Determining Comparability: In order to confidently compare two or more percents/rates, the numerators and denominators need to be compatible. Use data that have been collected with scientific rigor and consistency to ensure the integrity of the data. In addition, here are some other factors to consider:
Indirect Measures: Community indicators are not necessarily direct measures of the population's behavior but may be measures of the community service systems. Policy shifts, resource fluctuations, degree of centralization and standardization, local reporting practices and local service delivery differences affect indicator data. Other factors in the community-including availability, affordability and quality of service-may also affect indicator data. Local qualitative information regarding the role of the indicator in the context of its service setting must be obtained and applied when interpreting the results of indicator-based studies. For example, the youth arrest rate for driving while intoxicated is a direct measure of a law enforcement response to youth alcohol and other drug (AOD) use and is an indirect measure of your AOD use in the population.
Number: The number is the count of occurrences within a defined geographic area during a specified period of time. Numbers are used to determine the size of an occurrence in a particular location. Numbers, however, do not take the size of the population who could experience the occurrence into account. Therefore, numbers cannot show the probability of this occurring in the population. While numbers can compare the size of an occurrence within the same population group, numbers generally should not be used for comparisons, especially when comparison groups have differing population size or composition.
Small Numbers: Rates are sensitive to variations in the size of both the numerator (the number of occurrences) and the denominator (the estimated population who could experience the occurrence). KWIC reminds users that rates and percents based on fewer than 20 occurrences may exhibit significant variation from one year to the next. This variation is often referred to as "not stable."