Choosing data sources and the people they're applied to
If your indicators are specific, usually they will suggest the data source to measure it as well as the people in the target audience it should be applied to.
What people or program data will the indicator be applied to? Consider the following issues.
Examine the visuals below to explore issues related to each program. For background on programs, visit the Cases page on our full site.
Some or all of the target audience?
Remember that your target audience is usually stated in general characteristics and you may only involve part of that audience in your program as participants.
Illinois State Museum Changes
Because of high numbers involved, only randomly selected visitors and students will be interviewed after their visit to the Changes exhibit at the Illinois State Museum. For example, the interviewers might talk only to every fifth person who crosses a line at the exhibit.
West Dakota Library Rx
All WD residents are part of the target audience, but they cannot all be surveyed. A random sample of residents is assumed to be representatives of the whole group.
Some or all of the participants?
Springfield Library Summer Reading Program
All the participants are students in an urban library district. Should reading tests be monitored for all participants? At-risk students only? Those who complete a specific number of activities?
People other than the participants?
Springfield Library Summer Reading Program
Are parents likely to be better sources of information about their children (such as number of hours spent reading and attitude toward reading) than the children themselves?
Remember: Collecting data costs time and money. Collect only enough information to figure out if your program is successful, so be specific and concrete. Consider the difference in costs between collecting information about "children" and "children who need after-school tutoring." Or do you mean "children who participate in at least five tutoring sessions? If you expect 100 children to meet this criterion, should you say instead "a random sample of children who participate in at least five tutoring sessions"?