Research methods & design | AP Psychology Science practices

The study of psychology systematically examines human behavior and mental processes. Instead of relying on everyday guesses or personal intuition, psychology uses a rigorous framework of scientific inquiry. These frameworks guide psychologists as they form theories, design studies, analyze results, and apply findings responsibly.

For AP Psychology students, mastering these practices matters for exam success and for building critical thinking skills grounded in evidence.

Concept application

A key skill in psychology is applying different perspectives to analyze behavior and mental processes. You should be able to recall the main ideas of major approaches (biological, cognitive, behavioral, humanistic, and sociocultural) and compare how these different frameworks explain the same phenomenon.

Consider anxiety as an example:

From a cognitive perspective, anxiety may stem from persistent worry or negative thought patterns that exaggerate threats.
From a behavioral perspective, anxiety can be understood as a learned avoidance response that is reinforced over time.
From a biological perspective, anxiety may be linked to genetics, brain structures, or neurotransmitter activity.
From a humanistic perspective, anxiety may reflect a gap between one’s ideal self and actual self, or a barrier to achieving self-actualization.
From a sociocultural perspective, anxiety may be shaped by cultural expectations (for example, around achievement or interdependence) that influence stress responses.

Taken together, these perspectives create a richer, more complete explanation than any single approach alone.

You also need to recognize cognitive biases - systematic thinking errors that affect how people process information and make judgments - so you can guard against overly simple or distorted interpretations. Examples include:

Confirmation bias (favoring information that supports pre-existing beliefs)
Hindsight bias (believing an outcome was predictable after it occurs)
Overconfidence (overestimating one’s knowledge or accuracy)

Psychology also operates within cultural contexts that shape cognition, emotion, and behavior. Understanding cultural influences helps prevent misapplying psychological concepts in ways that are discriminatory or inappropriate.

For example, in collectivist societies, prioritizing group harmony may lead people to suppress personal desires for the community’s welfare. In contrast, individualistic cultures often emphasize independence and personal accomplishment, which can shape different definitions of success and different norms around mental health.

On the other hand, a behavior seen as assertive in one culture might be viewed as selfish in another. Similarly, many psychological theories were developed in Western, educated, industrialized, rich, and democratic (WEIRD) contexts and may not apply universally.

Recognizing these limitations supports a more inclusive and representative science.

Research methods and design

Psychological research uses different designs, each with strengths and limitations. The two main categories are experimental methodologies and non-experimental methodologies.

Experimental methodologies involve:

Manipulation of independent variables while controlling other factors (to study cause-and-effect relationships between variables).
Random assignment to create equivalent groups, which supports cause-and-effect conclusions. For example, to study sleep’s effect on memory, participants might be randomly assigned to different sleep durations before a memory test. If more sleep is consistently linked to better performance, a causal conclusion is more justified.

In contrast, when manipulating variables is unethical or impractical, researchers use non-experimental methodologies. These designs measure naturally occurring phenomena without researcher manipulation. This type includes:

Case studies: Involve a detailed examination of a specific subject within a real-life context.
Correlational studies: Identify relationships between variables without establishing causation. For instance, traumatic experiences may correlate positively with depression but do not prove that traumatic experiences cause depression.
Naturalistic observation: Behavior is recorded in natural environments to avoid laboratory artificiality.
Meta-analyses: Combine results from multiple studies to draw broader conclusions.

Fundamental principles in experiments (generally) involve:

Sample representation (whether the participants represent the population being studied). Participants can be selected through random sampling (where everyone has an equal chance of being chosen, which reduces bias) or convenience sampling (where participants are chosen based on proximity or ease of access, which can skew results).
Operational definitions clarify how concepts are measured so other researchers can replicate the study precisely. For example, stress might be measured using cortisol levels or questionnaire scores.
Measurement instruments (procedures for collecting data). These can be qualitative (non-numeric characteristics) or quantitative (numeric data that can be counted or measured on scales). One qualitative instrument is structured interviews (a set of predetermined questions). One quantitative instrument is Likert Scales (rating scales used to indicate agreement or disagreement with a statement).
Scientific process (a systematic approach for asking questions, forming hypotheses, conducting experiments, analyzing data, and drawing conclusions in a reproducible and objective way). This process includes practices such as peer review (experts evaluate a study before publication or acceptance) and replication (repeating studies to verify results and distinguish empirical findings from anecdotes).
Falsifiability (hypotheses must be written so evidence could show they are wrong).

How does random sampling differ from convenience sampling in psychological research?

(spoiler)

Random sampling gives every individual an equal chance of being selected, while convenience sampling chooses participants based on ease of access, increasing potential bias.

Principles unique to experimental methodologies include:

Selecting variables to study. Examples include independent variables (variables researchers manipulate to study their effects), dependent variables (variables measured to see the independent variable’s impact), and confounding variables (outside factors that may influence the relationship between the independent and dependent variables, distorting results).
Curating groups to study. These include experimental groups (receive the manipulation) and control groups (do not receive the manipulation, providing a baseline for comparison).
Within control groups, placebos may be used (a harmless treatment or substance that mimics the real treatment without active components). A placebo can produce a placebo effect (changes that occur because participants expect the treatment to work, not because of the treatment itself).
Determining whether the experiment is single-blind (participants don’t know whether they are in the control or experimental group) or double-blind (neither participants nor experimenters know group assignments). Single-blind designs help limit the placebo effect. Double-blind designs help prevent experimenter bias (researchers’ expectations influencing outcomes) and social desirability bias (participants responding in ways they think will be viewed positively rather than responding naturally).
Identifying whether the experiment involves sampling bias (the sample is not representative of the target population) or whether the findings have generalizability (the ability to apply results to a broader population).

How is a placebo effect distinct from experimenter bias in research design?

(spoiler)

The placebo effect comes from participants’ expectations influencing outcomes, while experimenter bias results from researchers’ expectations affecting results.

Techniques and nuances unique to non-experimental methodologies include:

Selecting a research technique such as survey techniques (collecting data through interviews or questionnaires). Surveys can be affected by social desirability bias or self-report bias (inaccurate reporting that leads to misleading conclusions).
Clarifying whether a relationship reflects causation (one variable directly changes another) or correlation (variables are associated, but one does not necessarily cause the other).
Evaluating whether correlational results involve a directionality problem (uncertainty about which variable influences the other) or a third variable problem (an unmeasured variable influences both variables, creating or inflating the relationship).

For AP Psychology students, mastering these practices matters for exam success and for building critical thinking skills grounded in evidence.

Concept application

Consider anxiety as an example:

From a cognitive perspective, anxiety may stem from persistent worry or negative thought patterns that exaggerate threats.
From a behavioral perspective, anxiety can be understood as a learned avoidance response that is reinforced over time.
From a biological perspective, anxiety may be linked to genetics, brain structures, or neurotransmitter activity.
From a humanistic perspective, anxiety may reflect a gap between one’s ideal self and actual self, or a barrier to achieving self-actualization.
From a sociocultural perspective, anxiety may be shaped by cultural expectations (for example, around achievement or interdependence) that influence stress responses.

Taken together, these perspectives create a richer, more complete explanation than any single approach alone.

Confirmation bias (favoring information that supports pre-existing beliefs)
Hindsight bias (believing an outcome was predictable after it occurs)
Overconfidence (overestimating one’s knowledge or accuracy)

Recognizing these limitations supports a more inclusive and representative science.

Research methods and design

Psychological research uses different designs, each with strengths and limitations. The two main categories are experimental methodologies and non-experimental methodologies.

Experimental methodologies involve:

Manipulation of independent variables while controlling other factors (to study cause-and-effect relationships between variables).
Random assignment to create equivalent groups, which supports cause-and-effect conclusions. For example, to study sleep’s effect on memory, participants might be randomly assigned to different sleep durations before a memory test. If more sleep is consistently linked to better performance, a causal conclusion is more justified.

Case studies: Involve a detailed examination of a specific subject within a real-life context.
Correlational studies: Identify relationships between variables without establishing causation. For instance, traumatic experiences may correlate positively with depression but do not prove that traumatic experiences cause depression.
Naturalistic observation: Behavior is recorded in natural environments to avoid laboratory artificiality.
Meta-analyses: Combine results from multiple studies to draw broader conclusions.

Fundamental principles in experiments (generally) involve:

Sample representation (whether the participants represent the population being studied). Participants can be selected through random sampling (where everyone has an equal chance of being chosen, which reduces bias) or convenience sampling (where participants are chosen based on proximity or ease of access, which can skew results).
Operational definitions clarify how concepts are measured so other researchers can replicate the study precisely. For example, stress might be measured using cortisol levels or questionnaire scores.
Measurement instruments (procedures for collecting data). These can be qualitative (non-numeric characteristics) or quantitative (numeric data that can be counted or measured on scales). One qualitative instrument is structured interviews (a set of predetermined questions). One quantitative instrument is Likert Scales (rating scales used to indicate agreement or disagreement with a statement).
Scientific process (a systematic approach for asking questions, forming hypotheses, conducting experiments, analyzing data, and drawing conclusions in a reproducible and objective way). This process includes practices such as peer review (experts evaluate a study before publication or acceptance) and replication (repeating studies to verify results and distinguish empirical findings from anecdotes).
Falsifiability (hypotheses must be written so evidence could show they are wrong).

How does random sampling differ from convenience sampling in psychological research?

(spoiler)

Random sampling gives every individual an equal chance of being selected, while convenience sampling chooses participants based on ease of access, increasing potential bias.

Principles unique to experimental methodologies include:

Selecting variables to study. Examples include independent variables (variables researchers manipulate to study their effects), dependent variables (variables measured to see the independent variable’s impact), and confounding variables (outside factors that may influence the relationship between the independent and dependent variables, distorting results).
Curating groups to study. These include experimental groups (receive the manipulation) and control groups (do not receive the manipulation, providing a baseline for comparison).
Within control groups, placebos may be used (a harmless treatment or substance that mimics the real treatment without active components). A placebo can produce a placebo effect (changes that occur because participants expect the treatment to work, not because of the treatment itself).
Determining whether the experiment is single-blind (participants don’t know whether they are in the control or experimental group) or double-blind (neither participants nor experimenters know group assignments). Single-blind designs help limit the placebo effect. Double-blind designs help prevent experimenter bias (researchers’ expectations influencing outcomes) and social desirability bias (participants responding in ways they think will be viewed positively rather than responding naturally).
Identifying whether the experiment involves sampling bias (the sample is not representative of the target population) or whether the findings have generalizability (the ability to apply results to a broader population).

How is a placebo effect distinct from experimenter bias in research design?

(spoiler)

The placebo effect comes from participants’ expectations influencing outcomes, while experimenter bias results from researchers’ expectations affecting results.

Techniques and nuances unique to non-experimental methodologies include:

Selecting a research technique such as survey techniques (collecting data through interviews or questionnaires). Surveys can be affected by social desirability bias or self-report bias (inaccurate reporting that leads to misleading conclusions).
Clarifying whether a relationship reflects causation (one variable directly changes another) or correlation (variables are associated, but one does not necessarily cause the other).
Evaluating whether correlational results involve a directionality problem (uncertainty about which variable influences the other) or a third variable problem (an unmeasured variable influences both variables, creating or inflating the relationship).