Statistics question: Is this evidence of the validity of astrology?

[Veritass]

It's been a long time since I took statistics, so I'm hoping one of you academic types can help me out here.

I work in a company that is mainly firmware and software engineers, with a few chip designers thrown in for good measure. There are 82 of us, so I would expect about 7 of us to have a birthday each month. They post a list of who is having a birthday each month.

I noticed that 3 people had a birth in October, 14 people had a birthday in November, and 3 people are having a birthday in December. Is this evidence that people born in November (roughly meaning Scorpios) are more likely to be engineering types, or does this fall in the expected variability?

---

p.s. I'm a Virgo, so I don't believe in astrology, because Virgos don't believe in astrology.

 

[Neomega]

p.s. I'm a Virgo, so I don't believe in astrology, because Virgos don't believe in astrology.

sweet..........

 

[MOTH]

82 people is too small of a sample to expect to have evenly distributed birthdays except by chance. I used to work in a group of 14 people and there were 2 birthdays on Feb 14 and 1 on Feb 15.

 

[Bartleby]

I'm a Taurus, so I don't believe in all that bull.

 

[Irish Caesar]

No, it means that their parents are more likely to celebrate Valentine's Day with more than just a card.

 

[nivi]

Birthdays are far from random.

 

[Veritass]

82 people is too small of a sample to expect to have evenly distributed birthdays except by chance. I used to work in a group of 14 people and there were 2 birthdays on Feb 14 and 1 on Feb 15.

Yes, and someone who is adept at statistics could probably tell us the probability of a group of 14 people having three birthdays in a two-day period. I submit that is quite a bit lower -- and statistically less significant -- than the group I originally asked about. I would be interested in the probability of both.

Maybe this is just a deceptive statistics thing, like you only need a group of 23 random people to make the odds greater than 50/50 that at least two of them share a birthday.

 

[nivi]

Maybe this is just a deceptive statistics thing, like you only need a group of 23 random people to make the odds greater than 50/50 that at least two of them share a birthday.

How does that work out?

 

[Neomega]

How does that work out?

If you pick one personas birthday, the odds are 23/356 that another person in the room shares a birthday.

However, if you are just looking for any person int he room to match any other person, than the probability goes up significantly.

 

[Veritass]

How does that work out?

Here's a link that explains it better. It helps to think that the probability that at least two people have the same birthday is the inverse of the probability that no one in the group shares a birthday. As you add people to the group, it gets less and less likely that each new person shares a birthday with nobody.

 

[Godwynn]

Is this evidence that people born in November (roughly meaning Scorpios) are more likely to be engineering types, or does this fall in the expected variability?

I'm a scorpio and my first year of college was for architecture. Until I realized how much I hated it.

 

[.Shane.]

Oh, lord no....

Its just normal statistical deviation that you'd see in a small data set. And you're an engineer?

 

[warpus]

Coincidence, obviously.

 

[Fugitive Sisyphus]

If you pick one personas birthday, the odds are 23/356 that another person in the room shares a birthday.

However, if you are just looking for any person int he room to match any other person, than the probability goes up significantly.

No.

With one person there is a 365/365 (i.e 100%) chance that there are no shared birthdays. With two people it is (365/365)(364/365), with three people (365/365)(364/365)(363/365). etc.

 

[Veritass]

Its just normal statistical deviation that you'd see in a small data set. And you're an engineer?

No, I'm not an engineer; I'm a programmer. I recognize that "real" engineers would never refer to programmers as "software engineers." You know what they say:

If architects designed buildings the way programmers designed software, the first woodpecker that came along would destroy civilization.​

 

[Masquerouge]

Here's a link that explains it better. It helps to think that the probability that at least two people have the same birthday is the inverse of the probability that no one in the group shares a birthday. As you add people to the group, it gets less and less likely that each new person shares a birthday with nobody.

I remember winning a lot of bets in High school thanks to that trick.
"hey, 10 bucks two people in this class have the same birthday!"
"no way, man, we're only 30 people!"
MWAHAHAHAHA!

 

[mangxema]

No, it means that their parents are more likely to celebrate Valentine's Day with more than just a card.

My thoughts exactly. My birthday comes nine months after my dad's birthday, almost to the day. I'm pretty sure he must have had a good time.

Totally different: For several friends of mine, their personalities are almost exactly as their zodiac signs would predict. Sometimes freakishly so, myself included.

 

[Phlegmak]

My thoughts exactly. My birthday comes nine months after my dad's birthday, almost to the day. I'm pretty sure he must have had a good time.

Totally different: For several friends of mine, their personalities are almost exactly as their zodiac signs would predict. Sometimes freakishly so, myself included.

Zodiac related predictions are so general, they can easily relate to anyone.

A skeptic of palm reading thought palm reading was bogus. He learned palm reading. He then read some people's palms and told them their fortunes. He made an amazing discovery. Palm reading works! The reason it worked is that the people in question shifted their view of themselves enough to fit what the palm reader was saying. Also, the palm reader was making general statements, wide enough to describe anyone.

 

[Turner]

It's fun to read my horoscope the day after and see how close it was.

I tend to not believe in it. However, I have noticed patterns where I get along better with certain signs, and not as well with others. :shrug:

 

[mangxema]

Zodiac related predictions are so general, they can easily relate to anyone.

If by this you mean horoscopes, then I agree. What I was referring to, though, was pages that are like, "This is what an Aquarius is probably like." And so far almost everything that I have read about Aquarians, I am. Even specifics. And I know that they are not so general that every category applies to everyone, because I have read about my friends signs and while they describe them well, they rarely are appropriate for me, certainly nowhere near to the degree that Aquarius does.