Chi squared (statistics problem)

[iamdjdz]

Hi all,

can anyone help me, I've an exercise to do for geography and I need a little help, the problem is I don't understand Chi squared testing and have read and re-read it loads of times. Anyway I have come up with an answer for one part of the exercise which is, "This means the value for chi squared is greater than the critical value of chi squared, 46.547 > 9.48773
so there is a significant difference between the observed data and that of a random expectation.
The H0 is rejected and the H1 is accepted."
Now I kind of see that as the two numbers are vastly different, 9.48.. being the value suiable from the Chi square table. The question is for the second part, I have the same table value, 9.48773 but a clser value 11.739. I just want to know if there is a significant difference or not and whether the null hypothesis can be rejected as before, or what should I say basically...

Long shot I know but normally this forums is handy for problems so umm anyone please...?

Got till 1pm.

Thanks

 

[Carzy]

If the value is greater than the critical value then there's always significance enough to reject the null hypothesis, doesn't matter whether it's over by 10 million or 0.1. Once you cross that line then it creates enough doubt with your model to need re-evaluating. And the difference of around 2 between the 11 and 9 values is actually quite a big difference a lot of the time in statistics...

 

[iamdjdz]

Ahh thanks very much, I wasn't sure if it was values over by a certain other value or all values over. I shall now be able to write my statement about the hypotheses and the write up correctly Thanks

 

[Dark88]

The higher the number over the critical value then the lower your P value, i.e the chance of your result happening at random. The crit value you've probably looked up is for 5%, there are other critical values for lower and higher percentages so you can determine how significant your results are.