the community site for and by developmental biologists

4 thoughts on “Make a difference: the alternative for p-values”

  1. I think the things mentioned are good if we have paired data. If the data is not dependent, then we cannot do any of the things mentioned. Anyway, it was a good read. Thanks for the post

  2. In case of paired data, it is possible to directly calculate the difference. Here, for independent data, the difference between the means (or medians) of two samples is calculated. This can be done (and is straightforward) – it’s a bit more complex if one wants to calculate the precision/uncertainty

  3. Very cool. I came up with the same approach and now am applying it for writing a scientific article. Yet, I wonder why this approach has not been widely recognized/used till now…
    I guess there is no very famous name on this, like “t-test”, and very few references can be found. Correct me if wrong. I would appreciate if anyone can share the first article that suggested this approach (or, is this blog the first one?).

Leave a Reply

Your email address will not be published. Required fields are marked *