When one looks at the quality and quantity of work performed, it appears that most so called multitasking is really better described as "scattered attention" - people do a weak job of doing several things intersperced rather than doing a good job at one thing after another. That's OK when you are sort-of watching TV, playing a game which can be paused and resumed, answering an occasional text message, and eating some pop tarts. If your job similarly needs only superficial and intermittant attention to do well, that may work (say, a non-busy receptionist?).

But people don't truly watch you-tube videos and compose a well structured document at the same time. At the very best, they could interleave the two such that, say, 30 minutes of you tube watching and 30 minutes of actual document writing can be done in 60 minutes total but split into many smaller interleaved chunks rather than two big chunks - in which case they are still being non-productive on the job for 30 minutes of that hour, albeit in smaller chunks. It's going to be like the jest that many tiny slices of cake have fewer calories than a single big slice of the same size. They are still going to take longer to get the work done than if they were not interleaving, and/or do a sloppier job.

And this is the best end of the scale - more often the frequent "context switches" actually make them less efficient at both tasks, taking 75 total minutes to watch 30 minutes of you tube and to write a document which would have taken them 30 minutes if they focussed on it.

A friend is a college professor tells me that a majority of incoming students today are very poor at written communications, even compared to a decade ago. They may think they as inherently great multitaskers did their homework at the same time that they talked with friends and watched TV, but frequently they only "got by" without learning quality focus of attention, and it shows later in life. Grade inflation (and recalibrating things like SAT scores) has masked the effects of this.

There are jobs which inherently require good interleaving of tasks - for example frequent interruptions and switching attention among many uncompleted tasks. Doing one thing well and getting to a logical pausing point before changing tasks may just not be possible. The best folks in these tasks are still not putting 100% attention on several tasks at once, they are just efficient at dividing their time such that each task gets a fraction of their attention interleaved with other tasks. But even with these people, if some of those tasks were personal and not work related, they have diverted that fraction of their total attention away from getting their jobs done. If its a tiny fraction, then it has little impact; if it's a large fraction then they may be cheating their employer of their expected contributions just as surely as somebody who focused on their job for 5 hours before taking off the last 3 (but charging for 8) - the "multitasking" myth just allows them to pretend it isn't happening.

One important insight into this is that research is showing that even people who think they multitask well actually don't - and in fact, they often accomplish less while multitasking than people who do not self describe as being good at it. One hypothesis is that people who think they are good multitaskers are on average just better at ignoring the discomfort of not doing a good job - ie: being a self described good multitasker really just means they are more comfortable with / less stressed by multitasking because they can numb out to the negative consequences better.

So if you think you are a good multitasker, the chances are that you are actually below average at *successfully* multitasking. If you think you can spend a good portion of your day browsing the internet for personal subjects and still give your employer 8 hours (or nearly 8 hours) of real attention, then you are almost certainly fooling yourself, and trying to fool your colleagues - or you have a low value job which requires hardly any competence or attention to do successfully.

Another friend in internal tech support tells me that his experience is that many young workers think they are just naturally great with technology because they grew up with it, and indeed they are fine at setting up a VCR or light configuration of a cell phone (which got them a lot of praise as a genius from grandma) - but often their tech knowledge still superficial, and they are uninterested in learning anything which they cannot master in 15 minutes via graphical user interfaces intended to make it simple. Obviously this doesn't include everybody and there are plenty of very bright young folks too - but the generalization about younger folks being naturals with technology (or multitasking) may often be more of a "strong narrative" than a reality.

Likewise, next time you try to buy, say, an auto part from some clerk who is simultaneously chatting with a friend by cell phone, see if you think you are getting their full attention and are just as well serviced as if they had not been "multitasking". It just doesn't happen, and it's time to replace the myth of multitasking well (as if several things get full simultaneous attention), replacing it with the concept that interleaving of tasks almost always comes at significant cost, tho the extra waste is smaller for some people than others.