Skip to main content
. 2018 Jun 19;11(4):82–94. doi: 10.1177/1937586718779223

Table 2.

Mean (M), Standard Deviation (SD), and Analysis of Variance (ANOVA) of Network Metrics by Unit Shape.a

Network Metricb Compact Circlec (n = 2) Compact Square (n = 7) Race Track (n = 9) Crossd (n = 6) ANOVA Result (df = 3)
M SD M SD M SD M SD
Node sizee 24.25 3.53 15.29 1.88 16.19 5.13 26.33 6.40 F = 8.08, p = .001
Densitye 0.35 0.07 0.40 0.04 0.43 0.06 0.30 0.08 F = 5.07, p < .01
Weighted densitye 0.21 0.07 0.26 0.03 0.27 0.05 0.17 0.05 F = 6.65, p < .01
Total degree centralitye 0.22 0.07 0.27 0.03 0.28 0.05 0.29 0.05 F = 6.74, p < .01
Betweenness centralityf 0.02 0.00 0.03 0.00 0.05 0.01 0.03 0.00 F = 9.56, p < .001
Eigenvector centralitye 0.26 0.02 0.34 0.02 0.35 0.06 0.26 0.04 F = 6.99, p < .01
Clustering coefficiente 0.50 0.06 0.53 0.04 0.50 0.04 0.42 0.05 F = 7.92, p = .001
Average distanceg 2.68 0.66 2.28 0.18 2.60 0.27 2.99 0.31 F = 6.47, p < .01
Diffusionh 0.71 0.05 0.67 0.05 0.87 0.12 0.77 0.03 F = 9.31, p < .001

a Note that n for each unit shape varies as shown in the table. bWith two exceptions (node size and average distance) that are actual counts, metrics are measured on a 0–1 scale. cTwo compact circles with short connecting corridor containing elevators and support space. dIn one unit, one arm was shorter than the others. eCross different from compact square and racetrack. fCompact circle different from racetrack and cross, compact square different from racetrack. gCross different from compact square. hCompact square different from racetrack.