Duration: 10/01/2026 to 09/30/2031

Administrative Advisor(s):

NIFA Reps:

Non-Technical Summary

Statisticians and data scientists play an important role in fulfilling the land-grant mission of institutions through consulting and collaboration. Consultations with agricultural researchers across a broad range of disciplines provide a fundamental service for research quality by ensuring rigorous experimental design and analysis, resulting in novel findings and the development of innovations that directly benefit U.S. agriculture. Keeping up with new developments in statistical analysis is difficult for most domain experts, so agricultural statisticians and agricultural experiment stations bridge the gap between domain specialists and cutting edge data analysis. Our target audiences are statisticians and data scientists working in agricultural research, and researchers who can benefit from our expertise in data analysis. The goals of this Project are to: (1) enable appropriate analytical choices in research applications; (2) provide continuing education to researchers; (3) elevate the overall visibility of agricultural statistics; and (4) develop novel statistical approaches to support research discovery and enhance sustainability and competitiveness of U.S. agriculture. These goals are realized through our annual meeting, collaborative arrangements and group efforts to produce educational materials representing an expert consensus on statistical methods important to agricultural research. At the annual meetings, we discuss analytical solutions we have developed, opportunities for further research and widespread misunderstandings among our colleagues that our group is well positioned to address. Our collaborative outputs in research and education help our institutional colleagues by ensuring high-quality data analysis and sufficient theoretical and practical background in statistical methodology to implement those methods.

Statement of Issues and Justification

Statisticians and data scientists who consult or collaborate with subject-matter scientists and actively participate in research at an Agricultural Experiment Station (AES) have a unique role in fulfilling the land grant mission of their institution. Consultations with faculty and students provide a fundamental service for research by ensuring rigorous experimental design, strengthening analysis, and enhancing the reliability of research outcomes that meet scrutiny under peer review.  These collaborations extend not only over a wide range of disciplines–animal and dairy science, agronomy, crop and soil science, entomology, food science and technology, horticulture, plant pathology, poultry science, and more–but also across fundamentally different analytic frameworks and data types.  Furthermore, these faculty often conduct research of their own and contribute to the advancement of applied statistics techniques in different fields of agricultural and life sciences.  As a result, statisticians and data scientists providing analytical expertise at an AES must have a broad understanding of statistical methodology that can be applied well beyond their domain-specific research areas. The NCCC-170 multi-state coordinating committee provides a mechanism to (i) bring together AES statisticians from academia, research stations, government, and industry, (ii) create a critical mass of expertise that strengthens our collective knowledge and ability to study emerging issues, and (iii) facilitate the development of educational resources and curricula for other statisticians and research scientists. The exchange of ideas and sharing of expertise strengthens the work of all project members and enriches the broader research efforts across their institutions and nationally.

The charge to remain current in the statistical profession is especially challenging given the development and widespread adoption of machine learning (ML) and generative artificial intelligence (AI). These continually emerging ML and AI techniques advance the analytical possibilities of designed experiments and observational studies, requiring rigorous, computationally intensive, and sometimes novel methodologies that ensure the quality of the data and the inferences drawn. In particular, ML and AI techniques are revolutionizing the scale at which high-throughput agronomic data can be obtained and analyzed. Because statistics is the analytical backbone of ML and AI, AES statisticians and data scientists have a critical responsibility to agricultural research to ensure that these innovative approaches are implemented and interpreted correctly. This requires them to keep abreast of the latest research, as well as to work to develop methodology appropriate to their consulting and collaborative responsibilities. Developing guidelines on best practices in the use of emerging statistical methodology in ML and AI is essential for promoting awareness and avoiding misleading uses of ML and AI in data analysis.

One ramification of the impact of AI on the day-to-day operations of AES statisticians is the widespread use of large language models (LLMs; e.g., ChatGPT, Gemini, Claude) for obtaining statistical advice. The use of LLMs for statistical advice can be beneficial because it may serve as a resource to broaden access to statistical advice, but it can also be detrimental given the unacceptably high prevalence of incorrect or misleading outputs coming from LLMs. Our group is uniquely positioned to address this concern by educating users on how to use and query LLMs to recognize and minimize incorrect or misleading advice related to statistical analysis and experimental design. For example, resources such as dedicated LLMs, built primarily with knowledge focused on statistical consulting, can also reduce the amount of incorrect or misleading advice suggested to the user. As these tools are continuing to be developed with state-of-the-art educational resources, our group can play a leading role in ensuring that these LLMs approaches will provide more reliable information.

Given that AES researchers often work in environments where time constraints or evolving methodologies make it difficult to stay current with statistical best practices, this committee provides an educational platform designed to support those needs and to assist agricultural researchers in advancing national research priorities. Our group is committed to promoting high-quality agricultural research, and collaboration between AES statisticians and scientists strengthens the ability of land-grant institutions to fulfill their agricultural research missions. As State and Federal appropriations fall short of the levels needed for research and competition for external funding increases, it is becoming even more important that research dollars be used as effectively as possible.

Other challenges faced by AES statisticians include the proper recognition of their efforts into the production of quality research. The infrastructure used to assess the traditional deliverables of subject-matter experts and tenure-track faculty does not necessarily reflect the scholarship and academic contributions made by AES statisticians. The adequate use of state-of-the-art statistical methodology must be properly articulated to administrators who may not be familiar with current advances in statistical practice. Failure to effectively communicate their advances can limit their professional impact and slow progress in agriculture. While some of our contributions and members have been recognized by professional societies, there is a need for ongoing efforts to enhance the visibility of agricultural statistics. These can occur through awards and nominations, both of which demonstrate the value of our academic and research efforts. Our goal is to advocate for more recognition within the broader field and at our home institutions.

In light of these challenges, it is critical that AES statisticians work cooperatively to determine the best approaches to common statistical problems. This will help guide future directions of statistical research and software development. Researchers and reviewers will have the necessary background to assess new developments in statistical methodology. This will ultimately underscore awareness of the irreplaceable scholarship and of our work to agricultural research and those involved in the evaluation, tenure, and promotion process.

This NCCC committee plays a vital role to US public agriculture by i) serving as a focal point for the development, implementation, and dissemination of sound statistical practice; and ii) providing a support network of direct relevance to AES statisticians and data scientists. Through the NCCC committee, members have direct access to their peers, which in itself is an indispensable resource for practical knowledge about agricultural statistics and statistical consulting. This NCCC’s function as a professional network raises its efficacy well above that offered by other more general venues. The NCCC annual meeting provides a valuable opportunity for the productive discussion of challenges and developments in agricultural statistics and practical considerations for application of emerging techniques. Furthermore, experienced members act as mentors to newer members, which provides a crucial opportunity for professional development in such a unique field. As a result, this NCCC committee improves the quality of research in the agricultural sciences at its participating institutions through educational outreach within and beyond its NCCC membership.

Objectives

Procedures and Activities

Since its inception in 1990 as NCR-170, NCCC-170 has addressed a series of statistical topics that impact AES research. During annual meetings, members discuss current research projects at their institutions that pose novel analytic challenges, discuss new methodologies, and sometimes reexamine established “best practices” in light of the new methodological development. These topics are compiled, and the members use this information to decide on a statistical topic to be addressed by the group when a critical mass of interest is reached. Once that decision has been made, a multi-year cycle begins.

The first and, if necessary, second year of a cycle is devoted to researching the chosen topics. Research and material development are carried out by a subcommittee of volunteers from the project membership. Members then present their progress at the annual meetings to obtain feedback from the entire group. Over the past 30 years, project members have dealt with statistical aspects of on-farm trials, experimental design, mixed models, spatial statistics, -omics data and bioinformatics, meta-analysis, predictive analytics, and generalized linear mixed models. Our group has developed specialized workshops to educate both statistical and subject matter audiences on topics such as mixed models, spatial statistics, Bayesian statistics, graphical models, meta-analysis, and generalized linear mixed models.

Workshops have been presented at the annual meetings of various professional societies, including the American Phytopathological Society, the American Society of Animal Sciences, the American Dairy Science Association, the Association of Veterinary Epidemiology and Preventive Medicine, the American Society of Horticultural Sciences, Aquaculture America, Agronomy Society of America, as well as at meetings of several local chapters of the American Statistical Association and the International Biometrics Society. A noteworthy outcome is the mixed models workshop, which has been offered approximately every other year since 1999 to the American Society of Animal Sciences and/or to the American Dairy Science Association. This workshop has been successful and continues to be updated and offered upon request. Other workshops originating from this group include a generalized linear mixed models for non-normal data, incorporation of spatial covariates into analysis, and systematic reviews/meta-analysis in agricultural research. Some of these workshops were also presented as full-day workshops accompanying the Conference on Applied Statistics in Agriculture, and NCCC-170 members have presented similar workshops at their home institutions. Currently, our group is working on an online resource/textbook aimed at helping other consulting statisticians and practitioners on common yet analytically challenging generalized linear mixed model (GLMM) settings. This collaboration blends many of our most senior group members with some of the most junior group members, to ensure the continuity of expertise across generations.

For the coming 2026-2031 cycle, we will work on the addition of generative AI tools into our normal workflow. Said addition of generative AI tools includes both facing new challenges and adapting to new approaches for spreading information and educational content. While the history of the NCCC-170 group has focused on delivering information to practitioners via peer-reviewed publications and workshops, the emergence of new technologies requires adaptation by statisticians. More specifically, generative AI tools have brought a new set of challenges and opportunities for the analysis of data in agricultural research. As mentioned in the justification, consulting statisticians are encountering clients seeking advice from LLMs as an alternative to original sources like textbooks or scholarly material with increasing frequency despite the risk of suboptimal results. Our clients are also seeking guidance from statisticians in the optimal use of LLMs for data analysis. Since LLMs have become a valuable asset in the scientist’s tool set, we are positioning our educational activities to address these realities.

We envision creating new instructional materials that leverage generative AI tools, while still relying on expert knowledge for handling complex problems. Specifically, we are committed to (i) developing and promoting guidelines and/or pipelines to optimize return of accurate results from LMMs for statistical analysis; and (i) flag areas where LLMs are likely to provide incorrect answers or advice, and (ii) contribute to the improvement of LMMs through development of custom models oriented towards statistical analysis and contribution of reliable content for LMM model building. The goal is to reach a broad audience across agricultural research disciplines that can complement on-site workshops. 

To enhance the visibility of agricultural statistics throughout the United States, we will advocate through three different mediums. First, we will provide mentorship and collaboration opportunities for colleagues at our institutions who are interested in incorporating agricultural statistics into research projects and programs. Second, we will continue to recognize NCCC-170 members for their contributions by promoting their accomplishments publicly and nominating them for professional awards. Such nominations are critical for agricultural statisticians to gain recognition of their contributions to agriculture, and to inspire the next generation of scientists to study applied statistics. Finally, we will educate other statisticians and administrators about the irreplaceable contributions of AES statisticians to scholarship and scientific discovery, as well as their role in the scientific community. Enhancing the visibility of the impact of agricultural statistics will help demonstrate its importance to administrators at Departments of Statistics at land grant and agricultural research institutes, and thus underscore the importance of keeping funding lines for agricultural statistician faculty open in the future.  For example, project members produced a position paper formally endorsed by the board of the American Statistical Association (ASA) in 2018 providing a conceptual framework and practical guidelines for evaluating scholarship by consulting statisticians collaborating with domain-specific scientists . The position paper currently can be accessed as an official statement on the ASA website: https://www.amstat.org/asa/files/pdfs/POL-Statistics-as-a-Scientific-Discipline.pdf. 

A book on applications of statistics to the agricultural and life sciences was published by group members and sponsored by the American Society of Agronomy, Crop Science Society of America, and Soil Science Society of America in 2025. This book, titled Applied Statistics in Biology: A Practical Guide Using SAS, R, and JMP, was authored by S.R. Bowley, E. van Santen, S. Riley, D.K. Michelson, and R.M. Hummel (project member names are bolded). In addition, group member W.W. Stroup, with M. Ptukhina and J. Garai, published the second edition of Generalized Linear Mixed Models: Modern concepts, Methods and Applications (CRC Press) in 2024. This and the first edition continue to be used by practitioners and statisticians.  Group members regularly publish materials that translate technical and complex statistical concepts for an applied agricultural audience (i.e., translational statistics), and provide guidelines for implementation of statistical models. 

A subcommittee from our group was invited to present their research advances on best practices for GLMM implementation in the 2025 annual Conference of Applied Statistics in Agriculture and Natural Resources (held in Gainesville, FL). The group presented their research with oral and poster presentations that spanned an entire morning and one afternoon session that addressed some common challenges in implementing GLMMs. 

• Best Practices for GLMM Implementation: Setting the Stage. Bello NM and WW Stroup. Oral presentation.
• Allowing Negative Variance Component Estimates in REML: Inferential Consequences for Fixed Effects and Type I Error Control. Poudel B, Howard R, Bello NM and WW Stroup. Oral presentation.
• Correlated Binomial Data: A Confluence of Analytic Challenges. Bello NM, Kogan C, Craig BA, Palmer DG, Durham SL and WW Stroup. Oral presentation.
• Mixed Models-Based Precision and Power Analyses for Design Comparisons Using PROC GLIMMIX and glmmTMB. Palmer D, Bello NM, Lacasa J and WW Stroup. Oral presentation.
• Multinomial Logistic GLMMs: A Patchy Landscape of Software Implementations. Read Q, Craig BA, Dixon PM and WW Stroup. Oral presentation.
• Using Randomized Quantile Residuals to Assess Model Fit of Generalized Linear Mixed Models. J Piaskowski. Oral presentation.
• Modeling Considerations and Challenges for Overdispersed Count Data in a Generalized Linear Mixed Model. Durham S, Bello NM, Palmer DG and WW Stroup. Poster presentation.
• Zero-inflated Models: Best Practices for Diagnostics and Model Fitting Across Software Platforms. Fair CG, Piaskowski J, Macchiavelli R, Lacasa J, Craig BA and WW Stroup. Poster presentation.
• Handling Within-Block Heterogeneity Using Smoothing Splines. Lacasa J, Durham SL, Bello NM and WW Stroup. Poster presentation.
• Analysis of Repeated Measures Revisited: Are Recommended Best Practices Applied Consistently Across and Within Software Platforms? Macchiavelli R, Bello NM and WW Stroup. Poster presentation.

Furthermore, project members have collaboratively written scientific articles on translational statistics addressing domain-specific disciplines on how to tailor modern concepts in experimental design to specific problems in the agricultural sciences. These articles are frequently published in peer-reviewed scientific journals. Here are some examples (names of project members are bold):

• Gerber EAE and BA Craig (2023). Residuals and diagnostics for multinomial regression models. Statistical Analysis and Data Mining: The ASA Data Science Journal. https://doi.org/10.1002/sam.11645
• Griffith EH., Sharp JL, Bridges WC,  Craig BA, Hanford KJ, and JR Stevens (2022).. “The academic collaborative statistician: research, training, and evaluation,” Stat. https://doi.org/10.1002/sta4.483
• Piepho H-P, and LV Madden (2022). How to observe the principle of concurrent control in an arm-based meta-analysis using SAS procedures GLIMMIX and BGLIMM. Research Synthesis Methods. 1-8. doi:10.1002/jrsm.1576
• Martin-Schwarze A, Niemi J, and Dixon P (2021). Joint modeling of distances and times in point-count surveys. Journal of Agricultural, Biological, and Environmental Statistics. 26, 289-305. https://doi.org/10.1007/s13253-021-00437-3
• Piaskowski J and W Price (2022). Incorporating Spatial Analysis into Agricultural Field Experiments. Online book: https://idahoagstats.github.io/guide-to-field-trial-spatial-analysis/
• Wisnieski LM, Sanderson W, Renter DG and NM Bello (2023). Inferential implications of normalizing binomial proportions in a structural equation model: A simulation study motivated by feedlot data”. Preventive Veterinary Medicine. 217:105963. doi: 10.1016/j.prevetmed.2023.105963. 

In addition to these group outputs, NCCC-170 discussions have also impacted project members' collaborations at their own institutions and have led to numerous refereed and non-refereed publications, participation in competitive grants, and presentations at professional meetings. These outputs are listed in detail in the corresponding annual project reports, available at https://nimss.org/projects/view/mrp/outline/18798/. Highlights from the 2025 report are listed later in this document. 

Expected Outcomes and Impacts

Projected Participation

View Appendix E: Participation

Educational Plan

A major focus of this project is education and outreach so our plans are directly embedded in our objectives and the strategies for meeting those objectives. Education occurs through (1) written resources providing expert guidance on statistical methods; (2) workshops on the application of statistical methods; and (3) collaborative research.  Members will educate their individual clientele of subject matter scientists and students in one-on-one and group settings as they deem appropriate. Workshops will continue to be offered to regional and national subject matter groups as well as journal editorial boards who may be interested in specific topics.  Finally, the annual meeting and subcommittee projects serve as key mechanisms for enhancing this group’s collective knowledge and expertise.

Organization/Governance

The Chair and Secretary are selected by the members present at the annual meeting. The meeting Host Chair for the next year is the member from the state in which the meeting will be held. The meeting Program Chair for the next year is a volunteer from among the members present at the annual meeting.

Literature Cited

This literature section contains selected publications from Project Annual Reports from 2021-2025 and project members as examples of the range of disciplines with which project members interact. Names of project members are bolded.

Brewer G, Kentaro M and K Hanford (2023). Measuring bee effect on seed traits of hybrid sunflower. Plants. https://doi.org/10.3390/plants12142662.

Clarke J, Cooper L, Poelchau M, Berardini T, Elser J, Farmer A, Ficklin S, Kumari S, Laporte M-A, Nelson R, Sadohara R, Selby S, Thessen A, Whitehead B, and T Sen. Data sharing and ontology use among agricultural genetics, genomics, and breeding databases and resources of the AgBioData consortium. Databases, November 2023. https://doi.org/10.1093/database/baad076

Coppock DL, Crowley L, Durham SL, Groves D, Jamison JC, Karlan D, Norton BE, and Ramsey RD (2022) Community-based rangeland management in Namibia improves resource governance but not environmental and economic outcomes. Nature Communications Earth & Environment 3:32. doi.org/10.1038/s43247-022-00361-5

Cramer EY, Ray EL, Lopez VK, Bracher J, Brennen A, Castro Ri- vadeneira AJ, Gerding A, Gneiting T, House KH, Huang Y, Jayawardena D, Kanji AH, Khandelwal A, Le K, Muhlemann A, Niemi J, [256 other authors], and G Reich. (2022) Evaluation of individual and ensemble probabilistic forecasts of COVID-19 mortality in the US. Proceedings of the National Academy of Sciences, 119(15), e2113561119. https://doi.org/10.1073/pnas.2113561119

Della Coletta R, Liese SE, Fernandes SB, Mikel MA, O Bohn M, Lipka AE, and CN Hirsch. Linking genetic and environmental factors through marker effect networks to understand trait plasticity. Genetics, Volume 224, Issue 4, August 2023, iyad103, https://doi.org/10.1093/genetics/iyad103

Ferreira REP., Balaguer MAL, Bresolin T, Chandra R, Rosa GJM, White HM and JRR Dórea. (2024) Multi-modal machine learning for the early detection of metabolic disorder in dairy cows using a cloud computing framework. Computers and Electronics in Agriculture 227: 109563.

Giordano N, Hayes D, Hefley TJ, Lacasa J, Beres B, Haag LA and R Lollato. (2023). Re-thinking wheat yield response to plant density: risk assessment of seed treatment and cleaning methods. Plant Methods 21, 38 (2025). https://doi.org/10.1186/s13007-025-01355-y

Griffith EH, Sharp JL, Bridges WC, Craig BA, Hanford KJ, and JR Stevens (2022). The academic collaborative statistician: research, training, and evaluation. Stat, 11(1), e483. https://doi.org/10.1002/sta4.48

Hensley C, Brye KR, Savin MC, Wood LS and E Gbur (2021). Earthworm density differences among tallgrass prairies over time in the Ozark Highlands. Agrosystems, Geosciences & Environment. 4:e20136.

Holder A, Gross MA, Moehlenpah AN, Goad C, Rolf M, Walker R, Rogers JK and D Lalman (2022). Effects of diet on feed intake, weight change, and gas emissions in beef cows, Journal of Animal Science 100(10). DOI: 10.1093/jas/skac257

Hu H, Rincent R and DE Runcie (2025). MegaLMM improves genomic predictions in new environments using environmental covariates. Genetics, 229(1), 1-41.

Khanal P and J Tempelman (2022). The use of milk Fourier transform mid-infrared spectroscopy to diagnose pregnancy and determine spectral regional associations with pregnancy in US dairy cows. Journal of Dairy Science 105(4):3209-3221.

Koebernick J, Gillen A, Fett R, Patel S, Fallen B, Pantalone V, Shannon G, Li Z, Scaboo A, Schapaugh W, Mian R and QD Read (2024). Soybean test weight in relation to genotype, environment, and genotype × environment interaction in the southern USA. Agronomy Journal. DOI: 10.1002/agj2.21551 

Lewers KS and BT Vinyard (2025). Strawberry Plant Propagation: Evaluation of Cultivars Using Different Growing Environments and Assessment Approaches, HortSci 60(5):667-673, https://doi.org/10/21273/HORTSCI18445-25 

Lewis M, Stock M, Black B, Drost D and X Dai. (2021). Improving Snapdragon Cut Flower Production through High Tunnel Season Extension, Transplant Timing, and Cultivar Selection. HortScience 56(10):1206–1212.

Martin-Schwarze A, Niemi J, and P Dixon (2021). Joint modeling of distances and times in point-count surveys. Journal of Agricultural, Biological, and Environmental Statistics. 26, 289-305. https://doi.org/10.1007/s13253-021-00437-3

Montesinos-López OA, Barajas-Ramirez EA, Montesinos-López A, Lecumberry F, Fariello MI, Montesinos-López JC, Ramirez Alcaraz JM, Crossa J, and R Howard (2025). Tuning Matters: Comparing Lambda Optimization Approaches for Ridge Regression in Genomic Prediction. Genes, 16(6), 618. https://doi.org/10.3390/genes16060618

Perttu RK, Ventura BA, Rendahl AK and Endres MI (2021). Public views of dairy calf welfare and dairy consumption habits of American youth and adults. Front Vet Sci. 8:693173, doi:10.3389/fvets.2021.693173

Piepho H-P and LV Madden (2022). How to observe the principle of concurrent control in an arm-based meta-analysis using SAS procedures GLIMMIX and BGLIMM. Research Synthesis Methods 13: 821-828.

Piaskowski J and W Price (2022). Incorporating Spatial Analysis into Agricultural Field Experiments. Gitbook: https://idahoagstats.github.io/guide-to-field-trial-spatial-analysis/

Rosario CMA, Miller RJ, Klafke GM, C. Coates C, Grant WE, Samenuk G, Yeater K, Tidwell J, Bach JS, de León AAP and PD Teel (2022). Interaction between anti-tick vaccine and a macrocyclic lactone improves acaricidal efficacy against Rhipicephalus (Boophilus) microplus (Canestrini)(Acari: Ixodidae) in experimentally infested cattle. Vaccine, 40(47), 6795-6801.

Sharp J, Griffith EH, Craig BA, Hanlon A, Peskoe S, and Van Mullekom J (2024). The Current Landscape of Academic Statistical and Data Science Collaboration Units with Examples, STAT, 13(3), e718.

Sutton KL, Skipper AL, Fair CG, and MR Abney (2024). Landscape factors affect relative abundance of rootworm species and pod injury in Georgia peanuts. Journal of Economic Entomology toae219. DOI: 10.1093/jee/toae219

Taghouti M, García J, Ibáñez MA, Macchiavelli RE and N Nicodemus (2021). Relationship between body chemical composition and reproductive traits in rabbit does. Animals, 11, 2299. https://doi.org/10.3390/ani11082299

Wisnieski LM, Sanderson W, Renter DG and NM Bello (2023). Inferential implications of normalizing binomial proportions in a structural equation model: A simulation study motivated by feedlot data”. Preventive Veterinary Medicine 2023 Aug; 217:105963. doi: 10.1016/j.prevetmed.2023.105963.

Attachments

Land Grant Participating States/Institutions

IA, IN, MN, PR, UT, WI

Non Land Grant Participating States/Institutions

University of Wisconsin-Madison, USDA-ARS/NE