test of association: chi - squared

Question 1

why should we choose chi - square as a statistical test

Answer 1

chi - square is a test of difference or association

the data is nominal and recorded as a frequency count of the categories

Question 2

what would be a worked example for the chi - square

Answer 2

we are looking for a DIFFERENCE in the ability to decentre in children aged 5 and children aged 8

there are two INDEPENDAND GROUPS of children which means that the design is unrelated

Finally, the level of measurement is nominal as data is collected in the form of frequencies in two categories: ability too decentre or not

Question 3

what are the hypotheses for the worked example

Answer 3

ALTERNATIVE HYPOTHESES: more 8yr old than 5yr are able to select a card that represents a perspective different from their own (directional one - tailed)

NULL HYPOTHESIS: there is no difference between the number of 5yr olds and 8yr old who can select a card that presents a perspective difference form their own

Question 4

what are the steps involved in chi - square

Answer 4

1. step 1: 2x2 contingency table
2. step 2: the table of expected frequencies
3. step 3: working out the value of x
4. step 4: the calculated and critical values

Question 5

how do you work out the calculated and critical values

Answer 5

To find the critical value, you need to calculate the degrees of freedom (df) by multiplying:
(rows-1)x(columns-1)=1
the rows and columns refers to the contingency table

Question 6

how do we work out if the results are significant or not

Answer 6

if the calculated value of x is more than the critical value then we can reject the null hypothesis and accept the alternative hypothesis