This article uses a nine-year period of work-life history data from the British Household Panel Survey (1991-1999) to examine married/cohabiting womens work trajectories. In particular, it tests some major contentions of Hakims (2000) preference theory. Both supportive and opposing evidence for the theory has been found. First, concurring with Hakims arguments, women who have followed a home-career path hold consistently more home-centred attitudes over time than women who have been committed to their employment careers. Moreover, it is found that presence of dependent children has little or no negative effect on a work-centred womans chance of being engaged in full-time work. But the findings could not rule out the possibility that womens employment careers are still constrained. The most work-centred women (as revealed in their gender role attitudes in the nine-year period), despite having been committed mostly to a full-time work, still have displayed a certain degree of discontinuity in their career pursuits. Finally, contrary to corollary of the preference theory, the relationship between gender role attitudes and womens participation in labour market work is reciprocal rather than unidirectional. That is, womens work orientation is endogenous to their labour market experiences.