Reading Comprehension Question Profile

2,000 question-answer pairs across 266 passages

Total Questions
2,000
266 passages
Dominant Type
What
46.8% of questions
Median Answer Length
2.0 words
Why averages 5.7
Typical Answer Position
38.2%
median passage position
G1 · Question Mix
What kinds of questions dominate?

Question wording is highly concentrated in a few interrogative forms. What questions alone account for 46.8% of the dataset, while the smallest category (Why) appears only 6 times.

The charts below show the distribution of interrogative forms and the overall spread of answer lengths.

Key takeaways
  • ●Median answer length is 2.0 words, so the typical target span is short and extractive.
  • ●Why questions have the longest answers on average at 5.7 words.
  • ●When answers tend to appear earliest in passages (median 29.7%).
G1.1 · Question-type distribution
G1.2 · Answer-length histogram
G2 · Answer Characteristics
How answer size and answer position vary by question type

Answer spans stay compact across most question types, but their location within the passage shifts. Comparing answer length and relative position helps separate concise fact lookups from less frequent, more elaborative prompts such as why.

G2.1 · Answer length by question type
G2.2 · Relative answer position by question type
G3 · Passage Concentration
Do a small number of passages generate most questions?

Question generation is somewhat top-heavy rather than dominated by just a handful of passages. The busiest passage has 20 questions, and the top 10% of passages contribute 20.2% of all questions.

Concentration checkpoints
11.3%of questions come from the top 14 passages (top 5%)
20.2%of questions come from the top 27 passages (top 10%)
37.1%of questions come from the top 54 passages (top 20%)
G3.1 · Questions per passage distribution
G3.2 · Top passages by question count
G3.3 · Cumulative concentration curve