AP Psychology | Science Practice 2 | Research Methods & Design

Research Methods and Design

Psychologists use systematic research to study behavior and mental processes. Well-designed studies help distinguish evidence-based findings from unsupported claims. The scientific method provides a structured way to form hypotheses, collect and analyze evidence, and draw conclusions.

Scientific Method and Research Design

• Form a hypothesis: Make a prediction, usually based on a theory.
  • A hypothesis must be falsifiable: there must be some possible result that could show it is wrong.
  • Vague ideas such as “positive energy improves productivity” are difficult to test because “positive energy” is not clearly measurable.
  • A more testable hypothesis identifies observable factors—for example, comparing job-satisfaction scores in offices with motivational quotes and offices without them.
• Identify and operationally define variables:
  • A variable is something measured, controlled, or manipulated in a study.
  • An operational definition states precisely how a variable will be measured or manipulated, translating an abstract idea into something observable.
• Consider confounding variables: These are outside factors that may affect results or make it appear that the studied variables are related when they are not. Researchers should identify and control them where possible.
• Measure the variables:
  • Qualitative measures collect non-numerical information, such as detailed responses in structured interviews.
  • Quantitative measures collect numerical data that can be analyzed statistically. Likert scales, which ask respondents to rate their agreement with statements, are one example.
• Review and replicate findings:
  • Peer review is an evaluation of a study by other experts, who assess its hypothesis, methods, analysis, and conclusions.
  • Replication means repeating a study using the same methods to see whether similar results occur.

Experimental Research

Experimental research manipulates a variable to test whether it causes a change in another variable.

Using the video’s sleep-medication example:

• Independent variable (IV): The factor researchers manipulate; it is the proposed cause. Here, it is whether participants receive the sleep medication.
• Dependent variable (DV): The outcome researchers measure. Here, it is how well-rested participants report feeling.
• Population and sample:
  • The population is the broader group researchers want to understand, such as people with sleep problems.
  • A sample is the subset who participate in the study.
  • A representative sample reflects key characteristics of the population, such as age, gender, and ethnicity.
• Sampling methods:
  • Random sampling gives each member of the population an equal chance of being selected. It can reduce sampling bias and help produce a representative sample.
  • Convenience sampling recruits people who are readily available and willing to participate. It is easier, but may overrepresent or underrepresent certain characteristics, creating sampling bias.
• Groups and assignment:
  • The experimental group receives the treatment being tested.
  • The control group does not receive the active treatment and provides a comparison.
  • A placebo resembles the actual treatment but has no active physical effect. It helps researchers distinguish the medication’s effects from participants’ expectations.
  • Random assignment places participants into groups by chance. Unlike random sampling, it concerns assigning participants to conditions and helps make groups comparable before treatment.
• Reduce bias:
  • Experimenter bias occurs when researchers’ expectations unintentionally influence how they conduct a study or interpret results.
  • Social-desirability bias occurs when participants respond in ways they think are socially acceptable or expected.
  • In a single-blind study, participants do not know whether they are receiving the treatment or a placebo.
  • In a double-blind study, neither participants nor the experimenters interacting with them know who receives the treatment during the study. Blinding helps reduce both participant and experimenter influence.

Strengths and limitations: Experiments can establish cause-and-effect relationships by manipulating one variable while controlling others. However, they may require substantial resources and strict control, which can be impractical or unethical. Highly controlled conditions may also make findings less applicable to everyday settings.

Non-Experimental Research

Non-experimental methods study behavior or relationships without manipulating variables. They are useful when experiments are impractical or ethically inappropriate, but generally cannot establish causation.

• Case study: An in-depth examination of one person or a small group.
  • Strengths: Provides detailed information, particularly about rare or unusual cases.
  • Limitations: Can be time-consuming, may not generalize to the wider population, and can be affected by researcher bias.
• Correlational study: Examines whether variables are related.
  • Correlation can reveal an association, but does not establish causation.
  • The directionality problem is uncertainty about which variable influences the other.
  • The third-variable problem occurs when another factor may explain the observed association. For example, academic stress might be related to both social-media use and anxiety.
  • Correlational studies can handle large data sets and provide useful preliminary evidence.
• Meta-analysis: Combines results from multiple studies to identify overall patterns.
  • It can help clarify mixed findings and provide greater statistical power.
  • Its conclusions depend on the quality of the studies included.
• Naturalistic observation: Observes behavior in its usual environment without interfering.
  • Strengths: Captures behavior as it naturally occurs and can be useful when an experiment would be unethical—for example, when studying bullying at school.
  • Limitations: Researchers have little control over events and may not be able to determine why a behavior occurred. People may also change their behavior if they know they are being observed.
• Survey: Collects responses from many people, often efficiently and at relatively low cost.
  • Limitations: Responses may be affected by self-report bias (inaccurate or dishonest reporting) and social-desirability bias.

Research Ethics

Research involving human participants or non-human animals must undergo an institutional ethics review. The video emphasizes these safeguards:

• Protect participants from physical and psychological harm.
• Protect participants’ privacy through confidentiality or anonymity, and avoid publishing identifying information.
• Use deception only when necessary and carefully.
• Debrief participants after the study: explain its true purpose, disclose any deception, and correct misunderstandings.

Speakers and Sources Featured

• Tim Steedman, presenter of Get Psyched with Tim Steedman.
• No other speakers or quoted sources are identified in the subtitles.