Day Care, Measuring Quality of Care

Researchers recognize that there are vast differences in the quality of care that children experience [Page 323]both in and outside their homes. Much of the literature confirms that children perform better on many dimensions when they have received care of higher quality. Measurement of quality has improved dramatically in the past two decades, and this has permitted significant progress in evaluating the effects of the quality of care on children's development.

Researchers and policymakers have developed what are known as process and structural measures of quality. Conceptually, structural and process measures differ to the extent that factors indexed by the structural measures promote high-quality interaction and care but do not guarantee it, whereas process measures try to quantify the actual care received by children. In practice, however, some of the most popular process measures include indicators of environmental features (including those that index safety, for example) as well as indicators of interactional quality.

Observational measures of the settings and of the interactions between care providers and children are often described as process measures, and the standardized measures developed by Thelma Harms and Richard Clifford (e.g., 1980) have been employed by most researchers in the United States. In addition to Harms and Clifford's measures, several other measures have also been used in major studies conducted in the United States. Scores on the various process measures are in fact highly correlated with one another, making it possible to use composite measures containing fewer items to measure quality than the complete measures do.

These measures have proven to be less useful as indices of the quality of care in Western Europe, however, perhaps because the quality of care available there is less variable and of higher average quality (Lamb, Sternberg, Hwang, & Broberg, 1992). Recognizing the need for more systematic and comprehensive measures that could be used internationally, Pierrehumbert and colleagues (1996) developed and validated measures of quality that could be used in countries where day care is of very high quality.

Instead of process variables, many researchers index quality using a number of structural indices, including measures of teacher training and experience, group size, teacher-child ratios, crowding, staff turnover, and the like. Most of these factors can be and often are regulated.

The adult-child ratio and the extent of teacher training are among the best and most widely used structural indices of high-quality center-based care, whereas group size, the degree of safety, and the appropriateness of care provider behavior best index the quality of family day care (Howes, 1983, 1997). The care providers' salaries also provide a valuable indirect measure of the quality of care, because salary levels predict the rate of staff turnover reasonably well. Howes has also introduced an important distinction between the conventional structural measures of quality (group size, adult-child ratio, care provider training) and more comprehensive and empirically derived measures, such as the number of care providers present at any given time, staff turnover, the number of settings experienced by each child, care provider sensitivity and involvement, and the provision of developmentally appropriate activities. Unfortunately, site- or care provider–specific measures of quality often fail to take account of the substantial frequency of moves by children from one setting to another. From an individual point of view, such instability may be developmentally disruptive.

...