Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2019 Mar;28(3):471-477.
doi: 10.1158/1055-9965.EPI-18-0797. Epub 2019 Jan 14.

Weighting Nonprobability and Probability Sample Surveys in Describing Cancer Catchment Areas

Ronaldo Iachan  1 , Lewis Berman  1 , Tonja M Kyle  2 , Kelly J Martin  1 , Yangyang Deng  1 , Davia N Moyse  1 , Deirdre Middleton  1 , Audie A Atienza  1
Affiliations

Weighting Nonprobability and Probability Sample Surveys in Describing Cancer Catchment Areas

Ronaldo Iachan et al. Cancer Epidemiol Biomarkers Prev. 2019 Mar.

Abstract

Background: The Population Health Assessment initiative by NCI sought to enhance cancer centers' capacity to acquire, aggregate, and integrate data from multiple sources, as well as to plan, coordinate, and enhance catchment area analysis activities.

Methods: Key objectives of this initiative are pooling data and comparing local data with national data. A novel aspect of analyzing data from this initiative is the methodology used to weight datasets from sites that collected both probability and nonprobability samples. This article describes the methods developed to weight data, which cancer centers collected with combinations of probability, and nonprobability sampling designs.

Results: We compare alternative weighting methods in particular for the hybrid probability and nonprobability sampling designs employed by different cancer centers. We also include comparisons of local center data with national survey data from large probability samples.

Conclusions: This hybrid approach to calculating statistical weights can be implemented both within cancer centers that collect both probability and nonprobability samples with common measures. Aggregation can also apply to cancer centers that share common data elements, and target similar populations, but differ in survey sampling designs.

Impact: Researchers interested in local versus national comparisons for cancer surveillance and control outcomes should consider various weighting approaches, including hybrid approaches, when analyzing their data.

PubMed Disclaimer

Conflict of interest statement

Conflict of Interest: The authors declare no potential conflicts of interest.

Figures

Figure 1
Figure 1. Dartmouth-Hitchcock Norris Cotton Cancer Center comparisons of weighting methods and national or state level data
Description: Figure 1 presents weighted results for Dartmouth-Hitchcock Norris Cotton Cancer Center which supported combining sample data for both probability and non-probability components of the hybrid sample. For two estimates, the figure shows results for two variables from the Dartmouth instrument that were selected from the Health Information National Trends Survey (HINTS) thus supporting comparisons with HINTS national data. The other two estimates present results for variables in the Dartmouth instrument selected from the Behavioral Risk Factor Surveillance System (BRFSS), which support comparisons with the BRFSS state data. RBG Colors Used in Figure 1:

  1. National HINTS: Gold: 255/192/0

  2. PSM+ Rake: Gray: 165/165/165

  3. PSM Only: Orange: 237/125/49

  4. Rake Only: Light Blue: 91/155/213

  5. VT, NH BRFSS: Green: 112/173/71

Figure 2
Figure 2. University of Kentucky – Markey Cancer Center comparisons of weighting methods and national or state level data
Description: Figure 2 presents results for University of Kentucky – Markey Cancer Center which required the probability and non-probability samples to be weighted separately. The first four measures in the figure present results for two variables selected from the HINTS survey, which support comparisons with HINTS national data. The last two measures present results for two variables selected from the BRFSS survey, which support comparisons with the BRFSS state data. RBG Colors Used in Figure 2:

  1. National HINTS: Dark Blue: 68/114/196

  2. NonProb, PSM + Rake: Gold: 255/192/0

  3. NonProb, PSM Only: Gray: 165/165/165

  4. NonProb, Rake Only: Orange: 237/125/49

  5. Probability: Light Blue: 91/155/213

  6. KY BRFSS: Green: 112/173/71

See this image and copyright information in PMC

References

    1. Observational Health Data Sciences and Informatics (2018). “Data Standardization”. https://www.ohdsi.org/data-standardization/. Last accessed on July 14, 2018.
    1. University of Michigan (2018). “Data Sharing for Demographic Research.” https://www.icpsr.umich.edu/icpsrweb/content/DSDR/harmonization.html. Last accessed on July 14, 2018.
    1. Angrisani M, Lee J (2012). Harmonization of Cross-National Studies of Aging to the Health and Retirement Study: Income Measures. RAND Corporation Working Papers.
    1. Gage-Bouchard EA, and Rawl S “Standardizing Measurement of Social and Behavioral Dimensions of Cancer Prevention and Control to Enhance Community Outreach and Engagement in NCI-Designated and Comprehensive Cancer Centers.” Currently in review for this focus issue. - PMC - PubMed
    1. Iachan R, Lu B, & Duffy T (2009, August). A comparison of imputation methods used in the Ohio Family Health Survey. Paper presented at the American Statistical Association meeting, Denver, CO.

Publication types