  ## Self assessment:

Which of the following describe data in the form of frequencies:

• The number of pregnant women falling into the following categories: non-smoker, light smoker, heavy smoker.

• The percentage of pregnant women falling into the following categories: non-smoker, light smoker, heavy smoker.

• The mean number of cigarettes smoked per day by 20 pregnant women.

• The hair colour (measured as red, brown, black or blonde) of 100 people from Great Britain.

• The number of leaves on 20 geranium plants grown for three weeks in sunlight.

• The number of geranium plants grown in sunlight for three weeks with between 0-10 leaves, 11-20 leaves, or ≥21 leaves.

For each of the following features, say whether it applies to a two-way Chi square test:

• Parametric test.

• Nonparametric test.

• Numbers plugged into the formula for the statistic are frequencies.

• Numbers plugged into the formula for the statistic are percentages.

• Numbers plugged into the formula for the statistic are means and standard deviations.

• Compares an observed frequency distribution with one based on a specified theoretical distribution.

• Should not be used if one or more expected values are zero.

• Should be used with caution if more than one-fifth of the expected values are less than 5.

• Should not be used when an item contributes to more than one cell in the frequency table.

• Used on unrelated data.

• Used for related data.

• Uses a statistic called the 'Pearson Chi-square'.

• Uses a statistic called the 'Likelihood Chi-square'.

• Reject the null hypothesis if the X2 value calculated is less than the critical value of X2 looked up in a table.