How can I improve my question on quantitative modeling of interpersonal skills?

Question

I am trying to improve my question Can interpersonal skills be modeled quantitatively? and would like some help from the community.

How can I improve the question so that it becomes worthy of upvotes or at least reopening?

A big concern is the fact that several people gave a close-vote reason of "because this might be better suited to cogsci.se", and this (is on topic somewhere else) is not even an acceptable reason to close on SE, so it is unclear what is is that is actually unacceptable about the question. Is the real problem that it is too broad? Is it too narrow (e.g. quantitative modeling is such a small part of the study of interpersonal skills that any answer would be trivial)? Is it fundamentally a concern about resource requests and the question would be acceptable if rephrased? Do people just object to the idea of quantifying interpersonal skills and no amount of editing would make it into a question they could appreciate?

I have modified the question to be less of a resource request, but am unsure of whether narrowing its scope would accomplish its purpose. Can someone recommend a narrower scope that would be acceptable, or can someone assist me in modifying it to an acceptable question?

Answer 1

Thanks for bringing your question to meta.

Are you looking for an existence proof -- whether any aspect of interpersonal interactions has been modeled in this way -- or are you interested in models for a particular type of interaction (and maybe even non-quantitative models might be helpful)? Both of those seem on-topic to me, in the same way that research into group behavior is on-topic on Community Building and research into salary negotiations is on-topic on Workplace. Most of our questions will be from people looking for a solution to a particular problem, which could be answered by reporting this type of research, but that doesn't make questions about the research itself off-topic.

The first version of your question seemed to ask for a "model of everything", which is broad and seems unlikely to me. So while I disagree with the close reason (it's too broad, not off-topic), I would have voted to close too. The current version is better, but I suggest you further refine the title and first sentence. If you are looking for an existence proof, try something like "does a quantitative model exist for any aspect of interpersonal skills?". If you're looking for models of a particular type, try something of the form "has the interpersonal behavior of western Internet users in online games been quantitatively modeled?" -- though you might ask the broader question of "...been researched?" and say in the body of the question that you're particularly interested in quantitative models.

Answer 2

Considering how vague and subjective the subject of interpersonal skills is, the existence of a quantitative model of any kind would be significant.

If something is subjective, then a quantitative model does not make it less subjective. You seem to be misunderstanding how quantitative models work.

Perhaps an example will help illustrate this. There are huge debates about what is good art and what is bad art. This debate is incredibly subjective, ie it requires you to have a personal opinion in order to participate. Imagine that someone wanted to create a mathematical model that ranked works of art. In building that model, the developer would have to give the model opinions on issues, such as whether the model thinks the color red is nice, or whether popularity should be used to judge whether art is good. So the model would still be subjective, someone could build a different model and it would be equally correct.

In addition, it is not even slightly clear what you are trying to model. Theres no such thing as a model of everything.

For example, "In theory, Smith (2015) defined a five-level Hierarchical Skills Progression Framework in Telephone Etiquette Skills from Level 1 Beginner to Level 5 Social Wizard Double Advanced Elite, but it will probably be until at least 2020 before you can just go and register for a test and get formally assessed or certified at any of the levels."

You're not looking for a quantitative model. You are looking for a textbook that will tell you what to do.