Sampling can help you reach a statistically valid conclusion about a data set from a relatively small number of samples. You can sample the entire data set, a subset of the data, or use global filters to perform conditional sampling.

Two commonly used methods of generating sampling sizes are the
Poisson and binomial distributions. *ACL* generates sample
sizes using the Poisson distribution, which does not require you
to know the size of the data set before you generate a sample size.

*ACL* can produce statistically valid attribute sample sizes
for most analyses. Exceptions may apply in the following situations:

You are sampling data sets of less than 1000 records

Your organization has in-house sampling experts who can define sample sizes precisely tailored to your needs.

Your organization has mandated the use of another sampling tool or methodology.

For typical data sets of a thousand or more records, the Poisson and binomial distributions generate nearly identical sample sizes. For populations of under a thousand records, sample sizes determined with the Poisson ratio tend to be slightly larger and therefore more conservative than sizes determined with the binomial distribution. This is because the binomial distribution adjusts the sample size downward for small populations but the Poisson distribution does not. With very small populations, the fixed sample size generated by the Poisson distribution can actually exceed the population size.

When calculating sample sizes in *ACL*, recognize that for
record sampling of small data sets, the sample size may be larger
than you need. This does not present an obstacle to analysis because
it is common practice to manually oversample small populations.