The JASP team is in the final stages of testing the upcoming version, which is scheduled for release next week. The upcoming version, 0.8.3., contains many new features. My personal favorites:
- JASP 0.8.3 includes frequentist logistic regression (hat tip to the hmeasure R package for the confusion matrix diagnostics; Anagnostopoulos, Hand, & Adams, 2012). Kudos to Erik-Jan van Kesteren for implementing an attractive GUI and a comprehensive set of analyses and plots. Figure 1 below gives an impression:
Figure 1. Screenshot of example output from frequentist logistic regression in JASP 0.8.3. On the right, the squared Pearson residuals plot allows users to check for scalar overdispersion; this is a cool feature that is missing in many other software packages (thanks to Dr. Dan Gillen from UC Irvine for attending JASPer Alex Etz to this option).
- JASP 0.8.3 includes frequentist hierarchical model testing, both for linear regression and for logistic regression. Thanks to the JASP users who repeatedly requested this feature on GitHub!
- JASP 0.8.3 includes progress bars for Bayesian ANOVAs and regression. Bayesian repeated measures ANOVA can take a long time, and the progress bar will tell you whether to write a quick Email, go for a cup of coffee, or –in extreme cases– watch an episode of your favorite sitcom. Progress bars were on my long-term wish list, but Tim de Jong produced the goods much sooner than I could have hoped.
- JASP 0.8.3 has an extended Preference menu that allows users to (1) specify missing values; (2) specify the number of decimals displayed in the tables.
Figure 2. The revamped Preference window in JASP 0.8.3. The left panel shows how to indicate missing values, and the right panel shows how to indicate the number of decimals to display.
- JASP 0.8.3 has effect sizes for post-hoc tests in ANOVA, confidence intervals for rank-based correlations, and the ability to change the y-axis label for the descriptives plot of repeated-measures ANOVA.
Figure 3. In JASP 0.8.3, change the y-axis for repeated-measures ANOVA at will.
- JASP 0.8.3 is leaner than before — the current installation is 230 MB. Future versions promise to be leaner still.
We are excited about the upcoming version.The network module needs a few more improvements before we are completely happy with it, and we plan to bring it out in an interim release.
Anagnostopoulos, C., Hand, D. J.,& Adams N. M. (2012). Measuring classification performance: The hmeasure package.
Like this post?
Subscribe to our newsletter to receive regular updates about JASP including our latest blog posts, JASP articles, example analyses, new features, interviews with team members, and more! You can unsubscribe at any time.