Skip to content
Tutorials Center
My learning

Topics

Lesson 10 of 44 ยท The Logic of Research

Chapter 2 Review: The Logic of Research

Review variables, operational definitions, theories, falsifiability and the four types of validity, with a recap table and a ten-question mixed quiz.

This chapter was about the reasoning inside a study. A good study links theory, hypothesis, variables and data in a chain that a skeptical reader can follow. Use this review to check your grip on each link.

A one-page summary of chapter 2: variable roles, operational definitions, falsifiability and the four validities.
Chapter 2 in one picture: define your variables, make a risky prediction, and check each kind of validity.

Recap table

Idea Key points
IV and DV IV is manipulated or presumed cause; DV is the measured outcome
Control and confound Controls are held constant; confounds vary with the IV and muddy the result
Operational definition A repeatable description of how a concept is measured or manipulated
Converging operations Several different measures of one concept, checked for agreement
Theory vs hypothesis A theory explains many observations; a hypothesis is a specific prediction
Falsifiability A claim must forbid some outcome to be informative
Internal validity Did X cause Y in this study?
External validity Will it generalize to other people, settings, times?
Construct validity Do the measures reflect the intended concepts?
Statistical conclusion validity Was the analysis sound?

Revisit the lessons if any row feels shaky: variables, theories and validity.

Check your understanding

Quick check ยท 10 questions

of 10 answered You scored / 10 ยท perfect!

  1. 1. In a study of whether music improves arithmetic speed, the time taken to finish the test is the

    It is the measured outcome. The music condition is the IV.
  2. 2. Which operational definition of 'aggression' is best?

    It names a procedure and a score that another researcher could repeat, though its construct validity can still be debated.
  3. 3. Why do researchers use converging operations?

    If different measures agree, the result is less likely to be an artifact of one measure.
  4. 4. A theory predicts that anxious people will perform worse on a timed task than calm people. What would count against it?

    A falsifiable theory forbids some outcome, here equal or better performance by anxious people.
  5. 5. Which Wason selection task cards test 'if vowel then even' among E, K, 4, 7?

    Only a vowel behind 7 or an odd number behind E could falsify the rule.
  6. 6. A school gives extra help to its lowest-scoring pupils and their scores rise. A skeptic's best first question is

    Without a comparable group, regression to the mean could explain the rise.
  7. 7. Which threat is reduced mainly by random assignment?

    Random assignment makes the groups equivalent on average at the start.
  8. 8. A tightly controlled lab experiment with an artificial task is most at risk for weakness in

    Control supports internal validity, while artificial settings can limit generalization.
  9. 9. Participants work out what a study is about and act accordingly. This is

    Demand characteristics threaten construct validity because behavior reflects guesses about the purpose.
  10. 10. Why is a theory that can explain any result weak?

    Informative theories take risks. One that fits everything has no testable content.

Summary

  • Know the roles of variables and write operational definitions that others can repeat.
  • Prefer risky, falsifiable predictions to all-purpose explanations.
  • Ask four validity questions of every study: internal, external, construct and statistical.

Next: Chapter 3 Overview: Measurement and Sampling