On the Q statistic with constant weights for standardized mean difference

Show simplified metadata
dc.contributor.authorBakbergenuly, Ilyas
dc.contributor.authorHoaglin, David C
dc.contributor.authorKulinskaya, Elena
dc.contributor.departmentDepartment of Population and Quantitative Health Sciences
dc.date2022-08-11T08:10:37.000
dc.date.accessioned2022-08-23T17:14:37Z
dc.date.available2022-08-23T17:14:37Z
dc.date.issued2022-01-30
dc.date.submitted2022-04-04
dc.description.abstractCochran's Q statistic is routinely used for testing heterogeneity in meta-analysis. Its expected value is also used in several popular estimators of the between-study variance, tau 2 . Those applications generally have not considered the implications of its use of estimated variances in the inverse-variance weights. Importantly, those weights make approximating the distribution of Q (more explicitly, Q IV ) rather complicated. As an alternative, we investigate a new Q statistic, Q F , whose constant weights use only the studies' effective sample sizes. For the standardized mean difference as the measure of effect, we study, by simulation, approximations to distributions of Q IV and Q F , as the basis for tests of heterogeneity and for new point and interval estimators of tau 2 . These include new DerSimonian-Kacker-type moment estimators based on the first moment of Q F , and novel median-unbiased estimators. The results show that: an approximation based on an algorithm of Farebrother follows both the null and the alternative distributions of Q F reasonably well, whereas the usual chi-squared approximation for the null distribution of Q IV and the Biggerstaff-Jackson approximation to its alternative distribution are poor; in estimating tau 2 , our moment estimator based on Q F is almost unbiased, the Mandel - Paule estimator has some negative bias in some situations, and the DerSimonian-Laird and restricted maximum likelihood estimators have considerable negative bias; and all 95% interval estimators have coverage that is too high when tau 2 = 0 , but otherwise the Q-profile interval performs very well.
dc.identifier.citation<p>Bakbergenuly I, Hoaglin DC, Kulinskaya E. On the Q statistic with constant weights for standardized mean difference. Br J Math Stat Psychol. 2022 Jan 30. doi: 10.1111/bmsp.12263. Epub ahead of print. PMID: 35094381. <a href="https://doi.org/10.1111/bmsp.12263">Link to article on publisher's site</a></p>
dc.identifier.contextkey28476733
dc.identifier.doi10.1111/bmsp.12263
dc.identifier.issn0007-1102 (Linking)
dc.identifier.legacycoverpagehttps://escholarship.umassmed.edu/qhs_pp/1470
dc.identifier.legacyfulltexthttps://escholarship.umassmed.edu/cgi/viewcontent.cgi?article=2474&amp;context=qhs_pp&amp;unstamped=1
dc.identifier.pmid35094381
dc.identifier.submissionpathqhs_pp/1470
dc.identifier.urihttps://hdl.handle.net/20.500.14038/46993
dc.language.isoen_US
dc.relation<p><a href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=pubmed&cmd=Retrieve&list_uids=35094381&dopt=Abstract">Link to Article in PubMed</a></p>
dc.rights© 2022 The Authors. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.
dc.rights.urihttp://creativecommons.org/licenses/by/4.0/
dc.source.journaltitleThe British journal of mathematical and statistical psychology
dc.subjecteffective sample sizes
dc.subjectheterogeneity
dc.subjectinverse-variance weights
dc.subjectrandom effects
dc.subjectEpidemiology
dc.subjectStatistics and Probability
dc.titleOn the Q statistic with constant weights for standardized mean difference
dc.typeJournal Article
dspace.entity.typePublication
html.description.abstract<p>Cochran's Q statistic is routinely used for testing heterogeneity in meta-analysis. Its expected value is also used in several popular estimators of the between-study variance, tau 2 . Those applications generally have not considered the implications of its use of estimated variances in the inverse-variance weights. Importantly, those weights make approximating the distribution of Q (more explicitly, Q IV ) rather complicated. As an alternative, we investigate a new Q statistic, Q F , whose constant weights use only the studies' effective sample sizes. For the standardized mean difference as the measure of effect, we study, by simulation, approximations to distributions of Q IV and Q F , as the basis for tests of heterogeneity and for new point and interval estimators of tau 2 . These include new DerSimonian-Kacker-type moment estimators based on the first moment of Q F , and novel median-unbiased estimators. The results show that: an approximation based on an algorithm of Farebrother follows both the null and the alternative distributions of Q F reasonably well, whereas the usual chi-squared approximation for the null distribution of Q IV and the Biggerstaff-Jackson approximation to its alternative distribution are poor; in estimating tau 2 , our moment estimator based on Q F is almost unbiased, the Mandel - Paule estimator has some negative bias in some situations, and the DerSimonian-Laird and restricted maximum likelihood estimators have considerable negative bias; and all 95% interval estimators have coverage that is too high when tau 2 = 0 , but otherwise the Q-profile interval performs very well.</p>
refterms.dateFOA2022-08-23T17:14:37Z
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
Brit_J_Math_Statis___2022___Bakbergenuly___On_the_Q_statistic_with_constant_weights_for_standardized_mean_difference.pdf
Size:
551.64 KB
Format:
Adobe Portable Document Format
Download
Collections
Population and Quantitative Health Sciences Publications