Since there are two spatial classes (geo-wide and geo-narrow), three temporal classes (evergreen, short-lived and bursty), and two socio-topical classes (social and topical), there are 12 possible combinations of classes.
In the spatial EC class, spatial correlation between local pixels is exploited; missing MBs are recovered by interpolation from neighbor pixels.
For each spatial lag class l = 1,…,6 and for each estimation m = KC or RS, we calculated the mean error [ME m(l)], the mean square error [MSE m(l)] and the mean absolute error [MAE m(l)] (Lee and Wentz 2008).
For each spatial lag class we calculated the percent change in MSE and MAE from RS to KC [e.g., % change in MSE(l) = {[MSEKC l) – MSERS l)]/MSERS l)} × 100, so that a negative percent change means that KC has a lower estimation error than RS].
These calculations were carried out using Moran's I [ 72, 73] for spatial distance classes (in metres), the dimension of which was 5,000 m for geographical distances (9 classes), and 102 m for altitude differences (9 classes).
In all three analyses, ten spatial distance classes were created using the equal-frequency method, which generates uneven distance intervals with roughly equal numbers of pairwise comparisons.
This novel recasts the spatial analogue of class warfare – ingrained in the British popular imagination of the skyscraper – as an emblem for the age of evolutionary psychology.
where N s is the maximum spatial distance within class, and the η is the weight.
These perceived class spatial and class or social boundaries are particularly visible during evenings, when labourers come home to rest after a long day of work, gathering in street corners to socialise with friends and acquaintances.
Very consistent spatial patterns of class change were found between transect pairs, suggesting high stability in the classification process.
