coe-staff: Dr. Perman Gochyyev -- Invited Colloquium

[Lisa Fortin]

College of Education Dean’s Invited Colloquium






Lord's Paradox and Consequences For Effects Of Interventions On Outcomes

Dr. Perman Gochyyev
October 30, 2019







Colloquium
When: October 30, 2019, 12:00-1:45 p.m.
Where: Lokey ED 176, College of Education


Student Seminar
When: October 30, 2019, 3:00-4:00 p.m.
Where: HEDCO 230T, College of Education







RSVP Colloquium<https://urldefense.com/v3/__https:/us7.mailchimp.com/mctx/click?url=https*3A*2F*2Foregon.qualtrics.com*2Fjfe*2Fform*2FSV_daOjWGoL5aV4fM9&xid=8e07797549&uid=109187122&pool=&subject=__;JSUlJSUl!5W9E9PnL_ac!TIJCfnv-jmTRkVAPht7EibSMKJl1I9u9lNA2KbDFp5raiAEYAR2fLxC6wAmFtAem9g$>



RSVP Student Seminar<https://urldefense.com/v3/__https:/us7.mailchimp.com/mctx/click?url=https*3A*2F*2Foregon.qualtrics.com*2Fjfe*2Fform*2FSV_da1XTifrB4IclRb&xid=8e07797549&uid=109187122&pool=&subject=__;JSUlJSUl!5W9E9PnL_ac!TIJCfnv-jmTRkVAPht7EibSMKJl1I9u9lNA2KbDFp5raiAEYAR2fLxC6wAnPhLNh8w$>



Abstract: The main focus of this talk is on the Lord’s paradox within the latent variable modeling framework. Lord (1967) describes a hypothetical paradox in which two researchers, analyzing the same dataset using different but defensible methods, come to very different conclusions about the effects of an intervention on outcomes. Lord’s paradox has re-emerged in many causal inference settings today around the issue of when it is appropriate to control for baseline status. Recent applications can be seen in areas such as estimating the local average treatment effect of special education services on student performance; the birth-weight paradox about the relationship between the birth weight and mortality rate of children born to tobacco smoking mothers; and potential over-medicalization of some mental health issues. This talk will discuss two main approaches for analyzing such data: (1) to regress the change from pretest to posttest on the treatment indicator; and (2) to regress posttest on treatment indicator and pretest. The talk will show how these two approaches can yield conflicting results—hence the paradox. I will elaborate on how approaches can be addressed within the structural equation modeling framework and provide examples from my research.













[https://gallery.mailchimp.com/1f6ab32a744b7c1084c7e4bd6/images/c4c32141-0242-4628-a10e-d31b5e8e2e08.jpg]

Perman Gochyyev, PhD, is a behavioral statistician and research center project manager at the University of California, Berkeley in the Berkeley Evaluation and Assessment Research (BEAR) Center. Dr. Gochyyev received his doctorate in Quantitative Methods and Evaluation from UC Berkeley in 2015 in the Policy, Organizations, Measurement & Evaluation (POME) area. His research focuses on latent variable and multilevel modeling, issues related to causal inference in educational and behavioral statistics, multidimensional and ordinal IRT models, and latent class models.




[https://gallery.mailchimp.com/1f6ab32a744b7c1084c7e4bd6/images/8fe5907e-8d3d-4cb7-9f6f-89d7932006ac.png]





[Facebook]<https://urldefense.com/v3/__https:/us7.mailchimp.com/mctx/click?url=https*3A*2F*2Fwww.facebook.com*2Fuoeducation*2F&xid=8e07797549&uid=109187122&pool=&subject=__;JSUlJSU!5W9E9PnL_ac!TIJCfnv-jmTRkVAPht7EibSMKJl1I9u9lNA2KbDFp5raiAEYAR2fLxC6wAmiahiyGw$>




[Twitter]<https://urldefense.com/v3/__https:/us7.mailchimp.com/mctx/click?url=https*3A*2F*2Ftwitter.com*2Fuoeducation&xid=8e07797549&uid=109187122&pool=&subject=__;JSUlJQ!5W9E9PnL_ac!TIJCfnv-jmTRkVAPht7EibSMKJl1I9u9lNA2KbDFp5raiAEYAR2fLxC6wAnxdEkUeg$>




[Link]<https://urldefense.com/v3/__https:/us7.mailchimp.com/mctx/click?url=https*3A*2F*2Fwww.instagram.com*2Fuoeducation*2F*3Fhl*3Den&xid=8e07797549&uid=109187122&pool=&subject=__;JSUlJSUlJQ!5W9E9PnL_ac!TIJCfnv-jmTRkVAPht7EibSMKJl1I9u9lNA2KbDFp5raiAEYAR2fLxC6wAnfAAGSdQ$>




[Website]<https://urldefense.com/v3/__https:/us7.mailchimp.com/mctx/click?url=https*3A*2F*2Feducation.uoregon.edu&xid=8e07797549&uid=109187122&pool=&subject=__;JSUl!5W9E9PnL_ac!TIJCfnv-jmTRkVAPht7EibSMKJl1I9u9lNA2KbDFp5raiAEYAR2fLxC6wAl2kbwpuQ$>













-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists-prod.uoregon.edu/pipermail/coe-staff/attachments/20191010/65b36cb9/attachment.html>