Glaciated Landscape Skills
- Geographical skills are working skills essential to developing a synoptic approach to answering questions but also observing the 'bigger picture' in geography
- It is important to be confident with a mixture of numerical quantitative skills and qualitative written communication skills
- Many of the skills are already outlined elsewhere in the revision notes
Understanding data
- In data analysis, variables are the amounts that have been measured - the number of drumlins or the size of erratics
- For each variable, a value is noted against each sample - drumlin a, drumlin b etc
- The data set is the collection of total values and is then analysed further
- The sample size and type of data, influence the choice of statistical test to use
- Spearman's rank would be used to test for correlation between two variables etc.
- Types of data can be grouped into:
- Nominal
- Ordinal
- Interval
- Ratio
- Nominal
- Also known as categorical data, its main purpose is to classify or group information
- Data is organised into distinct categories, but the categories have no numerical or quantitative meaning
- Examples of categories can include things such as dog, cat, blue, male and female etc.
- Or they can be labelled with numerical codes such as 1 for glacial 2 for periglacial etc.
- They can be summarised using percentages or frequencies e.g. 40% of periglacial pingos are open system
- Remember they have no order or mathematical relationship and performing statistical analysis is pointless
- Ordinal
- Ordinal data is a type of data that can be ordered or ranked into categories
- Examples include:
- Primary school, secondary school, sixth form or college, and university
- The categories show a clear progression/order on levels of possible education
- Very satisfied, satisfied, neutral, dissatisfied and very dissatisfied
- These categories show levels of satisfaction, but intervals between them may not be equal
- Primary school, secondary school, sixth form or college, and university
- Ordinal data allows ranking and comparing of vales, but doesn't provide information on size of the differences between the categories
- Statistical analysis and interpretation can be used such as calculating median, mode, or Spearman's rank; but ordinal data doesn't allow for mathematical calculations such as adding or subtracting
- Interval
- Interval data is the same as ordinal data, but the intervals between the categories is constant
- For example pH values of water; scale of temperature or time on a clock
- Interval data is therefore, a more precise measurement compared to nominal and ordinal data, but it does not include a true zero point
- Interval data allows for various statistical operations such as calculating mean, median mode, standard deviations, and conducting tests such as t and u-tests
- Interval data is the same as ordinal data, but the intervals between the categories is constant
- Ratio
- Unlike interval data, ratio data includes a true zero point, which allows for a more comprehensive analysis of the data
- The difference between any two consecutive values is the same throughout the entire range of the data
- The true zero point indicates a total absence of the value at that point
- This makes ratio data the highest level of measurement in terms of precision and mathematical operations
- Examples include:
- Weight: an object is weighed in kilograms and grams
- If a value of zero is recorded it means there was no weight to the object
- Distance: the distance travelled by a glacier for one day is recorded in centimetres
- A recorded value of zero indicates that the glacier travelled no distance
- Ratio data allows for a wide range of mathematical operations, including addition, subtraction, multiplication, and division
- Statistical analysis techniques applicable to ratio data include measures of central tendency (mean, median, mode), measures of dispersion (range, variance, standard deviation), and parametric tests
Evaluative skills
- Being asked to assess the impacts or causes of a range of factors is a common exam question
- When deciding if something is significant consider four things:
- Time - how long will it take for a strategy or impact to take effect?
- Scale - how many people will be affected?
- Cost - What will the cost be and to whom?
- A cost can be human or environmental - what benefits the environment may come at a cost to human activity
- Rather than considering whether something is expensive or cheap, think about whether it is worth the cost because of the benefits it will create
- It is important to remember that just because something is expensive that doesn’t mean it is the worst option
- Ethics - Does the strategy ensure dignity for local people and other stakeholders?
- This will allow for a well-rounded and substantiated argument in 9 mark and 20 mark questions
Photo analysis
- This is an important observational skill
- Look at the foreground, midground and background
- Consider the impact of the colours
- Think about what has not been included in the picture, what might be just out of frame?
Percentage and percentage change
- To give the amount A as a percentage of sample B, divide A by B and multiply by 100
- In 2020, 25 out of 360 homes in Catland were burgled
- What is the percentage (to the nearest whole number) of homes burgled?
- A percentage change shows by how much something has either increased or decreased
-
- In 2021 only 21 houses were burgled. What is the percentage change in Catland?
-
- There has been a decrease of 16% in the rate of burglaries in the Catland area
- Remember that a positive figure shows an increase but a negative is a decrease
Mann-Whitney U test
- Also known as the Wilcoxon rank-sum test, it is a nonparametric test used to compare two independent groups, population or samples to determine if there is a significant difference between their distributions
- It makes no assumptions of the data being normally distributed
- The test works by assigning ranks to the observations from both groups combined and considers all the values as a single pool
- Then, it compares the sums of the ranks for each group
- The test looks at whether the distributions of the two groups differ significantly based on the ranks
- A general outline of how the Mann-Whitney U test works:
- Combine the data from both groups into a single dataset
- Rank the combined data, assigning a rank to each observation (identical data are given an average rank)
- Then calculate the sum of the ranks for each group
- Use the U statistic (the smaller of the two sums of ranks) to determine the test statistic
- Compare the test statistic to the critical values in the Mann-Whitney U distribution or use a significance level to determine if the difference between the groups is statistically significant
- If the p-value is below the chosen significance level (often 0.05), the test concludes that there is a significant difference between the groups
- The Mann-Whitney U test does not make any specific distribution for the data and is effective in comparing ordinal or continuous variables between two independent groups
Worked example
- The following data was gathered, showing how questionnaire participants rated the quality of their service provision for two ski resorts in the Swiss Alps
- Ski resort A
- Ski resort B
- Ratings were given on a 0 to 5 scale
Participant No. | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 |
Ski resort A | 3 | 1 | 2 | 2 | 0 | 2 | 3 | 1 | 0 | 1 | 1 | 2 | 0 | 1 | 3 |
Ski resort B |
3 | 2 | 3 | 4 | 2 | 5 | 3 | 4 | 1 | 4 | 2 | 4 | 4 | 1 | 5 |
- Combine and sort the values of both samples into numerical order
- Keep a note of which sample refers to which ski resort
- If there are two of the same value, put ski resort A first - it doesn't really matter so long as you are consistent
A | A | A | A | A | A | A | A | B | B | A | A | A | A | B |
0 | 0 | 0 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 2 | 2 | 2 | 2 | 2 |
B | B | A | A | A | B | B | B | B | B | B | B | B | B | B |
2 | 2 | 3 | 3 | 3 | 3 | 3 | 3 | 4 | 4 | 4 | 4 | 4 | 5 | 5 |
- For every value for ski resort B, count how many ski resort A values comes before it in the list, then add these together to get a U₁ value
A | A | A | A | A | A | A | A | B | B | A | A | A | A | B |
0 | 0 | 0 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 2 | 2 | 2 | 2 | 2 |
8 | 8 | 12 |
B | B | A | A | A | B | B | B | B | B | B | B | B | B | B |
2 | 2 | 3 | 3 | 3 | 3 | 3 | 3 | 4 | 4 | 4 | 4 | 4 | 5 | 5 |
12 | 12 | 15 | 15 | 15 | 15 | 15 | 15 | 15 | 15 | 15 | 15 |
- U₁ = 8 + 8 + 12+ 12+ 12 + 15+ 15+ 15+ 15+ 15+ 15+ 15+ 15+ 15
- U₁ = 202
- Now repeat the process to count how many ski resort B vales come before A in the list, add together to get U₂
A | A | A | A | A | A | A | A | B | B | A | A | A | A | B |
0 | 0 | 0 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 2 | 2 | 2 | 2 | 2 |
0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 2 | 2 | 2 | 2 |
B | B | A | A | A | B | B | B | B | B | B | B | B | B | B |
2 | 2 | 3 | 3 | 3 | 3 | 3 | 3 | 4 | 4 | 4 | 4 | 4 | 5 | 5 |
5 | 5 | 5 |
- U₂ = 0 + 0 + 0 + 0 + 0 + 0 + 0 + 0 + 2+ 2+ 2+ 2 + 5 + 5+ 5
- U₂ = 23
- Using the critical value table, you can see if this result is significant or not - a copy will be given to you in the exam
- The extract below gives a critical value to 5% significance
n2 | 13 | 14 | 15 | 16 | |
n1 | |||||
13 | 45 | 50 | 54 | 59 | |
14 | 50 | 55 | 59 | 64 | |
15 | 54 | 59 | 64 | 70 | |
16 | 59 | 64 | 70 | 75 |
- The size of each sample is indicated by ?1 and ?2 (in this instance the samples size is the same for both resorts
- Both ?1 and ?2 are 15, giving a critical value of 64
- To determine significance, and not due to chance, the smaller U value must be equal or less than the table's critical value
- In this instance, U₂ = 23 and is therefore, less than the critical value of 64
- We can state with a 95% certainty that ski resort A has been rated significantly different to ski resort B by respondents of the questionnaire
- To find a reason why this might be, would be the next stage in an investigation