Introduction: Turning Data into Discovery

You’ve been out in the field, braved the weather, and collected a pile of numbers and notes. Now comes the exciting part: making sense of it all! Analysis and presentation is where you turn raw data into a geographical story. Whether you were measuring pebble sizes on a beach (Topic 2B) or interviewing residents about regeneration (Topic 4A), this stage allows you to prove your hypotheses and draw valid conclusions.

Don't worry if you find the "maths bit" or the "graphs bit" intimidating. This guide breaks down exactly what the Pearson Edexcel AS Level expects you to know, from simple charts to more advanced statistical tests.

Note: This chapter focuses on what to do with data once you have it. For how to design your enquiry or how to actually collect the data, see the chapters on "Fieldwork enquiry design" and "Primary data collection methods."

1. Presenting Your Data: Making it Visual

The goal of data presentation is to make patterns easy to see. Choosing the right method is vital—using the wrong graph is like trying to eat soup with a fork!

Mapping Your Findings

  • Dot Maps: These use dots to show the presence or quantity of a feature. They are excellent for showing spatial density (e.g., the number of independent shops in a regenerating town centre).
  • GIS (Geographic Information Systems): This is the "big data" way to present fieldwork. You can overlay different layers of information on a digital map, such as placing your own flood risk data over a base map of local housing.

Specialised Geographical Diagrams

  • Kite Diagrams: These are perfect for showing changes along a transect (a line across an area). In coastal geography (Topic 2B), you might use one to show how vegetation types change as you move from the shoreline across sand dunes. The "width" of the kite shows how much of a species is present.
  • Dispersion Diagrams: These show how "spread out" your data is. If you are comparing pebble sizes at two different locations, a dispersion diagram lets you see the range and the clusters of data points easily.

Choosing Your Scales

  • Linear Scales: These are your standard scales where each unit is equal (e.g., \(1, 2, 3, 4, ...\)).
  • Logarithmic Scales: These are used when your data covers a massive range (e.g., from \(1\) to \(10,000\)). Each step on the axis increases by a power of ten (\(1, 10, 100, 1000\)). This is useful for complex data sets where a linear scale would make small values invisible.

Quick Tip: Always remember the TULIPS rule for graphs: Title, Units, Labelled axes, Interval (consistent scale), Plotted correctly, and Source.

2. Quantitative Analysis: The Power of Numbers

Quantitative analysis involves using statistics to describe your data and test your ideas. You will need a calculator for these in your exam!

Descriptive Statistics (Central Tendency and Dispersion)

These describe the "average" and the "spread" of your data:

  • Mean: The average (sum of all values divided by the number of values).
  • Median: The middle value when data is in order.
  • Mode: The most common value.
  • Range: The difference between the highest and lowest value.
  • Interquartile Range (IQR): The range of the middle \(50\%\) of the data, which ignores extreme outliers.

Measuring Inequality (The Lorenz Curve and Gini Coefficient)

If your fieldwork was about Regenerating Places (Topic 4A) or Diverse Places (Topic 4B), you might need to analyse inequality.

  • Lorenz Curve: A graph that shows how "uneven" something is (e.g., the distribution of wealth in a town). The further the curve bows away from the \(45\)-degree line of equality, the more unequal the area is.
  • Gini Coefficient: A number between \(0\) and \(1\) calculated from the Lorenz Curve. A value of \(0\) means perfect equality; a value of \(1\) means perfect inequality.

Inferential Statistics: Testing for Relationships and Differences

Sometimes you need to prove that a pattern isn't just a lucky coincidence. This is where statistical tests come in:

  • Spearman’s Rank Correlation (\(r_s\)): This tests the association between two variables. For example: "As the distance from the CBD increases, does the quality of the environment also increase?"
    • A result of \(+1\) is a perfect positive relationship.
    • A result of \(-1\) is a perfect negative relationship.
    • A result of \(0\) means no relationship.
  • Student’s t-test: This tests the difference between the means of two sets of data. You might use this to see if the average sediment size on a "managed" beach is significantly different from an "unmanaged" beach.
  • Chi-squared (\(\chi^2\)): This is used to see if there is a significant difference between the observed frequencies (what you counted) and the expected frequencies. It’s great for categorical data, like types of land use.

Key Takeaway: You don't always need to calculate the entire formula from scratch in the exam, but you must understand what the result means and how to interpret "statistical significance."

3. Qualitative Analysis: Interpreting Words and Images

Not all geographical data is about numbers. Qualitative data (words, feelings, and pictures) is just as important, especially for human geography topics like Regenerating Places.

Handling Text and Interviews

  • Coding: This is the process of categorising qualitative data. If you interviewed \(20\) people about a new shopping centre, you might "code" their answers into themes like "Economic Benefit," "Traffic Concerns," or "Loss of Heritage." You can then count how many times each "code" appears.
  • Textual Evaluation: This involves looking at newspapers, social media, or oral accounts and identifying the attitudes and actions (A) of different players (P).

Handling Visual Data

  • Photo Annotation: Rather than just looking at a photo of a glaciated valley, you should annotate it to identify specific features like U-shaped valleys or moraine.
  • Sketches: Field sketches allow you to "filter" the environment, highlighting the most important geographical features while ignoring distractions (like a parked car in front of a historic building).

4. Identifying Errors and Data Quality

No fieldwork is perfect! To get top marks, you must be honest about the limitations of your data.

  • Measurement Errors: Did the tape measure sag? Was the clinometer held at the wrong height?
  • Sampling Bias: Did you only interview people on a Monday morning when students and workers were absent? This would make your data unrepresentative.
  • Data Misuse: Be careful not to claim that one thing causes another just because they are correlated. Just because an area has high crime and high regeneration spending doesn't mean the regeneration caused the crime!

Did you know? Using crowd-sourced data or "big data" from online maps can help reduce your own sampling bias by providing thousands more data points than you could collect alone.

Chapter Summary Checklist

Quick Review:
- Can you explain when to use a kite diagram vs. a dot map?
- Do you know that Spearman's Rank tests for relationships, while a t-test tests for differences?
- Can you define the Gini Coefficient in the context of inequality?
- Do you understand how coding helps simplify complex interview answers?
- Are you ready to identify sources of error in your own data collection?

Next Step: To see how to use this analysis to reach a final judgment, head over to the chapter on "Conclusions, evaluation and sources of error."