Research Data for: Video Information Trustworthiness and Quality Assessment Scale (VITAL)

This dataset provides the raw and processed data used for the development and validation of the Video Information Trustworthiness and Quality Assessment Scale (VITAL). The study was conducted on a total of 420 YouTube videos. The dataset is organized into two primary files to support the sequential validation process described in the associated research paper: 1. EFA Dataset (n=210): Contains the evaluation scores used for Exploratory Factor Analysis to determine the initial factor structure of the 10-item VITAL scale. 2. CFA Dataset (n=210): Contains a separate sample of video evaluations used for Confirmatory Factor Analysis to validate the one-dimensional structure. Data Collection and Scoring: The items for the VITAL scale were generated using a novel approach integrating Large Language Models (LLMs) such as ChatGPT, Gemini, and DeepSeek. Videos were evaluated by field experts using a 5-point Likert scale. The final 10-item scale demonstrated high internal consistency with a Cronbach's alpha of 0.903 and a McDonald's Omega of 0.911. This data supports the findings presented in the manuscript submitted to Computers & Education.