Species abundance statistical comparison assistance

#1
Hi,
I am new to statistics and needing to apply a relevant statistical test to the following question:

Is the presence/abundance of species X related to the presence/abundance of species Y, within the time period (yearly intervals) 2010 - 2016.

I have already spent hours looking into this myself, but am not generally inclined towards statistical analysis (this is my first time trying to put it into practice) and therefore am seeking some guidance.

Data has been collated to reflect the following:

2010 2011 2012 2013 2014 2015 2016
species X 2 1 1 2 8 6 20
species Y 43 68 75 26 89 36 45
*Data figures represent presence/abundance numbers of species X and Y, sampled with respect to the yearly intervals being the dependent variable.

Thinking that calculation of the Pearson's Correlation Coefficient to determine whether there is any linear relationship would be the way to go, I undertook manual calculations to return an r value of -0.008465. I do not know if I can use this - does it represent a very strong negative linear relationship or is too far outside -1; if it does represent a relationship, what does it mean being so negative?

I am thinking of replacing species X with another more prevalent species so that distribution between groups isn't so comparatively skewed, however even so, my general question will remain the same...

Is the Pearson's correlation coefficient a good test to use for the above (bold) general research question?

Considering data, would a paired t-test be another appropriate option? - If so, do you have any tips?

Are there any other statistical analyses that you wold recommend for appropriate analysis of my research question?

My task asks me to use simple statistical analysis to interpret some collated data.

Your assistance would be most appreciated.

Thank you.
Tam_123
 

rogojel

TS Contributor
#2
Hi,
I would definitely look at scatterplots first, to get an idea of the link between the two species. Then I would look at the Spearman correlation - the Pearson only looks at a linear relationship while the Spearman correlation is much more general - it is enough to have a monotonous realtionshi like high Xs are linked to high Ys - so it would be more appropriate.

regards
 

gianmarco

TS Contributor
#3
I do not quite follow you when you say 'presence/abundance'...can you clarify that?
I know about presence/absence, or about 'abundance' (i.e., counts).