2011
DOI: 10.1017/s0142716410000457
|View full text |Cite
|
Sign up to set email alerts
|

A tutorial for analyzing human reaction times: How to filter data, manage missing values, and choose a statistical model

Abstract: This tutorial for the statistical processing of reaction times collected through a repeated-measure design is addressed to researchers in psychology. It aims at making explicit some important methodological issues, at orienting researchers to the existing solutions, and at providing them some evaluation tools for choosing the most robust and precise way to analyze their data. The methodological issues we tackle concern data filtering, missing values management, and statistical modeling (F1, F 2, F1 + F 2, quas… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
1
1

Citation Types

0
54
0
3

Year Published

2012
2012
2023
2023

Publication Types

Select...
8

Relationship

1
7

Authors

Journals

citations
Cited by 64 publications
(57 citation statements)
references
References 23 publications
0
54
0
3
Order By: Relevance
“…Consistent with many chronometric investigations of cognitive processing (Lachaud & Renaud, 2011), the response time scores for a trial greater than 3 standard deviations from the mean (i.e., outliers) were replaced with the mean of the remaining scores (i.e., threat outliers replaced with non-outlier threat mean response times; neutral outliers replaced with non-outlier neutral mean response times). This was done separately for the high-FIS and low-FIS groups.…”
Section: Trimming Of Response Time Scoresmentioning
confidence: 99%
“…Consistent with many chronometric investigations of cognitive processing (Lachaud & Renaud, 2011), the response time scores for a trial greater than 3 standard deviations from the mean (i.e., outliers) were replaced with the mean of the remaining scores (i.e., threat outliers replaced with non-outlier threat mean response times; neutral outliers replaced with non-outlier neutral mean response times). This was done separately for the high-FIS and low-FIS groups.…”
Section: Trimming Of Response Time Scoresmentioning
confidence: 99%
“…Iako merenje na racio skali omogućava primenu složenih statističkih postupaka za obradu vremena reakcije, analize koje su najčešće u upotrebi uglavnom nisu prilagođene karakteristikama ove mere (Baayen & Milin, 2010;Lachaud & Renaud, 2011;Ratcliff, 1993;Salthouse & Hedden, 2002;Whelan, 2008;Wilcox, 1998). Pri obradi vremena reakcije, statističke analize se najčešće sprovode na uprosečenim vrednostima, za svakog ispitanika i/ili za svaki stimulus (Dibbets, Maes, & Vossen, 2002;Fields, Landon-Jimenez, Buffington, & Adams, 1995).…”
Section: Karakteristike Vremena Reakcijeunclassified
“…Međutim, istraživanja pokazuju da raspodela vremena reakcije ne zadovoljava praktično nijedan od pomenutih statističkih preduslova. Najozbiljniju (u smislu posledica koje ima po rezultate statističkih testova) i najviše istraživanu povredu uslova predstavlja odstupanje empirijske raspodele vremena reakcije od pretpostavljene normalne raspodele (Lachaud & Renaud, 2011;Whelan, 2008). Ublažavanje povrede ovog uslova, odnosno, aproksimacija normalne raspodele, dovodi i do delimičnog popravljanja ostalih metodoloških problema, kao što je heteroskedastičnost raspodele vrednosti.…”
Section: Karakteristike Vremena Reakcijeunclassified
See 1 more Smart Citation
“…This fact strongly suggests that they constitute an integrative part of children's cognitive processing 7 and that they can therefore not be discarded from the analyses. Facing missing values, and in the absence of an account for the reasons behind missing values, it is recommended that authors apply a replacement procedure or use a statistical technique that allows for the given incompleteness (see Lachaud and Renaud, 2011, for a tutorial applied to psycholinguistics data). As we will see in the next section, some of the observations reported are highly sensitive to how missing data are treated and therefore not interpretable as such.…”
Section: The Problem Of Missing Data: Critical Issues With Data Analysismentioning
confidence: 99%