I found it unbelievable as well so I tested these questions on college and graduate students. Answers were in range. CS majors literally believed that 40% of Americans are black and a third of the population lives in Texas.
What a person needs to answer correctly is sociocultural awareness.
I think this still shows a poor grasp of percentages/statistics. If you were to follow up the Texas question by asking in turn, “what percentage live in California?” " How about New York?" “And Florida?” “So then what percentage live in the 46 remaining states plus US territories?” you’d watch a classroom of students slowly realize they’ve way surpassed 100% thanks to their overinflated initial estimates. Or conversely, if you gave them a paper with a list of states and asked them to write down what percentage of Americans lived in each state next to the state’s name, it might not be accurate, but it would probably add up to 100% for significantly more respondents.
The over/under estimation problem is almost certainly worsened when you ask about a single demographic in isolation, rather than all possible demographics at once.
I found it unbelievable as well so I tested these questions on college and graduate students. Answers were in range. CS majors literally believed that 40% of Americans are black and a third of the population lives in Texas.
What a person needs to answer correctly is sociocultural awareness.
I think this still shows a poor grasp of percentages/statistics. If you were to follow up the Texas question by asking in turn, “what percentage live in California?” " How about New York?" “And Florida?” “So then what percentage live in the 46 remaining states plus US territories?” you’d watch a classroom of students slowly realize they’ve way surpassed 100% thanks to their overinflated initial estimates. Or conversely, if you gave them a paper with a list of states and asked them to write down what percentage of Americans lived in each state next to the state’s name, it might not be accurate, but it would probably add up to 100% for significantly more respondents.
The over/under estimation problem is almost certainly worsened when you ask about a single demographic in isolation, rather than all possible demographics at once.