Premed · Premed · Introductory Psychology
Lecture 2: Research Methods in Psychology
Introductory Psychology
Learning Objectives
By the end of this lecture, students will be able to:
- Describe the steps of the scientific method as applied to psychological research
- Distinguish between descriptive, correlational, and experimental research designs
- Identify independent, dependent, and confounding variables in experiments
- Explain key concepts of statistical analysis including measures of central tendency and significance
- Discuss ethical principles governing research with human and animal participants
Lecture Content
I. The Scientific Method in Psychology
Psychology uses the scientific method to systematically study behavior and mental processes. The process begins with identifying a question, often drawn from observation, theory, or previous research. From that question, a researcher formulates a hypothesis — a testable prediction about the relationship between variables — and defines those variables in measurable, concrete terms through operational definitions. The next step is to design and conduct a study using an appropriate method, after which the data are analyzed using statistics. Based on the results, the researcher draws conclusions that support, reject, or revise the hypothesis, and finally reports the findings in a peer-reviewed publication so that other researchers can attempt replication.
Underlying this process is the concept of theory, an organized set of principles that explains and predicts observed phenomena. Theories generate hypotheses, and data either support or challenge them. Good theories are falsifiable, parsimonious, and generative. Throughout every stage of research, critical thinking is essential: psychologists must avoid emotional reasoning, anecdotal evidence, and confirmation bias, and they must consider alternative explanations and weigh the quality of evidence carefully.
II. Descriptive Research Methods
Descriptive research methods aim to observe and describe behavior without manipulating variables. Naturalistic observation involves watching behavior in its natural environment without interference. This approach offers high ecological validity because it captures real-world behavior, but it is susceptible to observer bias, provides limited control, and cannot determine causation. Jane Goodall's studies of chimpanzee behavior are a classic example. Laboratory observation, by contrast, takes place in a controlled setting, offering greater control at the cost of reduced naturalness.
Case studies are in-depth investigations of a single individual, group, or event. They yield rich qualitative data and are especially useful for examining rare conditions — the cases of Phineas Gage and patient H.M. being among the most famous. The main limitation is that findings from a single case cannot be generalized to the broader population. Surveys and questionnaires collect self-report data from large samples and can assess attitudes, beliefs, behaviors, and demographics. However, they are vulnerable to wording effects (leading or ambiguous questions), social desirability bias (the tendency to respond in socially acceptable ways), and sampling problems. For results to be meaningful, the sample must represent the population of interest, ideally through random sampling, in which every member of the population has an equal chance of being selected.
<image>A four-panel comparison of descriptive research methods. Panel A: Naturalistic observation — a researcher with binoculars observing children on a playground, with a data recording sheet. Panel B: Case study — a detailed patient file with brain scans and interview notes for a single individual. Panel C: Survey — a questionnaire form with Likert scale items being distributed to a large group. Panel D: A table comparing advantages and limitations of each method in terms of control, generalizability, and depth.</image>
III. Correlational Research
Correlational research examines the relationship between two or more variables without manipulating them. The relationship is expressed as a correlation coefficient (r), which ranges from -1.00 to +1.00. A positive correlation means that as one variable increases, the other increases as well — height and weight are a common example. A negative correlation means that as one variable increases, the other decreases, as with stress and immune function. A zero correlation indicates no systematic relationship. The strength of the correlation is determined by the absolute value of r: the closer it is to 1.00, the stronger the relationship. Scatterplots provide a visual representation of correlational data.
The most critical limitation of correlational research is that correlation does not imply causation. For any observed correlation between variables A and B, there are at least three possible explanations: A causes B, B causes A, or a third variable C causes both. The classic example is that ice cream sales correlate with drowning rates — not because ice cream causes drowning, but because hot weather (the third variable) drives both. Illusory correlations, in which people perceive a relationship where none actually exists, are fueled by confirmation bias and a tendency to pay attention to memorable co-occurrences.
<image>Three scatterplot graphs displayed side by side. Panel A: A positive correlation (r = +0.85) with data points forming an upward slope, labeled "Study hours vs. Exam scores." Panel B: A negative correlation (r = -0.72) with data points forming a downward slope, labeled "Stress level vs. Immune function." Panel C: A near-zero correlation (r = +0.05) with data points scattered randomly, labeled "Shoe size vs. Intelligence." Below the three graphs, a diagram showing three possible causal interpretations of a correlation: A causes B, B causes A, or C causes both.</image>
IV. Experimental Research
The experiment is the only research method that can establish cause-and-effect relationships. In an experiment, the researcher manipulates the independent variable (IV) and measures its effect on the dependent variable (DV). Participants in the experimental group receive the treatment or manipulation, while those in the control group do not, serving as a baseline for comparison. Random assignment — randomly placing participants into experimental or control groups — ensures that the groups are roughly equivalent before the manipulation begins. It is important not to confuse random assignment (how participants are allocated to conditions) with random sampling (how participants are selected from a population).
Confounding variables are extraneous factors that change along with the independent variable and could therefore influence the dependent variable. Researchers use various controls to minimize confounds and bias. The placebo effect occurs when participants improve simply because they believe they are receiving treatment, and it can be addressed through single-blind procedures (where participants do not know their group assignment) or double-blind procedures (where neither participants nor researchers know the assignments). Experimenter bias, sometimes called the Rosenthal effect, occurs when a researcher's expectations subtly influence participant behavior. Experimental designs may be between-subjects, with different participants in each condition, or within-subjects (repeated measures), where the same participants are tested in all conditions. The within-subjects approach controls for individual differences but is vulnerable to order effects, which can be addressed through counterbalancing.
<image>A flowchart of an experimental design. At the top, a population pool is shown. An arrow labeled "Random sampling" leads to a sample. The sample is then split by "Random assignment" into two groups: Experimental Group (receives the independent variable manipulation) and Control Group (receives placebo or no treatment). Both groups are then measured on the dependent variable. Arrows lead to a comparison box at the bottom showing statistical analysis. Key terms (IV, DV, confounds, placebo) are annotated with callout boxes.</image>
V. Statistics in Psychology
Descriptive statistics summarize and organize data. Measures of central tendency include the mean (the arithmetic average, which is sensitive to outliers), the median (the middle score in a ranked distribution, which resists outliers), and the mode (the most frequently occurring score). Measures of variability include the range (the difference between the highest and lowest scores) and the standard deviation (the average distance of scores from the mean). In a normal distribution, sometimes called a bell curve, most scores cluster near the mean with symmetrical tails on either side.
Inferential statistics allow researchers to determine whether findings from a sample can be generalized to the broader population. Statistical significance indicates the probability that results are not due to chance, and the conventional threshold is p < 0.05, meaning there is less than a 5% probability that the result occurred by chance alone. Effect size, which is distinct from significance, describes the magnitude of the difference or relationship found. Replication — the ability of other researchers to reproduce findings — is essential for establishing reliability. Psychology has faced a "replication crisis" in recent years, as many classic findings have failed to replicate, prompting renewed emphasis on methodological rigor and transparency.
VI. Ethics in Psychological Research
The American Psychological Association (APA) has established ethical principles to protect research participants. Informed consent requires that participants be told the purpose, procedures, risks, and their right to withdraw from a study at any time. Deception is permissible only when scientifically necessary and when a thorough debriefing occurs afterward to explain the true purpose of the study. Researchers must also maintain confidentiality of participants' personal information and minimize both physical and psychological harm.
Institutional Review Boards (IRBs) review and approve all research involving human participants, weighing the potential benefits of the research against the risks to participants. Several landmark studies have shaped modern research ethics. Milgram's obedience studies caused significant psychological distress, Zimbardo's Stanford Prison Experiment subjected participants to extreme distress and raised questions about the loss of researcher objectivity, and the Tuskegee syphilis study — in which treatment was withheld from African American men — led directly to many of the research ethics regulations now in place.
Animal research must follow guidelines set by the APA and Institutional Animal Care and Use Committees (IACUCs). Such research is justified when the potential benefits outweigh the suffering involved, and researchers are expected to follow the 3 Rs: Replace (use alternatives to animal subjects when possible), Reduce (use the fewest animals necessary), and Refine (minimize suffering through improved procedures).
