Two principles should guide the choosing of ECCD models. First, construct models that fit national and local needs. The best model is one constructed on site, not one imported from elsewhere. Although models that have worked in other places can be extremely useful as sources of ideas and new ways of thinking, the specific conditions of a country or area make it difficult or impossible to import a model wholesale. We should expect models to respond to conditions and needs, not expect to mold conditions and needs to fit a particular model.

Second, consider a mix of models. In most settings a mix of complementary strategies and models will serve better than one strategy or model. This is so for several reasons:

• A child's development is influenced in different ways by the immediate environment, by the general community environment, by the environment provided through social institutions, and by the overall cultural environment. Dealing with each of these influences requires a different strategy.

• Different geographical and social conditions demand different strategies. For example, reaching children in an urban setting will demand a different model than reaching children in dispersed settlements. And reaching children of migrant parents will require a different model than reaching children in a stable location.

• Models for reaching parents and other adults who affect upbringing differ from those for reaching children directly, yet a project seeking long-term as well as short-term effects should include both.

What strategies and models are available?
The three most basic strategies for an ECCD project are supporting and educating caregivers, attending to children in centers, and promoting child-centered community development. Other strategies-such as strengthening institutions, disseminating knowledge, advocating for child development, and establishing legal and regulatory frameworks-essentially provide support to these more action-oriented approaches (see table 3.1 for a list of general strategies and corresponding models; the list is intended to be suggestive, not exhaustive).

For each general strategy, a range of models can be considered. And for each model, a range of options can be considered for putting it into practice. In educating parents and caregivers, for example, the home visiting model has used different visitors in different settings-health personnel, home economists employed in agricultural extension programs, and specially trained members of the local community chosen because of their success in bringing up their own children. Home visiting programs have sometimes been coordinated with mass media programs. Or they may be combined with a center-based program.

The menu of ECCD strategies and models is extensive. There is no one right way. If project discussions begin with the menu rather than with a particular model, the arrangement best suited for a setting will emerge in the course of the discussions.

Table 3.1 Complementary approaches and models for early childhood care and development programs

Program approach	Focus	Objectives	Models
Support and educate caregivers	Family Parents Other adults Siblings	Create awareness Change attitudes Improve or change practices	Home visiting Parental education Child-to-child Mass media
Attend to the child directly in centers	Child Age 0-2 Age 3-6 Age 0-6 Age 6-8	Survival Comprehensive development Socialization Rehabilitation Child care Transition to primary school	Home day care Integrated child development centers Workplace Preschools: formal, nonformal Primary school After-school programs
Promote child-centered community development	Community Leaders Promoters Residents	Create awareness Mobilize for action Change physical conditions	Education Technical or social mobilization
Strengthen institutional resources, capabilities	Institutions Professionals Paraprofessionals	Create awareness Improve skills Improve physical conditions	Training Provide plant or equipment
Develop policies, laws, regulatory frameworks	Frameworks Legislators, Policymakers, Bureaucrats	Increase awareness Create supportive laws, settings	Create alliances Legislate tax incentives, leave policies, and the like Set standards
Advocate for child development	Knowledge Policymakers Professionals Public	Create awareness Build political will Increase demand Change attitudes Establish national policies or laws	Social marketing Knowledge dissemination Create alliances Lobby

For descriptions of ways in which models have been applied see references in tab 8. Many examples of models can be found in The Twelve Who Survive (Myers 1992b) and in the World Bank publication Early Child Development: Investing in the Future (Young 1996).

What should be taken into account when choosing among models?

The choice of models for a program should take into account:

• The characteristics of the population the program is to serve.

• Information about current child care arrangements and the demand for different kinds of care.

• Information about models that have been tried in the country and elsewhere.

• Costs (this topic is taken up in tab 5).

The proposed beneficiaries. As discussed above, a first step in project planning is to make choices about the children (and thus the families and communities) who will be included in the project. These choices entail applying social criteria, defining the age group, and setting criteria related to the children's developmental status.

If the project is to focus on children under three years of age, serious consideration must be given to parental support and education options (such as support for building social networks that will reduce stress, or home visiting or parental education classes). Most younger children are still under the care of their mother or another family member, and many families are reluctant to put their very young children in centers under the care of another person.

But a program of adult education, no matter how good, will not meet the child care needs of many families. Some families will be unable to take care of younger children at home because of a need to work to survive and because of jobs that are incompatible with child care. To meet the needs of these families, a center-based program should be considered, even for children under age three. Such infant or nursery programs will require more intensive attention and a higher staff-to-child ratio than similar programs for older preschool children. In programs of child care and development for children under three, every effort should be made to include a strong health and nutrition component. These needs will likely bias the location of a project toward the health or welfare sector rather than education.

If the main project population of preschool children is those aged three and above, the demand for center-based care is likely to be high, both because of a need to free parents to work and because as children become older, they can benefit much more from interacting with one another. In this case some preference for the center-based strategy and related models emerges. To the extent possible, center-based programs should be combined with parental support and education.

If the focus is on educating adults, the model chosen must be at least as sound in its methods of adult education and interaction as it is in the information it provides about child care and development.

The choice of models should also take into account the cultural characteristics of the population the program is to serve. Some models will work in one cultural setting but not another.

Current arrangements and demand for child care. Before deciding on a center-based approach to child care and development, program managers should pay careful attention to child rearing arrangements. Where extended families or local arrangements for providing care function well, supporting those arrangements may be more appropriate than establishing centers.

Program managers should also be careful to define the real demand for child care and education. Potential demand, defined by the number of children in a particular age group or the number of families in an area, will always exceed real demand, because some families will always prefer to provide care and education themselves. Some families that are not part of the real demand are in a position to provide quality care and education; others are not, but for cultural reasons prefer to avoid outside interference. Others are skeptical about the kind of service being offered, but would be willing to participate in another kind of child care or education program. And others are simply uninformed. Thus the demand for child care and education must be differentiated. In some cases the first strategy may need to be to create demand. That leads to the next consideration in choosing a strategy or model.

Should a day care or child development model be chosen? There is no rule of thumb to guide the decision between home day care and larger community centers that bring children together in an institutional rather than a home setting. In general, day care in a home is thought to be more intimate, to provide greater personal attention to children, and to offer more flexible hours.

But the intimacy and personal attention caregivers provide depend on their personality and training and on the number of children for whom they are responsible. With an adequate ratio of staff to children and well-trained, loving caregivers, a larger community center can provide an environment as good as or better than that in a day care home. Moreover, conflicts may arise between natural mothers and home day care mothers because of a confusion of roles, something less likely to happen in a larger center. If community centers are borrowed, however, and not dedicated exclusively to child care, the environment may not be conducive to good care.

In sum, the choice between a day care and child development model will depend more on the availability of facilities, the personalities and training of the caregivers, and such factors as the need for flexibility than it will on whether care is given in a home or a community center.