Question

Indicate whether it is best assessed by using a confidence interval or a hypothesis test or whether statistical inference is not relevant to answer it. (a) What proportion of people using a public restroom wash their hands after going to the bathroom? (b) On average, how much more do adults who played sports in high school exercise than adults who did not play sports in high school? (c) In \(2010,\) what percent of the US Senate voted to confirm Elena Kagan as a member of the Supreme Court? (d) What is the average daily calorie intake of 20 year-old males?

Short answer

(a) Confidence interval, (b) Hypothesis test, (c) Statistical inference is not relevant, (d) Confidence interval

Step-by-step solution

Choice for (a)

The question asks for the proportion of people who follow a certain procedure, suggesting we are looking at a characteristic of a single group. Hence, the most appropriate test would be to construct a confidence interval for the proportion. This would give an estimate of the true proportion of all people visiting public restrooms who wash their hands afterwards.

Choice for (b)

This question is concerned with the difference in exercise levels between two groups of adults based on a characteristic: whether they played sports in high school or not. Therefore, a hypothesis test for difference in means would be the best method here.

Choice for (c)

This question refers to a single, known and specific historical event (the confirmation vote for Elena Kagan). As we first and foremost need to look at historical data to answer this, statistical inference is not relevant in this case.

Choice for (d)

The question seeks the average daily calorie intake for a particular group, 20 year-old males. Therefore, the best approach here would be to construct a confidence interval for the mean.

Key concepts

Confidence Interval

Understanding confidence intervals is essential in statistics, as they provide a range of values within which we can be certain, to a particular degree, that the true parameter value lies. For instance, in the exercise above, we use a confidence interval to estimate the true proportion of people washing their hands in a public restroom (part a) and the average daily calorie intake of 20-year-old males (part d).

A confidence interval is constructed around a sample statistic, such as a sample mean or proportion, and it indicates the reliability of this estimate. It's usually expressed in the form \( (\overline{x} - E, \overline{x} + E) \), where \( \overline{x} \) is the sample mean and \( E \) is the margin of error. The margin of error encompasses factors like sample size and variability, as well as the confidence level, typically selected at 90%, 95%, or 99%.

For better understanding, imagine we are trying to estimate the average score of a test in a classroom. If we calculate a 95% confidence interval and find it to be from 65 to 75, this means that we can be 95% certain that the true average score of all possible classrooms is between these two scores. Confidence intervals are valuable because they not only give an estimate of the parameter but also some idea about how uncertain we are about this estimate.

Hypothesis Test

When working with statistics, you will often need to determine whether there is enough evidence to support a specific claim. This is where the hypothesis test comes in to play. In our example (part b), we are comparing the average exercise amount between adults who played sports in high school and those who did not.

A hypothesis test evaluates two opposing statements about a population to determine which statement is best supported by the sample data. At its core, there is the null hypothesis (\(H_0\)), which is a statement of no effect or no difference, and the alternative hypothesis (\(H_A\)), which is what we suspect might be true instead.

In the case of determining if there is a meaningful difference in exercise levels, we formulate a null hypothesis that says there is no difference in the mean exercise amount between the two groups, and an alternative hypothesis that there is a difference (either more or less). Using statistical analysis, we can determine whether the observed data is significantly different from what the null hypothesis would predict, hence providing evidence for or against the alternative hypothesis.

Proportion

In statistics, a proportion is a type of ratio that tells us how many members of a group meet a particular criteria out of the total. It is essentially a measure of frequency. In the exercise question (part a), we want to determine what proportion of people wash their hands in a public restroom.

The proportion can be calculated using the formula \( P = \frac{x}{n} \), where \( x \) is the number of individuals with the characteristic of interest, and \( n \) is the total number in the group. If you are constructing a confidence interval for a proportion, you will typically also use a formula that includes the z-score associated with your chosen confidence level. This allows us to calculate the margin of error for the proportion, effectively creating a range in which we believe the true population proportion should fall. Understanding proportions is crucial because it sets the foundation for more complex statistical methods like regression analysis and hypothesis testing.

Mean Difference

The concept of mean difference is commonly used to compare the central tendency of two groups. It is the basis of many analytical procedures, including the t-test used in our example's exercise (part b). The mean difference is calculated simply by subtracting the mean of one group from the mean of another.

If you want to know whether the average amounts of something are significantly different between two groups, you subtract the average (mean) of one group from the average of the other. The result is the mean difference. When the hypothesis test is applied to the mean difference, it lets us determine if the two groups are different in a statistically significant way, beyond what might have been caused by random chance alone.

For example, to know how much more adults who played sports in high school exercise compared to those who didn't, you would take the average amount of exercise for the first group and subtract the average amount of exercise for the second group. If this mean difference is large enough and statistically significant according to the hypothesis test, we might conclude that there is indeed an effect of playing sports in high school on adult exercise levels.