Search results

This study aims to explore the evolutionary pattern of discussion topics over time in an online depression self-help community.

Using the Latent Dirichlet Allocation (LDA) method, the authors analyzed 17,534 posts and 138,567 comments posted over 8 years on an online depression self-help group in China and identified the major discussion topics. Based on significant changes in the frequency of posts over time, the authors identified five stages of development. Through a comparative analysis of discussion topics in the five stages, the authors identified the changes in the extent and range of topics over time. The authors discuss the influence of socio-cultural factors on depressed individuals' health information behavior.

The results illustrate an evolutionary pattern of topics in users' discussion in the online depression self-help group, including five distinct stages with a sequence of topic changes. The discussion topics of the group included self-reflection, daily record, peer diagnosis, companionship support and instrumental support. While some prominent topics were discussed frequently in each stage, some topics were short-lived.

While most prior research has ignored topic changes over time, the study takes an evolutionary perspective of online discussion topics among depressed individuals. The authors provide a nuanced account of the progression of topics through five distinct stages, showing that the community experienced a sequence of changes as it developed. Identifying this evolutionary pattern extends the scope of research on depression therapy in China and offers a deeper understanding of the support that individuals with depression seek, receive and provide online.

This article examines if there is evidence of racial or gender bias in email reference services in American public and academic libraries.

Using a two-by-two study design and an unobtrusive data collection, the authors conducted two studies in which the authors sent 1,960 email requests to 505 academic and public libraries. Requests in both studies differed in the perceived identity of the user as indicated by their name, and the counterbalanced method was utilized to control for intervening variables. Based on content analysis of the responses, the authors examined the statistical significance of the differences by race, gender and race by gender.

Overall, the authors found equitable service to users regardless of their race and gender; at times, however, there was evidence of favorable service to the White female in academic and public libraries and to the Black male in academic libraries.

There is little research into potential bias in email reference services in both academic and public libraries in the United States of America. Yet, following the rise of the Black Lives Matter Movement in 2020, there has been an increased focus on racial equality in library services and the American Library Association (ALA) Code of Ethics was modified accordingly. The authors' study makes significant contributions to the increasing body of research on racial and gender equality in online library services.