Scatter Graphs & Correlation Flashcards
All 14 cards in this deck
What are the steps to plot bivariate data on a scatter graph?
e.g. a table gives a baby of age months with weight kg
- Take the pair of values for one item:
- Go across to on the horizontal (age) axis and up to on the vertical (weight) axis
- Mark with a cross, then repeat for every pair in the table
What does positive correlation on a scatter graph mean?
As one variable increases, the other variable also increases.
What does negative correlation on a scatter graph mean?
As one variable increases, the other variable decreases.
How do you describe the relationship shown by a scatter graph in context (not just naming the correlation)?
e.g. rainfall and hours of sunshine, showing negative correlation
Write a sentence using both variables: "As the amount of rainfall increases, the number of hours of sunshine decreases."
What is an outlier on a scatter graph, and how do you write it down?
A point that does not fit the trend of the other points; write it as coordinates, e.g. .
True or false? If two variables show strong correlation, one must be causing the other to change.
False. Correlation does not indicate causation — a third factor may explain both.
What is a line of best fit?
A single straight line drawn by eye that passes through the middle of the plotted points, with roughly as many points above it as below it.
What should you do with an outlier when drawing a line of best fit?
Ignore it — the line follows the trend of the other points.
True or false? A line of best fit must start at the origin.
False. It is drawn to fit the points, so it need not pass through the origin or through any particular point.
What are the steps to estimate a value from a scatter graph?
e.g. estimate the maths mark of a student who scored in science
- Draw the line of best fit
- Draw a line up from on the science axis to the line
- Read straight across to the maths axis — that reading is the estimated maths mark
What is interpolation on a scatter graph?
Using the line of best fit to estimate a value inside the range of the data plotted.
What must you comment about an estimate made from a value beyond the range of the plotted data (extrapolation)?
It may not be reliable, because the point is outside the range of the data and the trend may not continue.
What are the steps to decide whether a scatter graph supports a claim about two variables?
e.g. a weatherman says "Temperatures are higher on days when there is more sunshine."
- Name the correlation shown: positive correlation between hours of sunshine and temperature.
- Check whether the claim matches that trend: the claim says both increase together, which matches.
- Say yes or no and give the trend as the reason: yes — as the hours of sunshine increase, the temperature increases.
True or false? To answer "Does the scatter graph support the claim that older babies are heavier?" you only need to write "yes".
False. You must also give a reason referring to the correlation or trend, e.g. as age increases, weight increases (positive correlation).