2016
DOI: 10.1007/s10869-016-9464-7
|View full text |Cite
|
Sign up to set email alerts
|

Estimating Group-Level Relationships: General Recommendations and Considerations for the Use of Intraclass Correlation Coefficients

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
5
0

Year Published

2017
2017
2023
2023

Publication Types

Select...
8

Relationship

1
7

Authors

Journals

citations
Cited by 11 publications
(5 citation statements)
references
References 65 publications
0
5
0
Order By: Relevance
“…This model and procedures hold considerable promise for studying other interrater agreement problems as well as other phenomena in group-level studies. For instance, for primary group-level studies and meta-analytic studies with group-level data, Bliese (1998) and Burke, Landis, and Burke (2016), respectively, argued for the use of ICC(2) values in correcting group-level correlations. Although these authors demonstrated the potential usefulness of such corrections with simulated and empirical data, their procedures did not allow for a more complete modeling of group level data and assessments of the accuracy of such reliability corrections.…”
Section: Discussionmentioning
confidence: 99%
“…This model and procedures hold considerable promise for studying other interrater agreement problems as well as other phenomena in group-level studies. For instance, for primary group-level studies and meta-analytic studies with group-level data, Bliese (1998) and Burke, Landis, and Burke (2016), respectively, argued for the use of ICC(2) values in correcting group-level correlations. Although these authors demonstrated the potential usefulness of such corrections with simulated and empirical data, their procedures did not allow for a more complete modeling of group level data and assessments of the accuracy of such reliability corrections.…”
Section: Discussionmentioning
confidence: 99%
“…In case of the multiple correlations from the same study were combined, the Mosier reliabilities of the composite measures were also computed (Schmidt & Hunter, 2015). In those few cases in which no reliabilities were reported, this study took the average reliability of the same variable from the set of primary studies, that is, an accepted method when conducting meta-analyses at the group level of analysis (M. I. Burke, Landis, & Burke, 2017).…”
Section: Methodsmentioning
confidence: 99%
“…In those few cases in which no reliabilities were reported, this study took the average reliability of the same variable from the set of primary studies, that is, an accepted method when conducting meta-analyses at the group level of analysis (M. I. Burke, Landis, & Burke, 2017). The average of the available alpha coefficients and ICC(2) values ranged from .78 to .93 and .63 to .76, respectively, for task-focused leadership and .78 to .93 and .75 to .77, respectively, for person-focused leadership.…”
Section: Meta-analytic Proceduresmentioning
confidence: 99%
“…Given that employees were nested in 26 countries, we evaluated the need for multilevel modeling by calculating the intraclass correlation (ICC(1) and ICC(2)) for the dependent variables (Burke, Landis, & Burke, ). The ICC(1) value is interpreted as the proportion of variance explained by group membership (i.e., country), and it was calculated using the following Bartko equation (see Bliese, ): ICC(1) = (MSB – MSW) / (MSB + ( n – 1) × MSW), where MSB is the between‐group mean square, MSW is the within‐group mean square, and n is the average number of members within groups.…”
Section: Methodsmentioning
confidence: 99%