We draw on gender theory and neo-institutional theory to examine the impact of workplace characteristics and family demands on negative job-to-home and home-to-job spillover. Our multivariate analyses of the 1997 National Study of the Changing Workforce data indicate that familysupportive workplace cultures reduce negative spillover in both directions, whereas the availability of company policies, such as dependent care benefits and flextime, do not. Our results also show that family demands

more »

... rease spillover more for women than for men. Our findings suggest that the atmosphere of the workplace is more important than the availability of company policies in reducing negative spillover. Ever since Kanter (1977) debunked the myth of separate spheres for work 1 and family, researchers have been examining the nature and extent of the job-family interface. One facet of this interface is the spillover from one domain to the other, whereby experiences in one domain moderate the experiences in the other (Barnett 1994). Spillover conceptually represents the process whereby behaviors, moods, stress, and emotions from one realm of social life affect those in another and vice versa (Williams and Alliger 1994; Frone, Yardley, and Markel 1997) . Spillover can be positive, but our concern is with the negative spillover, or workfamily conflict, people experience while trying to balance a job and family in contemporary society (Greenhaus and Parasuraman 1999; Grzywacz, Almeida, and McDonald 2002; Schieman, McBrier, and Van Gundy 2003) . Negative spillover occurs when demands from the two domains of job and home compete for an individual's time, energy, and attention (Small and Riley 1990) . For example, a worker whose child or elderly parent is ill may be less able to concentrate on the job. Similarly, a worker who is facing a tight deadline at work might have less time to help a child with homework or 108 The Sociological Quarterly 46 (2005) 107-135