Biweekly Student Collaboration Network Dataset for Early Prediction of Success

This dataset consists of biweekly nominations from students in a collaborative learning environment, collected over a specified period. Students were asked to nominate up to five peers with whom they collaborated most during each two-week period. The dataset captures these peer nominations across multiple intervals, allowing for the construction of dynamic collaboration networks. This structure facilitates the calculation of network centrality metrics, which can be analyzed to identify predictive indicators of academic success. The dataset includes:

• Student Identifiers: Anonymized IDs and fakenames for each student.
• Nomination Records: Each record includes the nominating student, the nominated peers (up to five), and the time interval of the nomination.
• Temporal Data: Each nomination event is timestamped, allowing for the exploration of evolving collaboration patterns.

The dataset is designed for educational data mining, particularly in analyzing peer network structures within collaborative learning contexts. It serves as a resource for exploring predictive indicators of academic performance based on social interactions and network positions. Researchers can apply various network centrality measures to understand how early nomination patterns correlate with student outcomes, offering potential insights for interventions aimed at enhancing collaborative learning.

本数据集由合作学习环境中的学生在指定时间段内提交的每两周一次的提名组成。学生被要求提名最多五位他们在每个两周期间内合作最为密切的同伴。该数据集记录了多个时间段的同伴提名，从而允许构建动态的协作网络。这种结构有助于计算网络中心性指标，通过分析这些指标，可以识别出学术成功的预测性指标。数据集包含以下内容：

• 学生标识：每位学生的匿名ID和化名。
• 提名记录：每条记录包括提名学生、被提名同伴（最多五位）以及提名的时间段。
• 时间数据：每个提名事件均附有时戳，便于探索协作模式的演变。

数据集旨在用于教育数据挖掘，特别是在分析合作学习环境中的同伴网络结构。它作为一项资源，用于探索基于社会互动和网络位置预测学术表现的指标。研究人员可以应用各种网络中心性度量，以了解早期提名模式与学生成果之间的相关性，为旨在提升合作学习的干预措施提供潜在见解。