Huiding Chen
Project Personnel, Vanderbilt University
2 active projects
Adjusting results using multiple crossproduct marginal distributions
Scientific Questions Being Studied
Using external open-to-public data source (e.g. US Census, NHANES), we want to better represent the U.S. population using AoU source.
Project Purpose(s)
- Methods Development
- Control Set
Scientific Approaches
We intend to utilize marginal and individual-level datasets which are open-to-public.
In order to include the populational information, the demographics provided by U.S. Census website (https://data.census.gov/profile/United_States?g=0100000US) will be gathered. The marginal information will be used as the finite population adjustment in raking method.
For individual-level data, the NHANES survey data from 2017 to 2020 will be imported to workspace (https://wwwn.cdc.gov/Nchs/Nhanes/2017-2018/P_DEMO.htm). This individual-level data will be used to calibrate the sampling probability/propensity score for AoU data in U.S. population.
Using the calibrated weights from the external sources, we want to compare the prevalence for several diseases by implenting regression to the U.S. population. The demographics and original disease status in AoU will be used comprehensively.
Methods: Raking, Iterative Proportional Fitting, GREG, Propensity Score
Anticipated Findings
By observing that some indexes derving from AoU database are different from those among the general U.S. population, we want to better adjust these indexes using multiple crossproduct of some demographics provided by other sources. We may making these observations more convincing(e.g. some diease risks) with the applied methods.
Demographic Categories of Interest
- Race / Ethnicity
- Age
- Sex at Birth
- Education Level
- Income Level
Data Set Used
Controlled TierResearch Team
Owner:
- Huiding Chen - Project Personnel, Vanderbilt University
Collaborators:
- Qingxia Chen - Other, All of Us Program Operational Use
Adjusting results using multiple crossproduct marginal distributions
Scientific Questions Being Studied
Using external open-to-public data source (e.g. US Census, NHANES), we want to better represent the U.S. population using AoU source.
Project Purpose(s)
- Methods Development
- Control Set
Scientific Approaches
We intend to utilize marginal and individual-level datasets which are open-to-public.
In order to include the populational information, the demographics provided by U.S. Census website (https://data.census.gov/profile/United_States?g=0100000US) will be gathered. The marginal information will be used as the finite population adjustment in raking method.
For individual-level data, the NHANES survey data from 2017 to 2020 will be imported to workspace (https://wwwn.cdc.gov/Nchs/Nhanes/2017-2018/P_DEMO.htm). This individual-level data will be used to calibrate the sampling probability/propensity score for AoU data in U.S. population.
Using the calibrated weights from the external sources, we want to compare the prevalence for several diseases by implenting regression to the U.S. population. The demographics and original disease status in AoU will be used comprehensively.
Methods: Raking, Iterative Proportional Fitting, GREG, Propensity Score
Anticipated Findings
By observing that some indexes derving from AoU database are different from those among the general U.S. population, we want to better adjust these indexes using multiple crossproduct of some demographics provided by other sources. We may making these observations more convincing(e.g. some diease risks) with the applied methods.
Demographic Categories of Interest
- Race / Ethnicity
- Age
- Sex at Birth
- Education Level
- Income Level
Data Set Used
Controlled TierResearch Team
Owner:
- Huiding Chen - Project Personnel, Vanderbilt University
Collaborators:
- Qingxia Chen - Other, All of Us Program Operational Use
Request a Review of this Research Project
You can request that the All of Us Resource Access Board (RAB) review a research purpose description if you have concerns that this research project may stigmatize All of Us participants or violate the Data User Code of Conduct in some other way. To request a review, you must fill in a form, which you can access by selecting ‘request a review’ below.
