Question

One way of checking the effect of undercoverage, nonresponse, and other sources of error in a sample survey is to compare the sample with known demographic facts about the population. The 2010 census found that 13.9%, or 32,576,000, of the 235,016,000 adults, aged 18 and over, in the United States identified themselves as of Hispanic origin. Is the value 13.9% a parameter or a statistic?
a. 13.9% is a parameter derived from a national Census that sampled 32,576,000 adults
b. 13.9% is a statistic derived from a national Census that sampled 32,576,000 adults
c. 13.9% is a parameter derived from a national Census that includes all individuals in a population
d. 13.9% is a statistic derived from a national Census that includes all individuals in a population.

Answer 1

Answer: Choice C

13.9% is a parameter derived from a national Census that includes all individuals in a population

======================================================

Explanation:

A census is a term in statistics that means that you survey or poll every individual in the population. In this case, the population is the set of people in the United States. As you can probably guess, a census takes a very long time, requiring lots of money and a lot of people to conduct it. The US census takes place every 10 years. The last of which happened in 2020. Despite the long time gap, census workers are always gearing up in the background getting ready for the next census on the horizon (since it's such a massive undertaking).

Because taking a census in statistics requires so much energy, it's often more effective to form a sample and try to see what the characteristics of that sample are like. If the sample is representative enough (ie is unbiased), then the sample statistics should be fairly close to the parameter counterparts. In this situation, we could maybe randomly select 10 states of the 50 total, and then ask everyone in those states what their ethnicity is. In this case, we're using a cluster sampling technique.