> For the complete documentation index, see [llms.txt](https://bdcatalyst.gitbook.io/biodata-catalyst-documentation/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://bdcatalyst.gitbook.io/biodata-catalyst-documentation/written-documentation/explore-available-data/understanding-data-harmonization-in-bdc.md).

# Understanding Data Harmonization in BDC

Data harmonization on NHLBI BioData Catalyst® (BDC) includes two primary approaches that researchers should understand when selecting datasets for analysis. 

Some datasets are harmonized directly by the original data generators or stewards, such as National Sleep Research Resource (NSRR), who apply domain-specific expertise and standardized protocols during data collection and curation. 

Other datasets are harmonized by secondary organizations or platforms, including members of the BDC consortium, which integrate and standardize data across studies to improve cross-study usability and interoperability. 

Because harmonization methods, variable definitions, and processing decisions may differ between these approaches, researchers should review the dataset summary information carefully before selecting data for analysis. Key details about harmonization processes, provenance, and any transformations applied should be clearly described within each dataset’s summary documentation to support transparent and informed use.