A Comparative Study of Tests for Homogeneity of Variances, with Applications to the Outer Continental Shelf Bidding Data

Conover, W. J.; Johnson, M. Eric; Johnson, Myrle M.

doi:10.2307/1268225

Cited by 313 publications

(256 citation statements)

References 0 publications

Supporting

Mentioning

242

Contrasting

Unclassified

Order By: Relevance

“…The results in the table confirm the sensitivity of Cochran's Test to non-normality, since much more than 1% of the results are significant when the underlying distribution is not normal. The results in Table 2 are also consistent with the results given in CONOVER et al (5), but are more applicable to a collaborative study setting. The results also confirm that the k-statistic in ASTM E691-2014 is much more prone to falsely identify outliers than Cochran's Test as discussed from a more theoretical perspective above.…”

Section: Cochran's Test and The K-statisticsupporting

confidence: 90%

“…It has long been known that the F-statistic used in testing whether variances are the same in two populations is quite sensitive to deviations of the underlying populations from normality (4). Since Cochran's Test and the k-statistic are derived from the F-Test, it would be expected that they are similarly or even more sensitive to deviations from normality and this was confirmed by CONOVER et al (5). Several approaches that are less subject to the sensitivity to deviations from normality seen with the F-test, Cochran's Test, or the k-statistic have been proposed for the general problem of testing whether all of the within-group-variances are the same (for example, see (6) and (7)).…”

Section: Introductionmentioning

confidence: 86%

“…As shown in Table 3, only L0 is both robust and close to the nominal significance level. It was further concluded in CONOVER et al (5) that procedures such as L0 and Lmed were among the best choices as regards statistical power. On the basis of these results, the L0 approach was adapted to derive an alternative to Cochran's Test.…”

Section: Levene's Test Simulationsmentioning

confidence: 97%

“…Lmed is often referred to as the Brown-Forsythe Test. The fourth, L0, was suggested in CONOVER et al (5) and examined in HINES and HINES (9).…”

Section: Levene's Testmentioning

confidence: 99%

See 3 more Smart Citations

Within-Laboratory Variance Outlier Detection: An Alternative to Cochran’s Test

Morton

2017

Beiträge Zur Tabakforschung International/Contributions to Tobacco Research

View full text Add to dashboard Cite

SummaryAn important step in the full definition of an analytical method is the characterization of the within and between laboratory variability. This is typically done through collaborative studies involving multiple laboratories. The statistical analysis of the results of collaborative studies is generally carried out using standardized protocols such as those given in ISO 5725-2 or ASTM E691-14.One aspect of the evaluation of collaborative studies is the identification of outlying laboratories which are then excluded from the variance calculation associated with the analytical method. Whether particular laboratories are identified as outliers can have a dramatic effect on the calculated variances.The generally recommended approach to identify laboratories with excessive within-laboratory variation is Cochran’s Test or something similar. However, Cochran’s Test is very sensitive to non-normality of the underlying statistical distribution. When the assumption of normality is violated, Cochran’s Test can wrongly identify laboratories as outliers at much greater than the nominally stated error rate, even for deviations from normality that are very difficult to detect analytically.In this paper, an alternative to Cochran’s Test, adapted from Levene’s Test, is proposed and shown to approximately maintain the stated error rate when the underlying distribution is not normal. This newly adapted test is recommended for future collaborative study analysis in place of Cochran’s Test.

show abstract

Section: Cochran's Test and The K-statisticsupporting

confidence: 90%

Section: Introductionmentioning

confidence: 86%

Section: Levene's Test Simulationsmentioning

confidence: 97%

“…Lmed is often referred to as the Brown-Forsythe Test. The fourth, L0, was suggested in CONOVER et al (5) and examined in HINES and HINES (9).…”

Section: Levene's Testmentioning

confidence: 99%

See 2 more Smart Citations

Within-Laboratory Variance Outlier Detection: An Alternative to Cochran’s Test

Morton

2017

Beiträge Zur Tabakforschung International/Contributions to Tobacco Research

View full text Add to dashboard Cite

show abstract

“…Prior to the analysis of the effect of the independent variable on the dependent variable we estimated the internal consistency degree of the items (Cronbach, 1951) and then we proceeded to the measures of the effect of the 10-month program of mental coaching (VI) with the Student test for two independent samples (Brown & Forsythe, 1974;Conover, Johnson, & Johnson, 1981;Olejnik & Algina, 1987). Table 2 shows that only 6 of 12 mental skills of the OMSAT-3 have been identified to achieve the statistical analysis: (Alpha min ≥ 0.62).…”

Section: Discussionmentioning

confidence: 99%

The Effects of Mental Imagery and Cardiac Coherence on Mental Skills of Tunisian Karate Players at School Age

Hamrouni¹,

Alem²,

Baert³

et al. 2015

APE

View full text Add to dashboard Cite

The aim of our research consisted in checking the influence of the mental coaching using the mental imagery and relaxation based on cardiac coherence, on the improvement of the cognitive, emotional and behavioural reactions of the Tunisian karate Elite at school age. Our study was about to check if the Tunisian karate Elite at school age who had undergone a mental coaching has a better mental profile than their counterparts of other sporting disciplines and karate players with usual training method. Our sample consisted on an experimental group (N = 24 athletes) and control group (N = 22 athletes), all aged from 16 to 19 years. The experimental group followed for 10 months a psychological and mental coaching. The control group continued to train normally for the same period of time. To assess the mental skills of the participants, we used the OMSAT-3 (Guelmami et al., 2014). Our study showed that a 10 month mental coaching was needed to improve 4 mental skills, i.e. "goal setting" "relaxation", "imagery" and "mental practices". Our findings sustained the importance of a three-dimensional mental coaching in relation to one basic skill (goal setting), a psychosomatic skill (relaxation) and two cognitive skills (imagery and mental practice).

show abstract