By Niels Hoven May 7, 2009

With over 2.5 million jobs lost in the past four months, it is an employer's market. I myself have a number of friends looking for work, a search no doubt made more difficult by the following fact: According to the 50,000 respondents of the General Social Survey, we young people are a self-admitted group of huge slackers.

It's understandable, however, that we have little time left for work, as apparently we spend all our time having sex and watching TV. (More on that later.)

Older workers work much harder than younger workers

Workers in their 20's are nearly 5 times more likely than workers in their 50's to describe their efforts as "only the minimum". Older workers, on the other hand, are 40% more likely to work "really hard".

So what are these younger workers doing with all their free time? Well, one thing they're doing is having sex. Remember this chart from Lose Your Job, Improve your Sex Life?

Frequency of sex declines dramatically with age

Oh, and then there's TV, of course. Time spent watching TV is at a near high for teenagers and decreases steadily with age, right up until retirement age when TV viewing spikes again. With all the time they're spending having sex and watching TV, it's no wonder that young workers don't have much time left for the office.

Everyone watches way too much TV. Three hours on average? Really? Young people and really old people watch the most, though.

It should be noted, however, that there is a cure to this idyllic lifestyle, and it's called marriage.

According to the survey responses, the stereotype of spouses who spend all their time at the office is more than just a cliché, it's a fact of life. Married respondents rate themselves as working far harder than never married respondents, though still not as hard as those whose marriages have ended. (For you nitpickers, the trend does hold even after controlling for age.)

Never-married people work less hard than married people. People from failed marriages work the hardest of all.

So what's the moral of this story? Well, there are several perspectives to consider.

As an employer, there's little you can do to make your young employees age faster. But maybe you can get them married off. Perhaps a subsidized matchmaking program?

If you're young and single, well, enjoy it while it lasts. And perhaps take your employer's new subsidized matchmaking program with a grain of salt.

And for those of you who are happily married? Well, we would never suggest that your employer is rooting for your marriage to fail just to get a few more hours of work out of you. But just to be on the safe side, maybe you shouldn't show them this article.


I appreciate these nice visualizations of GSS data. These images from Tableau are much more attractive than typical images of GSS data.

I would love to see some discussion of Tableau's strengths and limitations regarding categorical data analysis. Also, I would appreciate discussion of how one might use GSS weights in Tableau. These weights are designed to compensate for sampling strategies that make the GSS non-representative of the national population of adults. For example, weights correct for years with oversamples of African-Americans, for household size, and in recent years, for non-response bias. How feasible is it to utilize these GSS-supplied weights in Tableau analysis? Does the analysis posted thus far use weighted or unweighted data?

This site allows relatively easy access to GSS data, with convenient options to include or not include weights in analysis:

One final thought: it would be nice to identify which waves (years) of GSS data are used in each graph.

The key word in the above survey is "self-admitted". I think the data may indicate that young workers are just more honest/realistic about their work-effort than older folks! The real insight here is that "Older workers say they work much harder than younger workers" :)

Hi Conrad! Thanks for the thoughtful responses. In my stacked bar charts, I compute the percentages with a table calc, based off of the "Number of Records" measure. This essentially weights every single record equally.

One way to use weights is to include them as a measure, and then use the weights field rather than the "Number of Records" field.

With weighted averages rather weighted percentages, things get a little trickier, and I suspect you would have to create a calculated field to deal with them. I've asked our support team for a better answer and I'll let you know what I hear.

Instead of just looking at married / never married / no longer married, how about length of time married? And how about number of children? I wonder if people work a lot more when they have one or two kids, then work less as they have three or more.

Interesting point. I'll definitely factor length of marriage into a future visualization.

Yikes - too many assumptions of causality! For example, workers who don't like their jobs are likely to slack off and are also likely to retire earlier than workers who love their jobs. So it is only natural that older workers (who opted not to retire) are more committed to their jobs.

Maybe workers who love their jobs are more likely to overwork and die of a heart attack in middle age.