Introduction ALTHOUGH KINDERGARTENS in the United States have a history of more than 150 years, defining the most beneficial kindergarten program is an issue still to be settled. In early traditional kindergarten, the most beneficial program was identified as one that allowed a child to make a smooth transition from home to school. All that children needed to do in traditional kindergarten was to play and adjust themselves to a social setting (Nelson, 2000). These days, kindergartens in the United States emphasise academic achievement and school readiness, which had been the goal of the first grade curriculum in the past. American kindergarten children are expected to pay attention to highly structured and segmented academic instruction such as reading, writing, mathematics and science lessons (Clark, 2001). While there are debates among policy-makers and scholars over the benefits of full-day versus half-day kindergarten programs and the developmental goals of these programs, the large numbers of American kindergartens try to comply with the learning standards set by the state governments. Those guidelines delineate the general expectations for what young children should know and be able to do before they enter first grade (Cryan, Sheehan, Wiechel & Bandy Hedden, 1992; Elicker & Mathur, 1997). Moreover, the issue of high accountability of schools and the growing demands of a push for an 'early start' add to the importance of kindergarten children's academic performance (Vecchiotti, 2003). Along with the curricula of kindergarten programs, the length of kindergarten has also been an important policy issue in the United States. The number of all-day kindergartens in the US has substantially increased: a typical all-day kindergarten program runs five-six hours per day and five days per week, while a half-day runs two-three hours per day and three-five days per week. In 1999, 56 per cent of all kindergartens programs were all-day programs. Many kindergartens have switched from half-day to all-day programs in recent years, and many more are likely to do so (Elkind, 2001; Fusaro, 1997; Weast, 2001). The changing educational demands on young students explains the rapid growth of all-day kindergartens. Social reasons, such as a growing need for child care, have also provided an impetus for the growth of all-day kindergartens. Many more mothers are in the labour force, and all-day kindergarten is a safe and preferred alternative to other forms of child care (Boardman & Kelly, 2004). Educational policy-makers have also argued in favour of all-day kindergarten for low-income minority children (Magnusona, Ruhmb & Waldfogelc, 2007; Zvoch, Reynolds & Parker, 2008). There are many important policy issues embedded in the debate over the pros and cons of the all-day program. The shift from a half-day to a full-day kindergarten program can be extremery expensive. The change to an all-day program requires a substantial increase in the kindergarten budget for teachers' salaries and additional classrooms and other resources. Not only are the all-day programs more expensive, they also require more human resources, and make increased educational demands on young children. So it is important to critically examine the benefits of all-day versus half-day programs, and to gather data-based evidence on the effects, both short-term and long-term, of all-day programs on children's academic achievement. Although there is growing literature on the effects of all-day kindergarten and children's cognitive development, much of that work consists of policy briefs and concept papers. Proponents of all-day kindergarten claim, however, that those programs are beneficial and worth funding (Vecchiotti, 2003; Weast, 2001). Some studies have supported the benefits of all-day kindergarten for special groups of children and their families. For example, research shows that single-parent and dual-employment families prefer the convenience of all-day kindergartens (Clark & Kirk, 2000; Elkind, 2001), and the academic achievement of children who are at risk is higher for those in all-day kindergarten than for those in half-day kindergarten (Elicker & Mathur, 1997; Fusaro, 1997). …