Advanced Methods at Purdue (AMAP) in the Behavioral, Health, and Social Sciences

Other Related Events

Purdue Statistics Department Distinguished Theme Seminar Series 2022

Recent Advances in Statistical Inference, Fall 2022 

September 2, Friday (YouTube access)
10:30 – 11:30 a.m. EDT: On Dynamics-Informed Blending of Machine Learning and Game Theory, Professor Michael I. Jordan, University of California, Berkeley, Online.

September 7, Wednesday (YouTube access)
11:30 – 12:30 p.m. EDT: Conformal Prediction in 2022, Professor Emmanuel Candes - Stanford University, Online.

September 16, Friday (YouTube access)
10:30 – 11:30 a.m. EDT:  Miniaturizing Data Defect Correlation: A Versatile Strategy for Handling Non-Probability Samples, Professor Xiao-Li Meng, Harvard University, In person.

September 23, Friday (YouTube access
10:30 – 11:30 a.m. EDT:  Conformal Prediction Beyond Exchangeability, Professor Rina Barber, University of Chicago, In person.

More information can be found here.

Division D Summer Internship Opportunity 

Purdue provides a rich methodological environment for behavioral, health, and social scientists such as those who are affiliated with AMAP. However, there is a wide-ranging methodological community on campus who host talks, seminars, and other events that may be of interest. You can find a sampling of these below. 

NBME has additional opportunities for a part-time, fully remote summer internship position from June 6th to July 29th (dates negotiable). The internship opportunity would be for a maximum of 25 hours a week, although internships of 10 hours a week or 15 hours a week will be considered depending on the project match and the intern’s availability.

Interns will have the opportunity interact with other graduate students and NBME staff and will present completed projects or work-in-progress to NBME staff. The expected deliverables from the summer internship project are an internal research presentation, as well as a conference submission / presentation (AERA, NCME, AAMC, etc.) and/or a paper submitted for publication.

Requirements

  • Active enrollment in doctoral program in measurement, statistics, computer science, cognitive science, medical education, or related field; completion of two or more years of graduate coursework.
  • Must be authorized to work in the US for any employer. If selected, F-1 holders will need to apply for Curricular Practical Training authorization through their school’s international student office and have a social security number for payroll purposes.
  • Experience or coursework in one or more of the following: test development, IRT, CTT, statistics, research design, and cognitive science.
  • Skill in writing and presenting research.
  • Working knowledge of R; familiarity with R Shiny (only required for Project #2 listed below) 
  • Interns will be assigned to one or more mentors but must be able to work independently. 

Compensation: The pay is $37 an hour.

Research Projects

  1. Comparing the effectiveness of human raters and computer-based scoring. Human raters are required to answer short-answer questions, which can be costly, time-consuming, and prone to natural human error. When the expected response to a short-answer question is a specific phrase, there exist several approaches to quickly and effectively scoring short-answer items with very little human intervention. This project would compare several of these automated scoring methods to human-scored short-answer question responses.
  2. Building an R Shiny app to facilitate short-answer scoring. Using short-answer questions has several advantages over traditional multiple-choice items, but a barrier to their use is the time required to answer the questions. We are interested in building an R Shiny app that allows educators to more easily score short-answer questions utilizing some of the approaches mentioned above.
  3. Scoring of Complex Item Types. The measurement of complex mental processes or competencies often requires the use of item types that are different than standard MCQ items or dichotomously-scored free text response items. NBME has started a research program investigating the feasibility for new non-MCQ item types to reliably and validly measure aspects of clinical reasoning. The intern for this project will help explore scoring methods for these complex item types.

Candidates may apply by going to https://nbme.applicantpro.com/jobs/2388399.html.  A cover letter outlining experience and listing project interests by number, along with a current resume, are required. Applications will be reviewed on a rolling basis starting on Friday, May 27, 2022

A Series of Courses on Biostatistics for Health Care Researchers

The Department of Biostatistics in the School of Medicine and Richard M. Fairbanks School of Public Health will present a short course in biostatistics that is designed especially for health care researchers in the health sciences. This course will consist of three sessions. Sessions I and II will cover basic principles, design of medical research studies, standard statistical tests and data analyses, and data management. Session III will focus on more advanced topics, including multiple linear and logistic regression, survival analysis, longitudinal data and genetic analysis.

  • Participants will be able to learn
    • Recognize common study designs and statistical methods used in medical research.
    • Discuss complex study design and analysis with a statistician.
    • Specify basic concepts of data management.
    • Identify appropriate use of statistical procedures when given a common study design.
    • Implement simple statistical analyses under the guidance of a statistician.
  • Date: May 10 - 12 (Tue, Wed, & TR); 1:00 - 5:00 pm each day
  • Contents: Course will consist of three sessions
    • Attendance at Sessions I and II ($40)
    • Attendance at Sessions II and III ($40)
    • Attendance at Sessions I, II, and III ($60)
    • Suggested Instructional book: Basic and Clinical Biostatistics, 5th edition, © 2020, by White, ISBN# 978-1-260-46067-4
  • Further information: Biostatistics 

Myra Samuels Memorial Lecture 

 The Myra Samuels Memorial Lecture is named in memory of Myra L. Samuels, who was associate professor of biostatistics and epidemiology in Purdue's Department of Veterinary Pathobiology and associate director of Statistical Consulting in the Department of Statistics.

FSSA Day with The Data Mine: How healthcare data is utilized to drive data driven decision-making at FSSA? 

The event will be hosted by the Data & Analytics team from FSSA to showcase how healthcare data is being utilized to drive data-driven decision-making at FSSA. This event will serve two purposes: begin to recruit students for the Summer Internship and the upcoming Fall semester and further expand on collaborative efforts with Purdue. 

Women in Data Science Conference

The WiDS is a technical conference featuring outstanding women discussing their exceptional work in data science and related fields, in a wide variety of research domains. Visit the WiDS site for more information.

Integrative Data Science Initiative 

The IDSI is a University-wide initiative promoting robust interdisciplinary collaboration whose vision is to be at the forefront of advancing data science-enabled research and education by tightly coupling theory, discovery, and applications while providing students with an integrated, data science-fluent campus ecosystem. Visit their events page to find out about interesting data science talks across campus.

Statistics 

The statistics department at Purdue keeps a color-coded calendar of methods seminars and events across the University.

Digital Humanities/Digital Scholarship

Purdue Libraries also offer regular workshops on methods of interest to behavioral health and social scientists. Visit their site to see updates.

Additional Events

A "boot camp" on Machine Learning and Causal Inference presented by the Department of Statistics and the Krannert School of Management.