Is my data Parametric or non parametric?

Question

hi there,
i am doing a individual projects for one of my uni subjects and we have to come up with our own ideas of an experiment we can do, now i have chosen to do: The effect that environmental pollution has an effect of the biodiversity in a stream, i am using the kick sampling method and comparing the number of species found vs the level of pollution,

would my data be Parametric or non parametric?
sorry if this makes no sense, i have no clue when it comes to statistics!

cheers in advance!

Answer 1

I dunno, I guess your data will be nonparametric since you won't be using a statistical sampling plan and calculating averages and standard deviations. The occurence of a species even once in one sample will count as much as the occurence of some other species in every sample at the same locality.



What you'll be aiming for is a graph showing amount of pollution (however you measure it) as the independent variable (x-axis) and species richness as the dependent variable (y-axis). Sampling at each locality will produce one data point on the graph. If the plot line is straight, you might try calculating a correlation coefficient on a calculator. If your plot line isn't straight, you might try straightening it out by transforming the species richness data to logs and then doing the calculation. Or instead you might try plotting the untransformed data on semi-log paper.



Sampling always brings up a lot of questions. If you want to email me at my Y!/A address, that would be OK.

Answer 2

Parametric assumes the underlying distribution of the education has some style. it must be known, exponential or different countless distributions. Non-parametric do no longer make assumptions on the danger distribution of documents. a)b) and c) If the distribution of often occurring heights (or the different documents) are known or another universal style (you want a statistical try or a seen graph of the education), then it relatively is parametric ; in any different case, it relatively is non-parametric.

Answer 3

You have to look at the distribution of your data, if the distribution approximates a normal distribution use parametric tests, if not non-parametric. Ask somebody who knows at least a bit about stats to help you.

Answer 4

It is not the data that are parametric or non-parametric but the statistical routine that you use for analysis. There are several dozen statistical routines that you can use. Most, I think, are non-parametric.