2022
DOI: 10.1371/journal.pone.0271668
|View full text |Cite
|
Sign up to set email alerts
|

Interaction effect: Are you doing the right thing?

Abstract: How to correctly interpret interaction effects has been largely discussed in scientific literature. Nevertheless, misinterpretations are still frequently observed, and neuroscience is not exempt from this trend. We reviewed 645 papers published from 2019 to 2020 and found that, in the 93.2% of studies reporting a statistically significant interaction effect (N = 221), post-hoc pairwise comparisons were the designated method adopted to interpret its results. Given the widespread use of this approach, we aim to:… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
2
1

Citation Types

0
11
0

Year Published

2023
2023
2024
2024

Publication Types

Select...
9

Relationship

2
7

Authors

Journals

citations
Cited by 28 publications
(13 citation statements)
references
References 66 publications
(126 reference statements)
0
11
0
Order By: Relevance
“…Partial eta‐squared (normalηnormalp2) and its 90% confidence interval (CI) were computed as estimates of effect sizes for the ANOVAs' main effects and interactions (Lakens, 2013). In case of significant interactions, interpretation was based on the model estimated marginal means and 95% CI (although means and 95% CI calculated from the raw data are reported in the text and in the plots) (Garofalo et al, 2022). A statistical significance threshold of p < .05 was adopted.…”
Section: Methodsmentioning
confidence: 99%
“…Partial eta‐squared (normalηnormalp2) and its 90% confidence interval (CI) were computed as estimates of effect sizes for the ANOVAs' main effects and interactions (Lakens, 2013). In case of significant interactions, interpretation was based on the model estimated marginal means and 95% CI (although means and 95% CI calculated from the raw data are reported in the text and in the plots) (Garofalo et al, 2022). A statistical significance threshold of p < .05 was adopted.…”
Section: Methodsmentioning
confidence: 99%
“…However, this approach has several limitations, most prominently, it increases other sources of error, like the chance of false negatives (type II error). It has been argued that correcting for multiple comparisons is more appropriate when carrying out exploratory correlations (not hypothesis-driven) than when carrying out hypothesis-driven correlations, as in our case (for more on this, see 28 ). As this is a matter of debate, we have reported both corrected and uncorrected p-values.…”
Section: Discussionmentioning
confidence: 80%
“…A mixed‐factorial analysis of co‐variance (ANCOVA) examined how video type (single vs. consensus) and condition (Tower Building vs. Marshmallow Collecting) affected the total number of marshmallows collected by children with age as a covariate (see Figure 3). There was a significant interaction effect of video type on condition ( F [1, 64] = 4.59, p = .036, ηnormalp2 = 0.07), indicating that video type elicited more difference between conditions when observing a single model compared with when observing a group consensus (interpretation based on model estimated marginal means and 95% confidence intervals as shown in Figure 4, see Garofalo et al, 2022). There was no significant main effect of video type ( F [1, 64] = 0.36, p = .552) or condition ( F [1, 64] = 0.53, p = .469), nor a significant effect of age ( F [1, 64] = 1.63, p = .206).…”
Section: Resultsmentioning
confidence: 99%