Started By
Message

Graphing t-Test: Two-Sample Assuming Unequal Variances into Bell Curve

Posted on 11/20/15 at 3:19 pm
Posted by TigerDik86
Baton Rouge
Member since Oct 2011
2982 posts
Posted on 11/20/15 at 3:19 pm
I've searched online and am unsuccessful on finding anything definitive. I would be forever grateful if someone could show me how to graph into bell curve.

Here is my data

t-Test: Two-Sample Assuming Unequal Variances

Variable 1 Variable 2
Mean 17.4 7.2
Variance 56.8 19.7
Observations 5 5
Hypothesized Mean Difference 0
df 6
t Stat 2.607680962
P(T<=t) one-tail 0.02012269524
t Critical one-tail 1.943180274
P(T<=t) two-tail 0.04024539048
t Critical two-tail 2.446911846

non-gamer in-game deaths
20
18
8
13
28

Hardcore in-game deaths
10
12
4
5
15
This post was edited on 11/20/15 at 3:20 pm
Posted by GEAUXT
Member since Nov 2007
29248 posts
Posted on 11/20/15 at 3:20 pm to
Lol
Posted by StringedInstruments
Member since Oct 2013
18411 posts
Posted on 11/20/15 at 3:20 pm to
Kids these days, man.

Storming the president's office. Making absurd demands. Asking the OT to do their homework for them.

Bunch of wussies IYAM.
Posted by WG_Dawg
Hoover
Member since Jun 2004
86490 posts
Posted on 11/20/15 at 3:20 pm to
2drunk4dis
Posted by el Gaucho
He/They
Member since Dec 2010
53003 posts
Posted on 11/20/15 at 3:20 pm to
PEMDAS
Posted by Ingeniero
Baton Rouge
Member since Dec 2013
18286 posts
Posted on 11/20/15 at 3:26 pm to
Wait you only had 5 samples for each case? Do you even rule of three and law of large numbers bro?
Posted by CptBengal
BR Baby
Member since Dec 2007
71661 posts
Posted on 11/20/15 at 3:27 pm to
What are you trying to graph?

The distributions?

If so, plot the probability distributions.

This isn't hard.
Posted by Ash Williams
South of i-10
Member since May 2009
18147 posts
Posted on 11/20/15 at 3:30 pm to
Posted by nes2010
Member since Jun 2014
6761 posts
Posted on 11/20/15 at 3:34 pm to
I think excel can do this for you. You may have to activate the data analysis package.
This post was edited on 11/20/15 at 3:36 pm
Posted by UpToPar
Baton Rouge
Member since Sep 2008
22157 posts
Posted on 11/20/15 at 3:35 pm to
I read the title as Graphic-T test and was oddly curious.
Posted by gaetti15
AK
Member since Apr 2013
13365 posts
Posted on 11/20/15 at 3:36 pm to
quote:

What are you trying to graph?

The distributions?

If so, plot the probability distributions.

This isn't hard.

Posted by Sneaky__Sally
Member since Jul 2015
12364 posts
Posted on 11/20/15 at 3:36 pm to
lol
Posted by Black n Gold
Member since Feb 2009
15409 posts
Posted on 11/20/15 at 3:37 pm to
quote:

I read the title as Graphic-T test and was oddly curious.


I was sure this was an Affliction thread.
Posted by Pectus
Internet
Member since Apr 2010
67302 posts
Posted on 11/20/15 at 3:40 pm to
Here ARE my data.

So you want to show these data, tail and curve, in a spreadsheet?
Posted by stinkdawg
Savannah, smoking by the gas cans
Member since Aug 2014
4072 posts
Posted on 11/20/15 at 3:41 pm to
I'd just shade in "C" on the test.
Posted by FloridaMike
Member since Dec 2012
1524 posts
Posted on 11/20/15 at 3:42 pm to
The answer is 288, nerd
Posted by LSUTANGERINE
Baton Rouge LA
Member since Sep 2006
36113 posts
Posted on 11/20/15 at 3:43 pm to
Used to have to figure that by hand baw, with only the help of a regular non-scientific calculator.
Posted by sullivanct19a
Florida
Member since Oct 2015
5239 posts
Posted on 11/20/15 at 3:49 pm to
First, you should properly describe what in the world you're trying to do. That always helps.

Second, what do you mean 'graph into a bell curve'?

Third, you need to determine if those two samples are independent, then determine if they have equivalent variances, then determine the distribution of each.

Fourth, you need to look at the assumptions of the test or procedure you wish to do and whether the data satisfies them. Non-parametric tests are generally the way to go here, especially with small samples.

Fifth, it sounds like you're wanting one curve so you can plot a point on a the graph it rests on and compare it to another point, then maybe determine if those points are considered statistically different? If so, perhaps you should look at standardization. Repeat steps 3 and 4 in that context, if that's the case.
Posted by Epic Cajun
Lafayette, LA
Member since Feb 2013
32473 posts
Posted on 11/20/15 at 3:51 pm to
Plug that shite in SPSS baw.
Posted by anc
Member since Nov 2012
18073 posts
Posted on 11/20/15 at 3:53 pm to
This should work if you don't have SPSS

LINK
first pageprev pagePage 1 of 2Next pagelast page

Back to top
logoFollow TigerDroppings for LSU Football News
Follow us on Twitter, Facebook and Instagram to get the latest updates on LSU Football and Recruiting.

FacebookTwitterInstagram