Crossposted from Figuring Figuring.

I did a second survey that fixed some of the flaws of the first survey. The results from the second survey significantly color the interpretation of the results from the first survey given in the first “Conclusion and Discussion” section. Please continue reading past the section titled “Second Survey” to get a full picture of the results from all surveys.

Intro

A couple days ago a friend of mine on facebook asked about arguments in favor of mockery. They pointed out that they had noticed a lot of facebook posts mocking people for not wearing masks in the covid-19 era, and wondered whether this was an effective way to change people’s behaviors.

I said in the comment section of that post that I would make a survey that worked as follows. Roughly half of the survey takers would be randomly assigned to answer the following questions:

1. Do you think that mockery is an effective way to change people’s minds?
2. Do you think that mockery is an effective way to change people’s behaviors?

The other half would be randomly assigned to answer these questions:

1. Has being mocked ever caused you to change your mind about something?
2. Has being mocked ever caused you to change your habits or behaviors?

No survey respondent was permitted to see all four questions. The possible answers to each question were “Yes”, “No”, and “Not sure”.

I made this survey using GuidedTrack. I posted it on my facebook wall, and also posted it to Positly and paid people to participate.

A total of 145 people responded to any of the questions on positly. 74 were asked the first set of questions, and 71 were asked the second set of questions. A total of 66 people responded to any of the questions on facebook. 31 were asked the first set of questions, and 35 were asked the second set of questions.

Before I go on to tell you the results of the survey and the predictions me and some of my friends made, you might want to make your own predictions. I suggest you quickly scribble them down. Some particular questions you might want to make predictions about:

• Did more people answer yes to the first set of questions than the second set of questions, or is the reverse true, or were they about the same?
• Were facebook respondents (presumably people who are friends, or friends of friends of mind on facebook) more or less likely to say yes to the first set of questions?
• Were facebook respondents more or less likely to say yes to the second set of questions?
• What did I predict about the previous two questions?

There may be other fun questions to predict, and I’d be curious to hear how you did in the comments. Predictions from me and my friends coming up, so make sure you make your predictions beforehand. Again, I suggest that you write them down. You may also want to write down your reasoning beforehand.

_
_
_
_
_
_
_
_
_
__
_
_
_
_
_
_
____
__
_
_
__
_
_____
_
_
_
_
_
_
_
_
_
_
_
_
____
_
_
_
_
_

Predictions

Ok, last chance to make predictions before you hear some spoilers…

Alright.

I predicted that many more people would answer yes to the first set of questions (ie, the questions about whether mockery is effective) than to the second set of questions. I also predicted more people would say no to the second set of questions than to the first.

I’m not sure exactly what my theory was when I made that prediction—I made the prediction in the same comment that I suggested the survey, but I came up with two post hoc hypotheses that might explain the result I predicted. I do know that part of the reason I made that prediction is that mockery is fun, but admitting that fun is the main reason we do it rather than because of its positive effects on other people’s behavior feels kind of icky. So we use its effectiveness as an excuse.

One hypothesis is that we overestimate the effectiveness of mockery. This would make sense of the predicted result because it would be evidence that we all think mockery works on others, but none of us thinks it works on us.

The second hypothesis I made up to explain this predicted result was that while we know that mockery works on other people, we are hesitant to admit that it works on us, because that is a bit embarrassing. Perhaps people are also not that great at telling what actually caused them to change their minds or behaviors.

These hypotheses are not mutually exclusive.

In a private conversation, my friend predicted using similar reasoning that actually people would tend to answer the second set of questions (ie, those about how often we change our own minds as a result of mockery) affirmatively. Saying that you think mockery is effective feels kind of icky, but saying that you think you have never had your mind or behavior changed because of mockery seems kind of arrogant.

Seeing how such similar reasoning could be used to predict a totally different result made me feel a bit nervous.

Another friend of mine predicted that my facebook friends would be less likely to change their minds because of mockery than randomly selected survey participants. Positly users aren’t quite randomly selected, but they’re closer to randomly selected than whatever people happened to come across my facebook post.

Results

Sure enough, I was totally wrong.

Twice as many Positly respondents said that mockery has worked on them as said that mockery is effective. Positly respondents were slightly more likely to say that mockery is effective for changing behavior than for changing people’s minds, both for themselves and for others.

I think this is strong evidence against the hypotheses I suggested, and some evidence in favor of the hypothesis my friend suggested in conversation.

My facebook acquaintances were slightly less disproportionate. Only 1.2 times as many respondents said that mockery is effective on themselves as said that mockery is effective on others when it comes to behavior. However, when it comes to changing minds, still about twice as many said that mockery has worked on them as said that mockery is effective on others.

I was surprised by this, as I tend to think of myself as preferring people who do not use mockery, and not using mockery while also thinking it is effective is a hard pair of things for humans to do simultaneously.

Of the 35 facebook respondents that were asked the second set of questions, 37% said they had changed their mind because of mockery, 59% of Positly respondents said the same. This seems like decent evidence to me that my friend was right about my facebook acquaintances being less likely to change their minds because of mockery.

No respondent said that mockery was effective for changing their own, or other people’s minds and also answered that it was not effective for changing their own, or other people’s behaviors.

Other Responses

Here is a list of some of the things that people said they changed because of mockery. This was an optional part of the survey. Slightly edited for brevity, to protect anonymity, and avoid repeats.

• Economic beliefs
• Working out habits
• Avoiding people who mock them
• Writing about topics on fb
• My own appearance
• How good other people are
• Picking my nose in public
• Crying in public
• Basic cultural rules, like where to sit, how to join a conversation, etc.
• Philosophical or ethical beliefs
• Individualism as an ethical stance
• Fashion
• Music
• Using an old fashioned word
• Beliefs about what is socially acceptable
• Conversational habits
• Wearing briefs instead of boxers
• Stopped whistling
• Stopped/started wearing shorts
• Eating habits
• Lost weight
• Stopped playing sports
• Being late
• Hairstyle
• Stopped watching anime
• Mocked for being autistic, so changed the way I interact with people.
• Started wearing make up.
• Mocked for being outgoing, became less outgoing and confident.
• Started liking Trump
• Left Mormon religion
• Started thinking more before speaking
• Started brushing teeth more
• Stopped being conservative

Here is a list of some of the things that people said mockery was effective for changing in other people. This was also an optional part of the survey. The entries in this list have been edited as in the previous list.

• Weird opinions
• The way people think
• The way others dress
• Haircuts
• Mask wearing
• How often someone complains
• Making someone hide their opinions
• Arrogance
• Getting people to stop doing things around you
• Getting someone to stop writing things in public

I think these lists are similar enough in content to rule out another explanation of this data. You might have thought that people think mocking people is an effective way to get other people to change certain kinds of things, but when they think about what sorts of things they have changed themselves because of mockery, the two categories do not have much of an intersection. These lists make that seem unlikely to me.

Discussion and Conclusion (1)

These results seem like some evidence to me that people in general underestimate the effectiveness of mockery for getting others to change their minds. This is of course not necessarily an argument for using more mockery. I, for one, take the results of this survey to be a further reason that we should not mock people.

If you thought mockery was just some harmless fun you can have with your in group, as I sort of did, you might have thought that the costs to those being mocked are actually not that great. But it seems like mockery can make someone leave their religion, stop writing in public, change their political preferences, etc. I would strongly prefer for people to make decisions about those sorts of things using object level reasoning rather than reasoning about what will cause them to be mocked less. I will now much more than before see mockery as deliberate enemy action designed to interrupt other people’s cognition—not something to be taken as a joke, especially not in the context of conversations about important topics.

Second Survey

This section was written after getting the data for my second survey which was inspired by some criticisms of the questions in the first. Everything above was written before getting that data.

On the other hand, the questions I asked people in the original survey were not exactly analogous to each other. Firstly, people might have answered the first set of questions considering that although mockery is rarely effective for changing the behaviors or beliefs of those being mocked, it might work on bystanders who watch the mocking happen. Secondly, people answering the second set of questions with a “yes” might be thinking that “yes, mockery has ever caused me to change my mind” but that does not mean it is very effective.

To correct for this, I made a second survey. Half of respondents were asked the following two questions:

1. How often has mockery worked as a means of getting you to change your mind about something?
2. How often has mockery worked as a means of getting you to change your habits or behavior?

The other half were asked the following two questions:

1. How often does mockery work as a means of getting someone to change their mind about something?
2. How often does mockery work as a means of getting someone to change their habits or behavior?

The possible responses were: “very often” , “often”, “sometimes”, “rarely”, and “almost never”.

The survey was published on positly.

I will give you some room to make predictions before showing the results.

_
_
_
___
_
_
__
_
_
_
_
_
___
_
_
___
_
____
_
___
_
__
_
_
_
___
_
_
_
_
_
____
_
____
_
_
_
_
_
_
___
_
___
___
__
_
_
_
_
_
_
_

Second Survey Results

There were a total of 115 respondents. 57 were asked the first set of questions, 58 were asked the second set of questions. Here are the results compared across groups.

Mapping “very often” to 4, “often” to 3, “sometimes” to 2, “rarely” to 1, and “almost never” to 0, this gives a mean response for group 1 question 1 of 1.4562, and a mean response for group 2 question 1 of 1.5345, meaning that respondents overall thought that mockery was slightly more effective on others than on themselves.

Using the same mapping, the average response for group 1 question 2 was 1.5789, and the average response for group 2 was 1.6667. Again, respondents overall thought that mockery was slightly more effective on others than on themselves.

Discussion and Conclusion(2)

These results contradict my original interpretation of the first survey’s data. The second survey suggests that people are in general pretty well calibrated about the effectiveness of mockery, or perhaps slightly underestimate it. I conclude that much of the effect observed in the results of survey one’s data was caused by the two effects discussed at the beginning of the “Second Survey” section and not the result of people genuinely underestimating the effectiveness of mockery.

However, I think I am still going to take mockery more seriously than I did before, mostly because I still think this survey showed me that mockery is more effective than I thought it was. The list of personal examples people gave were fairly chilling. I also imagine people cave to mockery a lot more than they are able to notice or willing to admit on a survey. Furthermore, I don’t think it was a coincidence that it was mostly me and my weirdest friends who (incorrectly) predicted that people would say that mockery is much more effective on others than it is on themselves. Probably, we weirdos have grown numb to mockery’s sting, and fallen out of touch with what it feels like to be mocked for most people.

I Would like to Thank

Ozzie Gooen for inspiring me to make these surveys with his facebook post.

Frank Bellamy and Julia Kris Dzweiria for pointing out the assymetry of the questions in the original survey.

Beth Kessler and Aryeh Englander for useful discussion.

And Spencer Greenberg as well as the whole of the Positly and Guidedtrack teams for making it much easier to run surveys like these.