One person built this, and the tests back the numbers, not the byline

Christopher Campbell built this site, the model, and the documents it links to. He has no institutional affiliation, no co-author, and no organization behind the work. That is exactly why the disclosure below, and the model and data pages, exist: there is no institution’s name to lend the numbers credibility, so every number has to stand on being checkable instead.

The worked example running through every finding on this site (the $201,000 payor, the $570-a-week other parent, the three children) is my own child support order, computed under Massachusetts's 2025 worksheet. I am telling you this before you read a single finding because a reader should reasonably wonder whether an interested party has shaded a number in his own favor, and the honest answer to that is to let you check it rather than to ask you to trust me. The worksheet that produced these figures is at model/worksheet.py, checked against the official form's own calculation scripts on the model page. Run it on any numbers you choose, not just mine.

The project makes three checkable claims about how the worksheet computes

Each is checkable independently of the other two and of the author’s own case:

  1. The worksheet’s Section IV.C hardship test is computed on gross-derived income while the obligation is paid from net income, so it reports a lower percentage than the payor’s actual burden.
  2. Line 6b allocates child care on each parent’s income share measured before the base support order has already moved money between the two households: one calculation, two different income measures.
  3. The credit the worksheet gives for equal parenting time is the difference in two income shares with no parenting-time term in it, so the credit shrinks as the gap between the parents’ incomes widens rather than staying fixed to the time actually shared.

Each is laid out with its own figures and its own source files on the findings pages. All three are about how the worksheet computes, not about what any resulting number ought to be.

The project does not claim any order is wrong, or that Massachusetts is the worst state

License

Code is released under the MIT license (LICENSE-CODE.txt). The documents and figures are released under CC BY 4.0. A checkable error, reported as an issue or pull request against the repository, will be corrected and credited.

Check it yourself

Nothing on this page needs to be taken on trust: the code, the data, and the documents are all linked from here.