Welcome to Data Processing!

In this chapter, we are going to learn how to take a big pile of messy numbers and turn them into clear pictures and useful summaries. Think of a statistician as a storyteller—your job is to use data to tell a story that everyone can understand. Whether you're looking at house prices, exam scores, or how many goals your favorite team scores, the tools in this chapter will help you make sense of it all!

1. Picturing Your Data: Diagrams and Tables

Before we do any hard math, we usually start by drawing a picture of our data. Different data types need different types of pictures.

Tables and Tallies

The simplest way to organize data is a frequency table. We use tallies (those little bundles of five lines) to count things quickly without losing our place. Two-way tables are great when we want to compare two different categories at once (like "Year Group" vs. "Favorite Subject").

Stem and Leaf Diagrams

This is a clever way to show every single piece of data while still seeing the overall shape. Important: Your stem and leaf diagram must always be ordered (the leaves should go from smallest to largest) and it must have a key so people know what the numbers mean.

Example: If the stem is 2 and the leaf is 5, the key tells us if that means 25, 2.5, or 250.

Venn Diagrams and Pictograms

Venn Diagrams show how different groups overlap. Pictograms use little pictures to represent a set amount of data. Common Mistake: In a pictogram, always check the key! If one circle represents 4 people, a half-circle means 2 people. Don't just count the pictures.

Choropleth Maps and Population Pyramids

  • Choropleth Maps: These are maps where areas are shaded in different colors to show statistics (like a map of the UK where darker blue means more rain).
  • Population Pyramids: Two back-to-back bar charts showing how many males and females are in different age groups. They help us see if a population is "young" or "aging."

Quick Review: Every diagram needs a title, clear labels, and a key if necessary!

2. The "Average" Choice: Measures of Central Tendency

An average is a single number that represents a whole set of data. There are three main types you need to know:

The Three M's

  • Mode: The value that appears most often. (Memory Aid: MOde = MOst).
  • Median: The middle value when the data is in order. (Memory Aid: The median is the strip of grass in the middle of a road).
  • Mean: The "meanest" one because it makes you do the most work! Add them all up and divide by how many there are.

Averages for Grouped Data

Sometimes data is grouped (e.g., "10 to 20 mins"). We don't know the exact values, so we find an estimate of the mean. Step-by-Step: 1. Find the midpoint of each group (the middle of the interval). 2. Multiply the midpoint by the frequency. 3. Add these totals up. 4. Divide by the total frequency (the total number of items).

Which average should I use?

Selecting the right average is important: - Use the Mode for non-numerical data (like favorite color) or when you want to know the most popular item in a shop. - Use the Median if the data has "outliers" (random huge or tiny numbers) because the median isn't bothered by them. - Use the Mean when you want to include every single piece of data in your calculation.

Key Takeaway: If you add the same number to every value in a dataset, the mean, median, and mode will all increase by that same number!

3. How Spread Out is the Data? (Dispersion)

Two classes might have the same average test score, but one class might have everyone getting roughly the same mark, while the other has some geniuses and some who struggled. We measure this "spread" using:

  • Range: \( \text{Highest Value} - \text{Lowest Value} \). Simple, but can be ruined by one weirdly high or low number.
  • Quartiles: These split your data into four equal parts. The Lower Quartile (LQ) is 25% of the way in, and the Upper Quartile (UQ) is 75% of the way in.
  • Interquartile Range (IQR): \( \text{UQ} - \text{LQ} \). This shows the spread of the middle 50% of the data. It's much more reliable than the Range because it ignores outliers.

Box Plots

A box plot is a "five-number summary" in picture form. It shows: 1. Lowest value 2. Lower Quartile 3. Median 4. Upper Quartile 5. Highest value

Outliers

An outlier is a piece of data that just doesn't fit. Example: If everyone in a class is 1.5m tall and one student is 2.1m tall, that student is an outlier. Always check if an outlier is a mistake (like a typo) or just a very unusual result.

4. Skewness: Is it Wonky?

Skewness tells us if the data is symmetrical or if it's "leaning" to one side. - Positive Skew: The "tail" of the data points toward the higher numbers. In this case, usually \( \text{Mean} > \text{Median} > \text{Mode} \). - Negative Skew: The "tail" points toward the lower numbers. - Symmetrical: The data looks like a mirror image on both sides.

5. Scatter Diagrams and Correlation

We use scatter diagrams to see if there is a relationship (correlation) between two things (bivariate data).

Types of Correlation

  • Positive Correlation: As one goes up, the other goes up (e.g., height and shoe size).
  • Negative Correlation: As one goes up, the other goes down (e.g., age of a car and its value).
  • Zero Correlation: No relationship at all (e.g., your hair length and your math score).

Line of Best Fit

This is a straight line drawn through the middle of the points. Pro Tip: To draw an accurate line, first calculate the double mean point \( (\bar{x}, \bar{y}) \). This is the (mean of all x-values, mean of all y-values). Your line must pass through this point!

Correlation vs. Causation

Did you know? Just because two things are correlated doesn't mean one causes the other. Example: Ice cream sales and shark attacks are positively correlated. Does ice cream cause shark attacks? No! Both go up because it's summer and the weather is hot. This is called an association.

A time series is a graph showing how something changes over time. Because these graphs can be very "jumpy," we use moving averages to smooth them out.

For your exam, you may need to calculate a 4-point moving average. You take the first 4 values, find their mean, then move one step along and find the mean of the next 4, and so on. This helps you see the trend (the long-term direction) without getting distracted by seasonal ups and downs.

7. Index Numbers and Rates of Change

Index numbers are used to compare values over time relative to a "base year" (which is always set at 100). \( \text{Index Number} = \frac{\text{Current Value}}{\text{Value in Base Year}} \times 100 \)

You will also see rates of change, like the Crude Birth Rate. Don't worry: Complex formulas like this are usually given to you in the question! You just need to know how to plug in the numbers.

Example: \( \text{Crude Birth Rate} = \frac{\text{Number of births}}{\text{Total population}} \times 1000 \)

Key Takeaway: Always read the units carefully! If the formula says "per 1000," make sure your final answer makes sense in that context.

8. Misleading Graphs: Don't Be Fooled!

Sometimes graphs are drawn to trick you. Always check for: - Incorrect Scales: Does the y-axis start at zero? If it doesn't (a truncated axis), differences can look much bigger than they really are. - Distorted Sizing: In pictograms or 3D charts, the volume of the shapes can be misleading. - Missing Labels: If there's no key or units, the graph is useless!

Quick Review: Statistics is about being a detective. Look at the numbers, look at the picture, and ask yourself: "Does this actually make sense?"