Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

Supported by

Reliability of Analyses

Thanks for creating JASP. It has been so helpful.

I've been noticing some inconsistencies (minor, but still) when I re-run the same analyses. I was double-checking some work, and there were differences. I then re-ran the same data from the same file, and got different results. Why might this be the case?

For example, using the same file and same parameters I got:

When I ran the exact same file another time I got:

While the big picture is the same for both of these analyses, it makes me nervous to have different results from the exact same data. Any idea what is going on here?


Thanks!

Comments

  • This particular analysis includes a numerical routine. This means that unless you take an infinite number of samples (which takes an infinite amount of time) the results will change a little bit from one run to the next. The extent to which the results change is indicated by the "error %" column. You can also go to advanced options to increase the numerical stability; however, some variability will remain.

    [Note to Sebastiaan: we fixed your name in the next version]

    Cheers,

    E.J.

  • Thanks! This is reassuring.

Sign In or Register to comment.