Assessing the degree of urbanisation using a single-item self-report measure: a validation study

Figures & data

Table 1. Sensitivity, specificity, PPV, and NPV for different single-item self-report measure (SIDU) threshold scores for sample A and B.

Threshold	Sensitivity	Specificity	PPV	NPV
A: N = 36
1	1.00	0.00	0.81	NA
2	1.00	0.00	0.81	NA
3	1.00	0.00	0.81	NA
4	1.00	0.29	0.85	1.00
5	1.00	0.86	0.97	1.00
6	0.90	1.00	1.00	0.70
7	0.00	1.00	NA	0.19
B: N = 121
1	1.00	0.00	0.38	NA
2	1.00	0.13	0.41	1.00
3	1.00	0.23	0.44	1.00
4	0.98	0.36	0.48	0.96
5	0.93	0.67	0.63	0.94
6	0.76	0.87	0.78	0.86
7	0.00	1.00	NA	0.62

PPV: positive predictive value; NPV: negative predictive value. Bold faced values reflect the threshold score showing the most balanced combination of sensitivity and specificity.

Figure 1. Relationships between objective degree of urbanisation measures (left panels: surrounding address density, SAD; right panels: urbanisation grade categories) and the single-item self-report measure (SIDU), for two independent panels (A: N = 36, 65+, and B: N = 121, 55+). Correlation coefficients (r or r_s) and p-values are shown for each evaluated relationship. Dots on the right side of each red dashed red line represent those living in extremely urbanised (5) areas. SIDU scores are jittered to optimize visualization. *This coefficient reflects the relationship between $log\left(SAD\right)$ and the SIDU.

Table 2. Relationships between the single-item self-report measure (SIDU) and both the surrounding address density (SAD) and urbanisation grade categories, in Sample B.

	SAD ^a	Urbanisation grade categories
Age
<70 years old	n = 72, r(70) = 0.82 (95% CI 0.73–0.88), p < 0.001	r(70) = 0.79 (95% CI 0.69–0.87), p < 0.001
70+ years old	n = 49, r(47) = 0.84 (95% CI 0.73–0.91), p < 0.001	r(47) = 0.80 (95% CI 0.66–0.88), p < 0.001
Gender
Women	n = 71, r(69) = 0.82 (95% CI 0.72–0.88), p < 0.001	r(69) = 0.80 (95% CI 0.70–0.87), p < 0.001
Men	n = 50, r(48) = 0.84 (95% CI 0.73–0.91), p < 0.001	rs(48) = 0.77 (95% CI 0.63–0.87), p < 0.001

^aThe relationship between the SAD and SIDU was best described as an exponential function in this sample. Hence, the relationship between $log\left(SAD\right)$ and SIDU is reported here.