print share

CHIS Data Quality

CHIS 2003 study benchmarking key estimates for CHIS with those from major federal health surveys

One approach to assessing data quality is to compare results from one survey with those from another. This can be especially productive if one of the surveys is a well-established, high quality survey that can be used as a "gold standard" for benchmarking.

In collaboration with the National Center for Health Statistics (NCHS) and the Agency for Healthcare Research and Quality (AHRQ), we compared a variety of estimates from CHIS 2003 to specially weighted California extracts of the 2003 National Health Interview Survey (NHIS) and 2003 Medical Expenditure Panel Survey (MEPS). The NHIS clearly meets the threshold of a well established, high quality survey with a long history. Conducted in person, NHIS and MEPS achieve higher response rates than most telephone surveys (including CHIS); the response rates of 2003 NHIS, MEPS and CHIS were 72.8%, 64.5% and 33.5%, respectively.

Comparisons of numerous estimates from the three surveys did not reveal striking differences. Although many of the differences were statistically significant, this is mainly a function of the large CHIS sample size. Furthermore, no systematic patterns in the differences were evident in the results (i.e., for some estimates the CHIS sample appeared healthier, in others the NHIS, etc.). This benchmarking study finds no clear evidence that the quality of CHIS data is lower than that of NHIS or MEPS, and that the relatively lower CHIS response rate does not result in low quality data. (This study is nearing completion; when results have been published in a peer-review journal, they will be made available on this web site.)

See further documentation on this