Process and meta-analytic integration of the research unit's findings, and advancement of the theoretical model

点击翻译按钮获取中文摘要

Subproject M has a unique status within the research unit FOR 2385, as the project itself does not collect data in primary studies. Instead, the project is responsible for the meta-analytic integration of data and studies, the standardization of data collection, and coordinating the work on the conceptual framework as well as the support for the PhD students. So far, subproject M compiled a set of test instruments that are used in every subproject. In addition, subproject M conducted two meta-analyses with external studies from the research literature. The first of these meta-analyses has been accepted for publication already. Currently, subproject M is gathering and integrating data from the other subprojects in order to answer research questions that span across all the subprojects with a focus on the question of whether effects of scaffolding can be generalized. The results from the meta-analyses validated relationships between variables that are at the core of the conceptual framework of the research unit. For instance, the analyses revealed a medium-sized effect of various instructional support measures on diagnostic competences. Additionally, a large-sized effect of simulation-based learning environments on complex decision-making competences was found. Both meta-analyses showed that the effect of scaffolding is influenced by the prior knowledge of learners. These results motivated the research unit to focus on the adaptation of scaffolding in the second funding period. Therefore, subproject M will further develop the conceptual model of the research unit and conduct research about designing simulations in which scaffolding is adapted to the learners. The main research questions are: What is the relation between various learning prerequisites and types of scaffolding? Which adaptation strategies are most effective? Which process variables mediate the effects of adaptation? Which context variables have an influence? Subproject M contributes to answering these questions in a unique way: It tests the generalizability of the results by working on these questions in a way that spans across the other subprojects. To answer the questions, the meta-analyses with external studies from the research literature will be continued with an extended coding scheme. The integration of data from the whole research unit will be continued, too, using advanced statistical methods: Bayesian network analyses will be used to analyze process data and generalized mixed models will be used to integrate data. Subproject M will also keep its additional tasks of standardizing assessment instruments, collecting data, promoting open science principles, conducting workshops, and communicating results of the group.