Hi all, I have a vegetation survey data set and I'm trying to determine if there are significant differences in responding diversity variables when comparing between different sites, and different positions on the same site. I also want to see if there are any interactions between site and position that result in significant differences. I've set up my code as such:

```
summary(aov(nISI~Site*pos, newPDD)) #nISI is a diversity metric calculated at a plot-by-plot level
```

Now when I run the aov with Site and Position as interacting effects I get this:

```
> summary(aov(nISI~Site*pos, newPDD))
Df Sum Sq Mean Sq F value Pr(>F)
Site 6 42.51 7.085 7.687 **2.31e-06** ***
pos 1 4.02 4.015 4.357 **0.0405** *
Site:pos 6 11.65 1.941 2.106 0.0633 .
Residuals 70 64.51 0.922
---
```

In this case, both site and position seem to be significant effects when compared independently. However when I compare nISI between positions using a single-factor ANOVA and a t-test:

```
summary(aov(nISI~pos, newPDD))
t.test(nISI~pos)
```

The output is:

```
> summary(aov(nISI~pos, newPDD))
Df Sum Sq Mean Sq F value Pr(>F)
pos 1 4.02 4.015 2.775 **0.0996** .
Residuals 82 118.67 1.447
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
> t.test(nISI~pos)
Welch Two Sample t-test
data: nISI by pos
t = 1.6657, df = 81.439, p-value = **0.09962**
alternative hypothesis: true difference in means is not equal to 0
95 percent confidence interval:
-0.08501103 0.95954683
sample estimates:
mean in group Ag field mean in group Stream
2.159406 1.722139
```

In this case, I get p-values which are different and *not* significant. Can anyone explain what is going on here? What is the correct value to be using in this case? Why are there different values for the aov output for the same factor?

Thanks