Abstract
Addressing meat quality challenges accompanied by the increase of cattle size remains a focus of the beef industry. And using appropriate statistical analysis is vital for yielding trustworthy results for decision-making. The objectives of this research were to 1) assess the impact of an alternative fabrication method of the hot beef round on pH and temperature decline, color, and tenderness of cuts from the top round and the knuckle of heavy weight carcasses, 2) determine viability of aged beef semimembranosus samples stored at -20oC for myosin heavy chain separation through SDS-PAGE, 3) compare beef subjective color results generated using metric and ordinal models, and 4) introduce the use of Bayesian-framed ordinal model as an alternative way to analyze ordinal sensory data. Eleven heavy weight beef steers were slaughtered. The knuckle subprimal was partially peeled prior to rigor to expose the femur on alternating sies of each carcass, the adjacent side remained intact serving as control. The anatomical deep and superficial locations and length of retail display time had more impact on beef color than the fabrication method. Temperature and pH decline of the deep and superficial locations of the beef top round was monitored for 48 h postmortem. The alternative fabrication accelerated the chilling rate of the deep location of the top round, increased the pH decline at the superficial location, and elevated steak palatability attributes. In addition, steaks from the superficial location were more tender than those from the deep location. A portion of the 14d aged top round samples were stored in -20oC freezer for up to 20 months. Denaturing SDS-PAGE was performed to verify whether clear bands could be obtained from these samples kept at suboptimal conditions for prolonged time. Results showed relatively clear bands for quantification indicating that aged beef top round samples in long-storage remain a viable option for muscle fiber typing via SDS-PAGE. Subjective sensory data using hedonic or ordinal descriptive scales often appear numeric, however they represent ordered categories. This research used a set of beef subjective color data to compare the outcome and differences in interpretations of the frequentist metric and Bayesian ordinal models. Results suggested that evaluators should be modeled as a random effect regardless of statistical models used. The pairwise hypothesis testing using the Bayesian ordinal output supported the finding of metric ANOVA in the trait of color uniformity. Though similar findings were observed between the methodologies, ordinal models remain more appropriate because they respect the ordered categorical structure of ordinal data. In addition, the Bayesian framework provided probabilistic inferences from the posterior distribution, offering results that are more intuitive and informative for interpretation and decision-making.