# statistical test

Im working on an experiment for my A level course (UK) that will determine the effects of different respiratory substrates on yeast population growth. Im using 4 sugars; fructose, glucose, sucrose and lactose. Using a haemocytometer I'll find the size of the yeast population before the sugar is added and the growth 1 week after.

I have a choice of 3 statistical tests to choose from: the standard error test; the chi-squared test; and the Mann-Whitney U test. The standard error seems to be the best one to me, but i thought it could only be used for large samples ( at least above 10)?

https://brainmass.com/statistics/normal-distribution/statistical-test-experiments-13927

## SOLUTION This solution is **FREE** courtesy of BrainMass!

You can use the standard error test. The standard error test may be used for small sample sizes (less than 10) . What you have to do is to use t distribution instead of normal distribution (i.e. look up t values in Student's t distribution table instead of z value in normal distribution table).

You have not indicated how many samples you are taking. You can increase the number of samples for better statistical inferences.

Â© BrainMass Inc. brainmass.com December 24, 2021, 4:53 pm ad1c9bdddf>https://brainmass.com/statistics/normal-distribution/statistical-test-experiments-13927