ABSTRACT
What use is it to do meta-analyses? Many answers to that question have been formulated (Cooper, 1998; Hedges & Olkin, 1985; Hunter & Schmidt, 1996; Rosenthal, 1991), but one of the shortest may be: Meta-analyses are necessary for making policy and also for planning new research (Borenstein, Rothstein, & Cohen, 2000). They also help to get a grip on a certain research field, to find out which studies exist and what they have produced. In the case of preventive interventions, the number of studies has been increasing exponentially in the last decades. The idea that early interventions may be most effective in preventing less optimal or even deviant developmental pathways has led to an increase in efforts to shape and evaluate preventive interventions in the first few years of life, and even before the birth of the target child (e.g., Barnard et al., 1988, Huxley & Warner, 1993; Jacobson & Frye, 1991; Leitch, 1999; Luster, Perlstadt, McKinney, Sims, & Juang, 1996). In particular, the concept of a critical period of fast neuropsychological growth – as well as the possibly irreversible effects of impaired neurological development during the first three years after birth – has stimulated early interventions, although the scientific evidence for a critical period of neurological maturation is actually rather shallow (Fox, Leavitt, & Warhol, 1999; Shonkoff & Phillips, 2001).
