I have data from an ecological study on wasp abundance and diversity across a forest which I am trying to analyse appropriately.

We have abundance data from ten locations that were sampled twice, three months apart.

These ten sites can then be split by their vegetation,into two vegetation types:

  • Four sites = conifers dominant.
  • Six sites = angiosperms dominant.

I am trying to assess differences in wasp abundance between sampling periods and vegetation type. In particular, the between vegetation type analysis is most interesting to me but I do not want to have to pool the data from each sampling period whilst at the same accounting for temporal pseudo-replication.

I thought a repeated-measure ANOVA or mixed model may be appropriate given the nature of these data.

TIME is a within-subject factor and VEG TYPE is a between subjects factor.

However, as my sample size is small (n=10 per sampling period and n=4 and n=6 for each vegetation type), I am unsure whether to proceed down this way or look for a less complex analysis.

So far, I have applied a repeated measure ANOVA...

To help clarify the model I have created:

  1. ABUNDANCE = wasp abundance at each location - this was a count of number of wasps trapped.
  2. TIME = I have treated this as the repeated measure factor, we trapped the wasps at same location twice, three months apart.
  3. VEG = The vegetation community at each trapping location, either 1 = angiosperm type or 2 = conifer type
  4. SITE = Site identifier for each trap location 1:10.

I used the following code:



Error: SITE
          Df Sum Sq Mean Sq F value Pr(>F)  
VEG        1  4.676   4.676   6.379 0.0355 *
Residuals  8  5.864   0.733                 
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

          Df Sum Sq Mean Sq F value Pr(>F)  
TIME       1 0.7423  0.7423   7.237 0.0248 *
Residuals  9 0.9231  0.1026                 
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

WASP.ABUN=structure(list(SITE = structure(c(1L, 1L, 2L, 2L, 3L, 3L, 4L, 
4L, 5L, 5L, 6L, 6L, 7L, 7L, 8L, 8L, 9L, 9L, 10L, 10L), .Label = c("1", 
"2", "3", "4", "5", "6", "7", "8", "9", "10"), class = "factor"), 
    TIME = structure(c(1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 
    1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L), .Label = c("1", 
    "2"), class = "factor"), VEG = structure(c(1L, 1L, 2L, 2L, 
    2L, 2L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 2L, 2L, 1L, 1L, 2L, 
    2L), .Label = c("1", "2"), class = "factor"), iCHNA = c(8L, 
    4L, 5L, 4L, 5L, 2L, 9L, 4L, 17L, 4L, 17L, 13L, 2L, 3L, 0L, 
    0L, 10L, 6L, 2L, 2L)), .Names = c("SITE", "TIME", "VEG", 
"ABUNDANCE"), class = "data.frame", row.names = c(NA, -20L))
  • $\begingroup$ It is better now. However a few terms are not clear. pl. explain sampling period and WASPAV Site/Time $\endgroup$ – Subhash C. Davar May 30 '16 at 9:27
  • $\begingroup$ Sampling period = WASPAV$TIME; we sampled the wasps, trapped them, at the same locations twice. For Site/Time do you mean Error(Site/Time)? From what I'd seen online of other repeated ANOVAs this was the correct assignment of the error term for my data. Site is identifier of each trap location. $\endgroup$ – Liam May 30 '16 at 9:35
  • 1
    $\begingroup$ What is unclear? I believe I have explained everything properly $\endgroup$ – Liam May 31 '16 at 2:40
  • 1
    $\begingroup$ It may be useful to roll up some of the information added in comments into the main question with an edit (e.g. that "abundance" is a count of the wasps caught at each site); then I think it's clear enough. Thanks for your efforts @subhashc.davar! $\endgroup$ – Scortchi May 31 '16 at 12:33
  • 2
    $\begingroup$ Thank you for the advice @Scortchi, I have rolled up some of the info from the comments into the main question. And, thank you for all your help @subhashc.davar! $\endgroup$ – Liam May 31 '16 at 23:19

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.