Assessing library topics using sentiment analysis in R: a discussion and code sample

Abstract

This article discusses the use of R programing language for executing a sentiment analysis of tweets pertaining to library topics. This discussion is situated within the literature of marketing and management sciences, which is employing methods of machine learning and business intelligence to make informed decision-making, and library administration, which has expressed great interest in social media engagement within its literature but has yet to adopt these types of analysis. Presented in this article is a sample code with instructions on how users may execute it within R to retrieve and analyze tweets relevant to library services. Two examples created using the code (analysis of top librarians’ tweets and analysis of posts about major book publishers) are used to demonstrate the functionality of the code. The code presented in this article may be used by libraries to analyze tweets about their library and library-related topics, which, in turn, may inform management and marketing design.     

Effects of Multimedia versus Live Professional Development on Teachers’ and Students’ Performance Related to the Question Exploration Routine

In each of two studies, teachers were randomly assigned to either a Virtual Workshop (VW) group that used a computerized professional development program or an Actual Workshop (AW) group that participated in face-to-face professional development, including discussion, feedback, and collaboration. In both studies, teachers’ posttest scores related to their knowledge of the Question Exploration Routine and their plans for using it were significantly higher than their pretest scores. In Study 2, both groups’ posttest scores with regard to implementation and planning of the routine were significantly greater than their pretest scores. There were no significant differences between the groups at posttesting on any measure. The posttest knowledge scores of the whole groups of students and the subgroups of students with LD being taught by both groups of teachers were significantly higher than their pretest scores. All teachers indicated that they were satisfied with the training and the routine. VW teachers in both studies indicated that they were satisfied with the software program.