Table 1. Descriptive statistics for the clusters composed of at least ten cells.

	Min	Max	Median	Mean	SD	CV	Skewness
Dependent variable
${\tilde{n}}_c^{\mathrm{c}\mathrm{l}\mathrm{u}\mathrm{s}\mathrm{t}\mathrm{e}\mathrm{r}}\left(2010\right)$	22	42492718	623	47193.57	964925.07	20.446	35.962
Offset
$n_c^{\mathrm{c}\mathrm{l}\mathrm{u}\mathrm{s}\mathrm{t}\mathrm{e}\mathrm{r}}\left(2005\right)$	39	41403322	696	47086.07	946332.04	20.098	35.551
Independent variables
log(Area)	2.303	11.187	2.944	3.256	1.070	0.329	2.375
log(Density)	1.291	7.684	3.509	3.526	0.977	0.277	0.172
Roundness	0.066	1.273	0.397	0.434	0.200	0.460	0.877
Irregularity	1.746	2.685	2.336	2.327	0.181	0.078	–0.181
CL	0.110	0.720	0.296	0.301	0.078	0.260	0.710
Heterogeneity	0.342	3.190	1.052	1.119	0.387	0.346	1.308
Gender	0.236	0.694	0.523	0.520	0.027	0.051	–1.179
Age	39.1	47.1	44.4	44.6	1.240	0.028	–0.803
Tertiary	0.570	0.774	0.653	0.653	0.045	0.070	0.185
Area	10	72194	19	160.50	1896.4	11.816	26.795
Density	3.636	2172.5	33.40	56.19	86.59	1.541	10.636

There are N = 2689 clusters. The clusters whose Area was less than 10 were omitted from the calculation of the statistics shown in the table. By definition, all independent variables are calculated for the data in 2005. SD represents the standard deviation. CV represents the coefficient of variation. The non-standardized Area and Density variables are not used in the regression models but shown for completeness.