Asian Journal of Research and Review in Agriculture

Background: Studies of crop residues and agricultural by-products in sheep and goats report ingredient composition, complete-diet characteristics, and animal responses. These ev-idence types are related but not interchangeable, and study design affects whether reported outcomes can be combined.

Objective: To characterize a pilot evidence set on crop residues and agricultural or agro-industrial by-products used in small-ruminant feeding and assess the readiness of its extracted outcomes for quantitative synthesis or prediction.

Methods: Four source-identified reports contributed 25 numerical feed or diet descriptors and four qualitative animal-outcome notes. A separate extraction captured 11 outcome rows from two goat trials. We calculated 25 pairwise arithmetic treatment-minus-comparator con-trasts and relative differences, as well as six paired complete-diet composition differences. No contrast uncertainty was derived. We also fitted an unweighted additive ordinary-least-squares model to six treatment-by-period ADG group means from one trial. Ingredient, diet, theoretical feed-basket, and animal-response data remained distinct, and comparability and experimental dependence were assessed descriptively.

Results: The trials comprised a 60-day, three-arm randomized complete-block study in 12 goats and a two-period crossover in 10 goats. The source means yielded descriptive intake differences of -10.2 and -18.4 g/d for Gunun fermented-peel arms versus control, compared with +0.33 kg/d for the Romero ORG ration versus CTR. The ration bundles and designs differed, Romero paired-contrast variance was unavailable, and Gunun arms shared a control. No confidence intervals, new inferential tests, or pooled effects were produced. An additive ordinary-least-squares model summarised six Gunun arm-period ADG means (in-sample R² = 0.847; RMSE = 5.51 g/day); it summarizes observed means within that trial. Yeast-fermented cassava-peel neutral detergent fibre was reported as 37.6% of dry matter in a table and 36.7% in the text, an unresolved source discrepancy.

Conclusion: Within this pilot analysis, descriptive mapping and arithmetic contrasts char-acterize reported feed and animal outcomes. The observed contrasts are specific to the reported diets and trial designs; they do not estimate a common treatment effect. A within-study de-scriptive model summarised the Gunun ADG group means.