Breeding objectives that are clearly defined simplify selection decisions based on multiple traits, and are therefore essential in the development of efficient breeding strategies. The development of economic breeding objectives is driven by a need to define accurate selection criteria for traits affecting farm profitability (Simm, 1998), and to assign relative weightings to traits in selection indexes based on the relative economic importance of traits.
Traditional methodologies for deriving breeding objectives involve the use of profit functions, which calculate the impact on farm profit of changes in each trait. However, it is also important that breeding objectives reflect the farming philosophies of the breeders and commercial farmers for whom they are designed. The increasing importance of environmental (Olesen et al., 2000) and animal-welfare (Fisher and Webster, 2009; Nielsen et al., 2011) traits preferred by consumers, which may impact on market access has driven developed livestock industries to account for aspects of production systems beyond those that can be defined economically. Similarly, as developing countries build genetic-improvement programmes (e.g. Kosgey (2006)), breeding objectives must be defined for production systems where price and cost data are not readily available. A further challenge in defining breeding objectives in developing countries is that the value of an animal often encompasses intangible factors such as prestige, financing, insurance, or a role in cultural and ceremonial functions (Kosgey et al., 2004). Therefore, there is an increasing recognition of the need to incorporate the perceptions of industry stakeholders in breeding objectives.
This thesis developed, assessed, and compared traditional and novel approaches to the formulation of breeding objectives in sheep genetic improvement programmes. Traditional methodologies were used to derive breeding objectives for two sheep-producing industries, including the calculation of economic weights (EWs) for novel traits.
The internet-based software 1000Minds, originally designed to capture the preference of respondents in the healthcare system (Hansen and Ombler, 2009), has found application in a wide range of fields (e.g. Boyd et al. (2011), Golan et al. (2011), and Smith and Fennessy (2011)). The use of 1000Minds to define EWs for animal breeding programmes represents a novel aspect of the current study, and represents the development of a new tool in the derivation of breeding objectives.
Hazel (1943) stated that fluctuating, vague, and sometimes erroneous ideals often hindered the improvements produced by selection. Many of these ideals are not linked to production, but rather are linked to consumers’ perception of animal production systems, and may or may not have a direct economic impact on the production system. These ideals may, however, influence product acceptance by consumers and market access. Therefore, such drivers of the direction of selection are beyond those that can be defined economically in a traditional sense and these have to be taken into account in the construction of breeding objectives and selection indexes. This study has proposed and evaluated a method to enable traits that are not easily defined by economic models to be included in breeding objectives and reveals the opportunity to develop and refine breeding objectives. Such methods ensure that breeding objectives continue to meet the expectations of a changing market.
