The analysis of the Newcomer Tasks experiment (T230174) was complicated by the Variant A/B test (T238888). It also did not incorporate important developments such as Guidance. We therefore want to revisit it using a dataset of registrations from a period where Guidance was available and the feature set was stable. In this analysis we'll focus on the four main questions:
- Activation
- Retention
- Edit volume
- Revert rate
For all of these, we want to examine effects overall and for the Article & Article talk namespaces specifically. For edit volume and revert rate, we'll also want to limit it to the same time periods used for activation and retention, just like we did in the previous analysis.
The main goal of this analysis is to get reasonable estimates of the effect of the Homepage with Guidance during a period where features were stable. hopefully confirming the findings from NEWTEA.