Potential Problems with Sampling

AP Statistics· difficulty 2/5

Researchers compare reading scores between students who attend a private after-school tutoring program and those who do not. Tutored students score higher on average. Tutored students also tend to come from families with higher incomes.

Tutoring Reading score Family income Income tied to BOTH treatment and outcome --> Confounding

Why can we not conclude that tutoring caused higher scores?

  • A

    Family income is a potential confounding variable

    check_circle
  • B

    The study was double-blind

  • C

    The sample size is too small

  • D

    There was no placebo

Explanation

Income is associated with both treatment (tutoring) and outcome (scores), so its effect cannot be separated from the tutoring effect. This is confounding, fixable only with random assignment.

Want 10 more like this — adaptive to your weak spots?

Related questions