prefacing this with the fact I have a concussion and may or may not be making sense/thinking clearly at the moment. I'm working on rewriting an assignment that asks us to find significant differences between groups using ANOVA and then TukeyHSD as post-hoc analysis. When I originally did the analysis, I merely found that there were differences between groups and moved on; but apparently I am supposed to "specify the direction for each pairwise comparison". They're looking for the format of "food insecurity for males > females". My main question here is how do I do that? My understanding is that ANOVA checks if there is a difference between groups but does not give directional information. Should I be doing another analysis (which I don't think is the case) or am I missing something in the output?
I'm exploring using gender as a confounding factor for an insecurity score on a survey given to county residents. An Anova comparison showed a significant p-value, so I moved onto Tukey. But how am I supposed to figure out who has the larger insecurity score in the pairings from this? BTW The numbered groups are representative of a gender category (male, female, agender/non-binary)
TukeyHSD(aov(SHAPE_survey$Insecurity2~factor(SHAPE_survey$Gender3)))
Tukey multiple comparisons of means
95% family-wise confidence level
Fit: aov(formula = SHAPE_survey$Insecurity2 ~ factor(SHAPE_survey$Gender3))
$`factor(SHAPE_survey$Gender3)`
diff lwr upr p adj
2-1 -0.1294355 -0.4685115 0.2096405 0.6435152
3-1 1.3269409 0.1882313 2.4656506 0.0173732
3-2 1.4563765 0.3369656 2.5757873 0.0065052