Methodology is the science of methods, i.e., the ways in which psychology as a science acquires reliable and valid knowledge about human experience and behavior. It covers procedures from planning research to evaluating data and diagnosing clients.
For psychology not to remain mere speculation, it must base its claims on strict scientific rules. Methods are generally divided into research methods (aimed at discovering new general laws of how the mind works) and diagnostic methods (aimed at identifying the traits or state of a specific individual).
Basic Methods of Psychology
Psychology uses a wide range of approaches to collect data, from observation to long-term studies of large groups.
Introspection
Self-observation. A method where a person examines and describes their own internal experiences, feelings, and thoughts.
Extrospection
Observing the behavior of other people from the outside. It focuses on objectively observable and measurable manifestations.
Case Study
A detailed, in-depth study and description of one specific case (an individual, family, or situation). It is frequently used in clinical practice.
Longitudinal Research
Long-term research in which the same group of people is observed and tested repeatedly over several years (e.g., studying the development of intelligence in children from age 5 to 15).
Cross-sectional Research
A one-time research study where several different groups of people of various ages are tested simultaneously (e.g., comparing the memory of 20-year-olds, 40-year-olds, and 60-year-olds at the same moment).
Experiment and Variables
While observation allows us only to describe behavior, an experiment is the only research method that can prove causality (a cause-and-effect relationship between phenomena). The researcher actively manipulates conditions and observes the consequences.
Structure of an ExperimentVariables
Every experiment relies on working with variables:
Independent Variable
The cause. The variable that the experimenter deliberately manipulates to observe its effect.
Dependent Variable
The effect. The variable that the researcher measures. Its value depends on the changes in the independent variable.
Intervening Variable
A confounding factor. An external influence that was not part of the plan but could distort the results of the experiment (e.g., fatigue, noise, participants' mood).
Exam Trap
How to identify variables? Imagine an experiment investigating the effect of caffeine on students' memory.
Independent variable (cause): The administered dose of caffeine (the researcher determines who gets coffee and who gets water).
Dependent variable (effect): Performance on a memory test (what we ultimately measure and calculate).
Correlation vs. Causation
One of the most serious and common errors in science (and everyday life) is mistaking the relationship between two phenomena for a causal connection.
Relationships Between PhenomenaStatistics
Correlation
A statistical relationship between two phenomena. It indicates that two phenomena occur together or change in the same/opposite direction.
Causation
A cause-and-effect relationship. One phenomenon demonstrably and directly causes the other.
Golden Rule of Methodology
CORRELATION DOES NOT IMPLY CAUSATION!
If we find that two phenomena correlate, it does not mean that one causes the other. A classic exam trick: Statistics show a strong positive correlation between ice cream sales and the number of drownings. Does eating ice cream cause drowning? Never! Both of these seemingly dependent variables are actually driven by a hidden third (intervening) variable – hot summer weather, which makes people both buy ice cream and go swimming.
Psychodiagnostic Methods (Tests)
To measure individual differences, psychologists use specially constructed psychodiagnostic tests.
Classification of Psychological TestsPsychodiagnostics
We divide tests into three main categories based on what they measure and how they are constructed:
1. Performance Tests
Assess abilities, knowledge, or intelligence (e.g., IQ tests). They have objectively correct and incorrect answers. It is possible to achieve maximum performance.
2. Personality Tests – Objective Questionnaires
Measure personality traits, attitudes, and interests. They do not have 'correct' answers; the respondent usually rates how much they agree with a statement (e.g., Big Five questionnaire, MMPI).
3. Personality Tests – Projective Techniques
Present the respondent with an ambiguous (vague) stimulus. It is assumed that the client 'projects' their own unconscious conflicts, fears, and desires onto the ambiguous material.
For projective techniques, the following two tests are the most famous:
Thematic Apperception Test (TAT)
A projective test using images of ambiguous life situations. The client's task is to make up a story about the image (what happened before, what is happening now, how it will end).
Rorschach Test (ROR)
The most famous projective test using a series of cards with inkblots. The client's task is to state what they see in the symmetrical blots.
Basics of Statistics and Psychometrics
For a psychological test or research to be considered scientific and reliable, it must meet strict psychometric criteria and rely on statistical laws.
Basic Concepts of PsychometricsMeasurement Quality
Gaussian Curve
The normal distribution curve (bell curve). It graphically illustrates that the majority of the population has average values of a given trait (e.g., IQ) and that extreme values (geniuses and the mentally retarded) are rare.
Validity
An indicator of whether a test actually measures what it claims to measure. (For example, if a test is supposed to measure mathematical abilities but is written in such complex language that it actually tests reading comprehension, its validity is low).
Reliability
An indicator of how consistent and independent of chance the test results are. A highly reliable test should show the same (or very similar) results for the same person upon repeated measurement.
Key Takeaways
Experiment: The only research method that, through manipulation of the independent variable, proves cause and effect (causality).
Golden Rule: The mere existence of a statistical correlation between two phenomena does not prove that one causes the other.
Types of Tests: Diagnostic tests are divided into performance tests (IQ tests, have correct answers) and personality tests (questionnaires and projective techniques like the Rorschach test).
Psychometric Criteria: Every good test must have high reliability (measures consistently) and high validity (actually measures the phenomenon it was created for).
Normal Distribution: Most psychological traits in the population (including intelligence) correspond to a Gaussian curve.
An article appeared in a newspaper with the headline: 'Children who have more books at home get better grades in school. Buying books improves academic performance!' What gross methodological error did the author of the headline commit?
Correct Answer: They confused correlation with causation.Explanation: The fact that a large number of books and good grades occur together (correlate) does not mean that the books themselves cause better grades. It is likely that a hidden intervening variable is at play here – for example, a higher socioeconomic status and the education level of parents who both buy books for their children and support them more in their studies.
You have designed a new test to measure the level of anxiety. While verifying the test, you found that if a client takes the test on Monday, the result shows extreme anxiety, but if the same client takes it on Wednesday, the test shows they are completely calm, even though their actual psychological state has not changed. Which basic psychometric parameter is your test missing?
Correct Answer: Reliability.Explanation: Reliability expresses the precision and consistency of measurement over time. If the same tool shows completely different results every time when measuring the same, unchanging trait (under the same conditions), the test is unreliable (has low reliability) and cannot be used in practice.
