Assessing urban and rural neighborhood characteristics using audit and GIS data: derivation and reliability of constructs

Table 1 Standardized Cronbach's alpha by urbanicity and multi-group one-factor analysis^§ for a priori constructs, exploratory sample (n = 6,388).

Construct		Standardized Cronbach's Alpha^§		Goodness-of-fit for non-invariant model				Difference test^‡
		Urban	Rural	Number of parameters	CFI > 0.95*	TLI > 0.95*	RMSEA < 0.05*	p-value
Constructs from Caughy et al, 2001
Physical incivilities		0.57	0.59	40	0.960	0.951	0.002	<0.0001
Territoriality		0.22	0.34	34	0.211	0.096	0.005	<0.0001
Social spaces		0.31	0.24	38	0.664	0.614	0.042	<0.0001
Constructs from Pikora et al, 2002
Safety		0.39	0.41	24	0.774	0.661	0.006	<0.0001
Aesthetics	Streetscape	0.29	0.27	52	0.001	0.287	0.006	<0.0001
	Views	0.28	0.44	20	1	1	0	0.983
Functionality	All items together	0.18	0.19	70	0.791	0.791	0.004	<0.0001
Destinations		0.27	0.11	19	0.203	0.595	0.011	<0.0001

§ Data are weighted by the road segment's length.
‡ Ho: Invariance (i.e. all factor loadings and thresholds are the same by urbanicity) H₁: Non invariance (i.e. different factorloadings and thresholds by urbanicity)
Abbreviations: RMSEA root mean square error of approximation, CFI comparative fit index, and TLI Tucker-Lewis index
*These are "rules of thumb" guidelines to interpret these indices (see paper for more information).

ISSN: 1479-5868