5 ms·

Here's the primary source for the survey in question: http://blog.teamblind.com/index.php/2018/08/20/tech-workers-say-poor-leadership-is-number-one-cause-for-bu

by Sacho 8y ago

Here's the primary source for the survey in question: http://blog.teamblind.com/index.php/2018/08/20/tech-workers-say-poor-leadership-is-number-one-cause-for-burnout/ http://blog.teamblind.com/index.php/2018/08/20/tech-workers-... How useful is a survey like this? The survey does not define burnout to employees, leaving them to guess at the definition. Can employees reliably self-diagnose burnout and differentiate it from similar conditions - i.e. actual depression? Can they also reliably indicate the cause for the burnout, especially if the symptoms can overlap with the cause? Let's say you have burnout because of work overload, but you normally don't mind doing tons of work, just hate your boss. When the feelings of exhaustion/anger come up, will you point them at the work load or at your boss? Perhaps you could relabel it as a survey on workplace discontent, but even then, I don't think it's really well made. The problem with statistical data like this is that we can see the primary source and temper our enthusiasm for the accuracy of the survey, down the broken telephone line, it will be taken as gospel(see the wage gap/sexual harassment statistics as examples).

JohnBooty 8y ago
> The survey does not define burnout to employees, leaving them to guess at the definition. This sort of self-reporting is accepted in psychology. The way you determine a subject's happiness is to ask them how happy they feel. Think about how a more rigorous alternative might work. 1. We come up with some more precise definition of "burnout." Perhaps we define it as "an average of X 'frustration' events in the workplace over Y days, combined with a steadily declining sense of 'enthusiasm'" 2. OK, great, but now we need precise definitions of 'frustration' and 'enthusiasm.' Is 'frustration' the momentary frustration we feel when the compiler complains about something? Is it a barely-restrained feeling of wanting to smash one's keyboard over somebody's face? For how many seconds should that feeling be experienced to qualify? Likewise with 'enthusiasm.' Same problem we had in step 1. 3. Alternatively, we could monitor people's cortisol levels, blood pressure levels, and have them spend their workdays in an fMRI machine or something. Although even that would have limitations; that seems more likely to measure acute symptoms than something chronic like "burnout." > Can employees reliably self-diagnose burnout and differentiate it from similar conditions - i.e. actual depression? With a sufficiently large sample size this wouldn't matter. As you say, surely x% of respondents will confuse depression with burnout. However, the value of x% will surely be nearly the same at all companies.

by Sacho 8y ago

JohnBooty 8y ago

> The survey does not define burnout to employees, leaving them to guess at the definition. This sort of self-reporting is accepted in psychology. The way you determine a subject's happiness is to ask them how happy they feel. Think about how a more rigorous alternative might work. 1. We come up with some more precise definition of "burnout." Perhaps we define it as "an average of X 'frustration' events in the workplace over Y days, combined with a steadily declining sense of 'enthusiasm'" 2. OK, great, but now we need precise definitions of 'frustration' and 'enthusiasm.' Is 'frustration' the momentary frustration we feel when the compiler complains about something? Is it a barely-restrained feeling of wanting to smash one's keyboard over somebody's face? For how many seconds should that feeling be experienced to qualify? Likewise with 'enthusiasm.' Same problem we had in step 1. 3. Alternatively, we could monitor people's cortisol levels, blood pressure levels, and have them spend their workdays in an fMRI machine or something. Although even that would have limitations; that seems more likely to measure acute symptoms than something chronic like "burnout." > Can employees reliably self-diagnose burnout and differentiate it from similar conditions - i.e. actual depression? With a sufficiently large sample size this wouldn't matter. As you say, surely x% of respondents will confuse depression with burnout. However, the value of x% will surely be nearly the same at all companies.