Uses, Abuses and Misuses of Significance Tests in the Scientific Community: Won't the Bayesian Choice be Unavoidable?

Lecoutre, Bruno; Lecoutre, Marie-Paule; Poitevineau, Jacques

doi:10.1111/j.1751-5823.2001.tb00466.x

Cited by 47 publications

(25 citation statements)

References 111 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Finally, Bayesian probability theory sits within a broader hierarchy of methods for describing partial evidence (Hall, 2003); such methods may be better suited for representing and quantifying different forms of epistemic and aleatory uncertainties in geomorphology. Because of the institutional barriers present in the use of the NHST (Orlitzky, 2012), Lecoutre et al (2001) and Lecoutre (2006) argue that Bayesian methods are taught as a means to highlight the problems with the NHST. Whilst the Bayesian approach is not without problems, it importantly provides a means to express a probabilistic belief in a substantive hypothesis.…”

Section: Recommendations For Better Practicementioning

confidence: 99%

How significant (p < 0.05) is geomorphic research?

Hutton

2014

Earth Surf Processes Landf

View full text Add to dashboard Cite

The pervasive application of the Null Hypothesis Significance Test in geomorphic research runs counter to widespread, long running, and often severe criticism of the method in the broader scientific literature. The application of the methodology typically leads to a binary separation of evidence into ‘significant’ and ‘not significant’ results based on a p‐value that is dependent on the null hypothesis being true. The method should not therefore be used to provide statistical support for substantive hypotheses, and is unsuitable for scientific inference in open systems where confirmation can only ever be partial. Alternative approaches based on Bayesian statistics can importantly be applied to measure partial support for hypotheses, conditional on the available data. Though not without their own assumptions, wider application of such methods can help facilitate a transition towards a broader approach to statistical, and in turn, scientific inference in geomorphic systems. Copyright © 2014 John Wiley & Sons, Ltd.

show abstract

Section: Recommendations For Better Practicementioning

confidence: 99%

How significant (p < 0.05) is geomorphic research?

Hutton

2014

Earth Surf Processes Landf

View full text Add to dashboard Cite

show abstract

“…We judge all of these features to result in meaningful differences in proportion correct. Readers may wonder whether these results are statistically significant at p<.05 (or some other number), but, unfortunately, the p-value is a poor method for judging substantive meaningfulness (Abelson, 1997;Cohen, 1994;Howard, Maxwell, & Fleming, 2000;Krueger, 2001;Lecoutre, Lecoutre, & Poitevineau, 2001). Instead of p-values, we would argue that readers should study the differences in proportions correct within the nodes of each tree and judge for themselves whether these differences are meaningful.…”

Section: Dollar Symbol "$"mentioning

confidence: 99%

Examination of the legibility of isolated characters of onscreen typefaces

Chaparro¹,

Merkle²,

Fox³

et al. 2011

IDJ

View full text Add to dashboard Cite

This study reports on the legibility of digits, symbols, and letters from 20 onscreen typefaces. Participants were asked to identify individual characters after a brief exposure and performance was recorded. Classification tree analysis was used to examine the influence of specific character features on legibility. Features included attributes such as height, width, weight, stroke variation, contrast, and complexity. Results reveal the most influential features for six commonly confused characters (zero, one, "e", "l", ÷, and $). The use of classification tree analysis appears to be promising for typographers in determining the limits of typeface features and for practitioners who must choose an optimal typeface for a particular use.

show abstract

“…In addition, the problem of estimating activity coefficients from multidimensional uncertainty distributions is not an easy task. Computer-based techniques on basis of Bayesian statistics are available but are neither easy to implement nor easy to explain [51,52].…”

Section: Figmentioning

confidence: 99%

Extended traceability of pH: an evaluation of the role of Pitzer's equations

Meinrath

2002

Analytical and Bioanalytical Chemistry

View full text Add to dashboard Cite

The 2002 IUPAC recommendation on pH (provisional) has taken its own philosophy to provide a basis for comparable and traceable assignment of a value, from a measurement, to the quantity pH. Whereas the substituted 1983 IUPAC recommendation relied heavily on precisely prescribed experimental techniques and procedures, the current recommendation defines a hierarchical relationship between references for comparison (primary and secondary standards) and objective criteria on the comparison of measurements with these standards. The recommendation aims at a traceability chain from the national metrological institution (NMI) level down to field and laboratory measurements. Currently, however, the traceability chain is developed to the level of certified reference materials (CRM), namely the above mentioned primary and secondary standards. To complete the traceability chain, several theoretical and practical aspects have to be pondered. In part, the methods for comparative assessment of different options have yet to be developed. As an illustrating example of the complexity of issues to be considered in a further extension of the traceability chain is estimation of the doubt associated with Pitzer coefficients. The Pitzer equations for activity coefficient modelling are explicitly mentioned in the 2002 IUPAC recommendation on pH (provisional) as enabling possible improvement in the ionic strength extrapolations to zero ionic strength. An assessment of uncertainty of ternary Pitzer coefficients is given for the first time.

show abstract

Uses, Abuses and Misuses of Significance Tests in the Scientific Community: Won't the Bayesian Choice be Unavoidable?

Cited by 47 publications

References 111 publications

How significant (p < 0.05) is geomorphic research?

How significant (p < 0.05) is geomorphic research?

Examination of the legibility of isolated characters of onscreen typefaces

Extended traceability of pH: an evaluation of the role of Pitzer's equations

Contact Info

Product

Resources

About