coe-staff: Colloquium Invitation: Educational Leadership Tenure-Track Faculty Candidate, Cengiz Zopluoglu

[Carmen Cybula]

Good morning!

This is just a reminder about the colloquium today at 9:30 in Lokey Ed 115. We hope to see you there!

Best,
Carmen

From: coe-staff-bounces at lists.uoregon.edu <coe-staff-bounces at lists.uoregon.edu> On Behalf Of Carmen Cybula
Sent: Wednesday, November 27, 2019 1:56 PM
To: 'coe-staff at lists.uoregon.edu' <coe-staff at lists.uoregon.edu>
Subject: Re: coe-staff: Colloquium Invitation: Educational Leadership Tenure-Track Faculty Candidate, Cengiz Zopluoglu

Hi, all!

One slight correction: This colloquium is scheduled from 9:30-11:00 am on Monday, December 9.

Best,
Carmen

From: coe-staff-bounces at lists.uoregon.edu<mailto:coe-staff-bounces at lists.uoregon.edu> <coe-staff-bounces at lists.uoregon.edu<mailto:coe-staff-bounces at lists.uoregon.edu>> On Behalf Of Carmen Cybula
Sent: Tuesday, November 26, 2019 3:47 PM
To: 'coe-staff at lists.uoregon.edu' <coe-staff at lists.uoregon.edu<mailto:coe-staff at lists.uoregon.edu>>
Subject: coe-staff: Colloquium Invitation: Educational Leadership Tenure-Track Faculty Candidate, Cengiz Zopluoglu

Good afternoon!

Please mark your calendars for the following colloquium presentation by Dr. Cengiz Zopluoglu, a candidate for Educational Methodology, Policy, and Leadership's tenure-track faculty search (CV attached). The event will include a 1-hour presentation and a 30-minute Q&A session. Coffee and snacks will be provided.
________________________________

Monday, December 9, 10:00-11:30am, Lokey Ed 115: Cengiz Zopluoglu, PhD
Detecting Examinees with Item Preknowledge in Large-Scale Testing Using Extreme Gradient Boosting (XGBoost)
In the area of detecting fraud in testing, there have been few studies that have used machine learning methods to identify potential testing fraud. In this study, a brief overview of a recently developed machine learning algorithm, Extreme Gradient Boosting (XGBoost), is provided and the utility of XGBoost in detecting examinees with potential item preknowledge is investigated using a real data set that includes examinees who engaged in fraudulent testing behavior, such as illegally obtaining live test content before the exam. The predictive performance of XGBoost models with different inputs is evaluated using the area under the receiving operating characteristic curve and several classification measures such as the false-positive rate, true-positive rate, and precision. Machine learning models are frequently criticized for their black-box nature and difficulty in explaining the outcomes. This talk will also discuss ways to visualize the results from an XGboost model to ease the interpretation of predictive outcomes for this specific context and problem.
-----------------
Dr. Cengiz Zopluoglu earned his PhD in 2013 in Quantitative Method in Education from the University of Minnesota. After earning his PhD, he joined the faculty in the Research, Measurement, and Evaluation program in the School of Education and Human Development at the University of Miami. His research focuses on item response theory, statistical detection of fraud in large-scale testing, and nonlinear mixed-effects models. He teaches courses on Measurement and Psychometric Theory, Item Response Theory, General Linear Models, Categorical Data Analysis, and Data Analysis using R in Educational and Behavioral Research.

Carmen Cybula
Office Support Specialist
Educational Methodology, Policy, and Leadership
University of Oregon | College of Education
541-346-5171
Pronouns: she/her/hers

[cid:image001.png at 01D5AE6C.45268130]


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists-prod.uoregon.edu/pipermail/coe-staff/attachments/20191209/cd837a55/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image001.png
Type: image/png
Size: 18230 bytes
Desc: image001.png
URL: <http://lists-prod.uoregon.edu/pipermail/coe-staff/attachments/20191209/cd837a55/attachment.png>