Observational Studies versus Experiments

12.4: Observational Studies versus Experiments. Original editorial SAT preparation with three worked examples, ten flashcards, and thirty practice questions. Educator-review draft.

Lesson 4 of 63 parts Practice Flashcards

1/3

Observational studies

Objective: Recognize when researchers measure existing conditions.

Prerequisite skills: sample versus population and association. An observational study records outcomes without assigning the explanatory condition. It can reveal association but may contain confounding variables, so causation is not established from the design alone. Common error: calling a study an experiment just because it collects numerical data. Restrictions and interpretation: Observing existing behavior differs from assigning treatments; causal conclusions depend on study design.

Formulas

no assigned treatment ⇒ observational design

Worked examples

Try itResearchers record existing sleep habits and grades for 360 students without assigning sleep schedules. What is the design?Show answer

The researchers observe existing habits rather than assign them. The study is observational and may show an association, but other factors could influence both sleep and grades. Therefore the requested answer is Observational study.

Swipe sideways to move between sections

Observational studies

Objective: Recognize when researchers measure existing conditions.

Formulas

no assigned treatment ⇒ observational design

Worked examples

Try itResearchers record existing sleep habits and grades for 360 students without assigning sleep schedules. What is the design?Show answer

Experiments

Objective: Recognize assigned explanatory conditions.

An experiment deliberately imposes a treatment and measures outcomes. Random assignment distributes potential confounders between treatment groups on average; it is not the same as randomly sampling participants from a population. Common error: confusing random treatment assignment with random participant selection. Restrictions and interpretation: Observing existing behavior differs from assigning treatments; causal conclusions depend on study design.

Formulas

assigned treatment + randomized groups supports causal comparison

Worked examples

Try itA researcher randomly assigns 124 volunteers to use either a new practice method or the current one, then compares outcomes. What feature most supports a causal comparison?Show answer

Random assignment tends to balance preexisting differences across methods. Volunteering limits population generalization but does not remove the experimental treatment comparison. Therefore the requested answer is Random assignment to the two methods.

Control and comparison

Objective: Identify a comparison group’s role in an experiment.

A control or comparison group gives a baseline for the treatment contrast. It should differ in the assigned treatment while other procedures are held comparable. Without that baseline, improvement could reflect time, expectation, or another shared factor. Common error: attributing every before-after improvement to the treatment. Restrictions and interpretation: Observing existing behavior differs from assigning treatments; causal conclusions depend on study design. Desmos and method choice: Use Desmos for arithmetic or a graph/table check when that reduces work. Enter parentheses explicitly and verify the requested quantity and units. Use the algebraic steps shown here for an exact result; a rounded graph display is not proof of exact equality.

Formulas

treatment effect is assessed by comparing group outcomes

Worked examples

Try itIn a randomized 146-person study of a new training program, why should a group continue the existing program?Show answer

The existing-program group provides the baseline for comparison. Random assignment and otherwise comparable procedures make an outcome difference more informative about the new treatment. Therefore the requested answer is To provide a comparable baseline for the assigned treatment contrast.

You finished the reading!

Lock it in with a few practice questions, then move on to the next lesson.

Practice Next lesson

Previous lessonMargin of Error and Sample Size Next lessonRandom Sampling versus Random Assignment