Robustness of statistical methods when measure is affected by ceiling and/or floor effect

Abstract

Goals and methods A simulation study investigated how ceiling and floor effect (CFE) affect the performance of Welch’s t -test, F -test, Mann-Whitney test, Kruskal-Wallis test, Scheirer-Ray-Hare-test, trimmed t -test, Bayesian t -test, and the “two one-sided tests” equivalence testing procedure. The effect of CFE on the estimate of group difference and on its confidence interval, and on Cohen’s d and on its confidence interval was also evaluated. In addition, the parametric methods were applied to data transformed with log or logit function and the performance was evaluated. The notion of essential maximum from abstract measurement theory is used to formally define CFE and the principle of maximum entropy was used to derive probability distributions with essential maximum/minimum. These distributions allow the manipulation of the magnitude of CFE through a parameter. Beta, Gamma, Beta prime and Beta-binomial distributions were obtained in this way with the CFE parameter corresponding to the logarithm of the geometric mean. Wald distribution and ordered logistic regression were also included in the study due to their measure-theoretic connection to CFE, even though these models lack essential minimum/maximum. Performance in two-group, three-group and 2 × 2 factor design scenarios was investigated by fixing the group differences in terms of CFE parameter and by adjusting the base level of CFE. Results and conclusions In general, bias and uncertainty increased with CFE. Most problematic were occasional instances of biased inference which became more certain and more biased as the magnitude of CFE increased. The bias affected the estimate of group difference, the estimate of Cohen’s d and the decisions of the equivalence testing methods. Statistical methods worked best with transformed data, albeit this depended on the match between the choice of transformation and the type of CFE. Log transform worked well with Gamma and Beta prime distribution while logit transform worked well with Beta distribution. Rank-based tests showed best performance with discrete data, but it was demonstrated that even there a model derived with measurement-theoretic principles may show superior performance. Trimmed t -test showed poor performance. In the factor design, CFE prevented the detection of main effects as well as the detection of interaction. Irrespective of CFE, F -test misidentified main effects and interactions on multiple occasions. Five different constellations of main effect and interactions were investigated for each probability distribution, and weaknesses of each statistical method were identified and reported. As part of the discussion, the use of generalized linear models based on abstract measurement theory is recommended to counter CFE. Furthermore, the nec...

Stressors, coping and symptoms of adjustment disorder in the course of the COVID-19 pandemic – study protocol of the European Society for Traumatic Stress Studies (ESTSS) pan-European study

“…If the data will not follow a normal distribution, the data will be transformed with appropriate data transformation methods (e.g. linear square, cube root or logarithmic transformation, depending on the distribution of the skewed data) prior to data analysis (Šimkovic & Träuble, 2019 ).…”

Different measurements of bilingualism and their effect on performance on a Simon task

“…Furthermore, in particular because accuracy rates in our study were so high, the results should be interpreted with caution. The use of statistical analysis such as analysis of variance can be problematic when a ceiling effect is present and can lead to Type I errors (Šimkovic & Träuble, 2019).…”

Linking the Puzzle Pieces of the Past: A Study of Relational Memory in Children with Autism Spectrum Disorder

“…Data from the computer task was analyzed using generalized linear mixed effects models (GLMMs) by use of the glmer()‐ function (from the lme4 package) [Bates, Mächler, Bolker, & Walker, 2015]. GLMMs are recommended to counter ceiling and floor effects as might apply to the false alarm and corrected recognition ratio measures [Šimkovic & Träuble, 2019]. For hits and false alarms, a Poisson distribution with a log link was applied to the model to fit the count data, whereas a binomial distribution and the weight argument set to the total number of responses was applied to the proportion data from the corrected recognition ratio.…”