[Forum SIS] Seminario Joseph B. Lang

[manuela cazzaro]

 Annuncio di seminario:
(scusate per le informazioni precedenti incomplete)

*giovedì 15/03/2018 ore 11.00*

Aula Seminari (edificio U7, 4° piano, 4026)
Via Bicocca degli Arcimboldi, 8 – 20126 Milano
Dipartimento di Statistica e Metodi Quantitativi
Università di Milano-Bicocca

A RANDOMIZATION P-VALUE TEST FOR DETECTING
COPYING ON MULTIPLE-CHOICE EXAMS

Joseph B. Lang
University of Iowa - Department of Statistics and Actuarial Science -

Abstract
This talk is concerned with the statistical detection of copying on
multiple-choice exams.
As an alternative to existing permutation- and model-based copy-detection
approaches,
a very simple randomization p-value (RP) test is proposed. The RP test,
which is based
on an intuitive match-score statistic, makes no assumptions about the
distribution of
examinees' answer vectors and hence is broadly applicable. Especially
important in this
copy-detection setting, the RP test is shown to be exact in that its size
is guaranteed to
be no larger than a nominal alpha value. Additionally, simulation results
suggest that the
RP test is typically more powerful for copy detection than existing
approximate tests,
including the commonly used omega index test. The development of the RP
test is
based on the idea that the copy-detection problem can be re-cast as a
causal inference
and missing data problem. In particular, the observed data are viewed as a
subset of a
larger collection of potential values, or counterfactuals, and the null
hypothesis of ``no
copying" is viewed as a ``no causal effect" hypothesis and formally
expressed in terms
of constraints on potential variables.

Tutti gli interessati sono invitati a partecipare.

Per informazioni: manuela.cazzaro a unimib.it.




2018-03-08 17:58 GMT+01:00 manuela cazzaro <manuela.cazzaro a unimib.it>:

> Annuncio di seminario:
>
> A RANDOMIZATION P-VALUE TEST FOR DETECTING
> COPYING ON MULTIPLE-CHOICE EXAMS
>
> Joseph B. Lang
> University of Iowa - Department of Statistics and Actuarial Science -
>
> Abstract
> This talk is concerned with the statistical detection of copying on
> multiple-choice exams.
> As an alternative to existing permutation- and model-based copy-detection
> approaches,
> a very simple randomization p-value (RP) test is proposed. The RP test,
> which is based
> on an intuitive match-score statistic, makes no assumptions about the
> distribution of
> examinees' answer vectors and hence is broadly applicable. Especially
> important in this
> copy-detection setting, the RP test is shown to be exact in that its size
> is guaranteed to
> be no larger than a nominal alpha value. Additionally, simulation results
> suggest that the
> RP test is typically more powerful for copy detection than existing
> approximate tests,
> including the commonly used omega index test. The development of the RP
> test is
> based on the idea that the copy-detection problem can be re-cast as a
> causal inference
> and missing data problem. In particular, the observed data are viewed as a
> subset of a
> larger collection of potential values, or counterfactuals, and the null
> hypothesis of ``no
> copying" is viewed as a ``no causal effect" hypothesis and formally
> expressed in terms
> of constraints on potential variables.
>
> Tutti gli interessati sono invitati a partecipare.
>
> Per informazioni: manuela.cazzaro a unimib.it.
>
-------------- parte successiva --------------
Un allegato HTML è stato rimosso...
URL: <http://www.stat.unipg.it/pipermail/sis/attachments/20180309/548b85dc/attachment-0001.html>