Chi Square

Data Analysis through Statistics

quaLitative = what they’re like or categories

quaNitative = numbers

Chi Square

Problem = want to know everything about populations

Solution = take sample to represent the population

Purpose = take our sample data and process to see if it is significant or by chance to draw a prediction about the population

Process:

Step 1: Identify IV and DV

Step 2: Create a null hypothesis (Ho) and an alternative hypothesis (Ha)

• Ho = There is no difference or effect on the DV due to IV.

• Ha = There is a difference or effect on the DV due to IV.

Step 3: Determine the observed (o) and expected (e) values.

Step 4: Calculate X²

Step 5: Degree of freedom (df = n - 1); where n = # of categories in IV

Step 6: Compare X² to critical value (p)

• 95% confidence = 0.05 p-value

• 99% confidence = 0.01 p-value

Step 7: Draw a conclusion

• If X² > p then reject Ho. Data suggests there is a difference/effect on…

• If X² < p then fail to reject Ho. Data suggests there is no difference/effect on…

Example 1: Does age impact how many times heads is flipped?

Step 1:

• IV = age

• DV = # of heads flipped

Step 2:

• Ho = There is no effect on the # of heads flipped due to age.

• Ha = There is an effect on the # of heads flipped due to age.

Step 3:

Expected Date (e)

• 50%/5 → 1.5

Step 4:

X² = (3 - 5)² / 5 +(2 - 1.5)² / 1.5

X² = 0.8 + 0.17

X² = 0.97

Step 5: df - 1

Step 6:

Step 7: X² < p

Fail to reject null null hypothesis Ho. Our data suggests there is no effect on head flipping due to age.