Humans are excellent at telling you what they think. We are abysmal at telling you why we think what we think.
A great example of this came up recently regarding the trust that individuals have towards nonprofits. It’s an excellent survey from Independent Sector.
Some top-line data:
1) As of December, when the survey was done, 81% of Americans said they had trust in nonprofits to strengthen society. Other surveys have shown trust in nonprofits go up since COVID-19, meaning very high trust levels indeed.
2) People differentiate between nonprofits and philanthropy. Only68% of Americans trust philanthropy to make society better. So people like the organizations more than people giving the money, I guess? Similarly, people said nonprofits were more likely to address root causes than philanthropy (48% to 26%). People tend to associate philanthropy with the arts and nonprofits with communities. An interesting, if potentially false, distinction.
3) The more trust, the more people give. Likely not a shocker, but those with high trust in nonprofits volunteered five hours per month (median) versus three for those with low trust. And .38% of high trust folks gave $500+ per year, compared with .28% of low-trust folks.
But they also did a clever thing. They asked people what they thought was important to their trust in nonprofits. Then they also statistically derived what was actually driving people’s decisions.
On the vertical (Y) axis, you have what people said; on the horizontal (X), you have what actually mattered. So, in the upper right, you have “Clear mission” — people said it mattered and it did. In the lower left, you have “Political advocacy” — people said it didn’t matter and it didn’t.
So cast your eyes to the bottom right where it says “celebrity/public figure endorsements.” Very low on the stated effect: fourth worst, in fact. But it was third most influential overall.
This makes sense. You want to say that you make your philanthropic—I mean, giving, lest you think all my giving is going to the opera—decisions independently of anything else. I vote for who I want, not because of whom X endorsed. I’m not influenced by Y actor selling me life insurance on TV. I buy the basketball shoe necessary for my regime of throwing wadded up pieces of paper at the trash can (and missing), not because Z’s name is on them.
But back in the real world, these things matter. “Authority” was one of Cialdini’s basic seven influencers. What makes us think it doesn’t work on us?
Meanwhile, in the upper left, “no political influence” was rated highly by most, certainly more highly than celebrities or the #2 actual influencer: whether an organization engages underserved community.
And yet this one finished dead last in actually mattering. Free from political influence sounds like a good thing, but chances are there are organizations that aren’t free from political influence that you support; they just so happen to be influenced in the same way you are influenced. That could be party or an issue or demographic or what-have-you. So you rate it highly, but it something that you could even seek in some organizations.
This is important because of surveying. Too often, we ask people to analyze the thing they are bad at telling you — the why. Not only does this give you pretty hilariously incorrect answers, but some why questions can actually impact people’s overall ratings, making them less accurate.
So when you are surveying, stick to the things people know best: themselves and what they believe. The rest, you can figure out better with stats.