lalala
The way I learn best is by explaining the concepts I'm trying to hammer into my brain to other people. My roommate is out of town, dear lj, so you shall be my victim!
I will now attempt to explain the process of calculating a Chi square. Fucking fear ye.
A chi square is summed up in the expression x squared= the sum of (observed frequency-expected frequency) squared/expected frequency.
The first thing you have to do is calculate what's called your table marginals- the sums of all the rows and columns. You use those to establish your expected counts- by multiplying row total by column total and dividing it by grand total. Once you've repeated that for every step in the table, you calculate [observerd-expected]squared/expected for each square.
You add up the sum of all these values across the cells. You calculate your degrees of freedom (row-1)x(column-1). You go to your handily attached Chi Square distribution level, find your degrees of freedom in the row, and follow it along till you find the chi square statistic you got. At that point, you look up and see what probability value column you're in or between. In the class I'm in, the commonly accepted value is smaller than 0.05. If it is at that significance level, then you're good!
Fuck, that was garbled and horrible. I have an exam in this tomorrow and I'm sick as a dooooog.
I will now attempt to explain the process of calculating a Chi square. Fucking fear ye.
A chi square is summed up in the expression x squared= the sum of (observed frequency-expected frequency) squared/expected frequency.
The first thing you have to do is calculate what's called your table marginals- the sums of all the rows and columns. You use those to establish your expected counts- by multiplying row total by column total and dividing it by grand total. Once you've repeated that for every step in the table, you calculate [observerd-expected]squared/expected for each square.
You add up the sum of all these values across the cells. You calculate your degrees of freedom (row-1)x(column-1). You go to your handily attached Chi Square distribution level, find your degrees of freedom in the row, and follow it along till you find the chi square statistic you got. At that point, you look up and see what probability value column you're in or between. In the class I'm in, the commonly accepted value is smaller than 0.05. If it is at that significance level, then you're good!
Fuck, that was garbled and horrible. I have an exam in this tomorrow and I'm sick as a dooooog.