Quelle: Meine Analyse von 4M IRS Form 990-Einreichungen, verarbeitet aus digitalisierten XML-Rückgaben von IRS Exempt Organizations.
Tools: Python, Pandas für die Datenverarbeitung. Visualisierung erstellt mit rohem js.
Vollständiger Bericht und Ergebnisse: https://charitysense.com/insights/the-3-trillion-blind-spot
Von mtweak
6 Kommentare
I wonder what the breakdown is between “foundations” that provide grants to other nonprofits and actual nonprofit organizations that do have programs. I work at a nonprofit and I think you really have to examine individual nonprofits to see how their outcomes compare to their expenses and judge them on that basis. In aggregate there is too much lost detail to draw significant conclusions.
This is an advertisement for a “solution” but the “problem” as outlined is highly deceptive.
What library did you use to make that sankey ?
This is very interesting information to see and the write up put it into context well. One difficulty that was also implied by the write up, but never stated, is that transparency requires manpower that many nonprofits don’t have.
This is one of the real pluses to government provision or services (the main solution available for important services other than balancing inadequate funding/time with the utility of reporting)- government spending must be tracked and is able to be requested for those who want more info. And people are paid better wages with better hours to do that tracking.
Which number on the right means „this is where we did some good“? Member benefits? (I realize it’s probably not that simple to analyze.)
Here’s a link to OP’s previous attempt to post this, where comments noted that the analysis is terrible and also that **the whole thing is a veiled ad for a product that OP wants to sell you**. https://www.reddit.com/r/dataisbeautiful/comments/1rnmrqq/us_nonprofits_handle_3t_in_revenue_with_less/