In my experience with the stackexchange network, the format best supports questions that give a situation, actual behavior and desired behavior with the question being how to obtain desired behavior.

I see the same thing on this site. Looking at the highest voted questions, I see:

What to do if you are accidentally following someone?: Actual behavior: walking near people makes them nervous. Desired behavior: walk near people without causing tension.

How do I tell my new boyfriend that I make a lot of money?: Actual behavior: Boyfriend doesn't know how much money I make and it causes stress. Desired behavior: Tell my boyfriend I make more money without pushing him away.


However, I've seen a lot of comments and some meta posts talking about how 'How can I convince someone to do X' are bad questions.

Where is the line drawn with that type of question? To me, it seems almost any question on this site can be boiled down to some form of "here's my [actual behavior], how do I achieve [desired behavior]?" Which, since the questions are about interacting with other people, seem to become "How can I cause this behavior in some other person" which to me feels the same as 'How can I convince someone to do X.'

What is the difference between 'How can I convince someone to do X' questions and the questions this community supports? What are some hallmarks of the 'convince' questions that should be avoided (compared to their supported counterparts)?

NOTE: my question is different from Do we need to re-write "How can I get X to do Y" questions? in that I'm not asking how to "fix" a question of this form but rather how to avoid it altogether in the first place.

  • 1
    Upvotes aren't a good metric for question quality or an indicator that a question is a good fit for the site. Anyone with an association bonus can upvote a question on any site without needing to invest any time learning about what that site considers a good question.
    – sphennings
    Commented Mar 14, 2018 at 18:17
  • 4
    @sphennings that's fair, but SE communities evolve to fit the wants of the community itself. Upvotes are the manifestation of those wants. Maybe you can point me to some solid IPS questions of this community if you don't consider those I've posted to be up to snuff?
    – scohe001
    Commented Mar 14, 2018 at 18:24
  • 1
    People who drop in for a question or two aren't the community of a site. It's the people who are active on the site, participating in meta discussions, flagging, voting, editing, etc... that make up the community. I'm not commenting either way about the quality of those questions, I'm saying that upvotes aren't a good metric.
    – sphennings
    Commented Mar 14, 2018 at 18:25

3 Answers 3


After a wonderful chat with Tinkeringbell, Sphennings and Spiral Succulent, I've drawn the following conclusions.

It seems that the largest reason that questions are viewed as "convince" questions is because they focus more on how the asker wants the situation to change than how they can change it themselves. (both Tinkeringbell and Sphennings also put it very well in the chat!)

Here are some examples of questions that are too focused on the goal and thus could be called "convince" questions vs. alternate versions that would be acceptable on this site.

Convince Questions:

  1. How do I convince dad to buy me a new pc?

  2. How do I make people stop forcing drinks on me?

  3. How can I change my friend's political opinion?

IPS Acceptable Alternatives:

  1. How can I most effectively approach my dad about a new pc?

  2. How do I assert an alcohol-free lifestyle choice?

  3. How can I have a mutually respectful discussion with a friend of mine with different political opinions?

Remember that given the nature of interpersonal exchanges, nothing is guaranteed. Even with some ultra-amazing advice, the problem posed may not be solvable. Remember to allow for this when you ask (instead of being in the mindset of "How can I make this happen?" Try: "How can I most effectively approach this problem?").


I have migrated the majority of my original answer here, that answer encompasses my view of what is the defining principle between good and bad questions regarding changing the beliefs and opinions of others, here are the extracts from that which are actually an answer to this question.

The difference between a good and bad questions regarding changing the beliefs and opinions of others is that it respects the agency of other people.

A good question on this topic:

  • Focuses on the interpersonal aspect with words like "communicate"

  • Places the onus for action on the asker "how can I communicate"

  • Accepts that success is not guaranteed when measured by others' reactions/responses "how can I better communicate"

A question is bad when it seeks to (or appears to seek to) force another person into an action or opinion. Reading through the Meta post on the topic should help most question askers avoid approaching a scenario with/ creating question with this as a goal.

What to do with questions about "getting around" peoples' boundaries / autonomy

Measuring success of an interpersonal scenario by the post-interaction behavior of the other party should always accept the possibility of failure as there are no sure strategies for changing other people's views. There are sure strategies for improving your chances of success and questions about that personal improvement are what we want on IPS.

  • 1
    "You can only reliably change your own behavior" might be worth adding in parenthesis that even this isn't reliable if it's a deep habit or addiction!
    – scohe001
    Commented Mar 15, 2018 at 0:15
  • 1
    I think your three bullets under "Helping askers fix their questions" are solid points as answer to this question and good tips for future askers! However I'm not sure the focus on how to fix this fits my question. I do appreciate the time and effort you've put into this answer tho :)
    – scohe001
    Commented Mar 15, 2018 at 0:16
  • 2
    I edited this to uh, actually answer the question, hahaha. If you still have a downvote, please leave constructive feedback along with it Commented Mar 15, 2018 at 0:42

The goal of a convince question is to cause someone who isn't the OP to act in a specific way. Even with the skeevy techniques used by salespeople and pick up artists there is no way to make someone do something.

A good IPS question should be about the interaction between two people not the result of the action. Often times knowing the desired outcome is an important part of a good IPS question but more often than not that outcome is to have a productive conversation rather than getting your way.

  • 2
    I like your answer, but I don't know if it answers my question. It seems you're focusing on the purity of the intentions of the questions asked (which is fine if that's the answer to my question). I also agree that most questions on this site are seeking answers about facilitating a conversation with another person, but then so are questions asking to convince people to act a certain way. I don't feel you've made a distinction between "good" questions and "not so good" questions. Can you maybe add links to examples?
    – scohe001
    Commented Mar 14, 2018 at 18:22

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .