Collaborate

Text as Data collaborations use media, documents, and other text corpora to answer substantive questions across disciplines. Partners may contribute research questions, domain expertise, languages, corpora, human coding, validation, computational methods, or opportunities to compare workflows across fields.

Potential collaborations include corpus development, named-entity extraction, frame, sentiment, topic, discourse, or conversation analysis, LLM-assisted coding, method validation, training, joint publications, and conference panels. Strong proposals identify a substantive question, the text source, unit of analysis, language, and intended evidence or output.

Begin with a concise description of the research question, corpus, languages, expected unit of analysis, available data or expertise, intended outputs, participating researchers, and timeline. The lab will assess rights and access, sampling, coding design, validation, responsibilities, authorship, and a review schedule. Capacity varies, so timing and response expectations are confirmed during this initial fit review.

Start with the Text as Data Approach, method-centered Research Streams, text and media Databases, and Training modules. Related Publication pages show how these resources have been used and identify literature and contributors.

Contact Dr. Jacob Reidhead with the proposed question, corpus, languages, coding or extraction target, collaborators, outputs, and timing. The lab may suggest a methods session or small validation pilot before a larger commitment. Collaboration interest does not itself grant member or researcher access.