A good evaluation mechanism is another element needed to establish a successful information-sharing network (Wagner et al., 1997). Without evaluation tools, program participants are unable to determine whether the program is serving the needs of its intended clients and cannot make adjustments where necessary to do a better job. The lack of evaluation results can make it impossible to determine with any certainty whether information-sharing programs are effective in preventing or reducing juvenile delinquency (Isaacs, 1992).

Moreover, Wagner et al. (1997) suggest that both internal and external evaluations are necessary. An internal evaluation provides the participants in the network with a firsthand look at the way the program operates and builds capacity within the collaborating organizations for analyzing information and improving decisionmaking. An external evaluation provides for a degree of objectivity that an internal evaluation might not. Therefore, Wagner et al. (1997) set out six major goals that should be established when structuring a system evaluation process:

• Provide a rationale for the collaboration. This means making a commitment to a particular set of results that are meaningful for each member agency and the community at large. This commitment will provide a stronger basis for partners to continue to work together.

• Increase the program's effectiveness. In other words, track the program's progress toward a specific result. This tracking makes it more likely that the program will actually make a difference because the participants will know if they need to make changes to increase the program's effectiveness.

• Improve accountability to diverse groups. The evaluation should provide a detailed report of the results achieved or not achieved to help the partners tailor future efforts.

• Increase the program's credibility. An evaluation will provide documented results that will increase the program's credibility with funders and constituents who want to know that the money is being used efficiently and effectively. This may be particularly important given that one of the major impediments to the continued existence of many programs is the lack of ongoing and stable funding.

• Advocate for change in the system. Having ongoing documented knowledge of what is and is not working in the current system can provide a credible rationale for making fundamental changes in how the system works.

• Take into account the information needs of all concerned. Community participation is required to help build support for collaborative efforts. When an interagency information-sharing program enlists all groups that are affected by juvenile delinquency, every community resource can be used to its fullest potential. It is also important to stay flexible to address new information needs that arise within the partnership.

In addition to the goals set out for the evaluation, Wagner et al. (1997) suggest that the evaluation must have specific strategies for reaching these goals. The evaluation must be clear about the outcome desired by the partnership—in this case, a decrease in the amount of juvenile delinquency. Wagner et al. (1997) also assert that the evaluation must choose adequate measurement tools to assess the success of partnership activities for which data are available. These measurement tools must demonstrate a link between the activities of the partnership and its stated goals. If this is accomplished, the evaluation will necessarily meet the six goals outlined by Wagner et al. (1997) for the evaluation process. Thus, the evaluation should have at least two components—goals and measures that will be used to assess whether the partnership is reaching the goals. To successfully create these components, it is crucial to identify at the outset the assumptions that the partners are making about what will and will not work in a given community (Wagner et al., 1997). Proposed activities must be consistent with these assumptions. It is also important to continuously update the partnership's goals to reflect revised activities and to create short-term, interim, and long-term indicators for the continuous evaluation process.