Howdy, Stranger!

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

Supported by

Calculation of 95% CI and splitting by a variable

edited April 2017 in JASP & BayesFactor
  1. In Descriptives->Descriptive statistics->Statistics, there is no option for calculation of 95% CI. JASP can calculate mean and SD, but then I must use Excel or calculate it manually.
  2. I wanted to calculate the mean, 95%CI, SD etc of a dependent variable according to the values of an independent variable (e.g. gender). I didn't want to run any test, just to report the mean, 95%CI and SD separately for the males and females. There is no such option in Descriptive statistics. In plots, there is an option to create a boxplot after splitting by gender. Make this option (to split by a variable) more widely available.
  3. I tried t statistics. In t statistics there is an option to calculate 95%CI. However, 95%CI still does not appear in the table (VAR8 is the gender, and VAR25_1 is the dependent variable). It seems that this option is only for the "t value". I selected Descriptives plots and only there can I see the 95% CI, but it is only visual and the exact values are not available.

Comments

  • Good points. The 95% CI is an inferential measure and we didn't feel it belongs in descriptives. But I'll discuss this with the team. Making the "split by" option more general is highly desirable, we agree.
    E.J.

  • Well, on second thoughts, I can understand why you feel that it doesn't belong in descriptives (because it is often calculated using the student t distribution). However, I believe that there should be at least an option in t test to display the confidence interval, as I mentioned before (and not only visually in box plot).
    Thanks for your quick reply.

  • Thanks. We are working on it!
    E.J.

Sign In or Register to comment.