Contribution measurement

As a group writes a document or builds a deck, Dwixel attributes each contribution to the person who made it. The result is a live, per-member view of who is doing what, drawn from the work rather than from anyone’s opinion of it.

Measured from real work, not a survey

Nobody fills in a contribution form. The signal comes from the editing itself: authored text, edits and revisions, comments, and structural work, each attributed to a member as it happens. This matters because edit history can legitimately ground an assessment of participation, provided it is interpreted rather than blindly counted. ¹

Quality over volume

Counting edits rewards the wrong thing. Research on collaborative writing found that what reflects real, accepted contribution is the content that survives revision, not the number of edits a person made. ² Dwixel is built on that principle:

• ·Large pasted bursts are detected and de-credited, so dropping in a block of text (or AI output) does not register as authoring it.
• ·Contribution leans toward authored text that survives later revision.
• ·Editing and curation are counted as their own kind of work, not mistaken for fresh writing.
• ·Comments are weighted by substance rather than counted one for one.

Deterministic and explainable

There is no opaque model deciding a student’s share. The scoring rules are fixed and inspectable, which is the only honest basis for anything that touches a grade. It also addresses a documented risk of trace-based metrics: that the inferences drawn from them can be invalid if the method is a black box. ³

Students see their own data

Every student can see their own contribution profile and how it is calculated. Measurement is something done with the group, not to it, and the transparency is part of what makes it fair. Contribution is presented as a profile with the attributed history behind it, in the spirit of tools that show who did what over time rather than reducing it to one unexplained number. ⁴