Exploring the Big Data Paradox for various estimands using vaccination data from the global COVID-19 Trends and Impact Survey (CTIS).

Publisher: American Association for the Advancement of Science Country of Publication: United States NLM ID: 101653440 Publication Model: Print-Electronic Cited Medium: Internet ISSN: 2375-2548 (Electronic) Linking ISSN: 23752548 NLM ISO Abbreviation: Sci Adv Subsets: MEDLINE

Original Publication: Washington, DC : American Association for the Advancement of Science, [2015]-

COVID-19*/epidemiology ; COVID-19*/prevention & control ; COVID-19 Vaccines*/administration & dosage ; Vaccination*/statistics & numerical data ; SARS-CoV-2*/immunology ; Big Data*; Humans ; Female ; Male ; Adult ; Surveys and Questionnaires ; India/epidemiology ; Middle Aged

Selection bias poses a substantial challenge to valid statistical inference in nonprobability samples. This study compared estimates of the first-dose COVID-19 vaccination rates among Indian adults in 2021 from a large nonprobability sample, the COVID-19 Trends and Impact Survey (CTIS), and a small probability survey, the Center for Voting Options and Trends in Election Research (CVoter), against national benchmark data from the COVID Vaccine Intelligence Network. Notably, CTIS exhibits a larger estimation error on average (0.37) compared to CVoter (0.14). Additionally, we explored the accuracy (regarding mean squared error) of CTIS in estimating successive differences (over time) and subgroup differences (for females versus males) in mean vaccine uptakes. Compared to the overall vaccination rates, targeting these alternative estimands comparing differences or relative differences in two means increased the effective sample size. These results suggest that the Big Data Paradox can manifest in countries beyond the United States and may not apply equally to every estimand of interest.

MMWR Morb Mortal Wkly Rep. 2022 Mar 04;71(9):335-340. (PMID: 35239636)
J Community Psychol. 2022 Jul;50(5):2431-2442. (PMID: 34969152)
Educ Res. 2022 Dec;51(9):593-597. (PMID: 38603417)
Int J Nurs Stud. 2021 Nov;123:104058. (PMID: 34454334)
Lancet Infect Dis. 2020 May;20(5):533-534. (PMID: 32087114)
Proc Natl Acad Sci U S A. 2021 Dec 21;118(51):. (PMID: 34903656)
Vaccines (Basel). 2022 Aug 07;10(8):. (PMID: 36016160)
Nature. 2021 Dec;600(7890):695-700. (PMID: 34880504)
Sci Rep. 2023 May 24;13(1):8360. (PMID: 37225748)
J Surv Stat Methodol. 2020 Nov;8(5):932-964. (PMID: 33381610)
BMC Public Health. 2021 Apr 28;21(1):818. (PMID: 33910558)
J Off Stat. 2021 Sep;37(3):751-769. (PMID: 34566235)
Lancet Reg Health Eur. 2021 Aug;7:100130. (PMID: 34109321)
Lancet. 2022 Jul 2;400(10345):24. (PMID: 35780789)
Nature. 2021 Dec;600(7890):614-615. (PMID: 34880485)
MMWR Morb Mortal Wkly Rep. 2022 Feb 04;71(5):171-176. (PMID: 35113846)
Am J Epidemiol. 2022 Mar 24;191(4):570-583. (PMID: 34999751)
Nat Commun. 2022 Feb 1;13(1):636. (PMID: 35105869)
Vaccine. 2021 Oct 8;39(42):6256-6261. (PMID: 34544601)
Proc Natl Acad Sci U S A. 2021 Dec 21;118(51):. (PMID: 34903654)
Proc Natl Acad Sci U S A. 2021 Dec 21;118(51):. (PMID: 34903657)
Vaccine. 2021 May 18;39(21):2833-2842. (PMID: 33896661)
MMWR Morb Mortal Wkly Rep. 2021 Jun 25;70(25):922-927. (PMID: 34166331)
Lancet. 2021 May 8;397(10286):1725-1735. (PMID: 33901423)
J R Stat Soc Ser C Appl Stat. 2019 Nov;68(5):1465-1483. (PMID: 33304001)
Ann Appl Stat. 2023 Dec;17(4):2903-2923. (PMID: 38939875)
JAMA. 2021 Jun 15;325(23):2397-2399. (PMID: 34028495)
PLOS Glob Public Health. 2023 Nov 30;3(11):e0002601. (PMID: 38032861)
BMC Public Health. 2022 Oct 13;22(1):1911. (PMID: 36229804)

0 (COVID-19 Vaccines)

Date Created: 20240531 Date Completed: 20240531 Latest Revision: 20240811

20240812

PMC11314312

10.1126/sciadv.adj0266

38820165