Maths + Stats Flashcards
Bivariate Data
Data with 2 variables
Basically scatter graphs
What can we plot?
Socio-economic data - global and local Physical data - rivers and hydrology Fieldwork data - beefed up in new curriculum
What can students use Gapminder for?
Students can use the tools to create can scatter graph easily.
Students can use it for homework.
Positive correlation
As one goes up so does the other.
Negative correlation
As one increases the other decreases.
No correlation
No relationship, connection, or interdependence between the two variables.
How to describe the strengths of the correlation?
Strong vs Weak
Trend lines
- Drawn on by hand
- Illustrates general pattern that the graph shows
- Bigger picture of data
Line of Best Fit
- In exam they will be expected to draw this on to a graph.
- Shows the pattern
strong / weak positive/negative correlation. - Same number of points above and below the line
- Doesnt need to go through origin
Clusters
Are their clusters of data?
- What does that show?
e. g. Low/emerging/high income countries clustered together?
Anomalies
Data points not in the clusters / far from the other data.
Think why?
- What country is it?
- What are they doing differently?
What to remember about correlation?
Correlation does not equal causation.
Interpolate
Finding the value between two other values.
Extrapolate
Beyond what is on the graft and make a prediction.
Central Tendency
The tendency for the values of a random variable to cluster round its mean, mode, or median.
Spread
That is a measure the variability of values in a sample or the range.
Cumulative Frequency
This is a running total of frequency up to that point.
- Add the numbers of frequency up as you go through the data/table
- Can present as a cumulative frequency graph
Dispersion Diagrams
- Simple but students struggle.
- Show spread or range within one or possibly 2 sets of data.
- Visual way to see the degrees of dispersion or clustering of data.
How to plot a dispersion diagrams?
- Work out axis
For each time in the sample: - Plot where it occur on your Y axis
- If they have the same data they will appear in a row at that number
How can dispersion diagrams help us?
- Show us range
- Show us cluster
- Easy to see/calculate mean, mode, median
- Starting point for calculating inter-quartile range
Range
largest value minus smallest value
NB: It is the range of the data values, not of the frequencies.
A larger range means the data is less consistent. A smaller range means the data is
more consistent.
Mean
Add up all the values and divide by the number of values in the sample.
Mode
Most frequently occuring value.