Annual Survey of Public Pensions: State & Local Data

Data Processing: 2011

Editing

Editing is a process that ensures survey data are accurate, complete, and consistent. Efforts are made at all phases of collection, processing, and tabulation to minimize errors. Although some edits are built into the Internet data collection instrument and the data entry programs, the majority of the edits are performed after the case has been loaded into the Census Bureau’s database.

Edits consist primarily of two types: consistency and a ratio of the current year’s reported value to the prior year’s value. The consistency edits check the logical relationships of data items reported on the form. For example, if a value exists for the number of retirees receiving benefits because of age or length of service then there must be a value reported for the amount paid. The current year/prior year edits compare by item code the data reported for the current year with data reported for the prior year. If data fall out of acceptable tolerance levels, the item is flagged for review. For both types of edits, the edit results are reviewed by analysts and adjusted when needed. When the analyst is unable to resolve or accept the edit failure, contact is made with the respondent to verify or correct the reported data.

Imputation

Not all respondents answer every item on the questionnaire. There are also questionnaires that are not returned despite efforts to gain a response. Imputation is the process of filling in missing or invalid data with reasonable values in order to have a complete data set.

For both state and local government pension systems, the imputations were based on either a prior year annual survey or the most recent Census of Governments. All but six missing variables (Z90, Z95 and four relatively new variables: Z13, Z14, Z15, and Z16) were imputed using one of the following methods: cell median or donor distribution of Z81, cell mean, or reported prior year or census year data which was multiplied by a growth factor. If the non-respondent to Z90 and Z95 does not write in anything in the Other categories for Z90 and Z95, we impute those variables to be zero.

Sampling Error

The data that are provided come from a sample rather than a census of all possible units. The particular sample that was selected is one of a large number of possible samples of the same size and sample design that could have been selected. Each sample would have yielded different estimates. The estimated coefficients of variation, which are provided for each estimate, are an estimate of this sampling variability. In this tabulation, the coefficients of variation are expressed as percentages. The coefficient of variation is the standard error as a proportion of the magnitude of the estimate. In the tables, the coefficient of variation expresses the standard error as a percentage of the quantity being estimated.

State government pension systems are not subject to sampling error. Consequently, state and local government aggregates for individual states are statistically more reliable than the local government only estimates.

Revisions

The Annual Survey of Public Pensions: State- and Locally-Administered Data released data for Fiscal Year 2011 on November 15, 2012. Users should note that this release also includes revisions for Fiscal Years 2007, 2008, 2009, and 2010. The revised data are accessible on the survey’s website.

For further information, please refer to the 2011 Annual Survey of Public Pensions: State- and Locally-Administered Defined Benefit 2011 Survey Methodology.[PDF 124KB]