Data presentation and interpretation

Welcome to Data Presentation and Interpretation!

In this chapter, we are moving from simply looking at lists of numbers to "telling a story" with that data. Whether you are looking at how much people earn or how many goals a team scores, you need ways to summarize that information and spot patterns. For Paper 3, you need to know how to read graphs, calculate how "spread out" data is, and decide if a data point is just a weird mistake or a vital piece of information.

Don't worry if this seems tricky at first! Statistics is often just about applying common sense to mathematical formulas. Let’s break it down step-by-step.

1. Single-Variable Data: The Mighty Histogram

When we look at one type of data (like the heights of students), we call it single-variable data. The most important tool you'll use here is the Histogram.

The Golden Rule of Histograms

In a standard bar chart, the height tells you the frequency. But in a histogram, the Area represents the Frequency. This is a common place to lose marks, so remember this memory aid:

"Area is the Amount"

Calculating Frequency Density

To draw or interpret a histogram, we use Frequency Density on the vertical axis. The formula is:
\( \text{Frequency Density} = \frac{\text{Frequency}}{\text{Class Width}} \)

Analogy: Imagine you are spreading butter on different sizes of toast. If you have the same amount of butter (frequency) but a much wider piece of toast (class width), the layer of butter (frequency density) will be much thinner!

Connecting to Probability

Because the total area of a histogram represents the total frequency, we can relate this to Probability Distributions. If you scaled the total area to equal 1, the area of each bar would represent the probability of an item falling into that group.

Quick Review:
• Frequency = Area of the bar
• Frequency Density = Height of the bar
• Total Area = Total number of data points

2. Bivariate Data: Scatter Diagrams and Correlation

Bivariate data is just a fancy way of saying we are looking at two things at once to see if they are related (e.g., "Does more revision lead to higher marks?").

Scatter Diagrams and Regression Lines

We plot these on a scatter diagram. You might see a Regression Line (a line of best fit) drawn through the points. This line is used to make predictions.
• Interpolation: Predicting a value inside the range of your data (usually reliable).
• Extrapolation: Predicting a value outside the range of your data (very risky and often inaccurate!).

Correlation vs. Causation

This is a favorite topic for exam questions! Just because two things have a strong correlation (they move together), it doesn't mean one causes the other.

Did you know? Statistics show that ice cream sales and shark attacks both go up at the same time. Does eating ice cream cause shark attacks? No! The "hidden variable" is the weather—people do both more often when it's hot.

Distinct Sections of the Population

Sometimes a scatter diagram shows two different groups mixed together. For example, if you plot height vs. weight for a whole school, you might see two distinct "clouds" of dots—one for the younger children and one for the teachers. Recognizing these sub-populations is a key interpretation skill.

Key Takeaway: Correlation shows a relationship, but it never proves that one thing caused another to happen.

3. Measures of Central Tendency and Variation

We need numbers to describe the "middle" and the "spread" of our data.

The Mean (\( \bar{x} \))

The average. You calculate it by adding everything up and dividing by how many items there are:
\( \bar{x} = \frac{\sum x}{n} \)

Standard Deviation (\( \sigma \))

This measures how spread out the data is from the mean. A low standard deviation means the data is bunched close to the average; a high standard deviation means it's spread far and wide.

You need to be able to calculate it from summary statistics using this formula:
\( \sigma = \sqrt{\frac{\sum x^2}{n} - \left(\frac{\sum x}{n}\right)^2} \)

Simple Trick to remember the formula:
It's the "Square root of (Mean of the squares minus the square of the mean)."

Common Mistake to Avoid: When calculating the "square of the mean," make sure you calculate the mean (\( \bar{x} \)) first, and then square it. Don't confuse it with \( \sum x^2 \)!

4. Outliers and Data Cleaning

An outlier is a data point that is way off from the rest. It could be a very unusual result, or it could just be a mistake (like someone typing 150cm instead of 15cm).

How to spot an outlier

In the exam, they will usually give you a rule to identify them. Common rules include:
1. Any value more than 1.5 \(\times\) IQR (Interquartile Range) above the upper quartile or below the lower quartile.
2. Any value more than 2 standard deviations away from the mean.

Data Cleaning

Before you analyze data, you must "clean" it. This involves:
• Dealing with missing data: Deciding whether to ignore it or find the missing values.
• Correcting errors: Fixing obvious typos.
• Removing outliers: Only if you are sure they are errors or they will unfairly skew your results.

Choosing the Right Graph

You might be asked to critique a data presentation.
• Box Plots are great for showing outliers and comparing the "spread" of two different groups.
• Histograms are better for seeing the "shape" of the data (is it symmetrical or skewed?).

Key Takeaway: Always check your data for "weird" numbers before you start calculating. A single outlier can completely ruin your Mean and Standard Deviation!

Final Quick Review Box

1. Histograms: Area = Frequency. Use Frequency Density for the height.
2. Regression: Interpolation is safe; Extrapolation is "danger zone."
3. Correlation: Does not equal causation!
4. Standard Deviation: Measures spread. Use the "Mean of the squares minus square of the mean" formula.
5. Outliers: Use the 1.5 \(\times\) IQR rule or the 2rd standard deviation rule to find them.