Assessment of breeding practices, trait preferences of sheep farmers and sheep morphometric characterization in Makurung and Lenting Viillages of Lepelle-Nkumpi Local Municipality, Limpopo Province, South Africa

Abstract

Sheep farming is practiced country wide, including in dry areas of Northern Cape, Western Cape and Limpopo Province. This study assessed the existence of production objectives, breeding practices, trait preferences of sheep farmers and morphometric characterization of sheep in two selected villages (Makurung and Lenting) of Lepelle-Nkumpi local municipality of Limpopo Province. Using 306 sheep of different classes, morphometric measurements were taken to characterize and predict body weight. Data was analyzed using Chi-square tests, descriptive statistics, rank index, Pearson’s correlation, analysis of variance (ANOVA) and various data mining algorithms such as Multivariate Adaptive Regression Spline (MARS) and Classification and Regression Tree (CART). Furthermore, MARS and CART were employed as data mining algorithms to determine the goodness of fit in body weight prediction from morphometric measurements. Socio-economic status results indicated that the majority of sheep farmers in the two surveyed villages were males, and there was no significant difference (P>0.05) observed between the villages. All the sheep farmers from Makurung village had tertiary education as their highest level of education, while in Lenting village majority of farmers had secondary education as their highest level of education, and there was a highly significant (P<0.01) difference between the villages. Majority of the sheep farmers from the two selected villages had their age range from 41 - 49, with only Lenting village having few farmers (36.70 %) greater than 60 years of age. Production objectives indicated that Majority of the sheep farmers in Makurung and Lenting villages kept sheep for savings & investment (55.00 %) and meat (41.70 %). However, there was no significant difference (P>0.05) between the surveyed villages. Breeding practices indicated that a large proportion (90.00%) of sheep farmers in both villages practiced uncontrolled mating, and a highly significant difference (P<0.01) was observed between the villages. A larger proportion of sheep farmers knew about castration and culling practice, with few (36.70 %) having no knowledge about it. Rank and indices in selection of trait preferences of breeding rams looked at mating ability (0.291), body size (0.250) and growth rate (P >0.05), while for breeding ewes, twinning ability (0.289), mothering ability (0.181), and lambing interval (0.168). Correlation results of rams in Makurung village showed that BW had a highly significant correlation with RH, HG, RL, WH and BL, While with rams of Lenting village, BW had a positive highly significant correlation with HG, WH, and BL. With ewes in Makurung village, BW had a highly significant correlation with RH, HG, WH and BL. While in Lenting village, BW had a highly significant correlation with RH, HG, RL, WH and BL. MARS and CART results indicated that HG had a significant effect (P<0.01) on WH, followed by BL, RH and the village. Goodness of fit criteria results indicated there was a high r (0.953), Rsq (0.900), ARsq (0.887) and low SDR (0.306) in MARS model, showing that this model was the best as compared with CART. The findings of this study imply that sheep farmers in Makurung and Lenting villages can read and write effectively and, therefore, can make decisions based on the design of CBBP. It was concluded that farmers in the two selected villages had household heads as male, who were married with education level of secondary and higher with ages ranging from 41-49. Sheep were kept mainly for saving and investment, with farmers having shown their highest preference for mating ability in rams and twinning ability in ewes. Most sheep farmers were not controlling the mating, with majority practicing culling and castration. In both sexes, BL, WH and HG can be used as a selection criterion when determining BW of sheep. Furthermore, both MARS and CART suggest that HG alone can be used as a predictor of BW in sheep. The goodness of fit calculations suggests that MARS was the best model. This study recommended that farmers, researchers, agricultural extension workers and other stake holders must collaborate in designing and implementing a community-based breeding programme by considering the production objectives, trait preferences and breeding practices.