Table 1. Results of Kruskall-Wallis tests comparing arthropod abundance among sites, habitats and surveys.

Too few samples were available for a composite analysis of habitats.

Analysis *	No. traps /samples	No. ind.	W / p	Effect size	Remarks: No. sites; habitats; surveys; no. of samples or protocol considered (no. of samples) *
Sites: STI	630	1956	57.4 / <0.001	0.397	9; 4; 1; 41–92 traps
Sites: FOG	48	10777	12.8 / 0.078	-	8; 2; 1; 6 samples
Sites: FIT	275	7077	44.1 / <0.001	0.389	5; 4; 1; 53–57 samples
Sites: composite	1314	40771	779.3 / <0.001	0.280	8; 4; 1; BEA(20), BER(47), LIT(6), PIT(4), STI(41)
Habitats: STI	1150	3683	219.3 / < 0.001	0.484	9; 4; 4; 34–598 traps per habitat
Habitats: FIT	814	18092	19.5 / < 0.001	0.298	5; 4; 4; 6–535 samples per habitat
Surveys: FIT	814	18092	198.3 / < 0.001	0.641	5; 4; 4; 144–275 samples per survey
Surveys: LIT	96	14549	5.8 / 0.124	-	8; 2; 4; 12–48 samples per survey
Surveys: PIT	193	1288	23.4 / < 0.001	0.489	8; 1; 4; 27–95 samples per survey
Surveys: composite	403	39346	342.1 / < 0.001	0.507	9; 4; 4; FIT(12**), LIT(12), PIT(27), STI(8**)

* Codes of protocols: BEA = Beating, BER = Berlese-Tullgren, FITs = Flight-intercept traps, FOG = Fogging, LITs = Light traps, PITs = Pitfall traps, STIs = Sticky traps

** pooled by vertical transect at each site.