Introduction

Evaluating the impact of educational changes on children's capabilities can be done in a diversity of ways. Indeed, there is no unique way of using the CA (capability approach) for normative purposes. One possible route is to define a methodology based on the most striking characteristics of the approach and then to create different categories of assessment from the particular features of the targeted programs. When the CA is used as an informational space for normative valuations we have the approach at its best. This means that the approach is incomplete by necessity (in the sense that it depends on the use of particular substantive theories) and it is not used beyond its conceptual limits.

Among the most important features of the approach, as defined by Sen (1985; 1992; 1999) and Nussbaum (2000), may be mentioned:

• objective nature of functionings and capabilities, avoiding the use of subjective metrics and the problem of adaptive preferences;

• multidimensional assessment of capability-laden normative exercises, exploring the differential impact of actions and state-of-affairs on individuals' well-being and agency;

• emphasis on autonomy as a characteristic of individuals' advantage-point, providing a distinct perspective on how people are able to shape their well-being;

• comparable results, based on the principle of multiple realisability (Nussbaum, 2000: 77).

When addressing the issue of children's capabilities the importance of temporal dimensions in constituting the characterization of the autonomy and well-being of individuals must be noted. The same capability has different autonomy-values if belonging to different moments in time.