I have a dataset that contains a range of values. I have created a frequency distribution of the values, and have included the plot below. To my untrained eye, it appears that the frequency distribution plot illustrates two populations with approximately normal distribution: one larger population with mean at approximately 10^7 W, and a second, smaller population with mean approximately 2.5 x 10^8 W, with a threshold of 10^8 W that separates the two populations.

I am new to statistics and am having difficulty in searching for a statistical test that would be able to answer:

1) Can this dataset be described as a composite of two normally-distributed populations, with different means?;

2) Can this dataset be described as a composite of two non-normally distributed populations, with different means?

I have read of similar questions which recommend the Student's t-test. (For example, How to test for differences between two group means when the data is not normally distributed?; Separating two populations from the sample). I wonder if this is appropriate here, since I am not certain if my dataset is normally distributed, and since I have only one dataset from which I am trying to extract two populations. If you can suggest a statistical test that you feel would be appropriate to answer either of my questions, or advice, I would be grateful for your help! log-binned frequency distribution of data

  • $\begingroup$ Without access to your data, it is difficult to be able to answer for you. Can you provide sample data? $\endgroup$ – EngrStudent - Reinstate Monica Nov 2 '18 at 20:12

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.