Data: What makes sense to use?
Your common sense and knowledge of the situation will suggest what data to use. The kinds of data sources explained give the most common types. Remember that OBPE programs do not require formal research; although some programs may wish to get a consultant involved in evaluation.
No kind of data source is better than another. The data source chosen should depend on what is being evaluated.
Which data source should be chosen for the following? To indicate that children are developing a habit of reading: "# and % of Springfield students in the summer library reading program who spend at least an hour per day for independent reading for fun.
Anecdotes: Possible measure. If enough participants discuss their reading, you can get a good estimate of how much recreational reading is going on.
Survey: Good measure. Parents and children are the best source of information about behavior that takes place at home.
Participant project: Bad measure. This outcome is concerned with "normal" behavior, not specific projects.
Organizational records: Limited measure. Circulation records can show how many books are checked out. They might not be read…but it is something of an indication of behavior.
Which data source should be chose for the following? To indicate that West Dakota residents use public library databases as a preferred source of information: # and % of WD residents who say that they are likely or very likely to use the public library databases as one of their first 3 sources of health information.
Anecdotes: Limited measure. People who volunteer stories sometimes are not representative of the larger audience.
Survey: Good measure. A mail survey of randomly chosen citizens can validly measure general audience knowledge.
Participant projects: Bad measure. Most residents will be reached indirectly, through information providers, and so won't have any project.
Organizational records: Bad measure. Most citizens will access this information privately, even if they do so in a library.
Which data source should be chosen for the following? To indicate that Girl Scouts learn bird-identification skills: "# and % of Girl Scouts who correctly identify five birds common to the area on a field trip.
Anecdotes: Limited measure. Enthusiastic Girl Scouts are good, but this won't measure the quiet ones.
Survey: Weak measure. Some Girl Scouts might over-estimate their ability to identify birds.
Participant projects: Good measure, if the workshop ends with girls taking photographs and correctly identifying the birds. A very direct indicator.
Organizational records: Good measure, if the Girl Scout badge process includes a test of their identification skills. This will be an easy source of information.