Modeling contextual effects using individual-level data and without aggregation: an illustration of multilevel factor analysis (MLFA) with collective efficacy

Table 2 Intraclass Correlation Coefficients (ICC) for indicator variables in the Los Angeles Family and Neighborhood Study (LAFANS) n = 2594

	Intraclass correlation coefficient
	Total sample	Sample one	Sample two
Indicator variables	N = 2594	n = 1291	n = 1303
1…this is a close-knit neighborhood	0.083	0.112	0.121
2…there are adults that kids look up to	0.198	0.253	0.216
3…people around here are willing to help their neighbors	0.133	0.142	0.174
4…people in this neighborhood generally don’t get along with each other	0.149	0.148	0.178
5…adults watch out that kids are safe	0.085	0.112	0.089
6…people in this neighborhood do not share the same values	0.120	0.174	0.114
7…people in this neighborhood can be trusted	0.203	0.198	0.254
8…children were skipping school and hanging out on a street corner	0.104	0.131	0.125
9…children were spray-painting graffiti on a local building	0.262	0.299	0.273
10…children were showing disrespect to an adult	0.062	0.093	0.090

ICC refers to the proportion of variance in the indicator variable that is due to differences across neighborhoods. Neighborhoods were defined here as census tracts.
Items number 4 and 6 were reverse coded.

ISSN: 1478-7954