Skills

Tips: 

Use Ctrl + to zoom into the images

One of the AS geography exams is a skilled based exam. It is a 1 hour exam paper and is made up of structured skill and generic fieldwork questions - it accounts for 30% of your final AS mark. Here is a checklist of everything that you need to be able to do for it:

(I've left in examples of how to fill this table in)

Here, for some extra help, are descriptions and examples for each skill you need to know:

 

 

Friday 7th December 2014

 

Scattergraphs can be used to compare two variables, during the skills exam make sure you remember to do these things:

• Label your axes (the quantitative data on the x-axis)
• Create an appropriate scale
• Plot data very accurately
• Draw a line of best fit
• Title the graph

The correlation of the data is the connection between the variables, your line of best fit will show you what type your data has:

 

 

However, scattergraphs have limitations - who knew!

1. It is often difficult to plot lines of best fit by eye
2. Not all correlations are linear, they appear as curves making it difficult to fit a regression line
3. What might appear as a strong correlation could have been caused by chance (statistics can be misleading) or by a third variable that has not been considered

Worked example:

 

 

 

To find the strength of a relationship with a numerical value, we can use the Spearman Rank Correlation Coefficient. This can only be used if there are two sets of data with at least 10 pairs but less that 30. Using the data from the scattergraph example, here is how you work it out:

 

 

 

Don't worry you don't have to learn the formula, just how to use it.

 

Strengths of Spearman's:

• It gives you objective data
• It demonstrates a clear relationship
• You can see the accuracy of the data - whether it is significant or a coincidence
• It is less sensitive to anomalies because the data is ranked

Weaknesses of Spearman's:

• It does not tell you the type of relationship that exists e.g. a casual link
• Too many of the same rankings can effect the validity
• Human error can occur in calculations

Instead of Spearman's we can use the Pearson Product-Moment Correlation Coefficient. This is a more accurate measure because it does not use rankings so it is able to show the differences between values.