# chi square distribution, degrees of freedom, etc.

#### German35m

##### New Member
Hello,

I'm trying to make sense of something somebody said (I hope you can tell this is not a typical homework question?) and bc I couldnt seem to figure it out myself now came here.

If you're talking about a chi square Distribution, and used a scale from 1-10 and you assigned a score of 9.5 or 9.8 to something, how likely or unlikely would that observation be? an individual's height for example. 99.9%? 99.99%? 99.999% 99.9999%. I assume there are probability tables for this?

I guess you Need to know the degrees of freedom (which I guess are dependent on the number of observations, right?). I'd choose 1 df, first (to figure out the lowest possible probability?).

I hope I made any sense?

PS: I did read that only those who Show effort will receive help - and I did try to look it up (I learned that the df depend on the number of observations, I think - it's been a while), but really didn't understand it. And it isn't a homework question, either (Can you tell or do I come across like a Student?).

#### Dason

I don't really understand your question

#### German35m

##### New Member
thanks for the Reply, already. I'll try to think about it again and explain it better afterwards (hopefully). Sorry if it didn't make any sense!

#### katxt

##### Member
The important point is that the chi squared distribution applies only to count data. It can't be applied to heights, for instance. You count some things and then think "How unlikely is that?" For instance you catch 100 crabs and 66 are female. You say "I expected 50:50. How unlikely is it to get as bad as a 66:34 split?" This is a chi square question because it deals with count data. The degrees of freedom in this problem is 1, because the df is 1 less than the number of groups.

#### Dason

The Chi-square's most well known use is for count data but don't say that that is its only use. I don't want to say you're ignorant but never assume that a distribution only has a single use.

#### j58

##### Active Member
@German35m - Unfortunately, your question really doesn't make sense. What does the 1-10 scale have to do with the chi-squared distribution, and what does the scale have to do with an individual's height?

#### German35m

##### New Member
I just realized that obviously the 1-10 scale probably doesn't work for 'height'.

#### j58

##### Active Member
How do the measurements of whatever you're measuring relate to a chi-squared distribution?

Last edited: