Methods of Enquiry in Psychology
Introduction and Goals of Psychological Enquiry
Psychology, as a scientific endeavour, relies on formal and systematic observations to investigate human behaviour and mental processes. Like all scientific disciplines, it seeks to describe, predict, explain, and control the phenomena it studies. The methodology employed is what distinguishes psychology as a science. Because human behaviour is multi-faceted, psychologists utilize a diverse range of research methods, including observation, experimental research, correlational studies, surveys, psychological testing, and case studies. Gautam Buddha once noted that an idea put into action is more important than one that exists only as a concept, a sentiment that reflects the application-oriented nature of psychological enquiry.
The first goal of psychological enquiry is description. This involves attempting to describe a behaviour or phenomenon as accurately as possible to distinguish it from others. For instance, when studying study habits, a researcher must define and record specific behaviours such as attending classes regularly, submitting assignments on time, planning study schedules, and daily revision. Description requires painstaking recording to provide a proper understanding of the subject.
The second goal is prediction. Accurately describing behaviour allows researchers to identify relationships between that behaviour and other events or phenomena. This enables the forecasting of behaviour under specific conditions within a certain margin of error. An example is establishing a positive relationship between study time and academic achievement; if a child is known to devote more time to study, one can predict they will likely receive good marks. Prediction accuracy typically increases as the number of observed individuals grows.
The third goal is explanation, which aims to identify the causal factors or determinants of behaviour. Psychologists seek to understand the antecedent conditions—the conditions that lead to a particular behaviour—and those under which a behavior does not occur. This goal establishes cause-and-effect relationships between variables. For example, researchers may investigate why some children are more attentive in class than others by looking for specific determinants or conditions.
The fourth goal is control. If a researcher can explain為何 behavior occurs, they can potentially control it by altering its antecedent conditions. Control involves making a behaviour happen, reducing its occurrence, or enhancing its intensity. Psychological therapy is a prime example of bringing about changes in behavior through controlled treatment. Finally, the goal of application is to use research-generated knowledge to bring positive changes to people's lives and solve problems in various settings, such as using yoga or meditation to reduce stress and increase efficiency. Application also involves developing new theories to guide further research.
Steps in Conducting Scientific Research
Scientific research is defined by how it investigates rather than what it investigates. It must be objective, systematic, and testable. Objectivity means that independent researchers studying the same event using the same tools should reach the same conclusion. Systematic research follows a specific four-step procedure: conceptualisation of a problem, collection of data, drawing conclusions, and revising research conclusions/theory.
The first step, conceptualising a problem, begins with selecting a theme or topic. This is often narrowed down based on past research, personal observations, or experiences. In psychology, problems range from understanding our own behavior and feelings, such as joy or grief, to group behavior and organizational motivation. After identifying a problem, the researcher develops a hypothesis, which is a tentative answer. For example, one might hypothesise that "greater is the amount of time spent by children in viewing violence on television, higher is the degree of aggression displayed by them."
The second step is collecting data, which requires a blueprint or research design. This involves identifying participants (e.g., children, managers, or clinical patients), selecting methods (like observation or case study), choosing tools (such as interview schedules or questionnaires), and determining the procedure (individual or group administration). The researcher then proceeds with actual data collection.
The third step is drawing conclusions. This involves analyzing the collected data using statistical procedures and graphical representations, such as pie charts, bar diagrams, or cumulative frequencies. The goal is to verify the hypothesis and understand the data's meaning. Finally, the fourth step is revising research conclusions. The researcher evaluates whether the conclusions support the original hypothesis. If the theory is confirmed, the study is complete; if not, the researcher revises the hypothesis/theory and tests it again with new data. Research is thus a continuous process.
Alternative Paradigms of Research
A long-standing view in psychology suggests that human behaviour should be studied like physics or chemistry, assuming it is predictable, measured, and controlled through overt observations. However, a different method known as the interpretive paradigm has emerged. This view stresses understanding over explanation and prediction, arguing that human behavior is complex and variable, requiring different investigation methods than the physical world.
The interpretive tradition emphasizes how human beings give meaning to events and actions in specific contexts. For example, when someone experiences suffering due to a tsunami or a prolonged illness, objective measurement may be neither possible nor desirable. Reality is interpreted subjectively based on past experiences. The goal of this paradigm is to explore human experience without disturbing its natural flow, much like an explorer mapping uncharted wilderness with detailed descriptions. Psychological enquiry also aims at understanding the self through reflection on one's own experiences and insights.
Nature of Psychological Data
Data (the plural form of datum) relates to covert or overt behaviour, subjective experiences, and mental processes. Unlike other sciences, psychological data are not independent entities; they are located in physical and social contexts and tied to the methods and theories used for collection. Factors such as the time of the behavior, the persons involved (e.g., behaving differently in front of teachers versus friends), and the characteristics of the respondents (age, gender, rural/urban settings) all influence the quality of data. Data do not speak for themselves; reality is inferred by researchers who attach meaning based on context.
Psychologists collect various types of data: (i) Demographic Information (name, age, gender, education, income, etc.); (ii) Physical Information (ecological conditions, housing, transportation, etc.); (iii) Physiological Data (height, weight, heart rate, EEG, blood pressure, GSR, etc.); and (iv) Psychological Information (intelligence, personality, values, emotions, psychological disorders, etc.). This information can be crude categories (high/low), ordinal ranks (, , ), or scores (, , ). It also includes verbal reports, personal diaries, and field notes, which are analyzed using qualitative methods.
Observational Method
Observation is a powerful tool for describing behavior and differs from daily observation because it is scientific and systematic. The process involves three components: (a) Selection, where the psychologist chooses specific behaviors to observe rather than everything they encounter; (b) Recording, using tallies, field notes, symbols, photographs, or video; and (c) Analysis of Data, where researchers derive meaning from the recordings. Making good observations is a specialized skill requiring knowledge of whom, when, and where to observe.
Observation types include Naturalistic versus Controlled Observation. Naturalistic observation occurs in real-life settings like schools or hospitals without researcher interference. Controlled laboratory observation involves manipulating factors, such as the introduction of smoke in a lab experiment. Another distinction is Non-Participant versus Participant Observation. In non-participant observation, the researcher observes from a distance (e.g., via video or sitting in a corner), though their presence may change the subjects' behaviour. In participant observation, the researcher becomes part of the group, establishing rapport to be accepted as a member.
While observation allows for studying behavior as it occurs, it is labor-intensive, time-consuming, and subject to observer bias. Factors like the researcher's values and beliefs can influence interpretation—the principle that "we see things as we are and not as things are." To mitigate this, observers should record events as they happen without interpreting them immediately.
Experimental Method
Experiments are regulated procedures conducted to establish cause-effect relationships between variables in a controlled setting. A variable is any stimulus or attribute that varies and can be measured, such as the size of a room, height of individuals, or levels of intelligence. In an experiment, the Independent Variable (IV) is manipulated or altered by the researcher to see its effect, while the Dependent Variable (DV) represents the phenomenon the researcher wants to explain and is affected by the IV.
In the landmark experiment by Bibb Latane and John Darley () at Columbia University, students were placed in a room that filled with smoke. The IV was the presence or absence of others. The results showed that of students waiting alone reported the smoke. When in groups of three naive students, only reported it. If seated with two confederates (who ignored the smoke), only reported it. Here, the DV was the frequency of reporting the smoke.
Experiments involve Experimental Groups, where members receive the IV manipulation, and Control Groups, which are comparison groups where the manipulated variable is absent. To ensure accuracy, researchers must control Extraneous Variables. These include: (i) Organismic variables (anxiety, intelligence); (ii) Situational variables (noise, temperature, humidity); and (iii) Sequential variables (fatigue or practice effects). Control techniques include elimination of variables, holding them constant, matching participants, counter-balancing the order of tasks, and random assignment. While powerful for proving cause-effect, experiments can lack external validity due to their highly controlled nature and often cannot address ethical issues that prevent the manipulation of harmful conditions.
Field Experiments and Quasi Experiments
When a researcher needs high generalisability, they may conduct a field experiment in a natural setting, such as a school. Controls in the field are less stringent than in the lab, and these studies are often more expensive. In cases where variables cannot be manipulated for ethical or practical reasons (e.g., the effect of an earthquake on orphans), researchers use Quasi Experiments. In a quasi-experiment, the independent variable is selected rather than manipulated, using naturally occurring groups to form experimental and control sets.
Correlational Research
Correlational research determines the degree of association between two variables for prediction purposes without manipulation. The strength and direction of this relationship are expressed by a correlation coefficient, ranging from through to . A positive correlation occurs when both variables increase or decrease together (e.g., study time and achievement). A negative correlation occurs when one variable increases as the other decreases (e.g., study time and hours spent on leisure). A zero correlation indicates that no significant relationship exists between the variables.
Survey Research
Survey research studies opinions, attitudes, and social facts. It has evolved from simple baseline fact-finding (e.g., literacy rates) into a sophisticated technique for inferring causal relationships. A notable example is the Outlook Saptahik magazine survey () on Indian happiness, which surveyed people in eight cities (Mumbai, Delhi, Kolkata, Bangalore, Hyderabad, Ahmedabad, Jaipur, and Ranchi). Results showed were extremely happy, while believed money couldn't buy happiness. Survey techniques include personal interviews, questionnaires, and telephone surveys.
(1) Personal Interviews involve an interviewer asking an interviewee questions face-to-face or over the phone. Structured interviews use a predetermined list of questions called an interview schedule with close-ended responses. Unstructured interviews are flexible, allowing open-ended questions. Interview formats include individual-to-individual, individual-to-group (Focus Group Discussion), group-to-individual, and group-to-group.
(2) Questionnaires are low-cost self-report methods where respondents write answers to predetermined items. They use open-ended and closed-ended questions, often employing rating scales like , , or -point scales (e.g., Strongly Agree to Strongly Disagree). (3) Telephone surveys and mobile SMS surveys reduce time but often face issues with uncooperativeness or biased sample groups. While surveys can collect data from thousands quickly, they are limited by memory lapses or social desirability bias (participants saying what they think the researcher wants to hear).
Psychological Testing
A psychological test is a standardised and objective instrument used to assess an individual's standing relative to others on mental or behavioural characteristics. Objectivity implies that different researchers administering the same test will reach the same results. Standardisation involves a systematic construction process establishing reliability, validity, and norms.
Reliability refers to consistency in scores. Test-retest reliability measures temporal stability by re-administering the test after a period (e.g., days). Split-half reliability measures internal consistency by correlating scores from two halves of the test (e.g., odd versus even items). Validity answers the question of whether the test measures what it claims to measure. Norms represent the average performance of a group and allow for the interpretation of individual scores.
Tests are classified by: (a) Language: verbal, non-verbal, and performance-based; (b) Administration: individual (one-on-one, time-consuming) or group (efficient but prone to faking); and (c) Difficulty: Speed tests have a time limit and items of equal difficulty, while Power tests assess underlying ability through items of increasing difficulty without a time limit.
Case Study
The case study method involves an in-depth study of a particular case, such as an individual with unique characteristics (e.g., a psychological disorder), a small group (creative writers), institutions, or specific events (tsunami victims). It uses multiple methods, including interviews and observations, to provide rich, narrative descriptions. Significant developments in psychology, such as Freud's psychoanalytic theory and Piaget's theory of cognitive development, were built on case studies. However, generalising from a single case requires caution and should be supplemented by multiple data sources.
Analysis of Data
There are two primary methodological approaches to data analysis: quantitative and qualitative. Quantitative methods involve close-ended responses converted into numbers (e.g., for correct, for wrong). Statistical methods include measures of central tendency (mean, median, mode), variability (range, standard deviation), and correlation coefficients. Qualitative methods are used for complex human experiences that cannot be quantified, such as narrative analysis for life stories or content analysis for finding thematic categories in field notes, photographs, or informal talks. These methods are complementary rather than contradictory.
Limitations of Psychological Enquiry
Psychological enquiry faces several limitations: (1) Lack of True Zero Point: Unlike physical sciences, psychology has no absolute zero (e.g., no one has zero intelligence). Measurements are relative rather than absolute. (2) Relative Nature of Tools: Tests are context-sensitive. A test developed for urban environments may not suit tribal populations, and Western tests often require adaptation for Indian contexts. (3) Subjective Interpretation: Qualitative data depends heavily on researcher interpretation. It is recommended that multiple investigators work together and involve respondents in the meaning-making process.
Ethical Issues
Researchers must adhere to moral principles when studying humans: (1) Voluntary Participation ensures participants can choose to join or withdraw without penalty. (2) Informed Consent requires informing participants about the study's nature before it begins, including the use of any unpleasant stimuli or deception. (3) Debriefing involves providing necessary information to participants once the study ends to restore them to their initial mental state, especially if they were deceived. (4) Sharing Results is a moral obligation that fulfills participant expectations and can provide the researcher with new insights. (5) Confidentiality of Data Source protects the right to privacy by ensuring information is only used for research purposes; names are often replaced with code numbers that are destroyed after the study.