{"id":169,"date":"2015-07-21T11:20:49","date_gmt":"2015-07-21T16:20:49","guid":{"rendered":"https:\/\/wp.stolaf.edu\/cir\/?page_id=169"},"modified":"2015-07-21T11:38:48","modified_gmt":"2015-07-21T16:38:48","slug":"projects-for-2015-16","status":"publish","type":"page","link":"https:\/\/wp.stolaf.edu\/cir\/projects-for-2015-16\/","title":{"rendered":"Projects for 2015-16"},"content":{"rendered":"

 <\/p>\n

Determining the randomness of neural activity <\/strong><\/p>\n

Description<\/strong>: The activity of Head direction\u00a0(HD) cells in the rodent brain is thought to represent the neural correlate for a sense of direction. In some ways these neurons function as a \u201cneural compass\u201d as each individual neuron is only active when an animal\u2019s head points in a specific direction. Recent research in our lab has shown that HD cells appear to lose all directionality while the animals are unconscious under anesthesia, but these neurons quickly regain their directional signal upon the animal regaining consciousness. We have become increasingly interested in understanding how this directional signal is reinstated in the brain. One possibility is that the apparently random firing of HD cells under anesthesia is not truly random. Instead, there is some directional signal still preserved (albeit degraded) in the cell\u2019s firing activity that the animal uses to reinstate the directional signal upon regaining consciousness. Students working on this project will devise and implement ways of analyzing data from the firing activity of individual neurons for randomness.<\/p>\n

Domain Expert<\/strong>: Gary Muir (Psychology\/Neuroscience)<\/p>\n

 <\/p>\n

 <\/p>\n

Human Scent Project <\/strong><\/p>\n

Description<\/strong>: Our lab is involved in studying human scent profiles, in support of canine tracking dogs. Trained dogs can follow the scent of a person over a mile or more, up to several days after the person has left the trail. The current hypothesis is that the dogs are smelling the volatile organic compounds (VOCs) given off by dead epithelial (skin) cells shed by the person as he\/she walks. We are collecting epithelial cells from volunteers and determining in the lab what VOCs are given off by each person using gas chromatography-mass spectrometry (GC-MS). There are several aims of this work. The first is to show that each person has a unique scent profile that the dog could use to distinguish one person\u2019s trail from those of other people. We routinely detect 50 or more compounds emanating from human epithelial cells, in various quantities. The identities of many of these compounds are the same between people, but there are some differences observed as well. The amount of each compound can also vary between people. Aside from hopefully demonstrating that everyone has a unique scent profile, we also want to investigate if there are class characteristics in the profiles. Along with each sample, we are collecting data on gender, age, and ethnicity. We are interested in applying statistics, likely principle component analysis, to determine for example whether scent profiles of a given age group can be classified together, or if gender determination can be made based on the scent profile. If class characteristics can be identified, we would then want to build decision trees that might allow the age, gender, ethnicity, etc. to be determined based on the VOC scent profile observed.<\/p>\n

Domain Expert<\/strong>: Doug Buessman (Chemistry)<\/p>\n

 <\/p>\n

 <\/p>\n

Communities that vote: Electoral participation in low-income counties<\/strong><\/p>\n

Description<\/strong>: One of the most important debates in the study of political behavior is the question of why people vote (or often, why they do not).\u00a0 Explanations for non-voting are too numerous to enumerate in full \u2013ranging from the role of political institutions to campaign contact and mobilization efforts to childhood socialization and civic habits. While these explanations are all markedly different, each account of non-voting must take stock of the robust and enduring relationship between income and turnout.\u00a0 Specifically, the wealthy have vastly higher election turnout rates, a distortion which in turn leads to inequalities in political representation and public policy.\u00a0 This project seeks to understand the connection between income and voting by examining this connection in different contexts.\u00a0 Instead of looking at survey databases of voters as is often done (an approach which is valuable but has several limitations), this project will examine turnout at the county level, with a particular interest in a small handful of relatively poor counties that \u201coverperform\u201d in terms of political participation.\u00a0 In other words, while most poor counties have low participation rates, this is by no means universal.\u00a0 We will examine features that covary with high turnout in low income counties, including electoral competition, electoral institutions, mobilization efforts, and social capital. We hope to identify how the features of a community can facilitate (or hinder) higher electoral participation.<\/p>\n

This project involves overcoming several statistical challenges.\u00a0 Building a county-level measure of participation is not as straightforward as it might seem, given inconsistencies in election types, and differential felon disenfranchisement laws across states. We will also need to address ecological inference issues involved with understanding the behavior of low-income voters within counties.\u00a0 Relatedly, this project involves nested data structures (people within counties within states), which could present analysis challenges. Finally, we will be exploring a long list of covariates of voting behavior, each which involve unique challenges.<\/p>\n

Domain Expert<\/strong>: Chris Chapp \u2013 Political Science<\/p>\n

 <\/p>\n

 <\/p>\n

Is the Medical Imaging Slowdown Driven by Cost-Sharing or Information?<\/strong><\/p>\n

 <\/p>\n

Description<\/strong>: For several decades, hospitals rapidly increased their use of medical imaging, such as MRIs and CT scans conducted on patients.\u00a0 This added greatly to the cost of care, and some argued that the high use of imaging was unnecessary or driven by defensive medicine intended to protect doctors from lawsuits rather than helping patients.\u00a0 In recent years, there has been a slowdown in the use of medical imaging.\u00a0 Research on the slowdown suggests two possible reasons for it:
\n(a) Information<\/u>.\u00a0 Published medical research has questioned the benefit of excessive imaging for patients.\u00a0 Doctors may be responding to this evidence by recommending fewer scans.<\/p>\n

(b) Cost-sharing<\/u>. Insurance companies and Medicare have changed payment structures to require patients to pay a higher share of the bill for imaging.\u00a0 Patient demand, therefore, may be playing a role.<\/p>\n

This project seeks to better understand which of these two forces contributes more greatly to the imaging slowdown.<\/p>\n

 <\/p>\n

To test these hypotheses, this project will look at the timing of the imaging slowdown across different hospitals.\u00a0 From an economic standpoint, the imaging slowdown could be viewed as a medical technology that slowly (or quickly) diffuses across different hospitals.\u00a0 Hospitals that were the earliest to slow down their use of imaging are key players the imaging slowdown trend.\u00a0 What hospital characteristics make a hospital an \u201cearly adopter\u201d of less medical imaging?\u00a0 If the imaging slowdown is driven primarily by cost-sharing, then a hospital\u2019s payer mix would be an important factor associated with the timing of the hospital\u2019s slowdown.\u00a0 On the other hand, if information is driving the slowdown, then teaching hospitals, research hospitals, and hospitals with more technology adoption will be more likely to be early adopters.\u00a0 It is also likely that both factors may play a role.\u00a0 In fact, the working hypothesis is that both information-connectedness and changes in cost-sharing are necessary to drive early change.\u00a0 We can test this hypothesis using interaction variables between our measures of cost-sharing and information connectedness.<\/p>\n

Methodology<\/u>:\u00a0 The project would use an extensive patient-level data set on California patients. The project might employ hazard rate models and\/or structural change models to identify each hospital\u2019s timing of the slowing in medical imaging.<\/p>\n

Domain expert<\/strong>: Prof. Ashley Hodgson<\/p>\n

 <\/p>\n

 <\/p>\n

Analysis of Pre-health Profession Student Trajectories<\/strong><\/p>\n

Description:\u00a0<\/b>St. Olaf students seeking to pursue careers in health professions follow many paths.\u00a0 Under the direction of Professor Kevin Crisp, chair of the Health Professions Program, and Susan Kramer of the Piper Center, this project will work with data pulled from the MD and DO centralized application systems, the American Medical College Application Service (AMCAS) and the American Association Of Colleges of Osteopathic Medicine Application Service (AACOMAS).\u00a0 The plethora of information available from these sources be used to perform an analysis of pre-health profession experiences nationally. \u00a0We\u2019ll seek to model how sex, major, GPA, sciences GPA, MCAT, state of residence, and state of the medical school figure into acceptance.\u00a0 The new MCAT will pose a special challenge.\u00a0 This national analysis will then be linked to St. Olaf student experiences to support and inform pre-health advising at St. Olaf.\u00a0 In addition to pre-medical school experiences, experiences in Physical Therapy, Dental Health, Physician\u2019s Assistant, Occupational Therapy Public Health, and Health Care Administration will be analyzed as time permits.<\/p>\n

Domain Experts:<\/strong> Kevin Crisp and Susan Kramer<\/p>\n

 <\/p>\n

 <\/p>\n

Assessing a Professional Learning Community (PLC): Northfield Schools.<\/strong><\/p>\n

Description:\u00a0<\/b>In an effort to systematize their mode of delivering school-based professional development, Northfield public schools utilizes a unique model. Each Wednesday, all district schools have a one-hour late start to enable school teachers and staff members to meet in a Professional Learning Community (PLC). While there is anecdotal evidence that the Northfield PLC is having positive impacts on teacher practice and student learning, there has been no attempt to gather and disseminate any information regarding the development and implementation or effects of the PLC. Researchers on this project have collected and coded survey and interview data from Northfield Public School administrators, teachers, and parents regarding the history, implementation, and effects of the PLCs. Working with the CIR, it is hoped that a more thorough analysis of patterns in the data can be discovered.<\/p>\n

 <\/p>\n

Our main research questions for this CIR project are:<\/p>\n