1.1.2 – Other Random Sampling Methods There are many ways to randomly sample. Let’s review what we’ve already learned about, and compare and contrast it to a few more methods. The student council wants to conduct a survey about use of the school library during the first five minutes of an all-school assembly in the auditorium. There are 800 students present at the assembly. Here is a map of the auditorium. Note that students are seated by grade level and that the seats are numbered from 1 to 800. 1. Describe four different ways we could sample from this group. Randomly select from among all students – simple random sample Sample by row/grade level – stratified random sample Sample by column – cluster sample Sample as they enter – systematic random sample 2. Describe how to obtain a sample of 80 students using a simple random sampling. Use a random number generator to generate 80 unique random values from 1 to 800 and give the survey to the students in the corresponding seats. 3. Describe how to obtain a sample of 80 students using stratified random sampling. Explain your choice of strata and why this method might be preferred to simple random sampling. Stratify by grade level. For each grade, label the students from 1-200. Randomly generate 20 unique values from 1 to 200. Give the survey to those students. Repeat for each grade. Stratification reduces variability and guarantees proportionate representation in the sample. 4. Describe how to obtain a sample of 80 students using cluster sampling. Explain your choice of clusters and why this method might be preferred to simple random sampling. Each column is a cluster. Label the columns from 1 to 20. Randomly generate 2 unique values from 1 to 20. Give the survey to the 80 students sitting in those columns. Cluster sampling is much more efficient than finding 80 seats scattered about the auditorium. 5. Describe how to obtain a sample of 80 students using systematic random sampling. Explain why this method might be preferred to simple random sampling. Sample every 10th student. Generate a random integer from 1 to 10, call this k. Select the kth student to enter, and every 10th student after them. Systematic sampling is much more efficient than SRS because the individuals come to you instead of you going to them. Big Ideas: • • • Stratified Random Sample – Group by similar characteristics, SRS within each group Cluster Sample – Group by similar location, SRS of clusters Systematic Random Sample – Randomly select in the first k, and every k after Why use instead of SRS? • Stratified o Reduced variability o Proportionate representation • Cluster, Systematic o More efficient CHECK YOUR UNDERSTANDING A factory runs 24 hours a day, producing 15,000 wood pencils per day over three 8-hour shifts—day, evening, and overnight. In the last stage of manufacturing, the pencils are packaged in boxes of 10 pencils each. Each day a sample of 300 pencils is selected and inspected for quality. 1. Describe how to select a stratified random sample of 300 pencils. Explain your choice of strata. Because the quality of the pencils might be the same within each of the shifts, but differ across the shifts, use the shifts as strata. At the end of each 8-hour shift, label all the pencils produced during that shift 1 to 5000, where N is the total number of pencils produced on that shift. Generate 100 different random integers from 1 to 5000 and select those pencils for inspection. 2. Describe how to select a cluster sample of 300 pencils. Explain your choice of clusters. The boxes of 10 pencils could be used as a cluster sample because it would be relatively easy to select boxes of 10 pencils. At the end of the day, label all the boxes of pencils 1 to 1500. Generate 30 different random integers from 1 to 1500 and inspect all the pencils in the selected boxes. 3. Describe how to select a systematic random sample of 300 pencils. Select every 15,000/300 5 50th pencil that comes off the Start by randomly selecting a number from 1 to 50. Select that pencil and every 50th pencil thereafter until 300 pencils have been selected. 4. Explain a benefit of each of these three methods in this context. Stratified: We are guaranteed to inspect 100 pencils from each of the three shifts. Cluster: Simplifies the sampling process. Systematic: It is easier to find the selected pencils by selecting them as they come off the production line, which will guarantee that samples of pencils that were manufactured at regular intervals over the course of the day are selected.