3.5.1. How to prepare a Frequency Distribution

A frequency distribution is a comprehensive way to classify raw data of a quantitative variable. It shows how different values of a variable (here, the marks in mathematics scored by a student) are distributed in different classes along with their corresponding class frequencies.

A frequency distribution organizes raw data into a simpler table format. It breaks down a large amount of data into manageable groups called classes, and counts how many values fall into each class, known as the frequency. This helps in understanding how the data are spread out and highlights key features, such as where most data points are clustered.

Think of a teacher who needs to quickly assess student performance. Instead of looking at each student's mark individually, she groups the marks into ranges (e.g., 0-10, 10-20, etc.) and counts how many students scored in each range. This way, she can quickly see how many students did well and how many struggled, making it easier to understand the overall class performance.

While preparing a frequency distribution, the following five questions need to be addressed: 1. Should we have equal or unequal sized class intervals? 2. How many classes should we have? 3. What should be the size of each class? 4. How should we determine the class limits? 5. How should we get the frequency for each class?

These questions guide the process of building an efficient frequency distribution. The decision on whether to use equal or unequal classes can depend on the data's nature. Usually, having a manageable number of classes (about 6 to 15) is ideal. The size of the classes should reflect how the data clusters, and the class limits should be clearly defined to avoid confusion. Finally, frequencies can be determined through counting or tallying sensitive observations against each class.

Imagine you are sorting a collection of marbles by size. Before you start, you need to decide: do you want to divide them into tall and short (unequal) or maybe sizes of 0-5 cm, 6-10 cm, etc. (equal)? Also, how many groups will you have? Thinking of these questions helps you categorize the marbles effectively.

Class intervals are of two types: (i) Inclusive class intervals: where the values equal to the lower and upper limits of a class are included in the frequency of that same class. (ii) Exclusive class intervals: where an item equal to either the upper or the lower class limit is excluded from the frequency of that class.

Understanding the types of class intervals is crucial when preparing a frequency distribution, as it helps define how to assign data points to classes. Inclusive intervals mean that the boundary numbers belong to the class, while exclusive intervals only allow values within the range, excluding the boundary values themselves. This can affect the frequency count of each class.

If you're measuring the number of people who are aged 20-30, in an exclusive method, someone who is exactly 30 years old wouldn't be counted in that class. Conversely, in an inclusive method, they would be counted. Imagine organizing people into age groups at a party; it’s important to decide who fits where based on the intervals you choose.

Frequency refers to the number of values in a particular class. The counting of tally is made easier when four of them are put as // and the fifth tally is placed across them as |. Tallies are then counted as groups of five.

Class frequency is determined by counting how many data points fall within each class interval. Tally marks provide a visual method for counting, making it easy to keep track without losing count. After collecting tally marks, you can derive the total frequency for each class, which summarizes how many observations fit into each category.

Imagine you're counting how many times your friends choose different ice cream flavors. Every time someone orders a flavor, you make a tally mark. Once you have four marks for a flavor, you cross them with a fifth mark to make counting easy. When you're done, you can quickly see which flavor was the most popular.

The classification of data as a frequency distribution has an inherent shortcoming. While it summarizes the raw data making it concise and comprehensible, it does not show the details that are found in raw data.

Although frequency distributions simplify data and highlight key patterns, they inevitably lose the granularity of the information. For instance, if a class contains a wide range of values, all those values are summarized by a single class mark instead of being represented individually. This can lead to a loss of insight that could be critical for deeper analysis.

Think of a survey where you count how many students are under different score ranges in an exam. While it helps see the trend, you lose the individual scores of those students. If you’d like to know who struggled and who excelled, you’d miss out by only looking at the aggregated results.

--