Figure 4
A heatmap shows Spearman correlation coefficients among multiple features, with colors indicating strength and direction.The triangular heatmap titled “Spearman Correlation Coefficient Matrix”, displays pairwise correlations between numerous variables. The horizontal and vertical axes list the same set of features, including “A 1 underscore 1”; “A 3 underscore 1”; “A 5 underscore 1”; “A 7 underscore 1”; “A 7 underscore 1”; “A 11 underscore 1”; “W 1 underscore 1”; “W 4 underscore 1”; “W 6 underscore 1”; “W 9 underscore 1”; “W 12 underscore 1”; “W 12 underscore 1”; “E 1 underscore 1”; “E 3 underscore 1”; “E 5 underscore 1”; “E 7 underscore 1”; “Older than 35 underscore 1”; “K G C C underscore A f”; “K G C C underscore A w”; “K G C C underscore B S k”; “K G C C underscore B W k”; “K G C C underscore C s a”; “K G C C underscore D f a”; “project end use underscore Bridge”; “project end use underscore Highway, road, street”; “project end use underscore Multi-family dwelling”; “project end use underscore Other heavy construction”; “project end use underscore Powerline, transmission line”; “project end use underscore Refinery”; “project end use underscore Single family or duplex dwelling”; “project cost underscore 1 million to 5 million”; “project cost underscore 250 thousand to 500 thousand”; “project cost underscore 50 thousand to 250 thousand”; “project cost underscore Under 50 thousand dollars”; “project type underscore Demolition”; and “project type underscore New project or new addition”. Only the lower left triangular portion of the matrix is shown, with the upper right triangle omitted to avoid redundancy. Each cell represents the Spearman correlation coefficient between a pair of variables, encoded by color. A vertical color bar on the right ranges from negative 1.00 to 1.00 with an interval of 0.25, where deep blue indicates strong negative correlation, white or light gray indicates near zero correlation, and deep red indicates strong positive correlation. Most cells appear in light gray or faint shades, suggesting weak or negligible correlations across the majority of feature pairs. Scattered light orange or light blue cells indicate mild positive or negative relationships, respectively, but there are no large contiguous regions of strong correlation. The diagonal is not emphasized, as it would represent self-correlation.

Spearman correlation coefficient matrix. Source: Authors’ own work

or Create an Account

Close subscription notice
Close access options