What makes scientific research different from intuition or personal experience as a way of understanding behavior?
Choose an answer, then check it.
Psychology · Foundations
Research Methods
On this page 9 sections
In 30 seconds
Psychologists do not rely on hunches. They gather evidence with research methods The systematic, evidence-based approaches psychologists use to study behavior, including surveys, case studies, naturalistic observation, correlational studies, and experiments. Full entry → — systematic approaches such as surveys, case studies, naturalistic observation Watching and recording behavior as it occurs in its natural setting, while the observer stays unobtrusive. Full entry →, correlational studies, and experiments. Each method has strengths and blind spots, and the research question shapes which one fits. Findings come from samples, subsets of the people researchers ultimately want to describe. No single method gives a perfect picture, so good research combines methods, treating each as one lens on behavior rather than a mirror of the whole truth.
Why this matters
Every psychology claim you meet — in news, health advice, product marketing, or policy debates — rests on a method someone chose, and knowing how psychologists gather evidence tells you how much a finding can support. Academically, research methods are the backbone of the discipline: later courses on experiments, correlation, and statistics refine the toolbox this lesson introduces. Practically, these are everyday critical-thinking tools. When you hear that a survey A fixed set of questions administered to a group of people, often used to gather information from large numbers of participants quickly. Full entry → of two hundred people proves something about everyone, you can ask who was sampled and what the method can and cannot show. Looking ahead, the habit of asking which method fits which question, and what each method misses, transfers to any field that makes claims about people — which is most of them.
The college version
What research methods are
Research methods are the systematic approaches psychologists use to gather evidence about behavior and the mind. OpenStax's Psychology 2e textbook puts the working definition plainly: psychologists have many research methods available in their efforts to understand, describe, and explain behavior and the cognitive and biological processes that underlie it. The key word is evidence. The same textbook explains that scientific knowledge is empirical Grounded in observable, tangible evidence that can be checked, rather than in intuition or personal opinion. Full entry → — grounded in objective, tangible evidence that can be observed and checked — which is what separates research from intuition, rumor, or personal experience. Every method in the toolbox is a disciplined way of collecting that evidence, and each one makes different tradeoffs.
The toolbox: five methods
Surveys are fixed lists of questions given to many people, and they can be delivered on paper, online, or verbally. A campus recreation office might email every enrolled student twelve questions about how often they exercise and which facilities they use. The honest limitation is self-report: people may not answer accurately, so a student who skips the gym may still check the box for four workouts a week. Case studies focus intensely on one person or a few individuals. A memory researcher might spend two years with a single woman who recalls her childhood days in unusual detail, mapping how her memory is organized. The honest limitation is reach: someone that unusual may not represent anyone else. Naturalistic observation means watching behavior where it naturally happens, staying unobtrusive because people act differently when they know they are being watched. A team might stand at the end of a grocery aisle and record how shoppers pause, compare, and move on. The honest limitation is control: researchers cannot make the behavior happen on schedule, and observers can unconsciously see what they expect. Correlational studies measure variables as they naturally occur and ask whether they move together. Researchers might track students' nightly sleep and exam scores across a semester and find that more sleep goes with higher scores. The honest limitation is that a relationship is not proof that one thing caused the other — that question belongs to experiments. Experiments actively change one variable and measure the effect under controlled conditions, which is the strongest tool for causal questions, though the controlled setting can be artificial. Experiments get their own full lesson; here they are one tool among five.
Choosing a method: the question shapes the method
There is no single best method, only the method that fits the question. The Noba Project's research-methods chapter states the principle directly: which method you choose depends on the type of questions you are asking, as well as what resources are available. An original example makes it concrete. A researcher who wants to know how often first-year students visit the counseling center and how they feel about it chooses a survey, because it reaches many students cheaply and lets them answer privately. A researcher who wants to see what students actually do during orientation week chooses naturalistic observation, because she needs real behavior, not reported behavior. A researcher who wants to test whether a specific study-skills workshop raises grades needs an experiment. Time and money matter too: a quick survey can map a question before anyone invests in a costly long-term study.
Samples: who is studied
Psychologists almost never study everyone they care about. They study a sample A subset of individuals selected from a larger population to be studied; researchers generalize from the sample to the population. Full entry → — a subset of individuals selected from a population The overall group of individuals a researcher is interested in describing; the sample is drawn from this group. Full entry →, which is the overall group the researcher is interested in — and then seek to generalize the findings to that population. Representativeness is the catch: a sample should resemble the population it came from. If a survey about dining-hall hours is mostly answered by students who eat late dinners, the late-evening diners are overrepresented, and the results will not describe the whole campus. The same logic explains why a case study An intensive, in-depth examination of one person or a small number of individuals, often chosen because they have a rare or revealing characteristic. Full entry → of one extraordinary person cannot stand for everyone: the sample is real and detailed, but it does not represent the group. Big samples help, but a big, lopsided sample still misleads.
The honest note: lenses, not mirrors
Every method has blind spots, and pretending otherwise is how bad studies happen. Surveys buy breadth at the cost of honest self-reports; case studies buy depth at the cost of narrowness; naturalistic observation buys realism at the cost of control; correlational studies buy real-world patterns at the cost of causal proof; experiments buy control at the cost of artificial settings. The Noba chapter draws the honest conclusion: all methods have limitations, which is why the best research uses a variety of methods. Think of methods as lenses, not mirrors. A lens magnifies some features and distorts others; no single lens shows the whole object. Psychology builds trustworthy knowledge by pointing several lenses at the same question and checking whether they agree. That is also why one study is never the final word — and why asking which method produced a claim is a good reflex for anyone reading psychology.
Eli explains
The same idea, in plain words
Explain it like I’m 10
Psychologists want to know how people actually think and behave, but you cannot just trust what people say or what seems obvious. So they use research methods: organized ways of collecting evidence, and each method answers a certain kind of question. Want to hear from thousands of people quickly? Survey them. Want to understand one remarkable person deeply? Do a case study. Want to see real behavior? Watch people where it actually happens. Want to know whether two things move together? Use a correlational study. Want to test what causes what? Run an experiment. No method is perfect: surveys depend on honest answers, case studies focus on unusual people, watching gives little control, and correlations do not prove causes. That is why researchers mix methods and why one study is never the whole story.
Picture it like this
Think of research methods as different tools for fixing up a house. A hammer, a saw, and a level are all useful, but you would not saw a nail or level a board with a hammer. The job decides the tool. And no single tool builds a house alone — a good builder switches tools and checks the work from several angles. Psychologists do the same: the question picks the method, and solid conclusions come from combining them.
Where the picture stops working
The toolbox analogy makes methods sound interchangeable, but research choices are heavier than tool choices. A hammer does not change the wood it hits, while a method shapes what a researcher can see and what they miss. Tools are also neutral, whereas methods carry ethical rules about treating participants well, and a poorly run study can mislead people even when the method itself is a sensible one.
Worked example
Dr. Reyes wants to know whether students who use the campus tutoring center feel more confident in their courses. She starts with a survey: all 1,200 first-year students are emailed eight questions about tutoring use and confidence, and 300 reply. The 300 are her sample; the 1,200 are the population she wants to describe. Checking the replies, she notices they skew toward students who visited the center at least once — a sample that does not represent the whole class. So she adds a second method: during exam week she sits in the tutoring center's open study area and quietly records how students arrive, work, and leave, without interrupting. The survey gives her breadth, and the observation gives her real behavior. She does not claim the center causes confidence — that would take an experiment — but she now has two lenses on the same question, and each lens's blind spot is covered by the other.
Key takeaway
Research methods are the evidence-gathering tools of psychology: the question chooses the method, every method has a blind spot, and trustworthy conclusions come from representative samples and from methods that check one another.
Quick check
3 questions here, of 5 in this lesson’s practice set. Answers stay hidden until you check.
A psychologist spends two years following one stroke survivor, documenting her memory, mood, and daily routines in detail. Which method is she using?
Choose an answer, then check it.
A team wants to record how often commuters glance at their phones in a train station without influencing the behavior they are studying. Which method should they use, and why must the observers stay unobtrusive?
Choose an answer, then check it.
Study tools & related lessonsYou’ll learn to · Common mistakes · Easily confused · Key vocabulary · Related
You’ll learn to
- Define research methods as the systematic, evidence-based approaches psychologists use to study behavior, and explain what makes them empirical.
- Identify surveys, case studies, naturalistic observation, correlational studies, and experiments, and state one strength and one limitation of each.
- Apply the principle that the research question shapes the choice of method to a new scenario.
- Explain what a sample is and why a sample should represent the larger population a researcher wants to describe.
- Analyze why no single method gives a complete picture and why combining methods strengthens psychological evidence.
Common mistakes
Treating one study as the final answer.
Every method has blind spots, so a single study is rarely conclusive. Findings earn trust when different methods point the same way.
Assuming whoever answered a survey represents everyone.
Respondents are a sample, and claims about a population are only as strong as the sample's representativeness. A lopsided sample, even a large one, misleads.
Reading a correlation as proof of cause.
Correlational studies show that variables are related; they cannot show that one caused the other. Causal claims require experiments, which have their own lesson.
Expecting one method to answer every kind of question.
The question shapes the method: surveys for breadth, case studies for depth, observation for real behavior, experiments for cause. Matching method to question is the skill.
Easily confused
Survey vs. Case study
A survey trades depth for breadth, gathering thin data from many people; a case study trades breadth for depth, gathering rich data from one person or a few.
Naturalistic observation vs. Experiment
Observation keeps behavior natural but gives up control over when and how it occurs; experiments gain control by changing a variable deliberately, sometimes at the cost of an artificial setting.
Sample vs. Population
The population is everyone a researcher wants to describe; the sample is the subset actually studied, and findings generalize only as far as the sample represents the population.
Key vocabulary
- research methods
- The systematic, evidence-based approaches psychologists use to study behavior, including surveys, case studies, naturalistic observation, correlational studies, and experiments.
- empirical
- Grounded in observable, tangible evidence that can be checked, rather than in intuition or personal opinion.
- survey
- A fixed set of questions administered to a group of people, often used to gather information from large numbers of participants quickly.
- case study
- An intensive, in-depth examination of one person or a small number of individuals, often chosen because they have a rare or revealing characteristic.
- naturalistic observation
- Watching and recording behavior as it occurs in its natural setting, while the observer stays unobtrusive.
- correlational study
- A study that measures variables as they naturally occur and examines whether the variables are related to one another.
- sample
- A subset of individuals selected from a larger population to be studied; researchers generalize from the sample to the population.
- population
- The overall group of individuals a researcher is interested in describing; the sample is drawn from this group.
Sources & references
- Psychology 2e, Section 2.1: Why Is Research Important? — OpenStax (Rice University)
- 2.2 Approaches to Research — Psychology 2e — OpenStax / Rice University
- Research Designs — Noba Project (Christie Napa Scollon, Singapore Management University)
EliExplains lessons are original prose written from the open, credible references above. See Copyright & Licensing.
Researched 2026-08-22
Educational content only. It is not medical, legal or professional advice. Found an error? Tell us.