Data collection and sampling experiments

In a data collection experiment, you ask five classmates from a different class how many movies they saw at the theater last month. Then you use a systematic sampling technique: randomly pick one person on the class list, mark that name, move down four names, mark that name, and continue until you have marked 12 names. For each marked name, record the five data values, giving a total of 60 data values.

The lab has you build two relative frequency tables from the same 60 values. One gives every individual count its own row: 0, 1, 2, 3, 4, 5, 6, and 7 or more. The other collects those counts into intervals: 0-1, 2-3, 4-5, and 6-7+. In both, a row carries a frequency, a relative frequency (that frequency divided by 60), and a cumulative relative frequency.

Which grouping you build decides which questions you can answer. To find the percent of data that is at most two, you add the relative frequencies for 0, 1, and 2 -- only the per-count grouping lets you do that, because the interval 2-3 lumps 2 in with 3. For the percent that is more than two, subtract the cumulative relative frequency at 2 from 1, or add the relative frequencies from 3 up.

How grouping affects what you can answer