8. Scatter Diagrams
The answer to this question is similar to the one for: why do we bother working out averages
Why do we bother with Statistical Diagrams?
and measures of spread?.
We live in a world jam-packed full of statistics, and if we were forced to look at all the facts
and figures in their raw, untreated form, not only would we probably not be able to make any sense out of them, but there is also a very good chance our heads would explode.
Statistical Diagrams if they are done properly - present those figures in a clear, concise,
visually pleasing way, allowing us to make some sense out of the figures, summarise them, and compare them to other sets of data.
Example 1
Example 2
Example 3
Example 4
The Answers
1) This person has messed up their negative numbers. Remember, scales must go from smallest to biggest, from left to right, and down to up. 2) Classic mistake. Numbers must go on the lines, not between the spaces!
3) How many times have I seen this? The spaces around the centre (origin) are not equal. Look at the gap between 2 and 2. Deary me! 4) Inconsistent scales! Notice the numbers go up by 1 in the negatives and then 2 in the positives! Note: Another mistake in all of the diagrams is that the x and y axes are not labelled! Big Example Below is a table showing the number of pupils who fail to hand in their maths homework each day, and the minutes of yoga I need to do to calm myself down
Pupils missing homework Minutes of yoga 3 10 5 12 2 9 10 25 2 8 0 3 4 15 8 20 15 26 6 10 1 7 4 10
Draw a scatter diagram to show the information, add a line of best fit, and comment on the correlation
30 28 26 24 22 20 18 16 14 12 10 8 6 4 2
minutes of yoga
30 28 26 24 22 20 18 16 14 12 10 8 6 4 2
minutes of yoga
4. Correlation
The most important use of scatter diagrams is to determine the type (if any) of correlation between two variables Correlation is just a posh word for relationship. There are two categories of correlation that you need to be familiar with:
DIRECTION Positive line slopes upwards As one variable increases, so does the other Negative line slopes downwards As one variable increases, the other decreases No correlation line is close to horizontal No relationship between the variables STRENGTH Strong dots are close to each other Weak dots are far apart Tip: When deciding on the strength of correlation, I have a little rule: the longer it takes me to decide where to draw the line of best fit, the weaker the correlation
Strong negative
Weak Positive
No Correlation
Looking at our example, I would say there is a fairly strong, positive correlation. This is no surprise, because as the number of missing homeworks increases, so to does my need for yoga!
Question 1: If 7 pupils forget to hand in their homework, how many minutes of yoga might Mr Barton do? Following the red line up and across gives 16 minutes
Question 2: If Mr Barton does 28 minutes of yoga, how many pupils might have forgotten their homework? Following the purple line across and down 14.5 pupils