Experimental Design Flashcards
7 cards from real AP practice questions. Tap to flip, then mark Knew It or Still Learning — missed cards come back until you master them.
Read the first 7 Experimental Design flashcards as text
A researcher wants to test four new fertilizer formulas on tomato plants. She has 8 greenhouses available and blocks by location (north vs. south exposure), assigning two formulas to each block at random. This is an example of:
Answer: A randomized block design
Grouping greenhouses by sun exposure (block) and then randomly assigning treatments within each block is a randomized block design.
Which statement correctly describes the difference between random sampling and random assignment?
Answer: Random assignment allows causal inference; random sampling allows generalization
Random assignment (to treatments) supports causal conclusions; random sampling (from a population) supports generalization of results.
A placebo is used in an experiment primarily to:
Answer: Control for the psychological effect of receiving a treatment
A placebo controls for the psychological placebo effect — the tendency for people to improve simply because they believe they are receiving treatment.
In planning an experiment, a researcher increases the number of subjects per treatment group from 20 to 80. The primary statistical benefit is:
Answer: Increased power to detect a true treatment effect
Larger sample sizes increase statistical power, making it more likely to detect a real treatment effect if one exists.
A researcher studying stress reduction randomly assigns participants to yoga, meditation, or a control group. After 8 weeks, stress levels are measured. A key threat to this experiment's validity would be:
Answer: Participants dropping out of the yoga group at higher rates
Differential dropout (attrition bias) across treatment groups threatens validity because those who remain may differ systematically from those who left.
A student randomly assigns 10 of 20 subjects to a treatment group. Which of the following would make this a statistically valid simple experiment?
Answer: Measuring the response in both groups under identical conditions except for the treatment
A valid experiment measures the response in all groups under identical conditions, varying only the treatment, to isolate its effect.
An experiment tests a new reading intervention. Students are assigned to treatment or control based on their classroom. Within each classroom, assignment is random. This design is called a:
Answer: Cluster randomized design
A cluster randomized design randomly assigns intact groups (clusters like classrooms) rather than individuals to treatment conditions.