Presentation
Automated item generation (AIG): A future for more equitable measurement
September 2020
By: Sarah Miller, Sylvia Scheuring, Abby Javurek, Jake McMillen, Emily Bo, Mary Ann Simpson
Description
This presentation summarizes the work that’s been done at NWEA to implement automated item generation (AIG), in addition to traditional approaches, to leverage the future.
See the presentationTopics: Item response theory
Related Topics
A comparison of item parameter estimates in Pychometrik and the existing item calibration tool
This study evaluates the performance of Pychometrik, a new item calibration tool, against the existing tool for MAP Growth.
By: Wei He, Patrick Meyer, Russ Grandgeorge, Emily Bo
Topics: Item response theory
Calibration of Spanish MAP Growth Math tests
The calibration study was conducted to investigate the consequences of replacing the English MAP Growth Mathematics item parameters with the Spanish mathematics item parameters.
By: Shudong Wang, Xueming (Sylvia) Li
Products: MAP Growth, MAP Spanish
MAP Growth K-2 item fit analysis study: A follow-up study
This is a follow-up analysis to the May 2019 item fit analysis study to evaluate the fit of the items whose item difficulties were adjusted in the more recent 2020 follow-up scale alignment study.
By: Wei He
Products: MAP Growth
Topics: Item response theory
MAP Growth Kā2 item fit analysis study
The purpose of this study is to examine the fit of MAP Growth Kā2 items involved in a scale alignment study conducted by NWEA to realign the scales underlying the MAP Growth Kā2 and MAP Growth 2ā5 Reading and Mathematics tests.
By: Wei He
Products: MAP Growth
Topics: Item response theory
This technical report is written for measurement professionals and administrators to help evaluate the quality of the MAP Growth assessments.
By: Patrick Meyer
Products: MAP Growth
Topics: Measurement & scaling, Item response theory, Test design
Robust IRT scaling: Considerations in constructing item bank from tests across years
This study investigates the impact of three different IRT scaling and equating methods in building an item bank of tests from 23 years of a national licensure exam . The study focuses on several key psychometric issues including scaledriftandequatingerrors.
By: Jungnam Kim, Dong-In Kim, Furong Gao
Topics: Measurement & scaling, Computer adaptive testing, Item response theory
These studies are conducted based on assumptions under regular conditions for fixed test forms, such as no missing responses and normal distribution of unidimensional ability for a population.
By: Shudong Wang, Hong Jiao
Topics: Measurement & scaling, Computer adaptive testing, Item response theory