Resampling tests for meta-analysis of ecological data

Dean C. Adams, Jessica Gurevitch, Michael S. Rosenberg

Research output: Contribution to journalArticle

360 Citations (Scopus)

Abstract

Meta-analysis is a statistical technique that allows one to combine the results from multiple studies to glean inferences on the overall importance of various phenomena. This method can prove to be more informative than common 'vote counting', in which the number of significant results is compared to the number with nonsignificant results to determine whether the phenomenon of interest is globally important. While the use of meta-analysis is widespread in medicine and the social sciences, only recently has it been applied to ecological questions. We compared the results of parametric confidence limits and homogeneity statistics commonly obtained through meta-analysis to those obtained from resampling methods to ascertain the robustness of standard meta-analytic techniques. We found that confidence limits based on bootstrapping methods were wider than standard confidence limits, implying that resampling estimates are more conservative. In addition, we found that significance tests based on homogeneity statistics differed occasionally from results of randomization tests, implying that inferences based solely on chi-square significance tests may lead to erroneous conclusions. We conclude that resampling methods should be incorporated in meta-analysis studies, to ensure proper evaluation of main effects in ecological studies.

Original languageEnglish (US)
Pages (from-to)1277-1283
Number of pages7
JournalEcology
Volume78
Issue number4
StatePublished - 1997
Externally publishedYes

Fingerprint

meta-analysis
homogeneity
testing
bootstrapping
statistics
methodology
medicine
social sciences
method
test

Keywords

  • Bootstrapping
  • Meta-analysis
  • Randomization tests
  • Resampling statistics vs. standard methods
  • Statistical techniques

ASJC Scopus subject areas

  • Ecology

Cite this

Adams, D. C., Gurevitch, J., & Rosenberg, M. S. (1997). Resampling tests for meta-analysis of ecological data. Ecology, 78(4), 1277-1283.

Resampling tests for meta-analysis of ecological data. / Adams, Dean C.; Gurevitch, Jessica; Rosenberg, Michael S.

In: Ecology, Vol. 78, No. 4, 1997, p. 1277-1283.

Research output: Contribution to journalArticle

Adams, DC, Gurevitch, J & Rosenberg, MS 1997, 'Resampling tests for meta-analysis of ecological data', Ecology, vol. 78, no. 4, pp. 1277-1283.
Adams DC, Gurevitch J, Rosenberg MS. Resampling tests for meta-analysis of ecological data. Ecology. 1997;78(4):1277-1283.
Adams, Dean C. ; Gurevitch, Jessica ; Rosenberg, Michael S. / Resampling tests for meta-analysis of ecological data. In: Ecology. 1997 ; Vol. 78, No. 4. pp. 1277-1283.
@article{1a0ca802195f46ff9a0d839842522372,
title = "Resampling tests for meta-analysis of ecological data",
abstract = "Meta-analysis is a statistical technique that allows one to combine the results from multiple studies to glean inferences on the overall importance of various phenomena. This method can prove to be more informative than common 'vote counting', in which the number of significant results is compared to the number with nonsignificant results to determine whether the phenomenon of interest is globally important. While the use of meta-analysis is widespread in medicine and the social sciences, only recently has it been applied to ecological questions. We compared the results of parametric confidence limits and homogeneity statistics commonly obtained through meta-analysis to those obtained from resampling methods to ascertain the robustness of standard meta-analytic techniques. We found that confidence limits based on bootstrapping methods were wider than standard confidence limits, implying that resampling estimates are more conservative. In addition, we found that significance tests based on homogeneity statistics differed occasionally from results of randomization tests, implying that inferences based solely on chi-square significance tests may lead to erroneous conclusions. We conclude that resampling methods should be incorporated in meta-analysis studies, to ensure proper evaluation of main effects in ecological studies.",
keywords = "Bootstrapping, Meta-analysis, Randomization tests, Resampling statistics vs. standard methods, Statistical techniques",
author = "Adams, {Dean C.} and Jessica Gurevitch and Rosenberg, {Michael S.}",
year = "1997",
language = "English (US)",
volume = "78",
pages = "1277--1283",
journal = "Ecology",
issn = "0012-9658",
publisher = "Ecological Society of America",
number = "4",

}

TY - JOUR

T1 - Resampling tests for meta-analysis of ecological data

AU - Adams, Dean C.

AU - Gurevitch, Jessica

AU - Rosenberg, Michael S.

PY - 1997

Y1 - 1997

N2 - Meta-analysis is a statistical technique that allows one to combine the results from multiple studies to glean inferences on the overall importance of various phenomena. This method can prove to be more informative than common 'vote counting', in which the number of significant results is compared to the number with nonsignificant results to determine whether the phenomenon of interest is globally important. While the use of meta-analysis is widespread in medicine and the social sciences, only recently has it been applied to ecological questions. We compared the results of parametric confidence limits and homogeneity statistics commonly obtained through meta-analysis to those obtained from resampling methods to ascertain the robustness of standard meta-analytic techniques. We found that confidence limits based on bootstrapping methods were wider than standard confidence limits, implying that resampling estimates are more conservative. In addition, we found that significance tests based on homogeneity statistics differed occasionally from results of randomization tests, implying that inferences based solely on chi-square significance tests may lead to erroneous conclusions. We conclude that resampling methods should be incorporated in meta-analysis studies, to ensure proper evaluation of main effects in ecological studies.

AB - Meta-analysis is a statistical technique that allows one to combine the results from multiple studies to glean inferences on the overall importance of various phenomena. This method can prove to be more informative than common 'vote counting', in which the number of significant results is compared to the number with nonsignificant results to determine whether the phenomenon of interest is globally important. While the use of meta-analysis is widespread in medicine and the social sciences, only recently has it been applied to ecological questions. We compared the results of parametric confidence limits and homogeneity statistics commonly obtained through meta-analysis to those obtained from resampling methods to ascertain the robustness of standard meta-analytic techniques. We found that confidence limits based on bootstrapping methods were wider than standard confidence limits, implying that resampling estimates are more conservative. In addition, we found that significance tests based on homogeneity statistics differed occasionally from results of randomization tests, implying that inferences based solely on chi-square significance tests may lead to erroneous conclusions. We conclude that resampling methods should be incorporated in meta-analysis studies, to ensure proper evaluation of main effects in ecological studies.

KW - Bootstrapping

KW - Meta-analysis

KW - Randomization tests

KW - Resampling statistics vs. standard methods

KW - Statistical techniques

UR - http://www.scopus.com/inward/record.url?scp=0030613897&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0030613897&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:0030613897

VL - 78

SP - 1277

EP - 1283

JO - Ecology

JF - Ecology

SN - 0012-9658

IS - 4

ER -