eng TQMP Tutorials in Quantitative Methods for Psychology 1913-4126 2006-03-01 2 1 1 10 article Sébastien Hélie sebastien.helie@umontreal.ca 1 Université de Montréal Model selection is a complicated matter in science, and psychology is no exception. In particular, the high variance in the object of study (i.e., humans) prevents the use of Popper’s falsification principle (which is the norm in other sciences). Therefore, the desirability of quantitative psychological models must be assessed by measuring the capacity of the model to fit empirical data. In the present paper, an error measure (likelihood), as well as five methods to compare model fits (the likelihood ratio test, Akaike’s information criterion, the Bayesian information criterion, bootstrapping and cross-validation), are presented. The use of each method is illustrated by an example, and the advantages and weaknesses of each method are also discussed. http://www.tqmp.org/Content/vol02-1/p001/p001.pdf Modelisation Model selection Number of free parameters eng TQMP Tutorials in Quantitative Methods for Psychology 1913-4126 2006-03-01 2 1 11 19 article Dominic Beaulieu-Prévost dominic.beaulieu.prevost@umontreal.ca 1 Université de Montréal For the last 50 years of research in quantitative social sciences, the empirical evaluation of scientific hypotheses has been based on the rejection or not of the null hypothesis. However, more than 300 articles demonstrated that this method was problematic. In summary, null hypothesis testing (NHT) is unfalsifiable, its results depend directly on sample size and the null hypothesis is both improbable and not plausible. Consequently, alternatives to NHT such as confidence intervals (CI) and measures of effect size are starting to be used in scientific publications. The purpose of this article is, first, to provide the conceptual tools necessary to implement an approach based on confidence intervals, and second, to briefly demonstrate why such an approach is an interesting alternative to an approach based on NHT. As demonstrated in the article, the proposed CI approach avoids most problems related to a NHT approach and can often improve the scientific and contextual relevance of the statistical interpretations by testing range hypotheses instead of a point hypothesis and by defining the minimal value of a substantial effect. The main advantage of such a CI approach is that it replaces the notion of statistical power by an easily interpretable three-value logic (probable presence of a substantial effect, probable absence of a substantial effect and probabilistic undetermination). The demonstration includes a complete example. http://www.tqmp.org/Content/vol02-1/p011/p011.pdf Statistics Confidence intervals Null hypotesis tests eng TQMP Tutorials in Quantitative Methods for Psychology 1913-4126 2006-03-01 2 1 20 25 article Guy L. Lacroix guy.lacroix@umontreal.ca 1 Gyslain Giguère g.giguere@umontreal.ca 1 Université de Montréal In this tutorial, we demonstrate how to use the Aggregate and Restructure procedures available in SPSS (versions 11 and up) to prepare data files for repeated-measures analyses. In the first two sections of the tutorial, we briefly describe the Aggregate and Restructure procedures. In the final section, we present an example in which the data from a fictional lexical decision task are prepared for analysis using a mixed-design ANOVA. The tutorial demonstrates that the presented method is the most efficient way to prepare data for repeated-measures analyses in SPSS. http://www.tqmp.org/Content/vol02-1/p020/p020.pdf Statistics repeated-measure designs eng TQMP Tutorials in Quantitative Methods for Psychology 1913-4126 2006-03-01 2 1 26 38 article Gyslain Giguère g.giguere@umontreal.ca 1 Université de Montréal This paper aims at providing a quick and simple guide to using a multidimensional scaling procedure to analyze experimental data. First, the operations of data collection and preparation are described. Next, instructions for data analysis using the ALSCAL procedure (Takane, Young and DeLeeuw, 1977), found in SPSS, are detailed. Overall, a description of useful commands, measures and graphs is provided. Emphasis is made on experimental designs and program use, rather than the description of techniques in an algebraic or geometrical fashion. http://www.tqmp.org/Content/vol02-1/p026/p026.pdf Statistics Multidimensional scaling