The evaluation of a CCI usually takes place on two levels. Generally, you will require each grantee to evaluate at the site level. Beyond this, you may want to conduct a multisite evaluation to study the initiative as a whole. Although it might appear easy to combine the two levels--using a uniform set of data from all sites to evaluate both individual sites and the initiative as a whole--CCI evaluation is rarely that simple.

• The questions sites want to answer through the evaluation will vary depending on the interests of local stakeholders.
• The capacity for data collection will vary across sites.
• The multisite evaluation may depend on data that a site would not otherwise collect for its own evaluation.

Coordinating the multisite and local evaluations will make the best use of resources.

Design and award the multisite evaluation before you select sites for funding. That way you can involve the multisite evaluator in establishing the evaluation requirements for sites to ensure that multisite and local evaluations are as well coordinated as possible.

Select a multisite evaluator who is flexible and skilled in working with local communities. (See Guideline #2, What qualities should I look for when I select a multisite evaluator?)

Draw on data that is already being tracked by sites. In the design of the multisite evaluation, build on existing local capacity for tracking and using data.

Find out what data local stakeholders need to track, and incorporate it into the core elements of the multisite evaluation.

Use evaluation methodologies that make it easier to handle variation among sites. (See also Guideline #2.) Allow for variation among sites in how receptive they are to evaluation, their capacity to conduct evaluation, and the questions they want to answer. Be aware, however, that allowing too much variation across sites may be perilous. If sites differ widely in the questions they answer and the data they collect, you won't learn as much from the multisite evaluation. So you need to maintain a core of common purpose and approach.

Schedule meetings to guide sites through the development of local evaluation approaches. After the sites have formulated their individual logic models, schedule a meeting--for funders, evaluators, and sites - to reconcile inconsistencies between site logic models and the national logic model. Educate site representatives and local evaluators about the national evaluation model, highlighting how there can be alignment.See the example of a web-based tool created by a funder to help sites develop a logic model that is coordinated with the federal logic model.

Take advantage of cross-site conferences for evaluators to continue meeting with site representatives, reviewing evaluation progress and challenges.

See What Makes a Solution for information about integrating local and multisite evaluations through a hub and spoke model, pages 15, and 39-44.

Institute a process for regular data collection at all sites. There are several models for collecting data about initiative activities.

• Require each grantee to collect and report regularly on its own data, based on a common set of variables that address both systems change and programs.
• Arrange for a contractor to collect a common set of data at all sites.
• Ask a cohort of grantees to participate in the development of its own data collection system. (This will also encourage grantee buy-in to the system.)

Make sure that continuous feedback is built into contracts and grants with both multisite and local evaluators. Set standards for site evaluation, and specify continuous feedback as a deliverable.

Use the evaluator's ongoing reports and observations to make decisions about improvements, supports, and expectations. For example, evaluation results can help you estimate funding for a new round of initiatives or for sustainability of current sites. Use findings also to adjust timelines.

Communicate findings from the multisite evaluation back to sites on a regular basis.

Use evaluation TA to help each site establish a feedback loop. Help sites learn how to analyze their data, draw conclusions, and use the findings to inform decisionmaking.

Draw on the site's evaluation findings during regular meetings with the program officer. Use the site evaluation findings as a basis for discussion about progress and planning.