Search for: All records

Studying interactions faces methodological challenges and existing methods, such as configural diagramming, have limitations. This work demonstrates Epistemic Network Analysis (ENA) as an analytical method to construct configural diagrams. We demonstrated ENA as an analytical tool by applying this method to study dementia caregiver work systems. We conducted 20 semistructured interviews with caregivers to collect caregiving experiences. Guided by the Patient Work System model, we conducted a directed content analysis to identify work system components and used ENA to study interactions between components. By using ENA to create configural diagrams, we identified five frequently occurring interactions, compared work system configurations of caregivers providing care at home and away from home. Although we were underpowered to determine statistically significant differences, we identified visual and qualitative differences. Our results demonstrate the capability of ENA as 

When text datasets are very large, manually coding line by line becomes impractical. As a result, researchers sometimes try to use machine learning algorithms to automatically code text data. One of the most popular algorithms is topic modeling. For a given text dataset, a topic model provides probability distributions of words for a set of “topics” in the data, which researchers then use to interpret meaning of the topics. A topic model also gives each document in the dataset a score for each topic, which can be used as a non-binary coding for what proportion of a topic is in the document. Unfortunately, it is often difficult to interpret what the topics mean in a defensible way, or to validate document topic proportion scores as meaningful codes. In this study, we examine how keywords from codes developed by human experts were distributed in topics generated from topic modeling. The results show that (1) top keywords of a single topic often contain words from multiple human-generated codes; and conversely, (2) words from human-generated codes appear as high-probability keywords in multiple topic. These results explain why directly using topics from topic models as codes is problematic. However, they also imply that topic modeling makes it possible for researchers to discover codes from short word lists.