Formulating Social Psychological Investigations: Key Methods Explained

When researchers want to understand why people conform, how prejudice spreads, or what drives helping behavior, they don’t just observe and guess – they design systematic investigations. The way a social psychological study is formulated determines everything: what gets measured, how it’s measured, and what conclusions can legitimately be drawn. Getting the method right is not a technicality. It is the foundation of credible knowledge about human social behavior.

Theoretical vs. applied research: two purposes, one discipline
How data is collected in social psychology
Observations
Document studies
Questionnaires and interviews
The four primary research methods
The observational method
The correlational method
The experimental method
The ethnographic method
Analytical methods: making sense of the data

Theoretical vs. applied research: two purposes, one discipline

Every investigation in social psychology starts with a purpose. That purpose usually falls into one of two broad categories: theoretical research (also called basic or pure research) or applied research.

Theoretical research is primarily concerned with building knowledge – developing frameworks, refining theories, and expanding our fundamental understanding of social phenomena. It is not driven by an immediate real-world problem. A researcher studying the psychological mechanisms behind conformity, without any specific intervention in mind, is engaging in theoretical work. The goal is insight into how the social world operates.

Applied research, by contrast, targets a specific, immediate problem. It takes theoretical knowledge and puts it to work in natural, real-world settings. According to the Journal of Applied Social Psychology, applied research should bridge theoretically and practically oriented problems – translating abstract findings into actionable solutions. A study designed to reduce workplace discrimination or improve community mental health programs falls squarely into this category.

Importantly, these two types are not opposites – they exist on a continuum. Theoretical research often generates the insights that applied research later puts into practice, while applied problems frequently send researchers back to update or revise existing theories.

Before any analysis can take place, researchers must gather raw material. Social psychology uses several data-collection methods, each suited to different research goals.

Observations

Observational research involves making systematic, objective records of behavior – without manipulating or interfering with it. It comes in two main forms. In naturalistic observation, researchers observe people in their actual environments – a playground, a street, a classroom – as behavior unfolds organically. In controlled observation, a specific situation is created (often in a laboratory) so that the conditions under which behavior occurs can be more precisely tracked.

The observational method is particularly useful for generating hypotheses and describing social behavior. Its main limitation is that it cannot establish why behaviors are occurring or what causes them – only that they are.

Document studies

Researchers also make use of existing records – photographs, newspapers, social media archives, government statistics, historical texts. This archival approach shares the ecological validity of naturalistic observation because the data comes from real-world behavior rather than a constructed laboratory scenario. Examining how men and women are portrayed in magazine advertisements over decades, for instance, can reveal shifting social norms without ever recruiting a single participant.

Questionnaires and interviews

Surveys and questionnaires allow researchers to collect self-reported data on attitudes, beliefs, and behaviors from large groups efficiently. They are highly versatile, lending themselves to both theoretical investigations (measuring implicit attitudes toward outgroups) and applied ones (assessing the impact of a public health campaign). Interviews, whether structured or open-ended, add depth – allowing participants to elaborate on their experiences in ways that closed survey questions cannot capture.

The four primary research methods

Beyond how data is gathered, researchers must choose a broader methodological design. Social psychologists primarily rely on three analytical approaches: observational, correlational, and experimental – with ethnography representing a distinct qualitative tradition that cuts across these categories.

The observational method

As a formal research design, the observational method describes and documents social behavior as it occurs. It provides a picture of what is happening to a given set of people at a given time and how they are responding to the social situation. It is the most basic of the three main designs and is especially valuable when breaking into a new area of inquiry where little prior data exists.

Its core limitation is that it can only describe – it cannot establish relationships between variables or determine causation. Two variables may change together in the data without one causing the other.

The correlational method

The correlational method systematically measures two or more variables to assess the relationship between them. Correlational research is designed to discover relationships among variables and to allow the prediction of future events from present knowledge. If high school grades correlate with university grades, knowing one score lets us predict the other – even without understanding exactly why the relationship exists.

Correlations have a direction (positive, negative, or unrelated) and a strength, expressed as a coefficient between −1 and +1. A coefficient of 0 means no linear relationship; values closer to −1 or +1 indicate stronger relationships in either direction.

The critical caveat is one of the most important principles in all of research methodology: correlation does not imply causation. A third, unmeasured variable – sometimes called a “mystery variable” – could be driving both observed factors simultaneously. Children who play more violent video games may also be more aggressive, but testosterone levels, parental supervision, or socioeconomic stress could be influencing both variables at once. Only experiments can establish causation.

The experimental method

The experiment is the most powerful tool in the social psychologist’s methodological repertoire precisely because it can answer causal questions. An experiment is a controlled test of a hypothesis in which a researcher manipulates one variable – the independent variable – and measures its effect on another – the dependent variable.

What makes an experiment uniquely powerful is random assignment. Participants are randomly placed into different conditions, ensuring that the groups are statistically equivalent before the manipulation occurs. Any differences observed afterward can therefore be attributed to the independent variable, not to pre-existing differences between participants.

Experiments are typically conducted in laboratory settings where variables can be tightly controlled. This gives them high internal validity – meaning we can be confident that the independent variable is genuinely causing the change in the dependent variable. The tradeoff is external validity: the artificial control of a laboratory may not reflect how people actually behave in everyday life. This is why experimental findings are often followed by field studies to see whether effects replicate in natural settings.

Norman Triplett’s famous 1898 study – in which children reeled fishing line faster when paired with another child than when working alone – is widely described as the first experimental study in social psychology and an early illustration of what we now call social facilitation.

The ethnographic method

Ethnography takes a fundamentally different approach. Rather than measuring variables or testing hypotheses, it seeks to understand a group or culture from the inside. Ethnographic research utilizes broad observational methods – including action research, structured interviews, and participant observation – to gather rich information unattainable by classical experimentation.

The word ethnography comes from the Greek: ethnos (folk or people) and graphia (writing). Ethnographers immerse themselves in a community – sometimes for months or years – observing, participating, and interviewing to build what the anthropologist Clifford Geertz called a “thick description”: a rich, deeply contextualised account of how people think, relate, and make meaning within their social world.

What distinguishes ethnography from informal observation is its formalised, multi-method structure. Researchers gather data from interviews, observations, and documents simultaneously, then compare these sources to ensure reliability. This multimethod approach reduces the risks that come from relying on a single kind of data, allowing researchers to triangulate their findings.

Ethnography is particularly suited to studying behavior in everyday contexts – social dynamics within a school, power structures in a workplace, or the cultural logic of a close-knit community. It captures nuance and process in ways that a survey or a controlled experiment simply cannot. The classic Robbers Cave study by Sherif and colleagues (1961) illustrates how ethnographic tools can work alongside experimental designs: the researchers used quantitative methods to set up their conditions but relied on in-depth qualitative interviews and ethnographic descriptions to make sense of how conflict and cooperation unfolded between the two groups of boys.

Analytical methods: making sense of the data

Collecting data is only half the task. Researchers must also interpret it, and this is where analytical methods come in. These fall into two broad categories: statistical techniques and logical constructs.

Statistical analysis is central to quantitative research. It allows researchers to identify patterns, test hypotheses, and determine whether observed effects are likely to be real or the product of chance. Tools like regression analysis can reveal the strength of a relationship between variables, while correlation coefficients quantify direction and magnitude. Statistical methods also help researchers guard against drawing causal conclusions from data that only supports association.

Logical constructs complement statistical analysis by situating findings within theory. Researchers draw on existing frameworks, prior studies, and conceptual models to interpret what their data means – not just statistically, but conceptually. In the social sciences, basic and applied research are interdependent: the findings from one feed back into the other, ensuring that the field stays both theoretically grounded and practically relevant.

Together, these analytical tools are what turn raw observations into credible social psychological knowledge. No single method or analytical approach is universally superior. The best formulation of an investigation matches the method to the question – using experiments when causation needs to be established, correlational designs when prediction is the goal, observational or ethnographic methods when depth and context matter most.

What do you think? If you were designing a study to understand how social media use affects loneliness in young adults, which method – experimental, correlational, or ethnographic – would give you the most meaningful answers, and what would each approach reveal that the others couldn’t? And do you think the sharp distinction between theoretical and applied research still holds in an era when science is increasingly expected to address pressing social problems in real time?

References

PDF 📄

Table of Contents

Theoretical vs. applied research: two purposes, one discipline 🔗

How data is collected in social psychology 🔗

Observations 🔗

Document studies 🔗

Questionnaires and interviews 🔗

The four primary research methods 🔗

The observational method 🔗

The correlational method 🔗

The experimental method 🔗

The ethnographic method 🔗

Analytical methods: making sense of the data 🔗