Performance of five two-sample location tests for skewed distributions with unequal variances.

Ullevål Department of Research Administration, Oslo University Hospital, N-0407 Oslo, Norway.
Contemporary clinical trials (Impact Factor: 1.51). 08/2009; 30(5):490-6. DOI: 10.1016/j.cct.2009.06.007
Source: PubMed

ABSTRACT Tests for comparing the locations of two independent populations are associated with different null hypotheses, but results are often interpreted as evidence for or against equality of means or medians. We examine the appropriateness of this practice by investigating the performance of five frequently used tests: the two-sample T test, the Welch U test, the Yuen-Welch test, the Wilcoxon-Mann-Whitney test, and the Brunner-Munzel test. Under combined violations of normality and variance homogeneity, the true significance level and power of the tests depend on a complex interplay of several factors. In a wide ranging simulation study, we consider scenarios differing in skewness, skewness heterogeneity, variance heterogeneity, sample size, and sample size ratio. We find that small differences in distribution properties can alter test performance markedly, thus confounding the effort to present simple test recommendations. Instead, we provide detailed recommendations in Appendix A. The Welch U test is recommended most frequently, but cannot be considered an omnibus test for this problem.

  • [Show abstract] [Hide abstract]
    ABSTRACT: We replicate Andreoni (Quarterly Journal of Economics 110: 1–21, 1995)’s finding that agents behave more selfishly when taking from a public account than when giving to a public good. Based on a neutral language setting we add new insights into motivations to give or take in a linear public good setting: we find that Andreoni’s result is partly driven by the complete elimination of giving options in the taking frame. However, a pure extension of the action space into the taking domain also leads to a significant increase in selfish behavior.
    Experimental Economics 01/2014; DOI:10.1007/s10683-014-9411-2 · 1.36 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: This study examined the influence of heterogeneity of variance on Type I error rates and power of the independent-samples Student's t-test of equality of means on samples of scores from normal and 10 non-normal distributions. The same test of equality of means was performed on corresponding rank-transformed scores. For many non-normal distributions, both versions produced anomalous power functions, resulting partly from the fact that the hypothesis test was biased, so that under some conditions, the probability of rejecting H 0 decreased as the difference between means increased. In all cases where bias occurred, the t-test on ranks exhibited substantially greater bias than the t-test on scores. This anomalous result was independent of the more familiar changes in Type I error rates and power attributable to unequal sample sizes combined with unequal variances.
    Journal of Applied Statistics 01/2013; 40(1). DOI:10.1080/02664763.2012.740620 · 0.45 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: The impact of touch-screen technology on spatial cognitive skills as related to molecular geometries was assessed through 102 one-on-one interviews with undergraduate students. Participants were provided with either printed 2D ball-and-stick images of molecules or manipulable projections of 3D molecular structures on an iPad. Following a brief introduction to common molecular shapes, participants were assessed on their representational competence. In particular, learners were tested on their ability to match and construct molecular representations. Using the device for less than 15 min, iPad users exhibited increased ability to correctly identify related chemical representations relative to learners taught with a paper-based method. Even in the last stage of the experiment, without access to the iPad, a significant difference between the two populations was sustained, with iPad-based learners demonstrating significantly higher representational competence than learners using the paper-based method. These findings suggest that touch-screen devices such as the iPad serve as effective learning technology for development of visuospatial and representational competence skills.
    Journal of chemical education 11/2014; 91(11):1810-1817. DOI:10.1021/ed400674v · 1.00 Impact Factor

Full-text (2 Sources)

Available from
Jun 2, 2014