Milk was sampled from lactating dairy cows on 2 occasions in early lactation. Milks were cultured and identified from each cow from each cow on 2 days. Outcome of culture was classified as positive (growth of bacteria) or negative (no growth). Question was whether frequency of growth was equal on the 2 days.
You can use a McNemar test too which will compare the culture results (Y/N or 0/1) of the same animals. I'm a user of R.
The command is very simple in base R:
mcnemar.test(table(test$day1, test$day2))
# where the dataset is the contingency table of the cows test over the 2 days (eg ++ / +-/-+/--)
The null hypothesis is that the marginal probabilities for each outcome is the same (ie the test result does not change over time)
Basic tutorial and explanation of McNemar test can be found in https://datatricks.co.uk/mcnemars-test-in-r#:~:text=What%20is%20a%20McNemar's%20test,nominal%20instead%20of%20interval%20variables.
@Sebastien Buczinski... Can we use chi square test for the same set of data.
McNemar is indicated because the data are paired (same cows sampled twice) which is different from the Chi-square hypothesis
- Badges
- Science topic
- Similar topics
- Philosophy
- Esthetics
- Interpretation