The elusive “good” : how performance measures shape the US child support enforcement program, by Ana Carolina Macatangay

PhD thesis. University of St. Andrews.


“Performance measures do not just capture an organization’s ‘good performance’; they intervene in the valuation and construction of what counts as ‘good’ so it can be valued, enacted, and measured. This is the central theme of this thesis. Using a performativity lens, and drawing on valuation studies and the sociology of quantification, this research investigates the framing processes and mechanisms by which the federal performance measures, through a network of actors and devices, enable a U.S. child support enforcement agency to enact the ‘good performance’ that can be measured. Using ethnographic and case study methods, the study examines how actors and devices, through a framing process, qualify and negotiate the boundaries of their interactions to produce the ‘good performance’ that can be counted and measured. The boundaries they establish, however, produce overflows that potentially threaten the production of the qualified and quantified ‘good’. Such overflows emerge due to unexpected events, different versions of ‘good’, and conflicting frames of valuation of what counts as ‘good’. Based on the findings, the study offers a taxonomy of performative framing to explain how measures stimulate the organizing work of an agency so the ‘good performance’ can be organized, articulated, produced, and validated. It also offers a taxonomy of counter-performation to describe the uncertainties, multiplicities, and rivalries that the measured ‘good performance’ generates.”


“Individuals, organizations, and societies use categories to make sense of and interact with the world. Categories (e.g. car types by engine size or hotel quality by star rating), which have been ‘collectively crafted, sustained, and enforced’ (Fourcade, 2016: 176), often are held as fact and used for determining how things are similar and, by implication, different from others that have been excluded from a ‘category’. They enable the ordering of things based on a general principle of worth (Boltanski &Thévenot, 2006) and guide or format people’s choices and actions. They are used to assess, compare, or assign values (e.g. price or scores) to things, and to interpret the social and physical world. When things or entities are qualified into some type of ‘thing’, their qualities are reduced to a finite number that will unify them, establish their ‘equivalence’ (Desrosières, 1998), and turn them into realizable and recognizable categories (Diaz-Bone & Didier, 2016). The qualities that make the ‘thing’ (e.g. engine size for car type) become the parameter for sorting and grouping entities into a category to which other entities could be assigned through a process of encoding. As Desrosières (1998) explained, coding signifies a ‘reduction’, a ‘sacrifice’, as other qualities have been disregarded. But by doing this, entities could easily be coded to discrete categories and compared to others based on their selectively compressed qualities (Espeland & Lom, 2015) that had been articulated through the coding’s evaluative form. By making entities and their selective qualities visible, they become measurable ‘objects of reference’ (Centemeri, 2012) for interpretation, coordination and action, supported by huge ‘investments in conventional forms’ (Thévenot, 2015) – such as the use of indicators, statistics, ratings, etc. – to formalize the categories and classifications of information. It may be said therefore that while qualification defines categories, quantification formalizes them. The very forms that help name or construct them also reveal them; and by revealing them, they are made ‘real’ (Desrosières, 1998; Fourcade, 2016).” (p. 44)

Supervisors: Philip Roscoe and Shona Russell

Download link: