3.5.5. How should we determine the class limits?

Class limits should be definite and clearly stated. Generally, open-ended classes such as “70 and over” or “less than 10” are not desirable.

Class limits are the boundaries of the ranges that define a particular class in a frequency distribution. It’s important that these limits are clearly defined so that there's no confusion about which data falls into each class. For example, saying a class has a limit of '70 and over' can be ambiguous since there's no upper boundary. Instead, having well-defined limits helps to maintain consistency when organizing data.

Think of class limits like the boundaries of a playground that define where kids can play. Without clear boundaries, kids might wander into unsafe or restricted areas, just like data points might get misclassified without clear class limits.

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

Inclusive class intervals include both the upper and lower limits within the class. For instance, in the class interval '10 to 20', both 10 and 20 are counted. On the other hand, exclusive class intervals do not include these limits; thus, '10 to 20' means it includes values greater than 10 but less than 20. The choice between inclusive and exclusive depends on how the data is structured and the analysis requirements.

Imagine a party where there's an age requirement: if the rule states that it's only for ages '10 to 20', inclusive would mean anyone who turns 10 or is exactly 20 can enter. However, using exclusive rules would mean only those who are strictly older than 10 and younger than 20 are allowed. This finer distinction helps manage the population that can enter.

Given the range of the variable, we can determine the number of classes once we decide the class interval. Thus, we find that these two decisions are interlinked. We cannot decide on one without deciding on the other.

The range refers to the difference between the maximum and minimum values in your dataset. Once you have this range, you can determine how many classes you want to create. For example, if the range is 100 and you decide to create 10 classes, each class would span 10 units. It’s important to decide on both parameters simultaneously because changing one will affect the other in terms of data organization.

Consider organizing a bookshelf. If you have 100 books (your range), and you want to organize them in 10 sections (your classes), you will place about 10 books in each section. However, if you decide to only have 5 sections, you need to rearrange the books, as now each section needs to hold around 20 books. This interplay between the quantity of classes and their size is crucial in effective data organization.

In simple terms, frequency of an observation means how many times that observation occurs in the raw data. Class frequency refers to the number of values in a particular class.

Frequency is a key concept in statistics that represents the number of times a specific value or range of values appears in a dataset. In a frequency distribution, you count how many observations fall within each class interval. For instance, if 5 students scored between 40-50 marks, the frequency for that class would be 5. This helps to summarize large datasets into more manageable figures.

Imagine counting how many times different types of fruits appear in a basket. If there are 3 apples, 5 oranges, and 2 bananas, you can summarize that data by saying there are 3 apples, 5 oranges, and 2 bananas. This summary gives you an easier overview of what is in the basket, just like frequency counts provide clarity on the distribution of data in statistical analysis.

--