Description

The original source for this data set is the IPUMS project (RugglesSobek, 1997). The IPUMS project is a large collection of federal census data which has standardized coding schemes to make comparisons across time easy. The data is an unweighted 1 in 100 sample of responses from the Los Angeles -- Long Beach area for the years 1970, 1980, and 1990. The household and individual records were flattened into a single table and we used all variables that were available for all three years. When there was more than one version of a variable, such as for race, we used the most general. For occupation and industry we used the 1950 basis. Note that PUMS data is based on cluster samples, i.e. samples are made of households or dwellings from which there may be multiple individuals. Individuals from the same household are no longer independent. Ruggles (1995) considers this issue further and discusses its effect (along with the effects of stratification) on standard errors. The variable schltype appears to have different coding values across the years 1970, 1980, and 1990. There are two versions of this data set: 1. The Small Data Set The small data set contains a 1 in 1000 sample of the Los Angeles and Long Beach area. It was formed by sampling from the large data set. 2. The Large Data Set The large data set contains a 1 in 100 sample of the Los Angeles and Long Beach area.

Related Papers

  • Ke Wang and Shiyu Zhou and Ada Wai-Chee Fu and Jeffrey Xu Yu. Mining Changes of Classification by Correspondence Tracing. SDM. 2003. [link]
  • Stephen D. Bay and Michael J. Pazzani. Detecting Group Differences: Mining Contrast Sets. Data Min. Knowl. Discov, 5. 2001. [link]
  • Chris Giannella and Bassem Sayrafi. An Information Theoretic Histogram for Single Dimensional Selectivity Estimation. Department of Computer Science, Indiana University Bloomington. [link]
  • [link]

Related datasets