|Comparing SF 3 Estimates with Corresponding Values in SF 1 and SF 2|
As in earlier censuses, the responses from the sample of households reporting on long forms must be weighted to reflect the entire population. Specifically, each responding household represents, on average, six or seven other households who reported using short forms.
One consequence of the weighting procedures is that each estimate based on the long form responses has an associated confidence interval. These confidence intervals are wider (as a percentage of the estimate) for geographic areas with smaller populations and for characteristics that occur less frequently in the area being examined (such as the proportion of people in poverty in a middle-income neighborhood).
In order to release as much useful information as possible, statisticians must balance a number of factors. In particular, for Census 2000, the Bureau of the Census created weighting areas --geographic areas from which about two hundred or more long forms were completed-- which are large enough to produce good quality estimates. If smaller weighting areas had been used, the confidence intervals around the estimates would have been significantly wider, rendering many estimates less useful due to their lower reliability.
The disadvantage of using weighting areas this large is that, for smaller geographic areas within them, the estimates of characteristics that are also reported on the short form will not match the counts reported in SF 1 or SF 2. Examples of these characteristics are the total number of people, the number of people reporting specific racial categories, and the number of housing units. The official values for items reported on the short form come from SF 1 and SF 2.
The differences between the long form estimates in SF 3 and values in SF 1 or SF 2 are particularly noticeable for the smallest places, tracts, and block groups. The long form estimates of total population and total housing units in SF 3 will, however, match the SF 1 and SF 2 counts for larger geographic areas such as counties and states, and will be essentially the same for medium and large cities.
This phenomenon also occurred for the 1990 Census, although in that case, the weighting areas included relatively small places. As a result, the long form estimates matched the short form counts for those places, but the confidence intervals around the estimates of characteristics collected only on the long form were often significantly wider (as a percentage of the estimate).
SF 1 gives exact numbers even for very small groups and areas, whereas SF 3 gives estimates for small groups and areas such as tracts and small places that are less exact. The goal of SF 3 is to identify large differences among areas or large changes over time. Estimates for small areas and small population groups often do exhibit large changes from one census to the next, so having the capability to measure them is worthwhile.