11 pages, This proof-of-concept study explores the innovative application of Large Language Models (LLMs) for qualitative analysis of feedback from an Extension program, addressing the challenge of efficiently analyzing qualitative data. The study juxtaposes traditional human-led qualitative analysis with Artificial Intelligence (AI)-driven techniques, revealing the complementary strengths of human insights and AI efficiency. It underscores the potential of LLMs to enhance qualitative analysis while recognizing the need for human oversight to ensure depth and context accuracy. This research contributes to the fields of program evaluation and data analysis, offering a new paradigm for integrating advanced AI tools in qualitative research.
19 pages., Online via UI e-subscription., Authors examined impacts of efforts by Report for America (RFA) to strengthen the capacity of local news and increase trust from the perspective of two communities: a neighborhood on Chicago's West Side and a rural county in eastern Kentucky. Findings illustrated "the influence of place and power dynamics on how residents navigate trustworthiness factors." They also revealed lack of feedback loops to provide coverage for communities.
2 pages., In a preview of this issue about "messy data in conservation," the author links messy data to related topics in conservation and urges a trans-disciplinary embrace of messiness to accelerate conservation progress.
Eitzinger, Anton (author), Cock, James (author), Atzmanstorfer, Karl (author), Binder, Claudia R. (author), Läderach, Peter (author), Bonilla-Findji, Osana (author), Bartlin, Mona (author), Mwongera, Caroline (author), Zurita, Leo (author), and Jarvis, Andy (author)
Format:
Online journal article
Publication Date:
2019-03
Published:
Germany: Elsevier
Location:
Agricultural Communications Documentation Center, Funk Library, University of Illinois Box: 7 Document Number: D10292
13 pages., Via online journal., Farmers can manage their crops and farms better if they can communicate their experiences, both positive and negative, with each other and with experts. Digital agriculture using internet communication technology (ICT) may facilitate the sharing of experiences between farmers themselves and with experts and others interested in agriculture. ICT approaches in agriculture are, however, still out of the reach of many farmers. The reasons are lack of connectivity, missing capacity building and poor usability of ICT applications. We decided to tackle this problem through cost-effective, easy to use ICT approaches, based on infrastructure and services currently available to small-scale producers in developing areas. Working through a participatory design approach, we developed and tested a novel technology. GeoFarmer provides near real-time, two-way data flows that support processes of co-innovation in agricultural development projects. It can be used as a cost-effective ICT-based platform to monitor agricultural production systems with interactive feedback between the users, within pre-defined geographical domains. We tested GeoFarmer in four geographic domains associated with ongoing agricultural development projects in East and West Africa and Latin America. We demonstrate that GeoFarmer is a cost-effective means of providing and sharing opportune indicators of on-farm performance. It is a potentially useful tool that farmers and agricultural practitioners can use to manage their crops and farms better, reduce risk, increase productivity and improve their livelihoods.
Stebner, Scott (author) and Baker, Lauri M. (author)
Format:
Conference paper
Publication Date:
2016-02
Published:
USA
Location:
Agricultural Communications Documentation Center, Funk Library, University of Illinois Box: 162 Document Number: D08143
Notes:
Research paper presented in the Agricultural Communications Section, Southern Association of Agricultural Scientists (SAAS), in San Antonio, Texas, February 7-8, 2016. 24 pages.
