4

I am working on a farming experiment and I have two batches of chickens that are growing in a side-by-side trial. The difference between them is how they are fed. I need to select some of the chickens from each batch to send in for nutritional analysis of their meat.

How do I know how many chickens from each batch to send into the lab so that the sample sent is indicative of the entire batch? I was told to look at and report p-value, but I am not really a statistics guy.

Stephan Kolassa
  • 95,027
  • 13
  • 197
  • 357
Randy
  • 141
  • 2

1 Answers1

2

This is a straightforward question of determination based on a .

Essentially, you need to specify the alpha level you will use (typically 0.05), the power you wish (often 0.80), and a relevant effect size. Given these three input data and the assumption that you will calculate plain vanilla unpaired t tests, there is a number of sample size calculators online that will tell you the required $N$. If you have a more complex model, e.g., a regression with covariates, it is usually simplest to simulate.

What effect size to input into these calculations is usually the hardest decision. Often, people will look through the literature and use previously reported effect sizes, but because of publication bias, this (a) typically over-estimates the true population effect size, and (b) only accounts for statistical significance, not clinical significance. In other words, this approach may give you decent power to detect an effect size that is too large for the use that you want to put your analysis to, or too small.

The best recommendation is to use an effect size that you would be sorry to miss. For your poultry, changing the feed for all chickens in case your experiment comes out positive will likely entail some costs (either switchover costs, or the new feed may simply be more expensive), so you would require some minimum effect in the outcome, i.e., the nutritional value of the chicken. This minimum effect is the one you should power your analysis for.

Here is a very good earlier answer to a similar question.

If you do plan on using a complex model and are not a statistician, I recommend you think about retaining a statistical consultant. This may be cheaper than an overpowered study. And less frustrating than an underpowered one.

Stephan Kolassa
  • 95,027
  • 13
  • 197
  • 357
  • What is the relevant effect size? – Randy Jun 08 '18 at 20:45
  • Good point. I edited the post to add two paragraphs on the effect size. – Stephan Kolassa Jun 09 '18 at 06:07
  • 'The best recommendation is to use an effect size that you would be sorry to miss', very well put! – ReneBt Jun 09 '18 at 06:18
  • Maybe I should have instead asked for a link to a website or something that could explain more about my current experiment. I don't understand most of the terms you are using, and the link to the earlier answer has a lot of missing images. – Randy Jun 14 '18 at 03:35
  • I used http://www.anzmtg.org/stats/PowerCalculator/PowerTTest to try this out. I put in a power of 0.8 (a "large effect size" according to the help), a significance level of 0.05 (is that p-value?), and a sample type of "Two Samples" and alternative test of "One Sided". That calculated out to a sample size of 20.03. So does that mean I need 10 chickens from each batch? – Randy Jun 14 '18 at 03:39
  • 1
    No, the calculated sample size refers to each group, so you will need 20 chickens from each batch. – Stephan Kolassa Jun 14 '18 at 13:35