Lecture 04a Activity

Unit B · Chapter 04

An in-class activity. Nothing to turn in and no answer key – this one needs other people, which is why it happens in class rather than at home.

The idea. A population is defined by the question you are asking, not by the data you happened to collect – and the gap is far easier to see in someone else’s study than in your own.

Format. About 10 minutes · pairs · no laptop needed.

1. Write one (3 minutes)

On a half sheet of paper, describe a study from your own area in two or three sentences. Say what you want to know and exactly how you would recruit. Use no statistical vocabulary, and do not name the population – just the recruitment.

Put your name on it and hand it to your partner.

2. Take theirs apart (5 minutes)

Working from your partner’s paper alone:

  • Name the target population their question implies.
  • Name the sampled population their recruitment would actually produce.
  • Name one specific way a parameter could differ between the two.
  • Decide whether a larger \(n\) would close the gap you found, or leave it exactly where it is.

Hand it back. The author’s job is to say whether the gap is fair – and if it is, whether they would have caught it themselves.

3. Whole class (2 minutes)

  • Whose partner found a gap the author had not thought of? Two examples.
  • Show of hands: how many of the gaps found in this room would a larger sample fix?

If you have more time: two labs sample the same population and report means of 1.3 and 1.9. List everything that could cause that, then sort the list into “would shrink with more data” and “would not.”

If you were not in class

The hands almost never go up, and that is the point: a larger \(n\) shrinks sampling error and does nothing at all to a mismatch between the population you sampled and the one you meant to describe. A study of 10,000 undergraduates is still a study of undergraduates.

The reason this runs on someone else’s study rather than a textbook vignette is that the gap in a vignette is planted and findable, while the gap in your own design is invisible to you – you already know who you meant, so the recruitment sentence reads as though it says so. Your partner has only the words on the page, which is also all a reviewer will have.