Naslov
Towards an accessible assessment of reasoning: the relation of statistical reasoning and classic reasoning task performance

Autori
Valerjev, Pavle ; Dujmović, Marin

Vrsta, podvrsta i kategorija rada
Sažeci sa skupova, sažetak, znanstveni

Izvornik
XXVIII Scientific Conference Empirical Studies in Psychology / - Beograd : Institute of Psychology ; Laboratory for Experimental Psychology (LEP), 2022, 50-51

ISBN
978-86-6427-199-8

Skup
28. naučni skup Empirijska istraživanja u psihologiji = 28th Empirical Studies in Psychology Conference

Mjesto i datum
Beograd, Srbija, 31.03.2022. - 03.04.2022

Vrsta sudjelovanja
Predavanje

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
dual-process theory ; reasoning ; statistical reasoning ; probabilistic reasoning ; rationality

Sažetak
Classic reasoning tasks regularly require computerized administration, tight experimental control and are overall not accessible to researchers outside the field. While there have been attempts to develop reasoning assessments, these have resulted in comprehensive yet difficult to implement instruments. This study is part of a project with the aim of determining which key factors need to be covered in such an instrument, while being easily administered and accessible. Such an instrument should cover various aspects connected to reasoning performance, one of which is statistical reasoning. In order to assess how a quick and simple test of statistical reasoning (TSR) is related to performance in classic reasoning tasks, it was conducted alongside various standard tasks (Cognitive Reflection Test, Base rate bias, Linda problem bias, and the Covariation detection task). The TSR has previously been found to be a good predictor of success in base-rate tasks. Therefore, it is considered a good candidate as part of a future, easily administrable and widely available instrument within the hybrid dual processing approach to reasoning. The goal of this study was to get better insight in how performance on different reasoning tasks and the TSR test relate. The correlational study was conducted in two countries and languages (UK and Croatia ; NUK=298 ; NCRO=292) and the samples were equalized on demographic characteristics. Analyses on both samples showed quite robust relations between performance in the reasoning tasks and on the test. Regression analyses show that continuous scores of all the tasks are significant predictors for TSR scores (both uniquely and as a linear combination) in both samples (UK sample: R = .58 ; R2 = .34 ; Radj2 = .33 ; F(4, 290) = 37.27 ; p < .01 ; Croatian sample: R = .49 ; R2 = .24 ; Radj2 = .23 ; F(4, 284) = 22.15 ; p <.01). An alternative analysis was conducted by simplifying the scores on our classic reasoning tasks (Base rate, Linda problem, Covariation detection). The number of correctly solved tasks was computed for each participant which served as a grouping variable (0-3 correctly solved tasks). The ANOVA shows that there is a significant effect of reasoning scores on TSR scores (F(3, 586) = 45.42, p < .01, ηp2 = .19). We can conclude that one factor which needs to be a part of any future full reasoning assessment has to cover statistical reasoning as it is robustly related to overall performance, however it is not suitable as a sole measure of reasoning performance.

Izvorni jezik
Engleski

Znanstvena područja
Psihologija

POVEZANOST RADA