Post-stratified estimation: with-in strata and total sample size recommendations

  • Authors: Westfall, James A.; Patterson, Paul L.; Coulston, John W.
  • Publication Year: 2011
  • Publication Series: Miscellaneous Publication
  • Source: Canadian Journal of Forest Research. 41: 1130-1139.

Abstract

Post-stratification is used to reduce the variance of estimates of the mean. Because the stratification is not fixed in advance, within-strata sample sizes can be quite small. The survey statistics literature provides some guidance on minimum within-strata sample sizes; however, the recommendations and justifications are inconsistent and apply broadly for many different population structures. The impacts of minimum within-strata and total sample sizes on estimates of means and standard errors were examined for two forest inventory variables: proportion forestland and cubic net volume. Estimates of the means seem unbiased across a range of minimum within-strata sample sizes. A ratio that described the decrease in variability with increasing sample size allowed for assessment of minimum within-strata sample requirements to obtain stable estimates of means. This metric indicated that the minimum within-strata sample size should be at least 10. Estimates of standard errors were found to be biased at small total sample sizes. To obtain a bias of less than 3%, the required minimum total sample size was 25 for proportion forestland and 75 for cubic net volume. The results presented allow analysts to determine within-stratum and total sample size requirements corresponding to their criteria for acceptable levels of bias and variability.

  • Citation: Westfall, James A.; Patterson, Paul L.; Coulston, John W. 2011. Post-stratified estimation: with-in strata and total sample size recommendations. Canadian Journal of Forest Research. 41: 1130-1139.
  • Keywords: post-stratification, within-strata sample, estimates
  • Posted Date: May 20, 2011
  • Modified Date: June 17, 2011
  • Print Publications Are No Longer Available

    In an ongoing effort to be fiscally responsible, the Southern Research Station (SRS) will no longer produce and distribute hard copies of our publications. Many SRS publications are available at cost via the Government Printing Office (GPO). Electronic versions of publications may be downloaded, printed, and distributed.

    Publication Notes

    • This article was written and prepared by U.S. Government employees on official time, and is therefore in the public domain.
    • To view this article, download the latest version of Adobe Acrobat Reader.