Chappell Family Gallery Exhibit Usability Test - June 2016

Emily Daly and Thomas Crichlow conducted a usability test of the Duke University Chapel exhibit displayed in the Chappell Family Gallery on May 20, 2016. The test was conducted to learn how people interact with the exhibit. The test consisted of two general questions, three tasks, and brief followup questions; each test took approximately 15 minutes to complete.

Online Journal Titles Usability Test - June 2016

Emily Daly and Gordon Chadwick conducted a think-aloud usability study on June 20, 2016 in the Perkins Library at Duke University. The study tested users’ perceptions of a new search results page for online journal titles.

Research databases page usability test - August 2016

During the summer of 2016, Duke University Libraries staff began a project to update the way that research databases are displayed on the library website. The new research databases page is a customized version of the default A-Z list that Springshare provides for its LibGuides content management system. Duke Libraries staff made adjustments to the content and interface of the page. In order to see how Duke users navigated the new interface, usability testing was conducted on August 9th, 2016.

Duke Images LibGuide usability test - August 2016

Thomas Crichlow and Gordon Chadwick conducted a think-aloud usability study on August 1, 2016 in the Perkins Library at Duke University. The study tested users’ perceptions of a new LibGuides driven research guide for scholarly images by asking them to complete information seeking tasks and provide feedback.

A Bayesian Dirichlet-Multinomial Test for Cross-Group Differences

Testing for differences within data sets is an important issue across various applications. Our work is primarily motivated by the analysis of microbiomial composition, which has been increasingly relevant and important with the rise of DNA sequencing. We first review classical frequentist tests that are commonly used in tackling such problems. We then propose a Bayesian Dirichlet-multinomial framework for modeling the metagenomic data and for testing underlying differences between the samples. A parametric Dirichlet-multinomial model uses an intuitive hierarchical structure that allows for flexibility in characterizing both the within-group variation and the cross-group difference and provides very interpretable parameters. A computational method for evaluating the marginal likelihoods under the null and alternative hypotheses is also given. Through simulations, we show that our Bayesian model performs competitively against frequentist counterparts. We illustrate the method through analyzing metagenomic applications using the Human Microbiome Project data.