Histograms, stem and leaf diagrams and box plots
Studying this course? Keep your progress in one place.
Key facts
- In a histogram the area of a bar is the frequency, so plot frequency density .
- A stem and leaf diagram needs ordered leaves and a key; the median is the th value.
- A box plot shows the minimum, , median, and maximum; IQR .
- If the data are positively skewed; if less, negatively skewed.
- Compare distributions with a measure of location and a measure of spread, in context.
Histograms and frequency density
A histogram shows continuous grouped data: the area of each bar is the frequency, so the vertical axis shows frequency density.
A histogram displays continuous grouped data. Because class widths can differ, the area of each bar (not its height) is proportional to the frequency, so the vertical axis shows frequency density.
Bar heights are in proportion to frequency densities.
To estimate a frequency in part of a class, assume the data are evenly spread within the class.
- Frequency density
Worked example
60 calls have classes 0 to 2 (12 calls), 2 to 5 (21), 5 to 10 (15) and 10 to 20 (12). Find the frequency densities and estimate the number of calls between 5 and 8 minutes.
- 1
Frequency densities: , , , .
- 2
Assume the data are evenly spread within the class.
- 3
Between 5 and 8 minutes: width , density , so frequency .
In a histogram, what is proportional to the frequency of a class?
Stem and leaf diagrams
A stem and leaf diagram shows every data value in order; the median and quartiles are read from positions.
A stem and leaf diagram shows every data value. The stem is the leading digit(s) and each leaf is the last digit.
Leaves are written in order and a key is always given, for example 4|2 means 42.
For values: the median is the th value; is at position and at .
If the position is a whole number take the mean of that value and the next; otherwise round up.
- Median position
- Lower quartile position
- Upper quartile position
Worked example
A stem and leaf diagram shows 15 scores. Find the positions of the median and quartiles.
- 1
Median: , so the 8th value.
- 2
: , round up to the 4th value.
- 3
: , round up to the 12th value.
Which of these must a stem and leaf diagram always have?
Back-to-back stem and leaf diagrams
A back-to-back diagram puts two data sets on one stem, with leaves outwards on each side; find each side's median and quartiles separately.
A back-to-back diagram compares two data sets on one common stem, with leaves of one set to the left (reading outwards from the stem) and the other to the right.
Each side is ordered with its own key. You find the median and quartiles of each side separately.
Check both keys and read each side outwards from the stem.
- 1
Check both keys
each side has its own
- 2
Read outwards
leaves run away from the stem
- 3
Order each side
leaves in increasing order
- 4
Summarise each side
median, quartiles, IQR
Worked example
12 girls' times are 38, 40, 42, 44, 45, 46, 49, 51, 53, 55, 57, 62. Find the median, quartiles and IQR.
- 1
Median: th value, so .
- 2
: position , a whole number, so .
- 3
: position , so .
- 4
IQR .
12 values are in order. How do you find ?
Box plots and skewness
A box plot shows the five-number summary; comparing with tells you the skew.
A box plot shows the five-number summary: minimum, , median (), and maximum.
The box spans to with a line at the median, and whiskers extend to the extremes. The interquartile range is and the range is maximum minimum. Outliers, where identified, are plotted separately as crosses.
Roughly 25% of values lie in each of the four sections, so in a sample of 40, about 10 values are above .
Positive skew
Symmetrical
Negative skew
- Interquartile range
- Rangemaximum minimum
A distribution has , and . How is it skewed?
Comparing distributions
Make at least two comments in context: one about location (median) and one about spread (IQR or range), using numbers from both.
To compare two distributions, make at least two comments, each in the context of the question: one about location (median) and one about spread (IQR or range). Use numbers from both distributions.
For example, has median 47 and IQR 14; has median 45 and IQR 25. So 's typical mark is slightly higher and its marks are more consistent.
Finish with a conclusion if asked which performed better, and avoid vague words such as 'better' without evidence.
| Distribution X | Distribution Y | |
|---|---|---|
| Median (location) | 45 | 47 |
| IQR (spread) | 25 | 14 |
Distribution X
- Median (location):
- 45
- IQR (spread):
- 25
Distribution Y
- Median (location):
- 47
- IQR (spread):
- 14
Which pair of measures should you use to compare two distributions?
Try an exam question
The lengths of phone calls are summarised: – minutes, calls; – minutes, calls; – minutes, calls; – minutes, calls. A histogram is drawn. (a) Find the frequency density for the – minute class. (b) Estimate the number of calls between and minutes, stating an assumption.
[4 marks]
- [1]Frequency density for – is .
- [1]Frequency density for – is .
- [1]Calls between and minutes .
- [1]Assumes the calls are evenly spread within the class.
That's the notes covered.
Carry on to the next subtopic.
Try a question
Find the frequency density of the class 5 to 10 minutes.
1 mark, marked against the mark scheme
Find the frequency density of the class 5 to 10 minutes.
1 mark, marked against the mark scheme