eng TQMP Tutorials in Quantitative Methods for Psychology 1913-4126 2007-03-01 3 1 1 7 article George A. Michael george.michael@univ-lyon2.fr 1 Université Lyon 2 When investigating the deficits of a single patient, psychologists usually compare his/her performance in one or more tests to the performance of a control group. This can be done for any kind of variables, provided (i) that the design does not require the investigation of interactions between two or more factors, (ii) that the comparison between two or more individuals is not desired, and (iii) that the collection of the control data is possible. Yet, researchers are constantly interested in assessing interactions in the performance of an individual, and in the comparison of two or more individuals for investigating double dissociations or the efficiency of different methods of therapy, etc. They also may desire to investigate cases where only extremely simple and easy tasks can be performed, where ceiling effects are observed in the performance of the controls, and thus the case-controls comparison is impossible. The available statistical tools for the analysis of intra-individual or inter-individual performance (mainly with proportions) do not offer the possibility to assess interaction, they are not appropriate when some cells may contain 0 or 1 proportions, and when the sample size is small. Here, we present the Q’ test which may be used to test the hypothesis of equal proportions and proportion differences in 2 x K designs, offering therefore the possibility for researchers to investigate the main effects and interaction. This test can be used for any sample size and even when the data contains extreme proportions. Finally, a procedure of multiple comparisons described in this paper may be used to locate statistically significant sources of variance and differences. http://www.tqmp.org/Content/vol03-1/p001/p001.pdf Statistical test Test of proportions eng TQMP Tutorials in Quantitative Methods for Psychology 1913-4126 2007-03-01 3 1 8 13 article Denis Cousineau denis.cousineau@umontreal.ca 1 Université de Montréal Estimating parameters describing response time distributions is difficult. The most commonly used method for parameter estimation is the maximum likelihood method (ML). However, this method applied on the three-parameter Weibull distribution returns biased estimates and the amount of bias is unknown. A recent method, that we call nested maximum likelihood, was proposed by Gourdin, Hansen and Jaumard (1994). Due to its complexity, it has never been used and tested systematically. Here I compare it to the maximum likelihood method. The results shows that nested maximum likelihood is slightly better than ML. Although the gains are marginal, the method has important implications for future research in parameter estimation. http://www.tqmp.org/Content/vol03-1/p008/p008.pdf Parameter estimation Maximum likelihood method fre TQMP Tutorials in Quantitative Methods for Psychology 1913-4126 2007-03-01 3 1 14 25 article Louis Laurencelle louis.laurencelle@uqtr.ca 1 Université du Québec à Trois-Rivières Fait peu connu des chercheurs, voire des statisticiens appliqués, les tableaux de fréquences 2 × 2 appartiennent à trois catégories logiques différentes, selon que les effectifs totaux des rangées et des colonnes soient fixés d’avance ou non. Dans le modèle sans prédétermination, ou modèle 0, chaque élément peut s’inscrire dans n’importe quelle rangée ou colonne, comme lorsqu’on croise deux items dichotomiques d’un questionnaire. N’ayant trouvé aucun traitement spécifique de ce modèle dans la documentation, nous en proposons un, l’indice de concordance C, vérifié statistiquement par un simple test binomial. La comparaison des proportions obtenues dans deux groupes représente le modèle 1, les tailles de groupes étant prédéterminées : un test dû à Liddell et son approximation par le Khi-deux sont proposés, justifiés et validés. Enfin, dans le modèle 2, le total des réponses d’une sorte et de l’autre est connu d’avance ; pour ces cas, plus rares, le « test exact de Fisher » sied parfaitement, et le Khi-deux de Yates (avec correction de - 1/2 n) en est l’approximation par excellence. Les tests basés sur les rapports de vraisemblance ainsi que d’autres approximations sont aussi discutés. http://www.tqmp.org/Content/vol03-1/p014/p014.pdf Statistical test Tests of proportions