| View previous topic :: View next topic |
| Author |
Message |
scarneck
Joined: 18 Aug 2003
Location: Seoul
|
 Posted: Tue Feb 07, 2006 1:27 am Post subject: Statistics Help! |
 |
|
I haven't a real clue how to answer this question concerning stats:
1. What is the average (mean) difference in age between the women student and the women non-students?
Data given:
Variable:Women Students (n=412) Mean age: 20.7+-1.6
Women Non-Student: (n=219) Mean Age: 22.1+-1.7
I just added the standard deviations to the number, and took the difference and came up with 1.6 years.
Is this correct? I did not take stats and this course kicking my a$$. (Research Methods in Education)..
Is it that simple and I'm a complete moron? or is there more to it?
Any help?
Thanks
Seriously, I need help. |
|
| Back to top |
|
 |
Swiss James
Joined: 26 Nov 2003
Location: Shanghai
|
| Posted: Tue Feb 07, 2006 2:42 am Post subject: |
|
|
surely if the two populations are normally distributed then you can work out the mean difference just be subtracting one mean from the other?
So it really is as simple as
22.1 - 20.7 = 1.4
Can't see how any other method would be more accurate. |
|
| Back to top |
|
|
scarneck
Joined: 18 Aug 2003
Location: Seoul
|
| Posted: Tue Feb 07, 2006 3:03 am Post subject: |
|
|
Thanks that's what I did, but wasn't sure I was headed in the right direction.
I appreciate the reply |
|
| Back to top |
|
|
The Cosmic Hum
Joined: 09 May 2003
Location: Sonic Space
|
| Posted: Tue Feb 07, 2006 6:53 am Post subject: |
|
|
"There are 3 kinds of lies in the world...
lies
damned lies
and statistics."
...Benjamin Disraeli |
|
| Back to top |
|
|
denverdeath
Joined: 21 May 2005
Location: Boo-sahn
|
| Posted: Tue Feb 07, 2006 7:28 am Post subject: |
|
|
Take a look at this page:
http://www.netnam.vn/unescocourse/statistics/75.htm
...a little closer to what you have to do given the data you have.
Uh, maybe I'm doing sth wrong...it's been about 8 years since I took my last stats class. 
Answer: (-1.59, -2.13). Means that the women students are between 1.59 and 2.13 years YOUNGER than the women non-students.  Hope I'm right!
Last edited by denverdeath on Tue Feb 07, 2006 9:03 am; edited 1 time in total |
|
| Back to top |
|
|
joe_doufu
Joined: 09 May 2005
Location: Elsewhere
|
| Posted: Tue Feb 07, 2006 8:59 am Post subject: |
|
|
| "The world is full of numbers, and if sufficiently tortured, the numbers will talk." (in a HBS case study I read at b-school) |
|
| Back to top |
|
|
Swiss James
Joined: 26 Nov 2003
Location: Shanghai
|
| Posted: Tue Feb 07, 2006 5:28 pm Post subject: |
|
|
| denverdeath wrote: |
Take a look at this page:
http://www.netnam.vn/unescocourse/statistics/75.htm
...a little closer to what you have to do given the data you have.
Uh, maybe I'm doing sth wrong...it's been about 8 years since I took my last stats class.
Answer: (-1.59, -2.13). Means that the women students are between 1.59 and 2.13 years YOUNGER than the women non-students. Hope I'm right! |
But the mean difference is a number not a range, so how can this be correct? |
|
| Back to top |
|
|
joe_doufu
Joined: 09 May 2005
Location: Elsewhere
|
| Posted: Tue Feb 07, 2006 5:49 pm Post subject: |
|
|
| I think he's written the problem in a funny way, because I can't understand his question. Please write the problem for us exactly as it appears on your homework. |
|
| Back to top |
|
|
billybrobby
Joined: 09 Dec 2004
|
| Posted: Tue Feb 07, 2006 6:27 pm Post subject: |
|
|
| Swiss James wrote: |
| denverdeath wrote: |
Take a look at this page:
http://www.netnam.vn/unescocourse/statistics/75.htm
...a little closer to what you have to do given the data you have.
Uh, maybe I'm doing sth wrong...it's been about 8 years since I took my last stats class.
Answer: (-1.59, -2.13). Means that the women students are between 1.59 and 2.13 years YOUNGER than the women non-students. Hope I'm right! |
But the mean difference is a number not a range, so how can this be correct? |
The difference lies in whether or not we're talking about samples from 2 populations or the entire populations. because the sizes are indicated with a lower case n (assuming the OP didn't write that carelessly), i believe that means we are dealing with samples.
furthermore, i'm not familiar with that kind of notation. are the 1.6 and 1.7 standard devations? i figure if they go through the trouble to give you standard devation they are asking for an estimation of the mean difference based on samples like denverdeath is assuming, but the question doesn't clearly indicate that. |
|
| Back to top |
|
|
Swiss James
Joined: 26 Nov 2003
Location: Shanghai
|
| Posted: Tue Feb 07, 2006 7:03 pm Post subject: |
|
|
Look at it this way, if you had the raw data in front of you and someone asked you
"find the mean difference in the ages of the two populations"
what would you do?
If the population sizes were the same, you'd add up the first bunch, subtract the total of the second bunch- and bob's your uncle.
If the population sizes weren't the same, the best you can do is take the two means and subtract them.
There's no advantage to knowing the standard deviation |
|
| Back to top |
|
|
scarneck
Joined: 18 Aug 2003
Location: Seoul
|
| Posted: Tue Feb 07, 2006 10:38 pm Post subject: |
|
|
When I try to cut and paste the data given, it gets all jacked up. I can try to scan and paste the info as a .jpg
I really appreciate all the help. I'm starting to get some of it. It's a bit intimidating not having a clue on where to start. Some of it seems pretty straight forward if I don't read into it.
Hope this helps.
"one the average (mean), which group of women was older? Explain. |
|
| Back to top |
|
|
denverdeath
Joined: 21 May 2005
Location: Boo-sahn
|
| Posted: Tue Feb 07, 2006 11:13 pm Post subject: |
|
|
| Swiss James wrote: |
| denverdeath wrote: |
Take a look at this page:
http://www.netnam.vn/unescocourse/statistics/75.htm
...a little closer to what you have to do given the data you have.
Uh, maybe I'm doing sth wrong...it's been about 8 years since I took my last stats class.
Answer: (-1.59, -2.13). Means that the women students are between 1.59 and 2.13 years YOUNGER than the women non-students. Hope I'm right! |
But the mean difference is a number not a range, so how can this be correct? |
There's a range because that's where stats has to take into consideration std dev and variance...you know the typical stat, "guaranteed to be accurate within blah blah nine times out of ten"? The range is actually covering the statistical probability of error. I think I'm right with the answer...did similar questions in both pych and bus stats...just, brain's slightly fuzzy now.
OP, doesn't your text have any kind of examples or anything that can help you out? Most texts I've studied from have worked problems and lots of practice questions with answers and necessary charts and formulas. You only have to determine which formula to use when to find whatever it is that's necessary. |
|
| Back to top |
|
|
Swiss James
Joined: 26 Nov 2003
Location: Shanghai
|
| Posted: Tue Feb 07, 2006 11:34 pm Post subject: |
|
|
| denverdeath wrote: |
There's a range because that's where stats has to take into consideration std dev and variance...you know the typical stat, "guaranteed to be accurate within blah blah nine times out of ten"? The range is actually covering the statistical probability of error. |
I understand what standard deviation and variance are, but the question doesn't warrant mentioning them.
Sounds to me as though later questions are going to be taking into account the distribution of data- but this one is simple enough. |
|
| Back to top |
|
|
denverdeath
Joined: 21 May 2005
Location: Boo-sahn
|
| Posted: Wed Feb 08, 2006 1:13 am Post subject: |
|
|
| Okay, you may be right. Sometimes, the simple answer is the right(and better) one. I sent the same problem to a pal who works for CSC and compiles stats for StatsCan. Hopefully, he can set us(me) straight. |
|
| Back to top |
|
|
scarneck
Joined: 18 Aug 2003
Location: Seoul
|
| Posted: Wed Feb 08, 2006 3:58 am Post subject: |
|
|
Thanks DD and SJ for venturing where few dread go.
I'm going to work on the rest of the questions and submit them to the prof and see what happens.
Really, thanks. |
|
| Back to top |
|
|
|
FAQ
Search
Memberlist
Usergroups
Register
Profile
Log in to check your private messages
Log in
