This experiment identifies a discrepancy between "outcomes that benefit StackOverflow Inc." and "outcomes that benefit StackExchange Users", in a way that is going to make it hard for mods to trust any results of the site test.

Let me explain.

This question is posited as "would lowering voting thresholds increase engagement". But engagement, as in "users spending more time on the site", is not what most SEs need. Most of the SEs I participate in need more quality questions and answers.

So from a perspective of mods and other users of SE, what we'll want to know is "does being able to vote sooner result in users posting their first question/answer sooner, and how good are those questions/answers?" There is absolutely no reason to believe that it will, given that they haven't posted already, but I could be pleasantly surprised.

There are four possible outcomes for this test:

1. No change in behavior
2. Low-rep users do lots of voting but don't post
3. Low-rep users post more, but the posts are low-quality
4. Low-rep users post more, and the posts are high-quality

Of the above, outcomes 2 and 3 are actually harmful to many or most SE sites' usability.

In contrast to this experiment, I'd be interested in one where we lowered the threshold required to comment. I can see a much more direct line from commenting to answering/asking than from voting to posting.

Another experiment that we could try would be to eliminate the "canonical questions rule", which is one of the major things driving the decline in user posting activity. As an SE ages, it gets harder and harder to ask an actually "new" question, so why bother?

This experiment identifies a discrepancy between "outcomes that benefit StackOverflow Inc." and "outcomes that benefit StackExchange Users", in a way that is going to make it hard for mods to trust any results of the site test.

Let me explain.

This question is posited as "would lowering voting thresholds increase engagement". But engagement, as in "users spending more time on the site", is not what most SEs need. Most of the SEs I participate in need more quality questions and answers.

So from a perspective of mods and other users of SE, what we'll want to know is "does being able to vote sooner result in users posting their first question/answer sooner, and how good are those questions/answers?" There is absolutely no reason to believe that it will, given that they haven't posted already, but I could be pleasantly surprised.

There are four possible outcomes for this test:

1. No change in behavior
2. Low-rep users do lots of voting but don't post
3. Low-rep users post more, but the posts are low-quality
4. Low-rep users post more, and the posts are high-quality

Of the above, outcomes 2 and 3 can be actually harmful to many or most SE sites' usability; they are certainly not clearly "successful" outcomes.

In contrast to this experiment, I'd be interested in one where we lowered the threshold required to comment. I can see a much more direct line from commenting to answering/asking than from voting to posting.

Another experiment that we could try would be to eliminate the "canonical questions rule", which is one of the major things driving the decline in user posting activity. As an SE ages, it gets harder and harder to ask an actually "new" question, so why bother?