In general, the FHCS reports contain two major types of data comparisons:

1. Subgroup results-for example, males versus females or headquarters employees versus field employees.
2. 2004 results versus 2002 results.

Analysts tested whether there was a statistically significant difference between the positive percentages for each comparison (e.g., males versus females). When the results indicated there was less than a 5 percent chance that the difference between the two percentages was due to chance, they concluded the difference is reliable. The result is reported as "Yes" or an up (improved) or down (declined) arrow in the report columns labeled "Significant Difference."

Finding that a difference between two percentages is statistically significant does not necessarily mean the difference is meaningful. Government managers need to rely on their substantive understanding of the survey topic to decide if a statistically significant difference is important.