The correct answer is:
B. Open-ended classes

Explanation:

When constructing a frequency distribution for continuous data, the following elements are desirable:

1. All-inclusive classes (A): Every data point should fit into some class to avoid missing values.
2. Equal-width classes (C): Ensures that comparisons between intervals are meaningful and consistent.
3. Mutually exclusive classes (D): Each data point should fall into exactly one class to avoid ambiguity.

However:

• Open-ended classes (B) are not preferred because they do not provide clear boundaries (e.g., "40 and above"). This makes it difficult to accurately analyze or interpret data. Open-ended classes are typically used only when data collection constraints require them (e.g., extreme values).

---

Would you like more details or have further questions?

Here are 5 related questions you might find interesting:

1. What are the main types of frequency distributions?
2. How do you determine the appropriate class width for a dataset?
3. What are cumulative frequency distributions, and when are they used?
4. How can a histogram be derived from a frequency distribution?
5. What are some common issues when summarizing continuous data?

Tip: Always ensure that the class intervals in a frequency distribution do not overlap; this avoids confusion when categorizing data points.