We now turn our attention to the problem of determining whether sample data is a fit for a given distribution. These are often referred to as Goodness of Fit Tests.
In Kolmogorov-Smirnov Test for Normality we show how to employ the KS Test to determine whether data are normally distributed with a known population mean and standard deviation. In Lilliefors Test, we perform a similar test when the population mean and standard deviation are not known but needed to be estimated from the sample. Also, in Goodness Of Fit, we show how to use a chi-square test to determine the goodness of fit when data is in the form of a frequency table.
We now provide additional information about these and other useful goodness of fit tests for a variety of distributions. We also describe how to carry out such tests using the Real Statistics Resource Pack.
Topics
- One-Sample Kolmogorov-Smirnov Test
- Two-Sample Kolmogorov-Smirnov Test
- One-Sample Anderson-Darling Test
- Two-Sample Anderson-Darling Test
- One-Sample Chi-square Goodness of Fit Test
- Goodness-of-fit for Benford’s Law
- Real Statistics Data Analysis Tools
- Goodness-of-fit using the Characteristic Function
References
Wikipedia (2017) Goodness of fit
https://en.wikipedia.org/wiki/Goodness_of_fit
Hi Charles, I wonder if you can help me choosing the correct statistical test – I have data for the amount of time an animal spent engaged in 10 different categories/behaviours (e.g. resting, eating, moving etc) before and after an enclosure change. However, the amount of data points gathered before the change is not the same as after the change, so I am looking at percentages or proportions of time animals spent engaged before and after. Which test would be the most suitable to use in this instance?
I also have similar data for another species, where I have the same data for two different individuals, and I want to see if there is a difference between the animals in how much of their time is spent resting, eating, moving etc. Would a G test be suitable here?
Thanks so so much in advance
Hello Rachel,
What hypothesis are you trying to test?
Charles
Which One Sample Goodness of Fit Test is the best choose for small samples (less than 10 subjects)? Is the Exact binomial good selection?
Loo,
It depends on the nature of your data, but for a small sample the exact binomial test could be a good choice.
Charles