchelkowski
The project goal was to develop a mathematical model to predict patient outcomes. The team considered three years worth of data from the physical therapy department at HealthPoint. Variables such as clinical data, demographics, socioeconomic factors, and previous health conditions were considered. The findings show that socioeconomic factors and certain health conditions had the largest impact on physical therapy outcomes.
Progressive Insurance Company — Competitive Intelligence Analysis
Advisors: John Ramsay (Mathematics), Robert Kelvey (Mathematics)
Team: Zach Andreou, Aditi Chowbey, Zhen Guo
Progressive Insurance Company presented the College of Wooster AMRE team with a question within competitors analysis with respect to other leading auto insurance companies. The team was asked to identify specific risks that were receiving much higher or lower premiums that the market. The team developed two competitiveness metrics to use for this metric and the following comparitive analysis methods were used. Analysis methods included: Simple Linear Regression, Binary Logistic Regression, Random Forest, Decision Trees, Clustering, & Frequency Tables. The analysis and recommendations were submitted as part of a final report that identified key places were Progressive is outside the market as measured by the two metrics.
Admission Team
Advisors: John Ramsay (Mathematics), Robert Kelvey (Mathematics)
Team: Brett Stern, Shivam K.C.
The College of Wooster Admission Department wanted the AMRE team to determine if the college should go test optional.The team was given multiple datasets to clean and analyze in order to answer the research question.Additionally, the team was able to examine the factors that affect first-year retention.The major finding was the indication of SAT/ACT being valuable in predicting student success a the College of Wooster.
Mennonite Mutual Insurance
Advisors: Jen Bowen (Mathematics), Nathan Fox (Mathematics & Computer Science)
Team: Tom Kang, Judakwashe Ndebvudzemene (Ashesi University), Isaac Weiss
Mennonite Mutual currently gets loss costs through an industry-wide source.In order to better cusomize their pricing, they would like to replace this with estimates from their own claims data.The AMRE team analyzed historical data to determine a territory factor to be used in Mennonite Mutual’s pricing formula.Data Science techniques were applied to identify geographic region (by zip code) infuence on risk.The territory factor was also considered in the context of other risk factors available in the data set.
Goodyear
Advisors: Drew Pasteur (Mathematics), Nathan Fox (Mathematics & Computer Science)
Team: Immanuella Duke (Ashesi University), Jemal Jemal, David Pfeffer, Angelo Williams
The Research and Development team at Goodyear Tire and Rubber Company has contracted with AMRE team from the College of Wooster to utilize machine learning as a disciplinary approach in predicting the initial estimates for the parameters of the MACE (Multi-Stage Analysis and Constitutive Equation) model.The MACE material model is both nonlinear and implicit, and so it is difficult to fit the parameters that it requires.The previous implementation used an SAS JMP neural network, which was costly due to its paid licensure.Using a database of 200+ compounds with experimental data and MACE parameters, the team will build and train machine learning models in Python, using scikit-learn, Tensorflow, etc.AMRE’s approach utilizes free, open-source machine learning libraries to form the predictions.The goal is to approximate the high-accuracy existing model with accessible, open source software.
St. Mary of the Immaculate Conception
Advisor:Vikki Briggs (AMRE Coordinator)
Team:Abigail McFarren, Riley Smith
The AMRE team investigated enrollment and retention at St. Mary, a small Catholic school offering grades PreK through 8 in Wooster, Ohio.Data was collected from various comparable schools and a parish-wide survey was administered.The team researched enrollment and marketing strategies, compared enrollment trends with the general population and Wooster City Schools, and analyzed the parish-wide survey.The survey results identified factors that could help improve enrollment and retention.