Kabeer reviews two prevailing pure models used to describe decision-making behavior in women, primarily within the family. The two models are mainstream neo-classical economics and a sociological model, each emphasizes fundamentally different motives for decision-making behavior. However, through her interviews with women, Kabeer sees these two pure models as over deterministic, poor predictors of individual behavior, and do not permit the construction of researchable hypotheses. What she concludes is that both models represent end-points of a behavioral spectrum. For women (and the world at-large), actual decision making rules exist in a middle ground where each end-point informs and modifies the other. Kabeer continues by remarking that the methodological choices proposed by each model predispose the research towards static, non-transformative accounts of individual behavior. A more rigorous method must involve directly the researcher and the subject.

At one end is the rational fool. The neo-classical model of decision-making views individual rational choice within the family unit as a composite of utility between paid labor, unpaid domestic work, and leisure. The problem with this view is two-fold: first, it assumes every individual in the family derives the same utility from one family choice; second, the complexity of making a completely informed decision is impossible. Interviews show that individual choice is distorted from the neo-classical ideal by individual identity, household identity, and larger collective identities. Individual identity is presumed to be more value-laden, therefore making choice more resistant to market shifts. At the household level, individual choices are defined by an implicit contract, a contract crafted for unequal, gender based roles with the family. In such arrangements the range of choices typically favor the male, while the compliant wife secures her own self-interest. The field of choice is constrained further when viewed in the context of society. Though inequitable, options are spelled out according to cultural rankings such as gender, race, and class. Trade unions, feminist organizations, and political parties also contribute to the possible range of individual choice.

At the other end is the cultural dope. In this framework, individual choice is merely an automatic re-enactment of cultural norms, allowing no freedom for individual choice. This framework evolved largely out of two bodies of research. The first is the visible and empirical association of female inequality in Islamic societies where Purdah values severely circumscribe possible choices. The second is the role of the detached social scientist generally permitting two sweeping interpretations for individual choice: choices constrained by cultural norms or choice as an expression of independence (appealing to expatriate researchers). Though, neither option acknowledged the wisdom, the tongue-in-cheek sarcasm of village women. A more sophisticated approach to understanding choice is grounded, again, in the idea of the implicit contract. While family role may be culturally defined, there are rights and responsibilities that permit the negotiation of some choices within the family unit. The work of Bourdieu creates an even greater range of choice with his idea of habitus. Here, cultural norms create space for choice by delineating the possible actions from the impossible, and allows the individual to strategically use available resources to improve their own self-interest within those bounds. Here, agency is not the mechanical execution of culture; rather it is the creative interpretation of rules.

In the end, Kabeer builds out the idea of the middle-ground, a place where choices are less determined and that provides a theoretical basis for considering the transformative potential of human choice. She suggests this potential for women (and other oppressed classes) can be realized in two levels. First, the availability of paid employment for women so that they can better negotiate their positions within the family. Second, the development of a more finely grained knowledge that provides an accurate portrayal of individual choice. This requires a methodology that is sensitive to the pitfalls of personal testimonials and forces the researcher into a more involved role with the participants.