Central tendency and dispersion
In a minute
The , and mode describe the centre of a data set; the IQR and describe its spread. You find them with your GDC, estimate the mean of grouped data, and see what shifting or scaling the data does.
You'll be able to
- Find the mean, median and mode of a list or frequency table, and the mean both by formula and with your GDC.
- Estimate the mean of grouped data from mid-interval values, and find the .
- Find and the IQR with your GDC, and explain why a by-hand quartile can differ.
- Find the standard deviation and with your GDC, and interpret them in context.
- Say how adding to or multiplying every value changes the mean and standard deviation.
Before you start3 quick questions on what this builds on
Revise first: 1.2 Arithmetic sequences and series, 1.6 Rounding, bounds and percentage error, 4.1 Sampling, data quality and outliers and 4.2 Data presentation
Mean, median and mode
A podcast's last episodes lasted minutes. How long is a typical episode? Three different averages answer that.
Try it first, then reveal one step at a time.
- Mean: the total is and , so minutes.
- Median: in order, . With , average the th and th values: minutes.
- Mode: appears three times, more than any other length.
- Comment: the mean and median are close, so either describes a typical episode. The mode, , is the shortest length, so it doesn't.
In a frequency table, each value occurs times, so it adds to the total. This formula is in the booklet.
Here is the number of different values, and is the total frequency, not the number of columns.
Try it first, then reveal one step at a time.
- (a) , so nights.
- On the GDC, enter the values to in one list and the frequencies in another, then run one-variable statistics. Check , then read .
- (b) With , the median is the mean of the th and st values. The first guests stayed night, and guests to stayed , so both are .
- The median is nights. The mode is night, the value with the highest frequency, .
- The mean is above the median because a few long stays pull it up.
Grouped data: estimated mean and modal class
Sometimes you only know which each value lies in. Then you can only estimate the mean.
Treat every value in a class as if it sat at the class's mid-interval value: for the class .
Use the booklet formula with these as the . The result is an estimate, because the actual values are unknown.
Try it first, then reveal one step at a time.
- (a) The classes all have width , and the highest frequency is . So the modal class is .
- (b) The mid-interval values are .
- , so the estimate is minutes.
- (c) Only the class of each wait is known, so each wait is replaced by the middle of its class.
Try it first, then reveal one step at a time.
- Write down the lists you enter: mid-interval values and frequencies . That line earns the method mark.
- Run one-variable statistics with the frequency list attached. Check that : if it shows , the frequencies are missing.
- Read , and round only now: about cm.
- Interpret: the mean height of the sunflowers in the trial is estimated at about cm, inside the modal class, .
Quartiles and the IQR
A measure of dispersion says how spread out the data are. The is the simplest, but one extreme value can change it a lot.
The cut the ordered data into four roughly equal parts: about a quarter of the values lie below the lower quartile , and about three quarters below the upper quartile .
1. To find the quartiles by hand, put the data in order and find the median.
2. Split the data into a lower half and an upper half. When is odd, leave the median out of both halves.
3. is the median of the lower half, and is the median of the upper half.
The (IQR) is the spread of the middle half of the data, so a few extreme values have little effect on it.
Try it first, then reveal one step at a time.
- (a) In order: . The median is the th value, ; leave it out of both halves.
- Lower half , so . Upper half , so .
- seconds.
- (b) Keeping , the halves have values each: and , so her IQR is seconds.
- Both are correct for their method. Your GDC uses the method in (a).
Several methods for quartiles are in use, so a by-hand quartile and a technology quartile can differ slightly. Each is right for its method.
In the exam, find quartiles of with your GDC. In their usual settings, the three models in these notes leave the median out when is odd, as in (a).
Try it first, then reveal one step at a time.
- (a) Enter to in one list and the frequencies in another. Run one-variable statistics and check that .
- Scroll down to read and , so books.
- Check by counting: the lower half is the first values, and its th and th values are both .
- (b) The middle half of the students read between and books: a spread of books.
Standard deviation and variance
The IQR uses only the quartiles. The standard deviation uses every value: it measures how far the values typically lie from the mean.
In the exam you find with your GDC only. One small case by hand shows what it measures: the variance is the mean squared distance from the mean, .
Try it first, then reveal one step at a time.
- Mean: hours.
- Distances from the mean: . They add up to , so square them first.
- Squares: , total . So .
- hours: a typical night is about hours from the mean.
Your GDC shows two standard deviations. For the standard deviation of a data set, use , which divides by as above. Ignore (or ), which divides by .
Try it first, then reveal one step at a time.
- Enter the times in a list and run one-variable statistics. Check that .
- (a) minutes.
- (b) , so minutes. Keep the full value for (c).
- (c) minutes². Squaring the rounded would give instead.
- (d) The friends' times typically lie about minutes from the mean of minutes.
Adding to or scaling every value
Add to every value. The moves up by , but the gaps between values don't change, so neither does the .
Double every value instead. Every value and every gap doubles, so the mean and standard deviation both double.
So subtracting from every value lowers the mean by and leaves unchanged. The median changes like the mean, and so do the quartiles when ; the IQR and range change like .
Try it first, then reveal one step at a time.
- Every score has subtracted: .
- Mean: , so the golfer averages strokes over par.
- Standard deviation: subtracting a constant doesn't change any gap, so stays .
Try it first, then reveal one step at a time.
- Each price becomes .
- Mean: dollars.
- Standard deviation: the has no effect, so dollars.
- Adding to too would give , which is wrong.
Common mistakes
- Quoting from the GDC as the standard deviation of a data set. Use , which divides by .
- Dividing by the number of columns or classes. Divide by the total frequency .
- Leaving the frequency list off on the GDC. Check that it shows the total frequency as .
- Adding the constant to the standard deviation. Under , becomes ; the has no effect.
- Squaring a rounded standard deviation to get the variance. Square the GDC's full value.
- Interpreting a standard deviation without mentioning spread. Say how spread out or consistent the values are, in context.
Exam tips
- Write down the mode or modal class: no working needed. For Find the mean, show with numbers in, or your GDC lists, for the method mark.
- Interpret a standard deviation by comparing spread in context, such as less variation in waiting times. Comparing averages alone earns no mark.
- Explain why a grouped mean is an estimate: the actual values in each class are unknown, so mid-interval values stand in for them.
Key takeaways
Practise this
16 past-paper questions on Statistical measures and frequency tables in the AI HL bank, each with a mark scheme and a worked explanation.