Modern Robust Data Analysis Methods: Measures of Central Tendency.

Wilcox, Rand R.; Keselman, H. J.

doi:10.1037/1082-989x.8.3.254

Cited by 417 publications

(295 citation statements)

References 66 publications

Supporting

Mentioning

287

Contrasting

Unclassified

Order By: Relevance

“…Then, we present various robust statistical methods described by Wilcox (2012), who also provides R algorithms implementing them at his website. 4…”

Section: Robust Statistical Methodsmentioning

confidence: 99%

“…However, empirical studies summarised by Wilcox (2012) confirm that if distributions vary in shape, Type I errors may be incorrect.…”

Section: Wilcox Andmentioning

confidence: 99%

“…It also suggests that productivity is generally highest for organic projects and lowest for embedded projects, with semi-detached projects somewhere in between. 9 Using trimmed means and the algorithms produced by Wilcox, we can test whether there are significant differences among the trimmed means for the different modes and also whether there is a linear relationship between trimmed means (Wilcox 2012). Summary statistics of the COCOMO project productivity values are shown in Table 3.…”

Section: Robust Alternatives To T and F Testsmentioning

confidence: 99%

“…An important issue for the use of effect sizes in meta-analysis is that the variance of the effect size needs to be estimated. Effect size variances are often quite complex to calculate, but Wilcox's software provides standard errors for the Cliff's d and the probability of superiority (Wilcox 2012).…”

Section: Guidelines For Interpreting Effect Size Magnitudementioning

confidence: 99%

See 3 more Smart Citations

Robust Statistical Methods for Empirical Software Engineering

et al. 2016

View full text Add to dashboard Cite

There have been many changes in statistical theory in the past 30 years, including increased evidence that non-robust methods may fail to detect important results. The statistical advice available to software engineering researchers needs to be updated to address these issues. This paper aims both to explain the new results in the area of robust analysis methods and to provide a large-scale worked example of the new methods. We summarise the results of analyses of the Type 1 error efficiency and power of standard parametric and non-parametric statistical tests when applied to non-normal data sets. We identify parametric and non-parametric methods that are robust to non-normality. We present an analysis of a large-scale software engineering experiment to illustrate their use. We illustrate the use of kernel density plots, and parametric and non-parametric methods using four different software engineering data sets. We explain why the methods are necessary and the rationale for selecting a specific analysis. We suggest using kernel density plots rather than box plots to visualise data distributions. For parametric analysis, we recommend trimmed means, which can support reliable tests of the differences between the central location of two or

show abstract

“…Then, we present various robust statistical methods described by Wilcox (2012), who also provides R algorithms implementing them at his website. 4…”

Section: Robust Statistical Methodsmentioning

confidence: 99%

“…However, empirical studies summarised by Wilcox (2012) confirm that if distributions vary in shape, Type I errors may be incorrect.…”

Section: Wilcox Andmentioning

confidence: 99%

Section: Robust Alternatives To T and F Testsmentioning

confidence: 99%

Section: Guidelines For Interpreting Effect Size Magnitudementioning

confidence: 99%

See 2 more Smart Citations

Robust Statistical Methods for Empirical Software Engineering

et al. 2016

View full text Add to dashboard Cite

show abstract

“…However, as there is some research to suggest that males may experience more alexithymia (Levant et al, 2009), future studies should aim for more gender-balanced studies. All our variables were skewed which could have reduced power and the ability to detect certain relationships (Wilcox and Keselman, 2003). However, we dealt with the skewed data in a number of ways including bootstrapping making our results more reliable and robust.…”

Section: Limitationsmentioning

confidence: 99%

Sleep quality and paranoia: The role of alexithymia, negative emotions and perceptual anomalies

Rehman

Gumley

Biello

2018

Psychiatry Research

View full text Add to dashboard Cite

A B S T R A C TRecent evidence suggests that sleep problems are associated with psychotic like experiences including paranoia. However, the mechanisms underpinning this association are not well understood and thus studies modelling hypothesised mediating factors are required. Alexithymia, the inability to recognise and describe emotions within the self may be an important candidate. In two separate studies we sought to investigate factors mediating the relationship between sleep quality and paranoia using a cross-sectional design. Healthy volunteers without a mental health diagnosis were recruited (study 1, N = 401, study 2, N = 402). Participants completed a series of measures assessing paranoia, negative emotions, alexithymia and perceptual anomalies in an online survey. In study 1, regression and mediation analyses showed that the relationship between sleep quality and paranoia was partially mediated by alexithymia, perceptual anomalies and negative affect. In contrast, study 2 found that the relationship between sleep quality and paranoia was fully mediated by negative affect, alexithymia and perceptual anomalies. The link between sleep quality and paranoia is unclear and reasons for discrepant results are discussed. Novel findings in this study include the link between alexithymia and paranoia.

show abstract

Enlarged Perivascular Spaces in Infancy and Autism Diagnosis, Cerebrospinal Fluid Volume, and Later Sleep Problems

Garic,

McKinstry,

Rutsohn

et al. 2023

JAMA Netw Open

View full text Add to dashboard Cite

ImportancePerivascular spaces (PVS) and cerebrospinal fluid (CSF) are essential components of the glymphatic system, regulating brain homeostasis and clearing neural waste throughout the lifespan. Enlarged PVS have been implicated in neurological disorders and sleep problems in adults, and excessive CSF volume has been reported in infants who develop autism. Enlarged PVS have not been sufficiently studied longitudinally in infancy or in relation to autism outcomes or CSF volume.ObjectiveTo examine whether enlarged PVS are more prevalent in infants who develop autism compared with controls and whether they are associated with trajectories of extra-axial CSF volume (EA-CSF) and sleep problems in later childhood.Design, Setting, and ParticipantsThis prospective, longitudinal cohort study used data from the Infant Brain Imaging Study. Magnetic resonance images were acquired at ages 6, 12, and 24 months (2007-2017), with sleep questionnaires performed between ages 7 and 12 years (starting in 2018). Data were collected at 4 sites in North Carolina, Missouri, Pennsylvania, and Washington. Data were analyzed from March 2021 through August 2022.ExposurePVS (ie, fluid-filled channels that surround blood vessels in the brain) that are enlarged (ie, visible on magnetic resonance imaging).Main Outcomes and MeasuresOutcomes of interest were enlarged PVS and EA-CSF volume from 6 to 24 months, autism diagnosis at 24 months, sleep problems between ages 7 and 12 years.ResultsA total of 311 infants (197 [63.3%] male) were included: 47 infants at high familial likelihood for autism (ie, having an older sibling with autism) who were diagnosed with autism at age 24 months, 180 high likelihood infants not diagnosed with autism, and 84 low likelihood control infants not diagnosed with autism. Sleep measures at school-age were available for 109 participants. Of infants who developed autism, 21 (44.7%) had enlarged PVS at 24 months compared with 48 infants (26.7%) in the high likelihood but no autism diagnosis group (P = .02) and 22 infants in the control group (26.2%) (P = .03). Across all groups, enlarged PVS at 24 months was associated with greater EA-CSF volume from ages 6 to 24 months (β = 4.64; 95% CI, 0.58-8.72; P = .002) and more frequent night wakings at school-age (F = 7.76; η2 = 0.08; P = .006).Conclusions and RelevanceThese findings suggest that enlarged PVS emerged between ages 12 and 24 months in infants who developed autism. These results add to a growing body of evidence that, along with excessive CSF volume and sleep dysfunction, the glymphatic system could be dysregulated in infants who develop autism.

show abstract

Modern Robust Data Analysis Methods: Measures of Central Tendency.

Cited by 417 publications

References 66 publications

Robust Statistical Methods for Empirical Software Engineering

Robust Statistical Methods for Empirical Software Engineering

Sleep quality and paranoia: The role of alexithymia, negative emotions and perceptual anomalies

Enlarged Perivascular Spaces in Infancy and Autism Diagnosis, Cerebrospinal Fluid Volume, and Later Sleep Problems

Contact Info

Product

Resources

About