Behavioral Usability Evaluation of the Social Media Language Tool
Year: 2026
Type: article
Abstract: This study investigates how users cognitively interpret online identity, emotion, and generational communication styles within a theory-based dashboard designed to classify identity from textual data. The dashboard, operationalized as the Social Media Language (SL) Tool, is grounded in the Generational Virtual Textual Identity (GVTI) framework, which integrates emoji use, sentiment, personality traits, thematic classification, and generational communication patterns. To evaluate how users engage with these outputs, a mixed-methods behavioral usability study was conducted with nineteen participants, combining Likert-scale ratings, open-ended feedback analyzed through thematic analysis, and eye-tracking measures (heatmaps). This triangulated approach examined perceived interpretability and cognitive processing rather than predictive accuracy. Findings indicate that users focused primarily on identity labels and emojis as interpretive anchors, with gaze patterns converging with qualitative feedback. All Likert-scale items exceeded the predefined usability threshold, indicating high perceived clarity and trust in the framework outputs. However, participants also identified the need for domain-specific refinement of classification labels and additional interpretive aids, such as tooltips and contextual summaries. Rather than constituting theoretical validation, the results provide initial empirical support for the interpretive coherence of the GVTI framework and demonstrate the value of behavioral usability methods in evaluating identity-based media tools. By integrating qualitative, quantitative, and behavioral evidence, this study contributes to ongoing debates in media studies and human–computer interaction (HCI) regarding interpretability, transparency, and user-centered design in online dashboard environments.
Authors: Pratik Emmanuel, Sakirulai Olufemi Isiaq
Language: English
Subfield: Human-Computer Interaction
Field: Computer Science
Domain: Physical Sciences
Open Access status: hybrid
OpenAlex ID: W7161176574
FWCI:
Cites:
Cited by:
Related to: