MathBench > Statistical Tests

Chi-squared Tests

One small correction

The method I showed you on the last page was not quite right. For reasons that are difficult to explain without a degree in statistics, you need to SQUARE the raw deviation before dividing by the expected value. So we have the following sequence:

    Determine what you “expected” to see.
    Find out the difference between the observed and expected values (subtract)
    Square those differences
    Find out how big those squared differences are compared to what you expected (that is, divide)
    Add it all up. (This gives a value that we will call a chi-squared statistic.  Why "chi-squared" I hear you ask – this is another one of those instances when a degree in Statistics helps!)
arrow chi-square = sigma ( (o-e)^2) / e )
Place mouse on picture for more explanation

 

If the final chi-squared statistic is a big number, would this make you think that the data fits the model, or does not fit the model?

 

(To make this problem interactive, turn on javascript!)

I think I have the answer: Since the individual numbers you added
were deviations from the model predictions, a big chi-squared statistic means
that the data deviate a lot. In other words, the model is a bad fit.