1.8 Psychologists Use the Scientific Method

Five Steps in the Scientific Method

• Step one: Formulate a theory

• Step two: Develop a testable hypothesis

• Step three: Test with a research method

• Step four: Analyze the data

• Step five: Share the results and conduct more research

Goals of Psychological Research

• Describe what happens: Involves observing and documenting phenomena, behaviors, or mental processes. Through careful observation, researchers can establish a clear picture of an event or pattern.

• Predict when it happens: Once phenomena are described, researchers aim to forecast future occurrences. This involves identifying relationships between variables that allow for reliable predictions.

• Control what causes it to happen: If a phenomenon can be predicted, the next step is to identify its causes. By manipulating variables, researchers seek to influence or control the outcome.

• Explain why it happens: The ultimate goal is to understand the underlying mechanisms and reasons for observed behaviors or mental processes, providing a deeper insight into the phenomenon.

• Use a continuous, empirical cycle to achieve these goals: This iterative process ensures that findings are constantly refined and validated through ongoing research and evidence gathering.

The Scientific Method as a Continuous Cycle

• Psychologists follow the five steps to obtain objective information

• The method is revisited across the field of psychology and in various research contexts

• The process supports describing, predicting, controlling, and explaining psychological phenomena

• The scientific method consists of 5 steps forming a continuous cycle (no single study provides the final answer). This iterative nature means that research findings often lead to new questions, further hypotheses, and subsequent studies.

• Repetition and replication update and refine theories and findings; ongoing process. Even well-supported theories are continually tested and potentially modified as new evidence emerges.

• Researchers should determine what issues related to the topic need additional investigation to extend findings and test generalizability. This can involve exploring different populations, settings, or conditions to ensure the findings are robust and widely applicable.

Step one: Formulate a Theory

• Psychologists study interesting questions about thinking, feeling, or acting. These questions often stem from real-world observations or prior research gaps.

• A theory is an explanation of how a mental process or behavior occurs. It provides a generalized framework for understanding complex phenomena.

• The theory consists of interconnected ideas that explain prior findings and make predictions about future events. Good theories are parsimonious (simple), testable, and capable of generating multiple hypotheses.

• Formulate theory by reviewing prior research related to the topic. This ensures that new research builds upon existing knowledge and avoids unnecessary duplication.

• Example theory (from this chapter): 'Students learn better by repeated practice than by rereading information.' This theory offers a testable proposition about learning efficacy.

Literature Review and Theory

• Conduct a literature review using resources such as Google Scholar, PsycInfo, and PubMed. These databases provide access to peer-reviewed articles and scholarly works.

• Search by asking questions related to your research question or by keywords related to the topic. Effective search strategies involve using specific terms and Boolean operators (AND, OR, NOT) to narrow results.

• Findings indicate whether similar topics have been investigated and help refine theory. A thorough review can reveal existing evidence, contradictions, or gaps that your research can address.

• A clearly stated theory is the basis for Step 2 of the scientific method, as it provides the conceptual framework from which specific, testable predictions are derived.

Step two: Develop a Testable Hypothesis

• A hypothesis is a testable prediction about the relationship between variables. It translates the general principles of a theory into a specific, verifiable statement.

• Example: a prediction about the effect of a study technique on exam performance (e.g., some technique improves scores relative to another). This makes the abstract theory concrete and measurable.

• After theory, state a specific, testable prediction about what should be observed if the theory is accurate. Hypotheses must be falsifiable, meaning they can be proven wrong by data.

• A single theory is usually tested by several hypotheses across different studies. This allows for a comprehensive examination of the theory's various facets and implications.

• Each hypothesis tests an aspect of the theory by targeting one of the four goals of science: describing, predicting, controlling, and explaining. This ensures a multi-faceted approach to understanding the phenomenon.

• Example goals:

  • Describing: what study techniques students use or think are best. This might involve surveys or observational studies.

  • Predicting: relationship between time spent with a method and exam scores. This would involve correlational studies.

  • Controlling: vary study methods to see effect on outcomes. This is characteristic of experimental designs where variables are manipulated.

  • Explaining: how a technique yields higher scores. This delves into the cognitive processes or mechanisms at play.

• Example hypothesis: 'Students will perform better on an exam after repeated practice with information than after repeated reading of the information.' This hypothesis specifies the independent variable (study technique) and the dependent variable (exam performance).

Step three: Test with a Research Method

• Select the most appropriate research method to test the hypothesis. The choice of method is critical for effectively answering the research question.

• Collect data to evaluate the hypothesis. Data collection must be systematic and adhere to ethical guidelines.

• Example: an experiment comparing repeated practice vs repeated re-reading before an exam. This method allows for controlled manipulation of the study technique.

• After stating the hypothesis, choose the research method to test it. This decision is guided by the specific question, the variables involved, and the level of control required.

• These are described later in the chapter.

• The most appropriate method depends on the goal of your research and your hypothesis. For instance, if the goal is to establish causality, an experimental method is typically preferred.

• The research method you use also depends on how much control you need over manipulating and measuring certain factors in the study. Higher control is often needed in experimental research to isolate causal relationships.

• Each factor, called a variable, is something in the world that can vary and that the researcher can manipulate (change), measure (evaluate), or both. Variables are the building blocks of any scientific investigation.

Types of Research Methods

• The three main types of research methods: Descriptive, Correlational, Experimental.

  • Descriptive methods: Describe what is occurring. These methods are used to observe and systematically record behavior or mental processes. They provide a snapshot of a phenomenon without investigating relationships between variables or cause-and-effect.

  • Example: 'What study techniques do students believe are best?' (e.g., surveys, observational studies, case studies)

  • Correlational methods: Test the relationship between variables. These methods examine how two or more variables are naturally associated. They can identify patterns and make predictions but cannot establish causation.

  • Example: 'What is the relationship between how much time students spend using certain study methods and their later exam scores?' (e.g., surveys, archival data analysis)

  • Experimental methods: Investigate what causes an outcome. These methods involve manipulating one or more variables (independent variables) to observe their effect on an outcome variable (dependent variable), while controlling for other factors. This allows for conclusions about cause-and-effect.

  • Example: 'Which study technique results in the best exam scores possible?' (e.g., controlled experiments with random assignment)

• Other methods:

  • Case studies: In-depth examination of a single individual, group, or phenomenon. Useful for rare conditions or generating new hypotheses.

  • Observational studies: Researchers observe and record behavior in its natural setting without intervention. Can be naturalistic or participant observation.

  • Self-reports: Participants provide information about themselves through questionnaires, interviews, or surveys. Can be susceptible to biases (e.g., social desirability).

Variables and Experimental Design

• Variables: manipulated (independent) vs measured (dependent).

  • Independent Variable (IV): The variable that the researcher intentionally changes or manipulates (e.g., the type of study technique).

  • Dependent Variable (DV): The variable that is measured and is expected to change as a result of the independent variable manipulation (e.g., exam performance).

• Hypothesis targets manipulation of the study technique, making it the independent variable.

• Study groups:

  • Repeated reading: Participants read a passage, then reread it two more times before an exam. This serves as one condition of the independent variable.

  • Repeated practice: Participants read material and practice with it on two practice tests before an exam. This is the other condition, providing a comparison.

• Random assignment to control for confounds (e.g., GPA) to ensure similar academic ability across groups. Random assignment is crucial in experiments for minimizing pre-existing differences between groups, thereby increasing confidence that any observed effects are due to the independent variable.

• Measured outcome: exam performance on a later test. This is the dependent variable, quantifiable for analysis.

Step four: Analyze the Data

• Use appropriate statistical techniques to analyze the data. The choice of statistical test depends on the type of data, the number of variables, and the research question.

• Draw conclusions from the analysis, determining whether the evidence supports or refutes the hypothesis.

• If data do not support the hypothesis, discard or revise the theory and plan to test the revision. This is a critical part of the continuous cycle; negative results are still informative.

• Example outcome: repeated practice leads to better exam results than repeated re-reading, suggesting support for the hypothesis and theory.

• Steps: summarize raw data; then statistically evaluate whether differences exist between groups. Descriptive statistics (mean, median, mode) provide an overview, while inferential statistics help make generalizations.

• Assess whether differences reflect a meaningful/real effect (significant effect) due to the study technique. Statistical significance indicates that an observed effect is unlikely to have occurred by chance alone.

• Avoid post hoc hypothesis changes (no HARKing: Hypothesizing After the Results are Known): document hypotheses and methods before data collection and analysis (Lindsay, 2017; Vazire, 2018). HARKing is an unethical practice that can inflate false positives and undermine scientific integrity.

Step five: Share the Results and Conduct More Research

• Share findings with journals and at conferences. Dissemination of results allows the scientific community to review, critique, and build upon the research.

• Refine the theory based on results, make further predictions, and test those predictions. This closes the loop of the scientific method, leading to new inquiries.

• Aim to determine whether the technique improves learning across different people and situations. This addresses the generalizability of the findings.

• Poster sessions: Researchers present their work on large posters, allowing for informal discussions and questions from attendees.

• Conferences: Provide platforms for researchers to deliver oral presentations or participate in poster sessions, fostering networking and feedback.

• Journals: Publish in peer
  -reviewed journals with background, methodology, analyses, and interpretation. The peer-review process involves expert evaluation of the research for quality, validity, and significance before publication.

• Reporting all data increases confidence; avoid cherry-picking (unethically reporting only data supporting a hypothesis). Transparency in data reporting is essential for maintaining scientific credibility.

Ethical Considerations: preregistration and reporting practices

• Cherry-picking defined: selectively reporting data/analyses that support a hypothesis while omitting those that do not. This can lead to biased conclusions and misrepresentation of findings.

• Importance: full data reporting strengthens credibility of findings and allows other researchers to critically evaluate the study's conclusions. Preregistration of studies (registering hypotheses and methods before data collection) is a powerful tool against cherry-picking and HARKing.

Replication and Transparency in Psychological Research

• Replication: repeating the same procedures with similar participants to expect the same general results; important part of the scientific method; makes research ongoing and increases confidence in findings; references: Goodman et al. 2016; Vazire 2018. Replication is the cornerstone of scientific validity, ensuring that findings are robust and not simply due to chance or specific circumstances.

• Replication failures: some researchers cannot replicate previous findings; such attempts are important and continue to inform evidence related to theories; references: Collaboration 2015; Maxwell et al. 2015; Shrout & Rodgers 2018. Failures to replicate can highlight methodological issues, contextual dependencies, or even initial false positive findings, prompting re-evaluation of theories.

• Transparency and openness: best practices include reporting details of methods and analyses; provide free access to materials and data to enable replication; Morling & Calin
  -Jageman 2020. Open science practices enhance the trustworthiness and reproducibility of research.

• Generalizability: no single study definitively answers a research question; extend findings to diverse participants and settings to increase confidence in importance and generalizability; when many studies converge, findings are more trustworthy; note: repeated practice is more effective for studying than repeated reading (Dunlosky et al. 2013). Generalizability refers to the extent to which research findings can be applied to other situations and populations beyond those directly studied.

• Overall cycle: replication contributes to the ongoing cycle of the scientific method; see Figure 1.17 for the cycle of the five steps. Each replication attempt, successful or not, refines our understanding and leads to further research.

Tobacco Dependence Case: Hardening Hypothesis and Data Sources

• Introduction: 441{,}000 premature deaths per year due to cigarette smoking; leading cause of morbidity and mortality. This highlights the public health significance of understanding smoking cessation.

• Hardening Hypothesis (Hughes, 2011): Smokers who cannot quit may become more nicotine dependent; explains why declines in smoking may become less significant over time. This hypothesis suggests a changing demographic of smokers remaining, making population-level cessation harder.

• Data and Methods: NSDUH data (National Survey on Drug Use and Health), sponsored by SAMHSA; data merged from years 1985, 1988, 1990{-}2011 to analyze trends. Using large, longitudinal datasets like NSDUH allows researchers to track population-level changes and test hypotheses about long-term trends.

• Variables included: Respondent ID, Ever Used Cigarettes, Race, Gender, Year, Daily Smoker, Average cigarettes per day in last 30 days, Need more cigarettes to get the same effect (nicotine dependence), Age Category. These variables allow for comprehensive analysis of smoking patterns and contributing factors.

• Age Category coding (examples): This likely refers to how different age groups were defined and categorized for analysis within the NSDUH data.

Sage's Application of the Scientific Method

• Theory: caffeine influences one's ability to focus on and retain information. This is Sage's initial explanation for how a mental process (focus/retention) occurs.

• Literature review: search for related studies using keywords such as memory, attention, and caffeine. Sage would look for existing evidence to support or refine her theory.

• Hypothesis: people who consume more caffeine while studying will perform better on a test of the studied material. This is a specific, testable prediction derived from her theory.

• Method: design a study (e.g., an experiment) with at least two groups: one drinks caffeine while studying, the other does not. This experimental design allows Sage to manipulate the independent variable (caffeine consumption).

• Data collection and analysis: measure how well each group memorizes the list and determine whether the hypothesis is supported. Sage would use appropriate statistical methods to compare the groups' performance.

• If the hypothesis is not supported: revise the experiment to strengthen the study or revise the theory. This demonstrates the iterative nature of the scientific method.

• If the hypothesis is supported: the study provides evidence in favor of the theory, increasing confidence in its validity.

Sage's Hypothesis and Results

• If Sage's hypothesis is supported, she can feel more confident in her theory. A single study provides initial support, but not definitive proof.

• She will have additional studies to replicate the result, which is crucial for establishing the reliability and generalizability of her findings.

The Scientific Method — Key Steps

• Focus on a theory

• State a hypothesis

• Test with the research method

• Analyze the data

• Report results

• Embark on further inquiry

Sage's Next Steps

• Sage reports her results.

• She uses feedback from other researchers to plan the next study to test her theory, perhaps exploring different dosages of caffeine, types of memory tasks, or participant demographics.

Application Prompt

• How would the scientific method apply to a theory that interests you?

Key Concepts and Definitions

• Theory: a model of connected ideas that explains observations and makes predictions about future events. It's a broad explanation that generates testable hypotheses.

• Hypothesis: a specific testable prediction derived from the theory. It's a precise statement about the expected relationship between variables.

• Literature review: gathering related studies to inform the theory and hypotheses. This foundational step helps contextualize the research and identify knowledge gaps.

• Method: planning how to test the hypothesis (e.g., observation or experiment). This involves selecting the appropriate research design and procedures.

• Data analysis: examining results to determine support for the hypothesis and theory. This step involves statistical evaluation and interpretation of findings.