Using linguistic inquiry and word count software to analyze group interaction language data.

Abstract

Objective: Automated text analysis tools can uncover emergent group processes from communication data. The article presents an illustration aimed at novice users of how the Linguistic Inquiry and Word Count (LIWC) software tool reveals group processes from interaction data collected. Using a data set from published research, we apply the group interaction model (GIM) to illustrate how to operationalize a group interaction hypothesis from the variables provided by LIWC. Then, we guide the reader through the steps of preparing transcripts for analysis in LIWC, running the transcripts to get data from LIWC, and then merging spreadsheets into one data set for analysis. In addition, the article provides instructions and resources to manage the complexity of linguistic measures gathered at various time intervals. Method: We provide data preparation tips, instructions for using the fifth-generation version of the LIWC-22 tool to analyze group process data, and resources for combining the LIWC measures to reflect their time interval and to combine them into one file with timeinvariant group data. Results: In addition to providing accessible instructions, the article details several applications, such as naturalistic group data gathered across time and pre/post experimental design that increase statistical power. It also provides an annotated bibliography for further reading. Conclusion: The linguistic style of group communication often reveals group dynamics. Using the GIM, this article outlines how to uncover emergent group processes from communication data gathered at time intervals using LIWC-22. Highlights and Implications• The group interaction research landscape has transformed from one of paucity to "big" data abundance, calling for new group language analysis theory and methods.• We illustrate how emergent processes outlined in the group interaction model (GIM) may be converted into linguistic analysis hypotheses that are amendable to automated text analysis.• We discuss what the Linguistic Inquiry and Word Count (LIWC) is and how the variables LIWC-22 provides map onto GIM processes.• A data set from published articles is used to present a step-by-step example with tips to prepare transcripts, instructions on how to analyze transcripts with LIWC-22, and then instructions on how to merge the resulting data sets into a central data set for subsequent statistical analysis. • We provide Supplemental Materials detailing how to (a) use LIWC-22 to segment transcripts, (b) conduct an LIWC-22 analysis, and (c) use Python code to relabel LIWC data gathered across time and combine it with time-invariant data.This document is copyrighted by the American Psychological Association or one of its allied publishers.This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

'A Torch, a rope, a belly laugh’: Engaging with the multiple voices of support groups for people living with rare dementia

“…Firstly, we investigated how faithful the completed poems were to the source material (initial responses to the prompts). To do so, linguistic analysis and word count software (Linguistic Inquiry and Word Count-22 (LIWC-22)(42) was used to analyse language data (43)(44). It was important to determine if the completed poems re ected the words drawn from the source material in order to feel con dent that we could conduct a thematic analysis using the poems.…”

What Motivates the Vaccination Rift Effect? Psycho-Linguistic Features of Responses to Calls to Get Vaccinated Differ by Source and Recipient Vaccination Status

“…For the current analysis, we obtained the original dataset and analyzed the responses to the free-text question on participants’ opinion about the coronavirus vaccination. In line with recent recommendations [ 29 ], a research assistant blind to the studies’ hypotheses spell-checked and proof-read all comments. Obvious typographic errors were corrected but no other alterations were made to the texts.…”

“…In line with this view, linguistic analyses can identify ongoing psychological processes such as personality [ 26 ], relations between groups [ 27 ], and processes within groups [ 28 ]. Linguistic analyses have also proven useful in investigating a range of phenomena, including group interaction processes [ 29 ], communication [ 30 ], and societal divides [ 31 ]. Analyzing the words people use rather than their responses to a Likert-type scale has the advantage that participants can give free responses.…”

Verbal behavior and the future of social science.

“…10 Surprisingly, computerized language analysis research rarely takes social factors into account, despite its near-universally recognized importance (Boyd & Schwartz, 2021;Yeomans et al, 2023). However, recent innovations in conversational theory may help to energize and empower NLP scholars to integrate contextual features into psychological accounts of verbal behavior, including contemporary thinking about group settings as dynamic situations that demand individuals adapt their verbal behaviors in response to interpersonal and group needs and motives (Gerpott et al, 2019;Kane & van Swol, 2022;Yeomans et al, 2022).…”