Couldn't make it to this year's conference at Southern Methodist University?
Click on the videos below to view a few of the presentations from the 7th Annual TUESAP Conference at SMU.
Christopher Thompson, Ph.D., Brandi Semma and Eunkyeng Baek
Matthias von Davier, Ph.D.
Anthony Sparks, Ph.D. and Beth Adams, Ph.D.
2019 Location, Date & Schedule
Conference Schedule (Updated: 2/22/19)
Harold Clark Simmons Hall / Rooms 101 A & D
Southern Methodist University
6401 Airline Road, Dallas, Texas 75205
February 28, 2019 (Thursday) / Room 101A
- Pre-Conference Workshops by Stata (12:00 - 4:00 pm) (Updated: 2/20/19)
- Download Stata 30-day trial version
- Email email@example.com for username and password
12:00 - 1:00 pm | Data management in Stata
1:00 - 2:00 pm | Missing Data Imputations in Stata
2:00 - 2:15 pm | Coffee Break
2:15 - 3:15 pm | Graphics in Stata
3:15 - 4:15 pm | Macros, ereturn, and scalars in Stata
- Welcome Reception (4:30 - 6:30 pm)
March 1, 2019 (Friday) / Rooms 101 A & D
- Conference (8:00 am - 4:30 pm) (Updated: 2/25/19)
8:00 - 9:00 am | Registration & Light Breakfast
9:00 - 9:15 am | Opening Remarks
9:15 - 10:30 am | Keynote speech Dr. Matthias von Davier
10:30 am - 10:40 | Break
10:40 am - 12:00 pm | Paper Presentations - Part I
12:00 - 1:00 pm | Lunch
1:00 - 2:20 pm | Roundtable Presentations
2:20 - 2:35 pm | Coffee Break
2:35 - 4:15 pm | Paper Presentations - Part II
4:15 - 4:30 pm | Closing Remarks and Adjourn
Matthias von Davier, Ph.D.
- Distinguished Research Scientist, National Board of Medical Examiners (NBME)
- Executive Editor, Psychometrika
While Data Science Is What You Say, Regression Is What You Do (Mostly)
Data science, machine learning, and artificial intelligence - some may want you to believe - are the new kid on the block and the next big thing in psychometrics and educational measurement. Having ventured into deep learning for automated item generation (von Davier, 2018a) and high performance computing for latent variable modeling (von Davier, 2017) as well as computational approaches for network psychometrics (von Davier, 2018b). This presentation will review some of these applications as well as their potential and limitations. These methods are not a panacea, as most have well defined, but not unlimited scopes of application. The big news is that psychometricians can learn a few new tricks by building on what they already know. As Gelman said at the IMPS 2018, "... many times a statistician comes up with a seemingly new idea, it is found that this was already published in some psychometric journal 75+ years ago." While psychometrics will not be replaced any time soon, researchers and practitioners need to get used to some new terminology, and may need to learn how some methods that have originated in our field have evolved over time as computer science and other fields adopted them.
Since 2013, TUESAP has held annual conferences, providing opportunity for faculty and graduate students in quantitative methods and psychometrics programs in universities throughout the state of Texas to share their research outcomes and to build a research network with the ultimate goal to better inform applied researchers and research practitioners using quantitative research methodology and educational / psychological measurement.