Power Calculations

2020 Q3 Emergency NHIS Attachment C Power Calculations 6022020.xlsx

Changes to 2020 National Health Interview Survey in light of Novel Coronavirus (2019-nCoV)

Power Calculations

OMB: 0920-0214

Document [xlsx]
Download: xlsx | pdf
The table below shows the expected coefficients of variation (CV) for the estimates for adults and childrens and for smaller samples for subgroups. CV of 30% has been used as a threshold by NCHS for presentation criteria. Current NCHS presentation standards** moved away from CV and are based on exact confidence intervals but were established to be consistent with prior criteria. The current standards generally allow for more low prevalence estimates to be released.















CV were calculated assuming a design effect of 2.5 CV in RED are estimates that would not have met the CV critieria for presention.















Table 1. Sample Adults: Coefficients of Variation for estimation variables with varying expected prevalence by expected sample sizes according to covariates




























Expected NHIS sample size July, 2020-December 2021, accounting for a 33% reduction for 2020


36,000 32,000 22,000 18,000 12,000 5,500 4,500 3,200 2,000 1,000 500
Prevalence of estimation variable: Design Effect: 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5













Variables with prevalence 2% 2% 5.83 6.19 7.46 8.25 10.10 14.92 16.50 19.57 24.75 35.00 49.50
Variables with prevalence 3% 3% 4.74 5.03 6.06 6.70 8.21 12.12 13.40 15.89 20.10 28.43 40.21
Variables with prevalence 5% 5% 3.63 3.85 4.65 5.14 6.29 9.29 10.27 12.18 15.41 21.79 30.82
Variables with prevalence 7% 7% 3.04 3.22 3.89 4.30 5.26 7.77 8.59 10.19 12.89 18.22 25.77
Variables with prevalence 10% 10% 2.50 2.65 3.20 3.54 4.33 6.40 7.07 8.39 10.61 15.00 21.21
Variables with prevalence 20% 20% 1.67 1.77 2.13 2.36 2.89 4.26 4.71 5.59 7.07 10.00 14.14
Variables with prevalence 15% 15% 1.98 2.10 2.54 2.81 3.44 5.08 5.61 6.65 8.42 11.90 16.83
Variables with prevalence 25% 25% 1.44 1.53 1.85 2.04 2.50 3.69 4.08 4.84 6.12 8.66 12.25
Variables with prevalence 30% 30% 1.27 1.35 1.63 1.80 2.20 3.26 3.60 4.27 5.40 7.64 10.80
Variables with prevalence 40% 40% 1.02 1.08 1.31 1.44 1.77 2.61 2.89 3.42 4.33 6.12 8.66
Variables with prevalence 45% 45% 0.92 0.98 1.18 1.30 1.60 2.36 2.61 3.09 3.91 5.53 7.82














Examples of covariate subgroups associated with this sample size All Adults Any health insurance Age <65 (88%) Employed or recently employed (60%) One or more chronic conditions (50%) Has conditions that increase risk of complications from COVID-19 (33%) Current Age > 65 (16%); Office and Administrative Support Occupation (15%) Non-Hispanic Black (13%); Ever had cancer (12%) Uninsured adults (11%); Food Preparation and Serving Occupation (9%); Poor or fair health (9%) Asian (6%); Positive Diagnosis for COVID-19 (6%); Education and Library Occupation (5%) Multiple race (3%); Currently undergoing monitoring for cancer (3%) Moderate or severe COVID symptoms (2%); Currently undergoing treatment for cancer (2%)
File Typeapplication/vnd.openxmlformats-officedocument.spreadsheetml.sheet
File Modified0000-00-00
File Created0000-00-00

© 2024 OMB.report | Privacy Policy