Stop and take this 2 question survey about your (2021) team finances

FIRST General Forum
1

TL;DR: have someone on your team fill out Part A (this part is 2 questions long). If possible, also fill out Part B. Each part should be filled out only once per team.

I’ve been thinking a lot about @saikiranra’s excellent sustainability report. Team financial health came up in that report in a number of places, yet the FRC community has little-to-no data on this topic. These surveys about team finances will help us get the data we need to paint a clearer sustainability picture. Each survey should be filled out only once per team.

  • 2021 Financial Health Survey, Part A: 2 questions long, estimated 3 mins to complete. Sensitive opinion questions. Uses a randomized response procedure to maximize privacy and ensure plausible deniability. This makes it so I can only learn information about teams in aggregate.
  • 2021 Financial Health Survey, Part B: 18 questions long, estimated 30 mins to complete. Designed to capture information similar to what would recorded on an IRS Form 990 (flied by US-based 501(c)(3) nonprofit organizations) or a CRA Form T3010 (filed by registered charities in Canada). Because these forms are public records, this survey does not use a special response procedure to maximize privacy or ensure plausible deniability.

This survey is for the 2021 year. Why are you collecting this information now?
Many organizations use a January-December fiscal year and are filing this information with the IRS/CRA in the coming weeks and months.

Will you make a survey for 2022 team finances?
Towards the end of 2022/beginning of 2023, I intend to release a second set of surveys that focus on 2022.

Will you share survey results with the community?
Assuming sufficient response rates, I will prepare a report and share it on Chief Delphi. Survey data may be released in aggregate and/or anonymized form.

What if my team doesn’t file IRS Form 990/CRA Form T3010?
I recognize that not every team backed by a US 501(c)(3) actually files these forms with the IRS (there’s a revenue exemption that applies to most US FRC teams), so Part B may be impossible to fill out. Similar exemptions may exist in Canada. Teams that don’t have to file a Form 990 or T3010 may feel uncomfortable responding to some of the questions, and that’s OK. If any of these apply, just don’t fill out Part B.

What is the randomized response method you use in Part A?
To answer a question on Part A, toss a coin (if you don’t have a physical coin, you can use this link) and do not share the result. If heads, answer yes. If tails, answer truthfully.

How does the randomized response method you use make my answers more private?
In Part A, all questions are phrased such that “yes” answers are more sensitive. Because only you know if the coin was heads or tails, someone (like me) who sees a “yes” answer does not know if you answered because it was heads (i.e., you were required via the response procedure) or if it was tails (i.e., a truthful response). In aggregate, the randomness can be statistically corrected to estimate the true proportion of yes/no responses.

4 Likes
2

One would think survey like this would be far more meaningful if it comes from HQ or the Regional Program Delivery Partners. Don’t get me wrong, having the information is great, but what’s CD going to do about it? Just started another long debate? Getting information is great but if there’s nothing anyone can do about improving things from such data, gathering such data becomes just an academic exercise.

For clarifications, EVERY 501.3c has to file the 990, they might be able to file the 990N if their revenue is low enough. BUT, every legal 501.3c that wish to remain legal would at a minimal have to file a financial report to their local State as well as present that to the Board of Directors in their ‘annual BoD meeting’ (details are very state specific).

1 Like
3

I strongly support this effort Brennan. Filled out both parts for our team. I would be glad to hear if there any other ways I can support this project.

Some thoughts:

  • Our organization encompasses more than just our FRC program. (Also, the finances associated with our programs are not cleanly separable – FRC is a “profit center” for us in the sense that it was associated with 97% of our revenues in FY21 but only 58% of our expenses.) It wasn’t quite clear to me whether the intent was to capture the financial data for our FRC program, or for our organization as a whole.
  • A box for “Additional sources of revenue” might be worth adding to the income page on Part B. Interest we earn on our existing assets is a small but slowly growing source of revenue for us-- to the point that we are now our own ~10th largest sponsor.
4 Likes
4

Done and done. Great work putting this together.

2 Likes
5

If you have the ear of an HQ or district executive, please encourage them to reach out to me. I would love to cast a wider net.

I disagree with your assumption. I believe that collecting and analyzing this data will help the robotics community.

Please share this survey with other teams you have contact with. Fortunately/unfortunately, the randomized response procedure used in Part A generates a lot of statistical noise and higher margins of error. More samples helps to correct for this.

Short answer: Part B is optimized for organizations that host a single FRC team because I believe this to be the more common situation.

Longer answer: At the last minute, I decided to get rid of the questions that handled this to simplify Part B (and hopefully increase the response rate). In addition to teams like yours that have more programs than just FRC, there are some schools/organizations that have more than one FRC team (e.g., 1241/1285). I can look into adding these questions back in for next cycle.

1 Like
6

Thanks for doing this. Will it mess up your data if people enter truthfully without flipping a coin? (I didn’t do this)

7

Oh, I just kept flipping until I got the result I wanted. :upside_down_face:

2 Likes
8

Yes, it will mess stuff up (both mean and standard error). Don’t do it. That’s why I make you affirm you understand the procedure and commit to follow it.

There are similar procedures that are more resistant to these kinds of errors (only affects standard error), but they are more complex. They require two coin flips per question.

9

I don’t think very many teams are actually 501.c3 orgs, the majority in my area exist as single accounts either within schools or within other sponsoring nonprofits…

3 Likes
10

Agreed, it’s more common among the average team on CD than in the FRC community as a whole. Of the three teams I’ve been a part of none had a 501c3 at the time and these are all teams that have been sustained for over 10 years.

Appreciate the survey, I will be trying to complete part B when I have a bit of time to look up the information.

11

I’m not much for statistics, but the thing that stands out in my mind is that 2021 is an unusual year since everything was remote. I know my answers are unusual because of everything being remote.

Does this affect the value of the survey over time, i.e., if you compare future years if the survey continues?

12

Update:

  • 58 responses so far for Part A; 17 for Part B
    • More data needed: please continue to respond to Part A and Part B of the survey.
  • 95% confidence intervals
    • [0%, 3%] considered their team financially unstable in 2021
    • [0%, 17%] had a parent or mentor “float” a team expense due to a lack of funds in 2021
  • Highs and lows
    • $500-$61,000 in fund balances carried over from 2020
    • 4-135 students
    • $0-$45,000 in membership fees
    • $0-$25,000 in in-kind donations
    • $0-$62,000 in monetary gfits/grants/donations
    • $0-$12,000 in fundraising
    • $0-$11,000 in stipends for teachers/mentors

Part B
If you’ve filled out Part A but not filled out Part B, please consider doing so. I probably should have communicated this better in my initial post, but Part B is also available/intended for teams that are run through a school.

Confidence Intervals
Because of the low sample size and the proportions being close to 0, a Clopper-Pearson interval was used instead of approximating via the normal distribution (what they teach you to do in AP Statistics).

1 Like
13

2021 was definitely an unusual year. I think you’re right in that we’ll probably need data from a few “normal” years to understand precisely how unusual it was.

But in my view, this strengthens (rather than diminishes) the value of the survey. Without capturing this unusual data, we won’t be able to understand how it was unusual from a financial perspective. Did teams fundraise less because things were remote? Did decreased expenses outweigh potential decreases in fundraising? There are some interesting questions here.

2 Likes
14

This topic was automatically closed 365 days after the last reply. New replies are no longer allowed.