This dataset includes the raw data and R statistical software code needed to reproduce all statistical model outputs from the linked manuscript.

In this study, observations of fall armyworm (FAW; Spodoptera frugiperda) and corn earworm (CEW; Helicoverpa zea) damage, along with agronomic quality assessments, were made on different maize genotypes in 2022 and 2023. The goal was to characterize how resistance to FAW and CEW varies across genotypes and how this resistance correlates with agronomic quality. In the trial, three replicate plots of each genotype were planted each year. This dataset includes all the observations of FAW leaf damage, plant aspect, seed set, ear aspect, and ear rot (categorical scores) and CEW ear damage (continuous measurement) on each genotype. The R script includes the Bayesian generalized linear mixed models fit to the data, using the software Stan interfaced with R via the cmdstanr and brms packages. Cumulative logistic mixed models were fit to the categorical score data, and a hurdle Gamma model was fit to the ear damage data. The model posterior estimates were used to make comparisons between the genotype mean damage ratings. Pre-fit model objects are also included in this dataset.

Included files

• combined_2022_2023_data.xlsx: All raw data, in two spreadsheet tabs. The first tab includes leaf damage and ear damage data from 2022 and 2023. Up to 20 plants from each single-genotype plot were scored for leaf damage on a categorical 1-9 scale in each year. Up to 10 plants per plot were measured for ear damage, a continuous variable in mm units. Genotypes are identified with a numerical ID (entry column) and a name (genotype column). The second tab includes four different variables scored in 2022 on a categorical 1-9 scale, with a single value per plot: plant aspect, ear aspect, seed set, and ear rot.
• FAW_damage_rating_analysis.R: R script to import the data, fit Bayesian generalized linear mixed models, generate posterior estimates for genotype means and pairwise differences, generate comparison letter summaries, and produce figures and tables.
• get_comp_letters.R: R script defining custom function to generate comparison letter summaries from Bayesian model output.
• earrot_bfit.rds, ed_bfit.rds, ld_bfit.rds, plantaspect_bfit.rds, seedset_bfit.rds, earaspect_bfit.rds: R objects allowing the user to reload the pre-fit statistical models so that the outputs can be generated without refitting the models.