Hacker News Reader

Comment by wlesieutre | Hacker News Reader

Maybe it's just me, but the way these statistics are worded is setting off my skepticism alarms:

> Between 1986 and 1995, fatal traffic accidents rose 17% the Monday following the switch to Daylight Saving Time.

Accidents rose 17% that Monday? Does it mean 17% more than any other day, or just that the raw number for that Monday is 17% higher than it used to be? Because they worded it like the latter.

For all this page says, accidents were up 17% every day of the year over that decade.

EDIT: Wikipedia says total US traffic deaths were lower in 1995 than 1986, so I'll chalk this up as poor wording. https://en.wikipedia.org/wiki/List_of_motor_vehicle_deaths_i...

View on HN

>A 2014 study showed that the hour of sleep lost when switching to DST can increase your risk of a heart attack by 25%

I highly doubt that one less hour of sleep on one day of the year can affect that risk by so much. If that were the case, I assume the risk of heart attack would be much higher than it is now.

It's 25% higher on that day. For all that statistic tells you though, it might just be bringing forward a heart attack you would otherwise have had the following week.

Sure, and going out on a rainy day probably increases your chance of being struck by lightning by 10000%.

1.25*(a really tiny risk of a heart attack on a given day)

This same numerical trick happened when the WHO classified red meat as a carcinogen "in the same category as" cigarettes, and it was stated that it increased your risk of some cancer by ~50% (or something), but the baseline risk is so low that it's a lot less dramatic than it sounds.

This. When I asked a doctor, who was suggesting 'a certain med would reduce chances of some cardio incident by some 10s of percent' (I don't remember, specific details), as to what is the base percent if would be affecting, he sort of evaded the answer.

A 25% increase of a really low number is also a really low number, and quite frankly these numbers are low enough that it's all probably within a margin of error.

Yes, that's true. Just the way it's stated seems to imply that it's not specifically on that day but for an extended period of time.

It's also assuming that people get up an hour earlier. I'm sure a lot of people don't set alarms on Sunday and get a full night's sleep.

Not sure about the 17% in 1986-1995 thing, but here's a study of Colorado when it changed DST policy (I assume the date was changed) that found some effect: http://www.colorado.edu/economics/papers/WPs-14/wp14-05/wp14...

I'm not sure where the 17% is coming from either. The linked study talks about 300 deaths over a 10 year period. That's 30/year which, relative to the 30K or so automotive fatalities in the US per year is 0.01%--i.e. a almost vanishingly small percentage. It's actually not implausible to me that there would be some tiny effect associated with switching time/schedules even if it were only because of "OMG, I forgot to switch my clock and now I'm late for a meeting!" [EDIT: If you apply the 30 annual number against weekly fatalities you can get up into the few percent range. But you're really in the noise with these kind of numbers.]

It's like a case study in cherry-picking statistics.

Exactly. Here's one I cherry-picked showing the opposite:

http://www.degruyter.com/dg/viewarticle/j$002fbejeap.2007.7....

Right, so you found a study that shows there are more accidents at the end of winter when the roads are more icy than they are at the start of spring.

How about we forget the statistics and stop changing our clocks because people don't like losing an hour of sleep and disrupting their sleep schedule?

Or keep changing them because people like more light on summer evenings. :-)

[EDIT: And I would be perfectly fine with DST year-round but I understand why people in northern cities have issues with an extra hour of morning darkness for kids going to school and people driving to work.]

You don't have to change your clock for that. Just leave it on DST the whole year.

Then you have to get ready for work in the dark from October through March.

We have lights for that, and at the beginning Winter the Sun does not rise till 8am for places like Ann Arbor, Mi. So even with the adjustment for Winter you're still getting "ready" in the dark.

I'd rather have the extra hour in the Summer Time and be able to get a full 18 holes in after work everyday

I think that some people feel that way, but certainly not all, and the arguments against are not compelling enough.

I don't mind losing / gaining an hour of sleep. My sleep patterns have never been very precise, certainly not since having kids. I do like DST. That's just my opinion. I get that some people feel the opposite, no matter what your latitude.

I like that the Earth's axis is inclined and that summer is different to winter.

>My sleep patterns have never been very precise, certainly not since having kids. I do like DST.

And because it doesn't affect you it is your reason for imposing it on everyone else? I'm sure a majority of people dislike it. Supposing that everyone else be accustomed to imprecise sleep patterns as you are (which are most likely atypical) is not very fair.

I'm not imposing anything on anyone. I'm expressing my vote to the contrary of someone saying "people prefer it that way".

If you are sure a majority of people dislike it, then go and get it changed. Best of luck to you.

And saying I am "supposing that everyone else..." is a straw man. I specifically noted that other people feel the opposite to me. Back off, man :)

You're right, I didn't read your post fairly. Sorry about that.

Never trust a statistic you haven't cherry-picked yourself.

I was expecting this number to be followed by "compared to the previous Monday" each time it was used. But, alas, it was not.

The paper is actually linked, so I went ahead and skimmed this one.

From the abstract:

> Both models reveal a short-run increase in fatal crashes following the spring transition and no aggregate impact in the fall. Employing three tests, I decompose the aggregate effect into ambient light and sleep mechanisms. I find that shifting ambient light reallocates fatalities within a day, while sleep deprivation caused by the spring transition increases risk. The increased risk persists for the first six days of DST, causing a total of 302 deaths at a social cost of $2.75 billion over the 10-year sample period, underscoring the huge costs of even minor disruptions to sleep schedules.

I still have no idea where the 17% came from or what it's relative to. If anyone else finds it, please fill us in. Am I missing something, or did they borrow a made up statistic out of a clickbait image macro and link to a vaguely related paper as cover or something?

Reply on news.ycombinator.com