TABLE OF CONTENTS Data must be collecting using one or more of the methods outlined in this chapter. As the data are collected, several other issues need to be addressed and clarified.
This is a principle discussed earlier, but it is worth repeating. The data used in the analysis must be the most credible data available. If data are collected from more than once source, the most credible one is used, if there is clearly a difference. This leads to a guiding principle.
When collecting and analyzing data, use only the most credible sources.
It is rare for all the participants to provide data in a follow-up evaluation. The philosophy described in this chapter is to use only the available data for the total benefits. This philosophy is based on making every attempt possible to collect data from every participant, if at all possible. In reality, the return rate of questionnaires or the participation rate of other data collection methods will probably be in the 60 80% range. Below 50% should be considered questionable because of the extreme negative impact it will have on the results. This leads to a guiding principle:
If no improvement data are available for a population or from a specific source, it is assumed that little or no improvement has occurred.
Data should be tabulated and summarized, ready for analysis. Ideally, tabulation should be organized by particular evaluation levels and issues.
