Validation and comparison study of three urbanicity scales in a Thailand context

BMC Public Health

Table 3 Dimensionality, reliability, and criterion-related validity of three urbanicity scales

Test parameter		Novak et al.		Dahly and Adair		Jones-Smith and Popkin
Test parameter		Present^a	Previous^b	Present^a	Previous^b	Present^a	Previous^b
Dimensionality
	Number of factor(s)	0	1	1	n/a	2	1
Internal consistency
	Chonbach’s alpha	0.48	n/a	0.51	0.87	0.74	0.89
	Item scale correlations
	Population size	0.44	0.50	0.46	0.72
	Population density			0.45	0.64	0.71	0.47
	Economic activity	0.44	0.88			0.72	0.70
	Built environment	0.40	0.73
	Housing					0.76	0.80
	Sanitation					0.70	0.77
	Communication	0.44	0.80	0.52	0.70	0.72	0.68
	Transportation			0.49	0.56	0.72	0.40
	Education	0.52	0.85	0.54	0.48	0.71	0.67
	Health	0.37	0.62	0.41	0.75	0.69	0.62
	Markets			0.42	0.80
	Traditional market					0.72	0.61
	Modern market					0.73	0.75
	Social services					0.75	0.51
	Diversity	0.45	0.40			0.72	0.67
Criterion-related validity
Compared with urban–rural dichotomy classification
	Observed Agreement	64.43 %	88.10 %	71.88 %	n/a	74.86 %	74 %
	Expected Agreement	52.20 %	49.80 %	51.47 %	n/a	50.48 %	51 %
	Kappa Statistic	0.26	0.76	0.42	n/a	0.49	0.48
	p-value	<0.001	<0.0001	<0.001	n/a	<0.001	<0.05
Compare with four-category urban classification
	Spearman’s Correlation	0.37	0.84	0.45	n/a	0.58	0.75-0.78
	p-value	<0.001	<0.0001	<0.001		<0.001

ISSN: 1471-2458