Abstract
Correction for continuity is commonly used to improve the inference for binary data when the event of interest is rare or the sample size is small. A standard approach to reduce the bias in logit estimation is to add a small constant to both event and nonevent counts. The 0.5 adjustment is known as a correction rendering the estimation unbiased up to the order of \(K^{-1}\), where K is the size of a simple random sample. However, for general designs beyond simple random sampling, the bias in estimating the logit is no longer zero in order \(K^{-1}\). In this paper, we derive the formula of the correction factor that makes the first-order term of the bias vanish for general designs. We then apply it to estimate the logit when data are from ranked set sampling (RSS) embedded in a cluster randomized design (CRD). An RSS-structured CRD (RSS-CRD), introduced by Wang et al. (J Am Stat Assoc 111: 1576–1590, 2016), is a new two-stage design for more efficient estimation of treatment effect. We propose two methods to estimate the correction factors derived for RSS-CRDs. We numerically compare the proposed methods to those with the default factor 0.5 in terms of bias and mean squared error for estimating the treatment effect, and finally make recommendations to practitioners.
Similar content being viewed by others
Availability of data and material
Supplemental material for this article is available online.
References
Anscombe, F.J.: On estimating binomial response relations. Biometrika 43, 461–465 (1956)
Bhaumik, D.K., Amatya, A., Normand, S.L.T., et al.: Meta-analysis of rare binary adverse event data. J. Am. Stat. Assoc. 107(498), 555–567 (2012)
Blyth, C.R., Still, H.A.: Binomial confidence intervals. J. Am. Stat. Assoc. 78, 108–116 (1983)
Bradburn, M.J., Deeks, J.J., Berlin, J.A., et al.: Much ado about nothing: a comparison of the performance of meta-analytical methods with rare events. Stat. Med. 26(1), 53–77 (2007)
Chen, H., Stasny, E.A., Wolfe, D.A.: Ranked set sampling for efficient estimation of a population proportion. Stat. Med. 24, 3239–3385 (2005)
Cox, D.R.: The Analysis of Binary Data. Methuen, London (1970)
Dumbgen, L., Zamanzade, E.: Inference on a distribution function from ranked set samples. Ann. Inst. Stat. Math. 72(3), 157–185 (2020)
Fleiss, J.H.: Statistical Methods for Rates and Proportions. Wiley, New York (1981)
Frey, J., Wang, L.: Edf-based goodness-of-fit tests for ranked-set sampling. Can. J. Stat. 42(3), 451–469 (2014)
Frey, J., Zhang, Y.: Testing perfect rankings in ranked-set sampling with binary data. Can. J. Stat. 45(3), 326–339 (2017)
Frey, J., Zhang, Y.: An algorithm with applications in ranked-set sampling. J. Stat. Comput. Simul. 88(3), 471–481 (2018)
Frey, J., Zhang, Y.: Robust confidence intervals for a proportion using ranked-set sampling. J. Korean Stat. Soc. 50, 1009–1028 (2021)
Gart, J.J., Pettigrew, H.M., Thomas, D.G.: The effect of bias, variance estimation, skewness and kurtosis of the empirical logit on weighted least squares analyses. Biometrika 72, 179–190 (1985)
Grizzle, J.E.: Continuity correction in the \(\chi ^{2}\)-test for \(2\times 2\) tables. Am. Stat. 21, 28–32 (1967)
Haldane, J.B.S.: The estimation and significance of the logarithm of a ratio of frequencies. Ann. Hum. Genet. 20, 309–311 (1955)
Hamdan, M.A.: On the continuity correction. Technometrics 16(4), 631–632 (1974)
Li, L., Wang, X.: Meta-analysis of rare binary events in treatment groups with unequal variability. Stat. Methods Med. Res. (2017). https://doi.org/10.1177/0962280217721246
Newcombe, R.G.: Two-sided confidence intervals for the single proportion: comparison of seven methods. Stat. Med. 17, 857–872 (1998)
Ozturk, O.: Combining ranking information in judgment post stratified and ranked set sampling designs. Environ. Ecol. Stat. 19, 73–93 (2012)
Ozturk, O.: Two-stage cluster samples with ranked set sampling designs. Ann. Inst. Stat. Math. 71(1), 63–91 (2019)
Ozturk, O., Balakrishnan, N.: An exact control-versus-treatment comparison test based on ranked set samples. Biometrics 65(4), 1213–1222 (2009)
Pearson, E.: The choice of statistical test illustrated on the interpretation of data classed in a 2 \(\times \) 2 tables. Biometrika 34, 139–167 (1947)
Pirie, W.R., Hamdan, M.A.: Some revised continuity corrections for discrete distributions. Biometrics 28(3), 693–701 (1972)
Plackett, R.L.: The continuity correction in \(2\times 2\) tables. Biometrika 51, 327–337 (1964)
Sweeting, M., Sutton, A.J., Lambert, P.: What to add to nothing\(?\) use and avoidance of continuity corrections in meta analysis of sparse data. Stat. Med. 23(9), 1351–1375 (2004)
Terpstra, J.: On estimating a population proportion via ranked set sampling. Biom. J. 46, 264–272 (2004)
Wang, X., Ahn, S., Lim, J.: Unbalanced ranked set sampling in cluster randomized studies. J. Stat. Plan 187, 1–16 (2017)
Wang, X., Lim, J., Stokes, L.: Using ranked set sampling with cluster randomized designs for improved inference on treatment effects. J. Am. Stat. Assoc. 111, 1576–1590 (2016)
Wang, X., Wang, M., Lim, J., et al.: Using ranked set sampling with binary outcomes in cluster randomized designs. Can. J. Stat. 48(3), 342–365 (2020)
Yates, F.: Contingency tables involving small numbers and the \(\chi ^{2}\) test. J. R. Stat. Soc. 1, 217–235 (1934)
Zamanzade, E., Mahdizadeh, M.: Using ranked set sampling with extreme ranks in estimating the population proportion. Stat. Methods Med. Res. 29(1), 165–177 (2020)
Zamanzade, E., Wang, X.: Estimation of population proportion for judgment post-stratification. Comput. Stat. Data Anal. 112, 257–269 (2017)
Funding
This work was supported by the National Research Foundation of Korea (Grant no.: NRF-2017R1D1A1B03032073, NRF-2019R1F1A1056779)
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
The authors declare that there is no conflict of interest.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary Information
Below is the link to the electronic supplementary material.
Rights and permissions
About this article
Cite this article
Ahn, S., Wang, X., Wang, M. et al. On continuity correction for RSS-structured cluster randomized designs with binary outcomes. METRON 80, 383–397 (2022). https://doi.org/10.1007/s40300-021-00226-5
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s40300-021-00226-5