Cumulative Frequency (AQA GCSE Maths): Revision Notes
Cumulative frequency
What is cumulative frequency?
Cumulative frequency is a really useful concept in statistics that helps us understand how data builds up over time. Think of it as a running total - you're simply adding up frequencies as you move through your data groups. Instead of looking at individual frequencies, you're looking at the total number of observations up to each point.
The key idea is that each cumulative frequency value tells you how many items fall at or below a certain value. This makes it much easier to find important statistics like the median and quartiles from your data.
Creating a cumulative frequency table
Building a cumulative frequency table is straightforward once you understand the process. You start with your original frequency table that shows how many items fall into each group or class interval.
To create the cumulative frequency column, you begin with the first frequency value and then keep adding each subsequent frequency to your running total. For example, if your first three frequencies are 4, 9, and 20, your cumulative frequencies would be 4, then 4+9=13, then 13+20=33, and so on.
The beauty of this system is that your final cumulative frequency value should always equal the total number of items in your dataset. This gives you a great way to check your work - if the numbers don't match up, you know you've made an error somewhere.
Drawing a cumulative frequency graph
Creating a cumulative frequency graph requires careful attention to plotting the right points. Unlike regular frequency graphs, you need to plot your cumulative frequency values against the upper boundary of each class interval.
Start by adding your cumulative frequency column to your data table. Then, plot points using the highest value in each class interval on the x-axis and the corresponding cumulative frequency on the y-axis. For instance, if you have a class interval of 150-160, you would plot the cumulative frequency against 160.
Don't forget to plot zero at the lowest value of your first class interval. This ensures your graph starts from the correct baseline. Once you've plotted all your points, join them with a smooth curve. The resulting S-shaped curve is called an ogive, and it's your key tool for finding important statistical measures.
Finding vital statistics from your graph
Your cumulative frequency graph becomes a powerful tool for finding key statistical measures. The process might seem complex at first, but it follows a logical pattern once you understand the steps.
Finding the median
To find the median, you need to locate the middle value of your dataset. Go halfway up the cumulative frequency axis (so if your total is 120, go to 60), then move horizontally across to where you meet the curve. From that point, drop down vertically to read the value on the horizontal axis. This gives you an estimate of the median.
Finding quartiles
The quartiles divide your data into four equal parts. The lower quartile (Q1) is found by going one-quarter of the way up the cumulative frequency axis, across to the curve, then down to read the value. The upper quartile (Q3) is found by going three-quarters of the way up the axis and following the same process.
Calculating the interquartile range
The interquartile range (IQR) measures the spread of the middle 50% of your data. Simply subtract the lower quartile from the upper quartile. This gives you a measure of variability that isn't affected by extreme values, making it very useful for understanding your data's distribution.
Finding Statistics from a Cumulative Frequency Graph:
Step 1: Find the median
- Total frequency = 120, so median is at 120 ÷ 2 = 60
- Go to 60 on the y-axis, across to the curve, down to read x-value
Step 2: Find the quartiles
- Q1 is at 120 ÷ 4 = 30 on the y-axis
- Q3 is at 3 × 30 = 90 on the y-axis
- Follow the same process: across to curve, down to read values
Step 3: Calculate IQR
- IQR = Q3 - Q1
Understanding estimates from grouped data
It's crucial to remember that when you're working with grouped data, the values you read from your cumulative frequency graph are estimates, not exact values. This is because you don't know exactly how the data points are distributed within each class interval.
The grouped nature of your data means that the actual individual values could be anywhere within their respective class intervals. Your graph assumes an even distribution within each class, which may not reflect reality. However, these estimates are still very useful for understanding the overall pattern and characteristics of your dataset.
Using graphs for further estimations
Your cumulative frequency graph can help you estimate other useful information about your dataset. For example, you can estimate what percentage of values fall below or above a certain point, or determine how many items are greater than or less than a specific value.
To make these estimates, locate your target value on the horizontal axis, move up to the curve, then across to read the cumulative frequency. This tells you how many items fall at or below that value. You can then calculate percentages or find complementary values as needed.
To find what percentage of values are below a certain point, divide the cumulative frequency at that point by the total frequency, then multiply by 100.
Key Points to Remember:
- Cumulative frequency means adding up frequencies as you go along to get a running total
- Always plot cumulative frequency against the upper boundary of each class interval, and don't forget to start with zero at the lowest class value
- The median is found at the halfway point, while quartiles are at the quarter and three-quarter points on the cumulative frequency axis
- Values read from grouped data graphs are estimates because you don't know the exact distribution within each class interval
- Your cumulative frequency graph is a powerful tool for estimating percentages and making comparisons about your dataset