In 2021, three seperate studies on the fitness level of eleventh graders were conducted in the city of Los Angeles. In each study, every student in a group of eleventh graders took the same fitness test and received a score on it. The possible scores on the fitness test are the whole numbers from 0 to 100, inclusive. The distribution of the scores for each of the studies is shown in the table below.

$$ \begin{array} {|c|c|c|c|} \hline \text{Score range} & \text{Study I} & \text{Study II} & \text{Study III} \\ \hline 0-24 & 3 & 22 & 15 \\ \hline 25-50 & 33 & 53 & 34 \\ \hline 51-75 & 42 & 81 & 26 \\ \hline 76-100 & 22 & 44 & 25 \\ \hline \text{ Total number} \\ \text{of participants} & 100 & 200 & 100 \\ \hline \end{array} $$

The participants for the studies were selected as follows.

• For Study I, 100 eleventh graders were selected at random from all eleventh graders in Los Angeles.
• For Study II, 200 eleventh graders were selected at random from all eleventh graders in Los Angeles.
• For Study III, 100 eleventh graders from Los Angeles volunteered to participate.

No eleventh grader participated in more than one of the three studies. The results of which of the studies can appropriately be generalized to all eleventh graders in Los Angeles in 2021?

Summary Submit

Studies I and II are appropriate since we are taking a random sample of the eleventh graders from Los Angeles. Study III suffers from self selection bias; students who volunteer are not representative of the population. Think about the personalities of those who typically volunteer in class - you miss the silent majority!