Descriptive statistics
1. Central tendency
○ Mean, median, mode
2. Dispersion
○ Range, quantiles, interquartile range, variance, standard deviation
3. Shape
○ Skewness, kurtosis
4. Understand the meaning of the indicators
1) Central tendency
N
● ominal - can only use mode
● Ordinal - median, mode
● Interval - mean, median, mode
Level of Permissible operations Central tendency
measurement
Nominal Categorization Mode
Ordinal Ranking Median, mode
Interval Addition, subtraction Mean, median, mode
Ratio All arithmetic operations Mean, median, mode
2) Dispersion
R
● ange: the difference between the largest and smallest value
● Quartile: a type of quantile that divides the sorted dataset into 4 qual parts
○ Q1 - the median of the first half of the data
○ Q2 - the median of the full data set
○ Q3 - the median of the second half of the data
● Interquartile range (IQR) - Q3 to Q1
○ Usually only report IQR
● Why is IQR important
○ Median and range don't tell which class perform better
○ Ex:
1. Central tendency
○ Mean, median, mode
2. Dispersion
○ Range, quantiles, interquartile range, variance, standard deviation
3. Shape
○ Skewness, kurtosis
4. Understand the meaning of the indicators
1) Central tendency
N
● ominal - can only use mode
● Ordinal - median, mode
● Interval - mean, median, mode
Level of Permissible operations Central tendency
measurement
Nominal Categorization Mode
Ordinal Ranking Median, mode
Interval Addition, subtraction Mean, median, mode
Ratio All arithmetic operations Mean, median, mode
2) Dispersion
R
● ange: the difference between the largest and smallest value
● Quartile: a type of quantile that divides the sorted dataset into 4 qual parts
○ Q1 - the median of the first half of the data
○ Q2 - the median of the full data set
○ Q3 - the median of the second half of the data
● Interquartile range (IQR) - Q3 to Q1
○ Usually only report IQR
● Why is IQR important
○ Median and range don't tell which class perform better
○ Ex:
