Type IV errors and interactions.

Levin, Joel R.; Marascuilo, Leonard A.

doi:10.1037/h0033455

Cited by 51 publications

(28 citation statements)

References 19 publications

(13 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…7. Despite its widespread use, Rosenthal and Rosnow (1992) argued that the traditional method of interpreting interaction effects using means is misleading (see also Levin & Marascuilo, 1972). Because the means include both main effects and an interaction (in a 2 × 2 design), an experimenter can accurately understand the interaction only after removing the main effects from the means (and plotting these interaction residuals).…”

Section: Similar To Past Research On Attributions (Seementioning

confidence: 99%

Negotiating Biculturalism

Benet‐Martínez

Leu

Lee

et al. 2002

Journal of Cross-Cultural Psychology

820

238

View full text Add to dashboard Cite

The authors propose that cultural frame shifting—shifting between two culturally based interpretative lenses in response to cultural cues—is moderated by perceived compatibility (vs. opposition) between the two cultural orientations, or bicultural identity integration (BII). Three studies found that Chinese American biculturals who perceived their cultural identities as compatible (high BII) responded in culturally congruent ways to cultural cues: They made more external attributions (a characteristically Asian behavior) after being exposed to Chinese primes and more internal attributions (a characteristically Western behavior) after being exposed to American primes. However, Chinese American biculturals who perceived their cultural identities as oppositional (low BII) exhibited a reverse priming effect. This trend was not apparent for noncultural primes. The results show that individual differences in bicultural identity affect how cultural knowledge is used to interpret social events.

show abstract

Section: Similar To Past Research On Attributions (Seementioning

confidence: 99%

Negotiating Biculturalism

Benet‐Martínez

Leu

Lee

et al. 2002

Journal of Cross-Cultural Psychology

820

238

View full text Add to dashboard Cite

show abstract

“…Interpretation of interactions, especially three-way interactions, can be difficult, since the comparisons that are statistically significant require comparison of at least four cells for a two-way interaction and at least eight cells for a three-way interaction [22]. Thus, the Explanation X Case interaction does not allow one to say that Explanation is significantly (i. e. statistically) better than No Explanation for one case, while No Explanation is significantly better than Explanation for another case.…”

Section: Effects Of Explanationmentioning

confidence: 99%

The Impact of an Explanation Capability for a Computer Consultation System

Erdman¹

1985

Methods Inf Med

View full text Add to dashboard Cite

The impact of a computer consultation program including explanations for its decisions was examined. The consultation program was designed as a decision aid for non-psychiatrists regarding the treatment of depression. Both psychiatrists and non-psychiatrists were used to evaluate the system, with different results for the two groups. Explanations improved the non-psychiatrists’ attitudes towards the computer recommendations, while for psychiatrists the impact of explanation depended on the specific content of the recommendations and explanations. The results for psychiatrists were consistent with the view that explanations are important for computer consultation systems because they may help physicians decide whether or not the computer’s advice is reasonable.

show abstract

“…An explanation of differences between the variables does not answer the interaction question, but an explanation of differences between the differences of the variables under one condition compared with another condition does. The reader is encouraged to consult Marascuilo and Levin (1970) and Levin and Marascuilo (1972; for further information about interaction interpretations. If authors and editors are aware that interactions are frequently misinterpreted, they may be more likely to check their data or consult with statisticians before publishing their results.…”

Section: Othermentioning

confidence: 99%

An Examination of Statistical Interactions in the Special Education Literature

1986

View full text Add to dashboard Cite

This study investigated the prevalence in the special education literature of the Type IV error, an error encountered when the finding of a significant interaction in an analysis of variance model is followed by discussion that does not correspond to the correct conceptualization of an interaction. One hundred ninety-three studies were identified that included significant interaction effects based on a two-way ANOVA design. It was found that only 9% of the studies reviewed interpreted the significant interaction consistent with the accurate explanation of a differential or joint effect. These studies did not, however, include a post hoc analysis to determine the cause of the interaction. Seventy-seven percent of the studies with a significant interaction were explained by using the cell means comparison procedure, which does not correspond to the ANOVA model with interaction effects.

show abstract

Type IV errors and interactions.

Cited by 51 publications

References 19 publications

Negotiating Biculturalism

Negotiating Biculturalism

The Impact of an Explanation Capability for a Computer Consultation System

An Examination of Statistical Interactions in the Special Education Literature

Contact Info

Product

Resources

About