Doing a big survey on work, stress, and productivity. Feedback / anything you're curious about?

post by lionhearted · 2017-08-29T14:19:37.241Z · LW · GW · Legacy · 16 comments

In September, doing a big survey on work, stress, and productivity -- going to gather a bunch of possibly germane data, and then see what correlations stand out.

Current version is around 90% complete here --


Any feedback? Any data you'd be very interested in getting? We're basically guaranteed to get basic statistical significance / sample size, and might have respondents in the mid-thousands if things break right. What would you like to know? Feedback? Thanks.

Edit 7 September: it's now live here -- -- I answered a few of the top questions and read all the rest and incorporated some of the feedback. Thanks so much.


Comments sorted by top scores.

comment by gwern · 2017-08-29T15:23:04.903Z · LW(p) · GW(p)

Perhaps you could provide some more background? Who is this being run by? Where will it be advertising and sampling from? What data will be released? What analyses will be run, and what hypotheses are you particularly looking into? Does it build on any earlier surveys by reusing questions?

Replies from: lionhearted
comment by lionhearted · 2017-09-07T07:23:02.507Z · LW(p) · GW(p)

I'm running this with Kai Zau, my co-founder at Ultraworking. It's being promoted largely among people who have self-directed components to their work -- knowledge workers. Analyses, we'll see -- I'm an ok amateur statistician, and we plan to open up the whole data set, so people can go as far as they want with it. The lowest hanging fruit I'm looking for are correlations, especially unexpected ones, to investigate further. This is our first time doing it. Data released in October.

comment by ChristianKl · 2017-08-29T15:51:41.271Z · LW(p) · GW(p)

I would like a question about the amount of hours that the person works and sleeps per week (sleep can also asked as per day).

For the sake of open science it might be worthwhile to change "You may use my data for a research paper" to "My data can be used for a research paper".

Replies from: lionhearted
comment by lionhearted · 2017-09-07T07:24:18.911Z · LW(p) · GW(p)

Added, thanks.

comment by ChristianKl · 2017-08-29T15:19:21.164Z · LW(p) · GW(p)

Why don't you ask for the age in years, instead of providing existing batches?

Replies from: lionhearted
comment by lionhearted · 2017-09-07T07:24:10.620Z · LW(p) · GW(p)

Good question.

The biggest (counterintuitive at first) best practice I read about is fighting very hard to keep it under 9 minutes to reply, since fall-off rate seems to go up very quickly after that. It's faster for people to click entries without having to type things in manually. It means worse data on some things, but higher completion rate, in theory.

Replies from: ChristianKl
comment by ChristianKl · 2017-09-07T08:18:58.268Z · LW(p) · GW(p)

How do you know that this it is significantly faster? This means that the person has to read through the options before they click. A field with a number needs less reading.

comment by moridinamael · 2017-08-29T15:46:50.001Z · LW(p) · GW(p)

For a topic like this, I might not lump in "doctoral or professional degree." I imagine medical doctors, lawyers, people with PhDs in the liberal arts, PhDs in the sciences, and PhDs in engineering disciplines might all have materially different work/stress/productivity profiles.

Similarly you could separate "Engineering/CS" into two choices. Mechanical engineers and software "engineers" don't really do the same kind of work.

I am both a consultant and an employee, so I wouldn't know how to answer "what is your current role" unambiguously.

I would add some questions about commute time. Definitely an important factor when assessing stress and work satisfaction.

comment by ChristianKl · 2017-08-29T15:36:16.578Z · LW(p) · GW(p)

I would add "dance" as a choice for "Do you regularly participate in any of the following activities?"

comment by Friendly-HI · 2017-08-31T17:11:46.798Z · LW(p) · GW(p)

Background: My wife and I are both studying psychology, I'm doing my M.Sc. she does a very similar study on health risks at the workplace (an adaptive one, if general questions are answered "unfavorably" (i.e. indicator a health risk at the workplace is present), then a handful of more in-depth related questions to pinpoint the exact problem are given.

I didn't take the test yet, but just by reading ChristianKI's objections/suggestions (with which I agree) you need to really clarify the questions.

Here's a helpful piece of advice I've been given: Don't think about if the question is understandable, ask yourself if there is any chance at all that it could possibly be misunderstood by someone somewhere and change it as often as you need until that is no longer the case.

comment by NancyLebovitz · 2017-08-30T16:57:44.422Z · LW(p) · GW(p)

Discussion of different office set-ups. Open, own office, shared with one person, etc.

comment by Elo · 2017-08-29T21:39:07.405Z · LW(p) · GW(p)

Suggestions for books that might inspire good questions:

flow, flourish, deep work, gtd, 7 habits, secrets of productive people.

What is your biggest source of stress usually?
What is your biggest source of stress today?
relationships, overworked, financial problems, religion/spirituality/morality, health/exercise, other.

comment by Elo · 2017-08-29T21:31:17.934Z · LW(p) · GW(p)

Social media for news sources
Craft activities as a kid
Rather not say for income
How much do you donate?
Lizardman constant in various places
Expand the email question to include -1000, 5000, 10000, 50000
Regularity of exercise
How long is your commute?
Commute by (select as many as relevant): car, bus, walk, bike, taxi/uber, plane, ferry.

comment by ChristianKl · 2017-08-29T15:34:19.950Z · LW(p) · GW(p)

"If you're in business and your revenue is different than income, how has your revenue changed in in the last two years?"

I'm not exactly sure what "in business" means here. Does it mean the person owns the business in question?

comment by ChristianKl · 2017-08-29T15:27:48.671Z · LW(p) · GW(p)

"I go to sleep at the same time."

Different people are going to have a different idea of what "same time" means.

comment by ChristianKl · 2017-08-29T15:26:17.196Z · LW(p) · GW(p)

"How many colleagues do you interact with on a typical day?"

I'm not exactly clear what this question means. If I write an email that's read by 10 people, does that count as "interaction"? If I say "Hello" to 4 colleges while taking the elevator, does that mean I have interacted with 4 collegues?