Data analysis
Data analysis was carried out using Microsoft Excel (2019) and STATA
version 17 (StataCorp, College Station, Texas). Normality of variables
was tested using the Shapiro-Wilk test (Razali & Wah, 2011), alongside
visual inspection of histograms. Homogeneity of variance was tested
using Levene’s test (Nordstokke & Zumbo, 2010; Nordstokke et al.,
2011). The Shapiro-Wilk and Levene’s tests showed that the dependent
variables (adults/m2 and egg masses) did not follow a
normal distribution (p<.001) and had unequal variances
(p<.001). Therefore, the nonparametric Kruskal-Wallis H test
(Chan & Walmsley, 1997) (Vargha & Delaney, 1998) was used to determine
whether these variables differed across the different sections. A
pairwise post-hoc Dunn’s test with Bonferroni adjustments was done to
identify the sections which were significantly different, with respect
to the number of adults/m2 and egg masses. Only data
from Mwea was included in the analysis since no GAS were observed in all
the other schemes.