Dataset Information

A Framework for Anchor Methods and an Iterative Forward Approach for DIF Detection.

ABSTRACT: In differential item functioning (DIF) analysis, a common metric is necessary to compare item parameters between groups of test-takers. In the Rasch model, the same restriction is placed on the item parameters in each group to define a common metric. However, the question how the items in the restriction-termed anchor items-are selected appropriately is still a major challenge. This article proposes a conceptual framework for categorizing anchor methods: The anchor class to describe characteristics of the anchor methods and the anchor selection strategy to guide how the anchor items are determined. Furthermore, the new iterative forward anchor class is proposed. Several anchor classes are implemented with different anchor selection strategies and are compared in an extensive simulation study. The results show that the new anchor class combined with the single-anchor selection strategy is superior in situations where no prior knowledge about the direction of DIF is available.

SUBMITTER: Kopf J

PROVIDER: S-EPMC5978507 | biostudies-literature | 2015 Mar

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

A Framework for Anchor Methods and an Iterative Forward Approach for DIF Detection.

Kopf Julia J Zeileis Achim A Strobl Carolin C

Applied psychological measurement 20140825 2

In differential item functioning (DIF) analysis, a common metric is necessary to compare item parameters between groups of test-takers. In the Rasch model, the same restriction is placed on the item parameters in each group to define a common metric. However, the question how the items in the restriction-termed <i>anchor items</i>-are selected appropriately is still a major challenge. This article proposes a conceptual framework for categorizing anchor methods: The <i>anchor class</i> to describ ...[more]

PMID: 29880995

Similar Datasets

Project description:BackgroundCholera poses a significant global health burden. In Bangladesh, cholera is endemic and causes more than 100,000 cases each year. Established environmental reservoirs leave millions at risk of infection through the consumption of contaminated water. The Global Task Force for Cholera Control has called for increased environmental surveillance to detect contaminated water sources prior to human infection in an effort to reduce cases and deaths. The OmniVis rapid cholera detection device uses loop-mediated isothermal amplification and particle diffusometry detection methods integrated into a handheld hardware device that attaches to an iPhone 6 to identify and map contaminated water sources.ObjectiveThe aim of this study was to evaluate the usability of the OmniVis device with targeted end users to advance the iterative prototyping process and ultimately design a device that easily integrates into users' workflow.MethodsWater quality workers were trained to use the device and subsequently completed an independent device trial and usability questionnaire. Pretraining and posttraining knowledge assessments were administered to ensure training quality did not confound trial and questionnaire.ResultsDevice trials identified common user errors and device malfunctions including incorrect test kit insertion and device powering issues. We did not observe meaningful differences in user errors or device malfunctions accumulated per participant across demographic groups. Over 25 trials, the mean time to complete a test was 47 minutes, a significant reduction compared with laboratory protocols, which take approximately 3 days. Overall, participants found the device easy to use and expressed confidence and comfort in using the device independently.ConclusionsThese results are used to advance the iterative prototyping process of the OmniVis rapid cholera detection device so it can achieve user uptake, workflow integration, and scale to ultimately impact cholera control and elimination strategies. We hope this methodology will promote robust usability evaluations of rapid pathogen detection technologies in device development.

Dataset Information

A Framework for Anchor Methods and an Iterative Forward Approach for DIF Detection.

Publications

A Framework for Anchor Methods and an Iterative Forward Approach for DIF Detection.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure