English Deutsch Français Italiano Español Português 繁體中文 Bahasa Indonesia Tiếng Việt ภาษาไทย
All categories

I'm trying to do the students t-test, and i'm not sure what my degree of freedom is.

I'm comparing 2 sets of data, in one set of data i have 35 samples and in the other i have 31 samples. How do i know what my degree of freedom is??? PLEASE HELP??????

2006-12-30 03:53:24 · 4 answers · asked by me_girl_06 1 in Science & Mathematics Mathematics

4 answers

Your degree of freedom is the sum of the numbers of samples in each set, minus the number of sets you have.
Here, you haev 35 samples in one set, and 31 in the other, and you have two sets. So your df = 35 = 31 - 2 = 64

2006-12-30 04:52:33 · answer #1 · answered by Joni DaNerd 6 · 1 0

The degrees of freedom depends on what t-test you are performing.

If you are assuming equal variances, then you would use n1+n2-2 = 35+31-2 = 64.

However, if you are assuming unequal variances, then you would need to approximate the degrees of freedom. There are several ways to do it. Most statisicians believe that the Satterthwaite approximation is the best. That requires the use of the sample variances s1^2 and s2^2

df = (s1^1/n1 + s2^2/n2)^2/{(s1^2/n1)^2/(n1-1) + (s2^2/n2)^2/(n2-1)}

This estimate for the degrees of freedom might be hard to read in this format. Below is a link that has the equation in a better format. You would have to scroll down a little bit. It's short hand uses the Greek letter nu (ν)

If you are using a table, you would want to round this number down in all cases. So if you ended up with 51.87, you would use 51 degrees of freedom. However, on a computer, you should be able to use the decimal.

If you do not have the sample variances though, another approximation is to take the smaller of n1-1 and n2-1. So in this case it would be the smaller of 35-1 and 31-1, so 30.

Also keep in mind when you are doing this, the assumption of equal or nonequal varainces changes the calculation of the test statistic. The equal variances t-test would use a pooled sample variance; the unequal variance t-test would not. Be sure to know which one you are using.

2006-12-30 05:23:23 · answer #2 · answered by blahb31 6 · 1 0

Quantum Mechanics is all about the physics behind how atoms and molecules artwork, a level of freedom change into first developed even as physicists and chemists were attempting to comprehend the action of an electron in an atom. The degree of freedom objectives at attempting to comprehend how electrons are spaced out round an atom as a thanks to form the outer valence earrings.

2016-12-01 08:16:29 · answer #3 · answered by Anonymous · 0 0

Degrees of Freedom

1). Statistics. any of the unrestricted, independent rendom variables that constitute a statistis

2). Physics.
. ..a. Any of the minimun number of coordinates required to specify completly the motion of a mechinical system.

....b. Any of the independent thermodyamic variables, such as pressure, temperature, or composition required to specify a system with a given number of phases and comoponents.

Click on the URL below for additional information concernint degres of freedom

www.answers.com/topic/degrees-of-freedom

- - - - - - -s-

2006-12-30 04:23:29 · answer #4 · answered by SAMUEL D 7 · 0 2

fedest.com, questions and answers