Research Methods

Psychology is a science. As such, information about human behavior is collected in systematic, objective, and replicable ways, primarily through experiments and correlational studies.


Experiments examine cause-and-effect relationships between at least two variables. They have three essential characteristics:

1. One or more variables are manipulated,

2. The effects of the manipulated variable(s) on other variables are measured,

3. Extraneous factors are held constant so that cause-and-effect conclusions can be drawn.

Independent variables are the variables manipulated by the experimenter, while dependent variables are the ones that are measured. Independent variables are viewed as the cause, while the dependent variables are the effects.

The group that receives an active level of the independent variable (the treatment) is known as the experimental group, while the one that does not receive any treatment is known as the control group. The behavior of the control group sets a standard that the behavior of the experimental group can then be compared to.

When experimenters randomly assign different participants to each condition, they ensure that the experimental and control groups are equivalent at the start of the study. Alternatively, participants can be exposed to all of the conditions and counterbalanced in the order in which conditions are presented.

An experiment’s causal conclusions (internal validity) can be confounded when the independent variable becomes mixed up with an uncontrolled variable. Internal validity is weakened by

1. Demand characteristics – cues that tip participants off as to how they should behave

2. Placebo effects – when the mere expectation of receiving a treatment produces a change in behavior

3. Experimenter expectancy effects – subtle ways in which a researcher’s behavior influences participants to behave in a way consistent with the hypothesis being tested.

Using a double-blind procedure prevents placebo effects and experimenter expectancy effects from biasing research results. This is when neither the participant nor the experimenter knows what group the participant is in.

Correlational Research

Correlational research measures the relationship between naturally occurring variables. A positive correlation occurs when increases in variable X correspond to increases in variable Y. Two variables are negatively correlated when increases in variable X correspond to decreases in variable Y. Causal conclusions cannot be drawn from correlational data, because

1. Variable X may cause Y,

2. Variable Y may cause X, and

3. A third variable (variable Z) maybe be the true causes of both X and Y.

If two variables are correlated, knowing the scores of one variable will help you predict the scores of another. A correlation coefficient describes the strength of a relationship. It ranges from -1.0 to 1.0, with higher absolute values representing stronger relationships.


Surveys involve administering questionnaires or interviews to many people, asking them about their opinions, attitudes, or behaviors. Most surveys study a subset of people, called a sample, which is randomly drawn from the larger population of people the research is interested in. The goal of a survey is to produce a representative sample that reflects the characteristics of the larger population. Researchers can use surveys to determine the frequency with which people do things, and the relationships among those behaviors and attitudes.

Case Studies

Case studies involve an in-depth study of a person, group, or event. Freud's psychoanalytic theory was developed from a series of case studies.

Naturalistic Observation

Naturalistic observation gathers information about animal and human behavior in real-life settings. Rich descriptions of behavior can often be obtained, and relationships between variables can also be examined. Researchers must be careful to avoid influencing the participants being observed by remaining unobtrusive. They must also interpret their observations in an objective manner so that they can achieve a high agreement among observers as to what is happening. Agreement among observers is a measure of "inter-judge," "inter-rater," or "inter-observer" reliability.

Next Article:Click here for Research Methods in Psychology.

Share This Articles