On Judging the Significance of Differences by Examining the Overlap Between Confidence Intervals

By N. Schenker, J. Gentleman. 2001

1,153 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/557e2210ff97a69c601c94f8f3993455ba05cbfc

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Statistics and Probability

Citations in the corpus

Prevalence of mental illness in immigrant and non-immigrant U.S. Latino groups. (2008, background): “lapping confidence intervals because examining the overlap of confidence intervals is a more conservative approach to testing for significance (24).”
Effectiveness of Bowl Trapping and Netting for Inventory of a Bee Community (2011, background): “If their conÞdence intervals do overlap, a formal significance test is needed to determine whether the curves are, in fact, not signiÞcantly different (Schenker and Gentleman 2001).”
2007 National Survey of Mental Health and Wellbeing: Methods and Key Findings (2009, neutral): “This is a conservative approach to assessing differences [23].”
Inference by eye: confidence intervals and how to read pictures of data. (2005, background): “Schenker and Gentleman (2001) reported that many medical researchers take zero overlap of CIs as equivalent to statistical significance at the .05 level.”
Hearing Gender: Voice-Based Gender Classification Processes and Transgender Health Inequality (2019, result): “…overlap, the average marginal effects are still significant at the p < .05 level, and the proportion of confidence interval overlap is .23, which is considered a significant finding despite the evidently high levels of variance present in the sample (Cumming 2009; Schenker and Gentleman 2001).”
Consequences of shoot density and stiffness for ecosystem engineering by benthic macrophytes in flow dominated areas: a hydrodynamic flume study (2008, background): “Differences in <u–> were considered significant (p 0.05) if the 95% CI did not overlap (cf. Schenker & Gentleman 2001).”
The correspondence of family features with problem, aggressive, criminal, and violent behavior: a meta-analysis (2010, methodology): “Because they may be more conservative than standard methods, significance testing using 95% confidence intervals are prone to Type II error (Payton et al. 2003; Schenker and Gentleman 2001 ).”
Inference by eye: Reading the overlap of independent confidence intervals (2009, background): “The rule that 95% CIs just touching is equivalent to statistical significance was discussed by [16-18], and also by Schenker and Gentleman [ 2 ], who reported finding more than 60 articles in health sciences journals that had used the rule.”
Are Survey Weights Needed? A Review of Diagnostic Tests in Regression Analysis (2016, background): “…whether the associated confidence intervals (CIs) overlap, concluding that the weights make a significant difference if the CIs do not overlap.4 The authors base their argument on Schenker & Gentleman (2001), even 3Note that their derivation applies to structural equation models in general.”
Deadwood enrichment in European forests – Which tree species should be used to promote saproxylic beetle diversity? (2016, neutral): “Signi ﬁ cant differences in gamma-diversity betweentreespecieswerejudgedbynon-overlappingcon ﬁ denceinter-vals (Schenker and Gentleman, 2001).”
Births: final data for 2005. (2007, background): “That is, the difference between two rates may, in fact, be statistically significant even though confidence intervals for the two rates overlap (139).”