Section 1.5 Notes: The Scientific Method and Research Methods

Section 1.5.1 The Scientific Method

Psychology is defined as the scientific study of behavior and mental processes. Across its focus on behavior and mental processes—and their relation to mental disorders—the field consistently relies on empirical research conducted through the scientific method. Many outside the discipline may assume psychology eschews strict science, but the treatment methods used by mental health professionals are grounded in empirical research and the scientific method. The scientific method is a systematic approach to gathering knowledge about the world. The key descriptor here is systematic, meaning there is a set, repeatable way to proceed.

The method is described with a sequence of steps, though sources vary on the exact number. A preferred framing in this material includes:

  • Step zero: ask questions and be willing to wonder. This is the initial mindset that drives inquiry.
  • Step one: generate a research question or identify a problem to investigate.
  • Step two: attempt to explain the phenomena we wish to study.
  • Step three: test the hypothesis.
  • Step four: interpret the results.
  • Step five: draw conclusions carefully.
  • Step six: communicate our findings to the broader scientific community.

Science rests on three cardinal features that recur throughout the book: observation, experimentation, and measurement.

  • Observation is the process of knowing about the world through firsthand experience. In clinical contexts, observable behaviors often reveal the presence of a mental disorder. Examples include depression (withdrawal from activities), social anxiety (avoidance of social situations), schizophrenia (concern about others watching), and dependent personality disorder (reliance on trusted companions for major decisions). These behaviors can be observed by clinicians, patients, families, or friends.
  • Experimentation is the ability to make causal statements by isolating and manipulating variables. In an experimental design, one variable is manipulated while others are controlled to observe effects on a second variable. A common hypothetical study involves bipolar disorder treatment. Imagine three groups: a control group receiving no treatment, a control group receiving an existing proven treatment, and an experimental group receiving a new treatment. The manipulated variable is the type of treatment (no treatment, older treatment, newer treatment). The expected outcomes are:
    • No treatment: no change in bipolar symptoms.
    • Older treatment: a general reduction in symptoms.
    • Newer treatment: the same or better reduction in symptoms compared to the older treatment.
      The key question is whether the newer treatment offers additional value, such as cost savings, even if symptom reduction is similar.
  • Measurement is how we evaluate outcomes, typically by quantifying symptoms before and after treatment. A common approach in drug trials is a pre–post design:
    D = ext{Score}{ ext{post}} - ext{Score}{ ext{pre}}
    where $D$ represents the change in symptom severity. This pre–post testing framework provides a concrete way to assess whether a treatment yields measurable improvement.

These ideas connect to foundational principles: making careful observations, designing experiments to test causal claims, and using precise measurements to quantify change. The real-world relevance is clear in clinical decision-making, where treatments are selected based on evidence that they produce reliable, measurable improvements and, ideally, cost-effective benefits.

Key implications and practical notes

  • Placebo effects are addressed in experimental design to ensure observed changes are due to the treatment itself rather than expectations. In drug studies, a placebo (an inert sugar pill designed to resemble the real medication) is used to blind participants and reduce expectancy biases.
  • Communication of findings to the broader scientific community is essential for replication, critique, and cumulative knowledge building. This ensures that what works in one study may be validated or refined in subsequent research.

Section 1.5.2 Research Methods

Psychology uses five main research designs to study behavior and mental processes. Each design has unique strengths and limitations, and researchers often employ multiple methods to obtain a comprehensive understanding. Below, each design is described with its typical procedures, advantages, and drawbacks, followed by examples drawn from the transcript.

1.5.2.1 Naturalistic and Laboratory Observation

Naturalistic observation involves studying behavior in its natural environment (e.g., home, school, forest) without manipulating conditions. The researcher records behavior as it occurs, often using multiple judges to enhance reliability in measurement. The advantage is that behavior is observed in real-time without experimental interference, preserving ecological validity; the downside is that behavior may take a long time to appear, and the presence of the observer can influence behavior (the observer effect).

Laboratory observation occurs in a controlled setting (the laboratory). The researcher can use sophisticated equipment and standardized tasks (e.g., parent–child interactions, toy play, meals, or brief separations) to obtain precise recordings for later analysis. The main advantage is greater control and precision, but a major disadvantage is potential artificiality: participants may alter their behavior because they know they are being observed.

1.5.2.2 Case Studies

Case studies provide a detailed description of one person or a small group, historically exemplified by Sigmund Freud’s psychoanalytic work. The primary advantage is depth: rich, nuanced descriptions of a particular case can yield insights, generate hypotheses, and illuminate rare or unusual conditions. The main drawbacks are limited generalizability to broader populations and susceptibility to researcher bias in what is included or omitted from the narrative. Nevertheless, case studies can spark novel ideas about abnormal behavior and are especially valuable for studying unusual disorders or rare conditions when large samples are not feasible.

1.5.2.3 Surveys / Self-Report Data

Surveys involve questionnaires (often with scales) designed to assess a psychological construct such as parenting style, depression, locus of control, or sensation seeking. They may be administered on paper, by computer, or via interviews (structured or unstructured).

Advantages include the ability to collect large amounts of data quickly and efficiently. Drawbacks include potential social desirability bias—respondents answering in a way they think is more acceptable—and dishonesty. The design of questions and the mode of administration can influence responses. Surveys can be used to gather data on sensitive topics, but researchers must consider issues like framing, reliability, and validity.

1.5.2.4 Correlational Research

Correlational studies examine relationships between two variables (or two sets of variables) without manipulating them. A key statistic is the correlation coefficient, denoted by $r$, which quantifies the strength and direction of the relationship:

-1.00 \, \leq \, r \, \leq \, 1.00

  • A negative correlation indicates that as one variable increases, the other tends to decrease (e.g., as parental rigidity increases, child attachment may decrease).
  • A positive correlation indicates that as one variable increases, the other tends to increase (e.g., warmer parenting with greater child attachment).

Advantages of correlational research include the ability to study relationships between virtually any variables. A major limitation is that correlation does not imply causation; two variables may be related due to a third variable or coincidence. A light-hearted example from the transcript compared making peanut butter and jelly sandwiches with being attracted to someone sitting nearby; the two variables are not causally related, illustrating spurious correlations.

A special form of correlational research is epidemiology, which measures the prevalence (how widespread) and incidence (new cases) of a disorder within a population. Epidemiological studies help identify patterns and risk factors at the population level; see references to foundational sections for definitions.

1.5.2.5 Experiments

Experiments are the quintessential controlled tests of hypotheses. They involve manipulating one variable (the independent variable, IV) and measuring its effect on another variable (the dependent variable, DV). Experiments typically include a control group (which does not receive the manipulation) and an experimental group (which does). Random assignment assigns participants to groups with equal probability, helping to balance confounding variables.

Key concepts include:

  • IV: the manipulated variable.
  • DV: the measured outcome.
  • Control group vs. experimental group: enables causal inference by providing a baseline for comparison.
  • Placebo: in drug studies, a sham treatment administered to control for expectancy effects while keeping participants blind to their assignment.

The transcript also highlights single-subject experimental designs, which are useful when large samples are not possible. A prominent example is the ABAB design (also called reversal design): A represents baseline phases, B represents treatment phases, and the pattern ABAB allows researchers to assess whether the behavior changes systematically with the introduction and withdrawal of the intervention.

Concrete examples cited include:

  • Cutler, Miles, and Carson (1998) used social stories to reduce tantrum behavior in a 12-year-old with autism, fragile X syndrome, and intermittent explosive disorder, demonstrating that precursors to tantrum behavior decreased during the intervention (B) and increased when it was withdrawn (A).
  • Balakrishnan (Balakrishnan, 2017) used an ABAB design to examine social stories as a social learning tool for children with autism spectrum disorder (ASD). Baseline (A) was followed by treatment (B), then a return to baseline (A), and a final treatment phase (B). All participants showed improved positive peer interactions during treatment phases, with reductions in interactions when treatment was withdrawn, supporting the conclusion that social stories increased positive peer interactions in children with ASD.

Single-subject designs reduce potential confounds by employing multiple strategies to isolate effects, and they are particularly valuable when studying individuals with rare or heterogeneous conditions.

1.5.2.6 Multi-Method Research

No single research method is perfect; each comes with strengths and limitations. To obtain the most accurate and comprehensive picture of what affects behavior and mental processes, psychologists typically employ multiple methods across different stages of a study. This approach is known as multi-method research. By combining observations, self-reports, experiments, and other designs, researchers can triangulate findings, improve validity, and offset individual method biases. In practice, multi-method research enhances the reliability and applicability of conclusions in psychology and clinical practice, reflecting the field’s emphasis on converging evidence from diverse sources.