Data Collection and Basic Charts

Ever wondered how surveys actually work? Frequency tables are your starting point - they simply count how many times each value appears in your data. Think of it like counting votes for your favourite pizza topping.

Bar charts are brilliant for comparing different categories. The key rules are dead simple: same width bars, gaps between them, and always label your axes. Your data goes on the x-axis, and how often it occurs goes on the y-axis.

You'll also see compound bar charts (data stacked like building blocks) and comparative bar charts (data standing side by side like friends in a photo). Pie charts show how your data breaks into slices of a whole - just divide 360° by your total to work out each slice's angle.

Quick tip: For pie charts, if you have 30 people surveyed, each person equals 12° (360 ÷ 30). Easy!

Understanding Different Types of Data

Not all data is created equal, and knowing the difference will save you loads of confusion. Qualitative data describes things (like hair colour or favourite subject), while quantitative data deals with numbers you can actually measure.

Here's where it gets interesting: discrete data can only take specific values (like the number of pets you have - you can't have 2.5 cats!). Continuous data can be any value within a range (like your height, which could be 165.3cm or 165.34cm).

Primary data is what you collect yourself for your own research project. Secondary data is information someone else gathered that you're now using. Think of primary data as cooking from scratch, and secondary data as using a ready-made sauce.

Pictograms use symbols to represent data values - they're like emoji for statistics! Just remember to include a clear key showing what each symbol represents.

Remember: Grouped data bundles information into categories, making large datasets much easier to handle and analyse.

Averages and Statistical Measures

The mean is what most people call the average - add everything up and divide by how many values you have. For grouped data, you'll need to use midpoints and multiply by frequencies, but don't worry, it's just an extra step.

The median is your middle value when everything's in order. If you have an even number of values, take the average of the two middle ones. For tables, use the formula $n+1$/2 to find which position holds your median.

The mode is simply the most common value - the one that appears most often. You might have more than one mode or no mode at all. The range shows how spread out your data is by subtracting the smallest from the largest value.

Line graphs connect dots with straight lines to show how data changes over time. They're perfect for tracking trends like temperature changes or your exam scores throughout the year.

Pro tip: Smaller ranges mean more consistent data - think of it as measuring how reliable or predictable your information is.

Advanced Data Analysis Tools

Two-way tables organise data using two different categories, like gender and handedness. Fill them out step by step, making sure all your rows and columns add up correctly - it's like solving a number puzzle.

Box plots show five key values: minimum, lower quartile, median, upper quartile, and maximum. They're brilliant for comparing different datasets at a glance. When comparing box plots, always mention both the average (using medians) and the spread (using range or IQR).

Correlation means two things are connected somehow. Positive correlation means both increase together (like study time and test scores). Negative correlation means one goes up while the other goes down (like temperature and heating bills).

Causality is when one thing actually influences another. Just because two things correlate doesn't mean one causes the other - ice cream sales and drownings both increase in summer, but ice cream doesn't cause drowning!

Key insight: Strong correlation means the connection is clear and obvious, while weak correlation means there's a connection but it's not very reliable.

Scatter Graphs and Data Patterns

Scatter graphs plot two variables against each other to spot patterns and relationships. They're like detective tools for data - you're looking for clues about how two things might be connected.

The line of best fit is a straight line that best represents your scattered data points. It helps you make predictions and see the overall trend, even when individual points are all over the place.

Outliers are the rebels of data - values that don't fit with the rest. They stick out like a sore thumb and can sometimes reveal interesting stories or errors in data collection.

Frequency polygons work brilliantly for continuous data that's been grouped. You plot the midpoint of each group against its frequency, then join the dots. They're especially useful when you're dealing with large amounts of data that would be messy in other chart types.

Important: When you see "no correlation", it means there's no clear linear relationship between the variables - they're basically doing their own thing independently.