Today we released a new version of the CHAT dataset. There were a handful of tweaks made to the primary CHAT dataset, including the removal of an identifiable variable (recruitment site) and clarification of the household income-related variables. The latter issue was raised by an NSRR user who was attempting to incorporate socioeconomic status into her analysis. We agreed that the existing set of income-related variables were confusing and difficult to distinguish, so we cleaned up some of the definitions in our data dictionary.
The EEG spectral analysis datasets were also updated as part of this release. Members of the core NSRR team made improvements to the EEG signal processing pipeline, including the addition of ECG contamination checking, before re-running the analysis for CHAT. EEG spectral analysis results were also generated for non-randomized CHAT subjects.
Please reach out to support@sleepdata.org if you have questions about this release!
Here is the CHANGELOG for version 0.5.0:
income
variable (duplicate)siteid
variable (replaced by clusterid
)par5
and par5_rc
) definitionsref4
(race) to non-randomized subject dataset