From Exposure to Elaboration: A Longitudinal Dataset on Short-Video Learning and Conceptual Retention

Description 1. Overview This dataset contains longitudinal data designed to examine how short-video microlecture exposure contributes to conceptual retention through cognitive and motivational learning processes. The study adopts a process-oriented framework in which learning outcomes are understood as the result of dynamic interactions between behavioral engagement, perceived instructional alignment, motivational states, and deep cognitive processing over time. 2. Research Context Data were collected from undergraduate students across three measurement waves within a single academic semester. The dataset integrates both self-reported measures and performance-based assessments, allowing for the simultaneous analysis of subjective learning experiences and objective learning outcomes in digital learning environments. 3. Core Constructs The dataset includes six primary constructs capturing different dimensions of the learning process: Short-video Exposure (EXP): Behavioral engagement with microlecture videos, including viewing frequency, completion, and attentional focus. Content Personalization (PER): Perceived alignment between instructional content and individual learning needs, prior knowledge, and learning preferences. Digital Media Self-Efficacy (DMSE): Learners’ confidence in navigating, evaluating, and managing digital learning environments. Situational Interest (INT): Immediate motivational engagement during learning activities, including attention, enjoyment, curiosity, and perceived meaningfulness. Elaboration (ELAB): Deep cognitive processing strategies, including self-explanation, prior knowledge integration, and conceptual comparison. Conceptual Retention (RET): Performance-based learning outcomes measured through recall, explanation, integration, and transfer tasks. Each retention dimension is scored from 0 to 25, with total retention ranging from 0 to 100. 4. Longitudinal Structure The dataset follows a three-wave structure: T1: Baseline measures, including exposure, personalization, and self-efficacy T2: Intermediate learning outcomes and cognitive processing T3: Final learning outcomes and retention This structure enables the examination of temporal dynamics, including lagged relationships and developmental learning processes across time. 5. Measurement Approach The dataset combines: Likert-scale survey responses (1–7 scale) capturing perceptions and experiences Objective performance-based assessments evaluating conceptual understanding This dual-measurement approach supports more robust analysis of learning processes by linking internal states with observable outcomes. 6. Analytical Potential The dataset is suitable for advanced statistical analyses, including: Structural Equation Modeling (SEM) Longitudinal mediation analysis Cross-lagged panel modeling Growth and repeated-measures analysis The design allows researchers to investigate how early exposure and engagement translate into long-term learning outcomes through intermediate cognitive mechanisms. 7. Research Significance This dataset provides a comprehensive empirical foundation for studying learning processes in digitally mediated environments, particularly within short-video microlearning formats. It supports research on how engagement must be transformed into sustained cognitive processing in order to produce durable and transferable learning outcomes over time.数据集说明 1. 概述 本数据集包含纵向数据，旨在通过认知与动机性学习过程，探究短视频微课程（short-video microlecture）接触量对概念留存的影响。本研究采用过程导向框架，将学习成果理解为随时间推移，行为投入、感知教学匹配度、动机状态与深度认知加工之间动态交互的结果。 2. 研究背景 数据采集自单个学期内完成三轮测量波次的本科生群体。本数据集整合了自我报告测量与基于表现的评估，可同时分析数字学习环境中的主观学习体验与客观学习成果。 3. 核心构念（Core Constructs） 本数据集包含六个核心构念，覆盖学习过程的不同维度： - 短视频接触量（Short-video Exposure, EXP）：学习者对微课程视频的行为投入，涵盖观看频次、完成度与注意力集中度。 - 内容个性化适配度（Content Personalization, PER）：感知到的教学内容与个体学习需求、先验知识及学习偏好的匹配程度。 - 数字媒体自我效能感（Digital Media Self-Efficacy, DMSE）：学习者在使用、评估与管理数字学习环境方面的自信心。 - 情境兴趣（Situational Interest, INT）：学习活动中的即时动机投入，涵盖注意力、愉悦感、好奇心与感知到的意义性。 - 精细加工（Elaboration, ELAB）：深度认知加工策略，包括自我解释、先验知识整合与概念对比。 - 概念留存（Conceptual Retention, RET）：通过回忆、解释、整合与迁移任务评估的基于表现的学习成果。 每个留存维度的评分范围为0至25分，总留存得分区间为0至100分。 4. 纵向研究结构 本数据集采用三波次研究结构： - T1：基线测量，涵盖接触量、个性化适配度与自我效能感。 - T2：中期学习成果与认知加工。 - T3：最终学习成果与概念留存。 该结构可用于考察时序动态特征，包括跨时段的滞后关联与发展性学习过程。 5. 测量方法 本数据集结合了两类测量手段： - 李克特量表（Likert-scale）调查结果（1-7分量表）：用于收集感知与体验数据。 - 基于表现的客观评估：用于评估概念理解水平。 这种双测量方法可通过将内部状态与可观测成果相联系，为学习过程分析提供更稳健的支撑。 6. 分析潜力 本数据集适用于多种高级统计分析，包括： - 结构方程模型（Structural Equation Modeling, SEM） - 纵向中介分析 - 交叉滞后面板模型 - 增长模型与重复测量分析 该数据集的设计可支持研究者探究早期接触与投入如何通过中间认知机制转化为长期学习成果。 7. 研究意义 本数据集为数字媒介化学习环境（尤其是短视频微学习模式）中的学习过程研究提供了全面的实证基础。其可支撑相关研究，探究如何将学习投入转化为持续性认知加工，从而最终获得持久且可迁移的长期学习成果。