Unsubstantiated claims on Research & Impact page
The Research & Impact page contains the following 5 claims:

Goal: I tried my best to track down the scientific evidence for these claims by reviewing the four articles linked on the Research page:

Results: I was only able to find evidence for 1 of these 5 claims.
Recommendation: Please cite sources and/or revise this page to avoid misinformation.
**********************************************************************************
1) What is the evidence for the claim that "90% of people feel better after talking to listeners"?
My guess: Unable to find in articles linked on Research page. (It's completely possible that it's there and I missed it.)
2) What is the evidence for the claim that "97% of people view their listener positively"?
My guess: Egregiously inaccurate claim based on a misreading of the following research:

Baumel A, Schueller SM. Adjusting an Available Online Peer Support Platform in a Program to Supplement the Treatment of Perinatal Depression and Anxiety. JMIR Ment Health 2016;3(1):e11, https://doi.org/10.2196/mental.5335. URL: https://mental.jmir.org/2016/1/e11
Issues with this claim:
- Proportion of 97% is based on item-level responses (N=36) to a four-question section of a survey rather than participant-level responses (N=9)

- Subject sample is very small (N=9) and non-representative (i.e. only women with perinatal depression or anxiety)
- Listener sample is non-representative (i.e. only verified listeners who completed the perinatal mood disorder training)

3) What is the evidence for the claim that "80% of people believe listeners can help people with mental health issues"?
My guess: Misleading claim based on the following research:

Amit Baumel, Christoph U. Correll, Michael Birnbaum. Adaptation of a peer based online emotional support program as an adjunct to treatment for people with schizophrenia-spectrum disorders. Internet Interventions. Volume 4, Part 1, 2016, Pages 35-42, ISSN 2214-7829, https://doi.org/10.1016/j.invent.2016.03.003. URL: https://www.sciencedirect.com/science/article/pii/S2214782915300282
Issues with this claim:
- Subject sample is very small (N=10) and non-representative (i.e. only hospital outpatients diagnosed with schizophrenia-spectrum disorders)
- Listener sample is highly non-representative (i.e. only listeners with at least 2 months of experience, successful completion of NAMI training module, at least 20 reviews, and an average rating of at least 4.5)

4) What is the evidence for the claim that "81% of users consider 7 Cups as a helpful service"?
My guess: Unable to find in articles linked on Research page. (It's completely possible that it's there and I missed it.)
5) What is the evidence for the claim that "70% of users feel support provided by 7 Cups listeners is just as or more helpful than that provided by psychotherapy"?
My guess: Reasonable claim based on the following research:

Amit Baumel (2015) Online emotional support delivered by trained volunteers: users’ satisfaction and their perception of the service compared to psychotherapy, Journal of Mental Health, 24:5, 313-320, https://dx.doi.org/10.3109/09638237.2015.1079308. URL: https://www.researchgate.net/publication/283050936_Online_emotional_support_delivered_by_trained_volunteers_Users_satisfaction_and_their_perception_of_the_service_compared_to_psychotherapy
Info about this study:
- Survey of N=866 7 Cups members (from convenience sample of N=2900 who saw a banner displayed on the 7 Cups community page)

@QuietMagic Yes, it's interesting 
My feeling about it is that the information is presented for advertising purposes, not in order to advance scientific knowledge, so the standards of truthiness that are applied are the standards of the advertising industry.
Even in apparently scientific research papers it's often found that the data in the paper doesn't support the conclusions in the abstract, the conclusions in the abstract don't support the assertions in the press release, and the assertions in the press release don't support the headlines in the media.
By the way, I don't agree that 5 is a reasonable claim either, except for advertising purposes. The members were selected from those who visited the Community page, and selected again from those who have had psychotherapy, guaranteeing bias in 7 Cups favour.
Charlie
More great questions!
Throwning in my support toward this thread. These things are important!
@7CupsCommunity
Just wanted to say thank you for updating this. The stats look more reasonable.
https://www.7cups.com/about/

The "100%" comes from Baumel & Schueller (2016) and the other four %'s come from Baumel (2015).