Ho1 : no
difference between location in terms of bites/hr (A
= B)
Ho2 : no difference between
repellents in terms of bites/hr (C
= D = N)
Ho3 : no interaction between these
main effects in terms of bites/hr
source of
sum of degrees of mean
F ratio P
variation
squares freedom
square
~~~~~~~~
~~~~~~ ~~~~~~~~~ ~~~~~~~
~~~~~~ ~~~~~~~~~
all groups
330
5
location
202
1 202
22.96 <<0.01
repellent
112
2 56
6.37 <0.01
LxR inter
16
2
8 0.91
>0.05 ns
residual
475
54
8.796
Total
805
59
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Interaction
F ratio = MS(inter)÷
MS(resid) = 0.91
F(0.05, 2,54df) >3.15 .... unable
to reject Ho
Location F ratio = MS(locat)÷ MS(resid)
= 22.96
F(0.01, 1,54df) >7.08
.... reject Ho
Repellent F ratio = MS(repel)÷ MS(resid)
= 6.37
F(0.01, 2,54df) >4.98
.... reject Ho
Interaction. Very little interaction was observed (not
stat.sig), so results will be reported simply for main effects without any
important interaction between them. F: 0.91, 2,54 df.
P>0.05
Location. Strong evidence that bites/hr were recorded a much higher rate for
wetlands (mean: 38.7) than for forest areas (mean: 14.0). Difference
is statistically significant at 0.01 and the null hypothesis can be
rejected. F: 22.96, 1,54 df. P<0.01
Repellent.
Strong evidence that the three materials did not have the same effect. The
control group produced the highest bites/hr (mean: 42.5), roughly three
times higher than the "DEET" group (mean 16). The "new" material
was positions between the two (mean: 20.5) Difference is statistically
significant at 0.01 and the null hypothesis can be rejected. F: 6.37,
2,54 df. P<0.01