Malomane Dorcus Kholofelo, Norris David, Banga Cuthbert B, Ngambi Jones W
Department of Agricultural Economics and Animal Production, University of Limpopo, Private Bag X1106, Sovenga, 0727, South Africa.
Trop Anim Health Prod. 2014 Feb;46(2):331-5. doi: 10.1007/s11250-013-0492-2. Epub 2013 Oct 18.
Body weight and weight of body parts are of economic importance. It is difficult to directly predict body weight from highly correlated morphological traits through multiple regression. Factor analysis was carried out to examine the relationship between body weight and five linear body measurements (body length, body girth, wing length, shank thickness, and shank length) in South African Venda (VN), Naked neck (NN), and Potchefstroom koekoek (PK) indigenous chicken breeds, with a view to identify those factors that define body conformation. Multiple regression was subsequently performed to predict body weight, using orthogonal traits derived from the factor analysis. Measurements were obtained from 210 chickens, 22 weeks of age, 70 chickens per breed. High correlations were obtained between body weight and all body measurements except for wing length in PK. Two factors extracted after varimax rotation explained 91, 95, and 83% of total variation in VN, NN, and PK, respectively. Factor 1 explained 73, 90, and 64% in VN, NN, and PK, respectively, and was loaded on all body measurements except for wing length in VN and PK. In a multiple regression, these two factors accounted for 72% variation in body weight in VN, while only factor 1 accounted for 83 and 74% variation in body weight in NN and PK, respectively. The two factors could be used to define body size and conformation of these breeds. Factor 1 could predict body weight in all three breeds. Body measurements can be better selected jointly to improve body weight in these breeds.
体重和身体各部位的重量具有经济重要性。通过多元回归从高度相关的形态特征直接预测体重是困难的。进行了因子分析，以研究南非文达（VN）、裸颈（NN）和波切夫斯特鲁姆科伊科伊（PK）本地鸡品种的体重与五个线性身体测量指标（体长、体围、翅长、胫粗和胫长）之间的关系，旨在确定那些定义身体形态的因素。随后使用从因子分析得出的正交性状进行多元回归以预测体重。测量数据来自210只22周龄的鸡，每个品种70只。除PK的翅长外，体重与所有身体测量指标之间均具有高度相关性。方差最大化旋转后提取的两个因子分别解释了VN、NN和PK总变异的91%、95%和83%。因子1在VN、NN和PK中分别解释了73%、90%和64%，并且在VN和PK中除翅长外的所有身体测量指标上均有载荷。在多元回归中，这两个因子在VN中解释了体重变异的72%，而在NN和PK中仅因子1分别解释了体重变异的83%和74%。这两个因子可用于定义这些品种的体型和形态。因子1可预测所有三个品种的体重。可以更好地联合选择身体测量指标以提高这些品种的体重。
