I have multiple datasets (from different sites/ samples) showing decreasing hydrocarbon (HC) concentration in soil over time (as the soil is treated). Each data set has different time intervals (i.e. one data set has recordings of HC concentration every week for 10 weeks, while another has recordings of HC concentration every ~3 months for 8 years) and different starting and finishing HC concentrations (i.e. one data set starts at 8470 ppm TPH while another starts at 650 ppm TPH).
I want to statistically compare the percentage reduction in HC concentration over time, to find which site is most efficient at hydrocarbon bioremediation. However, I am unsure of which test/ tests would be best to do this.
The data sets are either non-normally distributed (Anderson-Darling, p
To my mind, Student's t-test should be the most appropriate.
Student's t-Test is one of the most commonly used techniques for testing a hypothesis on the basis of a difference between sample means. Explained in layman's terms, the t-test determines a probability that two populations are the same with respect to the variable tested.
A very simple example: Let’s say you have a cold and you try a naturopathic remedy. Your cold lasts a couple of days. The next time you have a cold, you buy an over-the-counter pharmaceutical and the cold lasts a week. You survey your friends and they all tell you that their colds were of a shorter duration (an average of 3 days) when they took the homoeopathic remedy. What you really want to know is, are these results repeatable? A t-test can tell you by comparing the means of the two groups and letting you know the probability of those results happening by chance.
Dear Tom Hallam,
There are various ways of testing the Differences between the two sets of data. However, for unbalanced and different time frame data can be tested with mean/proportion differences, and t- or z-test. In this context, I give below some links to help you to meet your requirements of studies.
https://dataz4s.com/statistics/comparing-two-proportions/
https://courses.lumenlearning.com/introstats1/chapter/comparing-two-independent-population-proportions/
(in the second link please Go ONLY with the Example, since there is a small error in the formula of The pooled proportion calculation)
http://www.quantitativeskills.com/sisa/statistics/t-test.htm
https://stattrek.com/hypothesis-test/difference-in-proportions.aspx
Hope the above will help you. Good luck.
Regards.
Percentage values should first be converted into arcsine transformation to conform to normal distribution of the data. Once converted, for a comparison of two means a t-test is the best fit for testing statistical differences.
I can solve your problem just send to me your row data, and before and statistical analysis if your data with percentage value suppose transfer them by Angular transformation using special tables
contact me for your case
Thank you Bijay Singh , Samithamby Senthilnathan ,and Anoop Kumar Srivastava for your help!
I have one concern with using the t-test, my data is non-normally distributed so would a t-test be appropriate? as to my knowledge t-test is a parametric test and assumes normally distributed data?
Are there any non-parametric alternatives to the t-test that would be more appropriate for my data?
Thank you Alisheikh Atta and Duraid K. Al-Taey , I think I will try arcsine / angular transformation to normalise my data followed by a t-test to compare the statistical differences.
- Badges
- Science topic