What Hypotheses do “Nonparametric” Two-Group Tests Actually Test?

@article{Conroy2012WhatHD,
  title={What Hypotheses do “Nonparametric” Two-Group Tests Actually Test?},
  author={Ron{\'a}n M. Conroy},
  journal={The Stata Journal},
  year={2012},
  volume={12},
  pages={182 - 190}
}
  • R. Conroy
  • Published 1 June 2012
  • Psychology
  • The Stata Journal
In this article, I discuss measures of effect size for two-group comparisons where data are not appropriately analyzed by least-squares methods. The Mann–Whitney test calculates a statistic that is a very useful measure of effect size, particularly suited to situations in which differences are measured on scales that either are ordinal or use arbitrary scale units. Both the difference in medians and the median difference between groups are also useful measures of effect size. 
Confidence interval estimation for treatment effects in cluster randomization trials based on ranks
  • G. Zou
  • Mathematics
    Statistics in medicine
  • 2021
TLDR
Primary emphasis is given to confidence interval estimation in trials with a small number of large clusters, to obtain placement values based on overall ranks and arm-specific ranks prior to application of the ratio estimator, cluster-size-weighted means and mixed models for adjusting clustering effects.
Statistical considerations for outcomes in clinical research: A review of common data types and methodology
TLDR
A brief, yet comprehensive overview of common data types in clinical research and appropriate statistical methods for analyses, which include continuous data, binary data, count data, multinomial data, and time-to-event data is provided.
It’s Personal: The Impact of Victimization on Motivations and Career Interests Among Criminal Justice Majors at Diverse Urban Colleges
Abstract This article extends a small but significant body of work on the motivations of criminal justice students to enter the major and to pursue a criminal justice career (Krimmel & Tartaro 1999;
Gender Gaps in Equity Crowdfunding: Evidence from a Randomized Field Experiment
Although prior research in traditional equity financing shows that male founders are preferred, emerging evidence in low-stakes crowdfunding (e.g., rewards-based crowdfunding) indicates that female
Leisure Time instead of a Monetary Bonus ? A computerized real effort task to elicit productivity levels
  • Economics
  • 2018
Recently, many big industries are doubting whether bonus schemes are the most optimal way to reward employees. Also, leisure time has gained high importance amongst the newest generations. This paper
People You Care about in and out of the System: The Impact of Arrest on Criminal Justice Views, Choice of Major, and Career Motivations
Abstract Borrowing from intersectionality theory, this study aims to understand how experiences of arrest – alone or in combination with victimization and criminal justice ties – inform students’
Experiences of Computer Science Transfer Students
TLDR
It is found that while transfer students were more diverse demographically, the support of the university for transfer student orientation tended to mitigate social shocks of transferring, but this did not eliminate gaps in academic performance.
...
...

References

SHOWING 1-10 OF 17 REFERENCES
A probability-based measure of effect size: robustness to base rates and other factors.
TLDR
The probability-based measure A, the nonparametric generalization of what K. O. McGraw and S. Wong (1992) called the common language effect size statistic, is insensitive to base rates and more robust to several other factors (e.g., extreme scores, nonlinear transformations).
A common language effect size statistic.
Some of the shortcomings in interpretability and generalizability of the effect size statistics currently available to researchers can be overcome by a statistic that expresses how often a score
Parameters behind “Nonparametric” Statistics: Kendall's tau, Somers’ D and Median Differences
So-called “nonparametric” statistical methods are often in fact based on population parameters, which can be estimated (with confidence limits) using the corresponding sample statistics. This article
Individual Comparisons by Ranking Methods
The comparison of two treatments generally falls into one of the following two categories: (a) we may have a number of replications for each of the two treatments, which are unpaired, or (b) we may
Using the ROC curve for gauging treatment effect in clinical trials
TLDR
This work adapts recently developed methods for receiver operating characteristic (ROC) curve regression analysis to extend the Mann-Whitney test to accommodate covariate adjustment and evaluation of effect modification.
Probabilistic index: an intuitive non-parametric approach to measuring the size of treatment effects.
TLDR
P(X > Y) is proposed as an alternative index, its correspondence with well‐known non‐parametric statistics, compare it to the standardized mean difference index, and illustrate with clinical data.
On a Use of the Mann-Whitney Statistic
For m = n an equivalent statistic had been proposed and studied earlier by Wilcoxon [2]. The main aim of these studies was to develop a test of the hypothesis that X and Y have the same probability
Estimates of Location Based on Rank Tests
A serious objection to many of the classical statistical methods based on linear models or normality assumptions is their vulnerability to gross errors. For certain testing problems this difficulty
A Critique and Improvement of the CL Common Language Effect Size Statistics of McGraw and Wong
McGraw and Wong (1992) described an appealing index of effect size, called CL, which measures the difference between two populations in terms of the probability that a score sampled at random from
SOMERSD: Stata module to calculate Kendall's tau-a, Somers' D and median differences
The somersd package contains the programs somersd, cendif and censlope, which calculate confidence intervals for the parameters behind rank or "nonparametric" statistics. The program somersd
...
...