Difficulty: Foundational | Prerequisites: None
This is the methodological backbone of the entire course. Every later topic in PSYCH 1100, from neuroscience to consciousness, rests on your ability to evaluate how evidence was gathered and whether conclusions are warranted. If you can distinguish correlation from causation and know why random assignment matters, the rest of the midterm topics become much easier to reason about.
Psychological research follows the scientific method: form a theory, derive testable hypotheses, collect data using one of several methods (descriptive, correlational, or experimental), analyse the results with statistics, and share findings through peer-reviewed publication. The method you choose determines what conclusions you can draw, and only true experiments with random assignment let you claim causation. Ethics oversight (IRB for humans, IACUC for animals) ensures research protects participants throughout.
Confirmation bias
The tendency to seek out, notice, and remember information that confirms what you already believe, while ignoring contradictory evidence. Think of it as your brain's built-in "yes-man" that filters reality to match your expectations.
Theory
A well-substantiated explanation of some aspect of the natural world, built from repeatedly tested and confirmed hypotheses. In simple terms, a theory is a big-picture framework that organises what we know and generates new predictions.
Hypothesis
A specific, testable prediction derived from a theory. Think of it as one concrete question you can actually run a study to answer.
Peer review
The process by which other experts in the field evaluate a researcher's work before it is published. In simple terms, it is a quality check by people who know the subject well enough to spot errors.
Replication
Repeating a study to see whether the same results emerge. If findings replicate, confidence in them grows; if they do not, the original result may have been a fluke.
Case study
An in-depth investigation of a single individual, group, or event. Think of it as a detailed portrait rather than a wide-angle photograph.
Naturalistic observation
Observing behaviour in its natural setting without interference from the researcher. In simple terms, you watch what people (or animals) do when they do not know they are being studied.
Self-report
A data-collection method in which participants provide information about themselves, typically through questionnaires or interviews. The obvious limitation is that people may not report accurately.
Sample
The subset of individuals selected from a larger population to participate in a study. Think of it as the slice of the pie you actually measure.
Population
The entire group of individuals that a researcher wants to draw conclusions about. The sample is supposed to represent this group.
Generalization
The extent to which findings from a sample can be applied to the broader population. If your sample is not representative, your results may not generalise.
Correlation
A statistical relationship between two variables, indicating that changes in one tend to accompany changes in the other. Crucially, correlation does not establish that one variable causes the other.
Positive correlation
Both variables move in the same direction: as one increases, the other tends to increase as well. Example: hours of study and exam scores often show a positive correlation.
Negative correlation
The variables move in opposite directions: as one increases, the other tends to decrease. Example: hours of sleep deprivation and cognitive performance.
Zero correlation
No systematic relationship between the two variables. Knowing one tells you nothing about the other.
Correlation coefficient
A number between -1 and +1 that quantifies the strength and direction of a correlation. Values near +1 or -1 indicate strong relationships; values near 0 indicate weak or no relationship.
Scatterplot
A graph that plots each participant's scores on two variables as a single dot, revealing the pattern (or lack thereof) in the relationship.
Directionality problem
When two variables correlate, you cannot tell from the correlation alone which one influences the other. Does A cause B, or does B cause A?
Third variable problem
An unmeasured variable may be responsible for the apparent relationship between the two measured variables. In simple terms, something you did not account for could be driving both.
Independent variable (IV)
The variable the experimenter deliberately manipulates to observe its effect. Think of it as the thing you change on purpose.
Dependent variable (DV)
The variable the experimenter measures to see whether the manipulation had an effect. Think of it as the outcome you are watching for.
Experimental group
The group of participants exposed to the independent variable (the treatment or condition of interest).
Control group
The group of participants who are not exposed to the independent variable, serving as a baseline for comparison.
Random assignment
Assigning participants to experimental or control groups using chance (e.g. a coin flip), so that pre-existing differences between people are spread evenly across conditions.
Placebo
An inert treatment (e.g. a sugar pill) given to the control group so that participants believe they are receiving the real treatment.
Placebo effect
A measurable change in behaviour or health that occurs simply because a participant expects the treatment to work, even though the treatment itself is inert.
Single-blind procedure
An experimental design in which the participants do not know whether they are in the experimental or control group, but the researcher does.
Double-blind procedure
Neither the participants nor the researchers interacting with them know who is in which group. This reduces both participant expectancy effects and experimenter bias.
Confounding variable
Any variable other than the IV that differs between the experimental and control groups and could explain the results. Confounds make it impossible to draw causal conclusions.
Operationalization
Defining a concept in terms of the specific procedures used to measure or manipulate it. For example, "stress" might be operationalized as cortisol level in saliva.
Meta-analysis
A statistical technique that combines results from many studies on the same topic to identify an overall pattern. Think of it as a study of studies.
Publication bias
The tendency for journals to publish studies with significant, positive results more often than studies with null or negative results.
File drawer problem
Studies that fail to find significant results often go unpublished (they sit in the researcher's file drawer), skewing the published literature toward positive findings.
Cross-sectional study
A study that compares different groups of people (e.g. different ages) at a single point in time.
Cohort effects
Differences between groups in a cross-sectional study that arise because the groups grew up in different historical periods, not because of age itself.
Longitudinal study
A study that follows the same group of individuals over an extended period, measuring them at multiple time points.
Mixed longitudinal design
A design that combines cross-sectional and longitudinal approaches, tracking multiple cohorts over time to separate age effects from cohort effects.
Reliability
The consistency of a measure: does it produce the same results under the same conditions? Think of it as whether your bathroom scale gives you the same reading twice in a row.
Validity
The extent to which a measure actually captures what it claims to measure. A scale that always reads 5 kg too high is reliable but not valid.
Frequency distribution
A summary of how often each value (or range of values) occurs in a data set. Often displayed as a histogram or bar chart.
Mean
The arithmetic average: add all scores, divide by the number of scores.
Median
The middle value when scores are arranged in order. Less affected by extreme scores (outliers) than the mean.
Mode
The most frequently occurring score in a data set.
Standard deviation
A measure of how spread out scores are around the mean. A small standard deviation means scores cluster tightly; a large one means they are spread widely.
Normal distribution (normal curve)
A symmetrical, bell-shaped distribution where most scores cluster near the mean and fewer scores appear at the extremes.
Null hypothesis
The default assumption that there is no effect or no difference between groups. Statistical tests evaluate whether the data provide enough evidence to reject this assumption.
Statistical significance
A result is statistically significant when it is unlikely to have occurred by chance alone (conventionally, when p < .05). This does not tell you how large or practically important the effect is.
Institutional Review Board (IRB)
A committee that reviews and approves research involving human participants to ensure ethical standards are met.
Institutional Animal Care and Use Committee (IACUC)
A committee that reviews and approves research involving animal subjects to ensure humane treatment.
Informed consent
The ethical requirement that participants be told about the nature of a study, its risks and benefits, and their right to withdraw at any time, before agreeing to take part.
Debriefing
A post-study explanation given to participants, revealing the true purpose of the research and any deception that was used.
Everyday reasoning relies on personal experience, anecdote, and intuition. Science adds systematic observation, measurement, and controls to reduce error.
The core difference: science deliberately tries to prove itself wrong (falsification), whereas casual observation tends to confirm what we already believe.
People naturally seek evidence that supports their existing views and dismiss evidence that contradicts them.
In research, confirmation bias can lead investigators to design studies, select data, or interpret results in ways that favour their hypothesis. Peer review and replication exist partly to catch this.
Evaluating claims by examining evidence, questioning assumptions, and considering alternative explanations.
Key habits: ask what evidence supports the claim, whether the source is credible, and whether other explanations could account for the finding.
A theory is an overarching explanation supported by a large body of evidence (e.g. evolutionary theory, cognitive dissonance theory). Theories are never "proven" in an absolute sense; they are supported, refined, or replaced.
A hypothesis is a specific, testable prediction derived from a theory. If the hypothesis fails, the theory may need revision.
Publication is how scientists share findings with the broader community. Without it, knowledge stays private.
Peer review subjects a manuscript to scrutiny by other experts before publication. It catches methodological flaws, questionable interpretations, and gaps in reasoning.
Replication means repeating a study (ideally by an independent lab) to see if the results hold. Replication failures have prompted a "replication crisis" conversation across psychology and other sciences.
Descriptive methods observe and document behaviour without manipulating anything. They are good for generating ideas and describing patterns, but they cannot establish cause and effect.
Deep dive into one individual, group, or event.
Pros: rich detail, useful for rare conditions (e.g. Phineas Gage), can generate hypotheses.
Cons: findings may not generalise to others, researcher interpretation can be subjective, no control over variables.
Watching behaviour in its natural environment without intervention.
Pros: high ecological validity (behaviour is genuine), useful for studying things that cannot be ethically manipulated.
Cons: observer may influence behaviour if detected (reactivity), no control over extraneous variables, hard to replicate exactly.
Collecting self-report data from many people using questionnaires or interviews.
Pros: efficient way to gather large amounts of data quickly, can cover topics that are hard to observe directly.
Cons: self-report can be inaccurate (social desirability bias, poor memory), wording of questions can bias responses.
The population is everyone you want your conclusions to apply to.
The sample is the subset you actually study.
Generalization is the leap from sample findings to population-level conclusions. It depends on having a representative sample; if your sample is biased (e.g. all first-year university students), your findings may not hold for the wider population.
Correlational studies measure two or more variables and look for systematic relationships between them. They tell you whether variables are associated and how strongly, but they cannot tell you whether one causes the other.
Positive correlation: both variables increase (or decrease) together. As X goes up, Y goes up.
Negative correlation: as one variable increases, the other decreases.
Zero correlation: no consistent pattern between the variables.
Scatterplot: each data point represents one participant's scores on both variables. The overall pattern reveals the direction and strength of the relationship.
Correlation coefficient (r): ranges from -1.00 to +1.00. The sign tells you direction; the absolute value tells you strength. An r of +0.85 is a strong positive relationship; an r of -0.10 is a weak negative one.
Two specific problems prevent causal claims:
Directionality problem: if A and B correlate, A might cause B, or B might cause A. Example: depression correlates with social isolation, but does depression cause people to withdraw, or does isolation cause depression?
Third variable problem: an unmeasured variable C might be causing both A and B, creating a spurious correlation. Example: ice cream sales and drowning rates both rise in summer, not because ice cream causes drowning, but because hot weather (the third variable) drives both.
The experiment is the only method that can establish cause and effect, because the researcher controls the independent variable and holds everything else constant.
The independent variable (IV) is what the experimenter manipulates (the presumed cause).
The dependent variable (DV) is what the experimenter measures (the presumed effect).
Example: testing whether caffeine improves reaction time. IV = caffeine dose. DV = reaction time in milliseconds.
Experimental group: receives the treatment (e.g. the caffeine).
Control group: does not receive the treatment, or receives a placebo. Serves as the baseline for comparison.
Any difference in the DV between groups can be attributed to the IV, provided the experiment is well controlled.
A placebo is a sham treatment (sugar pill, saline injection) that looks identical to the real treatment.
The placebo effect is improvement that occurs because participants believe they are being treated. This is why control groups need placebos: you are comparing the real treatment against the expectation of treatment.
Single-blind: participants do not know which group they are in, but the experimenter does. Reduces participant expectancy effects.
Double-blind: neither participants nor the experimenters who interact with them know group assignments. Eliminates both participant expectancy and experimenter bias.
Participants are placed into groups by chance (coin flip, random number generator).
Purpose: to distribute pre-existing individual differences (age, personality, health) evenly across groups so they do not confound the results.
Random assignment is what makes the experiment causal. Without it, group differences might reflect pre-existing differences, not the IV.
Any variable other than the IV that systematically differs between groups and could explain the outcome.
Good experimental design aims to eliminate confounds through random assignment, standardised procedures, and controlled conditions.
Translating an abstract concept into a concrete, measurable procedure.
Example: "aggression" might be operationalized as the number of times a child hits a Bobo doll in 10 minutes. Different operationalizations of the same concept can yield different results, which is why replication with varied measures matters.
A statistical technique that pools the results of many studies addressing the same question.
Gives a more reliable estimate of an effect than any single study, because it increases statistical power and averages out quirks of individual samples.
Journals preferentially publish studies with significant (p < .05) results.
Studies with null results often go unpublished, sitting in researchers' file drawers.
Consequence: the published literature may overestimate the size of real effects, because the failures to replicate never see print.
Cross-sectional studies compare different groups (e.g. 20-year-olds vs. 60-year-olds) at the same moment. Quick and inexpensive, but vulnerable to cohort effects: differences may reflect the era people grew up in, not age itself.
Longitudinal studies follow the same individuals over months, years, or decades. They track genuine change over time but are expensive, slow, and suffer from participant drop-out.
Mixed longitudinal designs combine both approaches: follow several different-age cohorts over time. This lets researchers separate true age effects from cohort effects more efficiently than either method alone.
Reliability = consistency. A reliable measure gives the same result each time under the same conditions.
Validity = accuracy. A valid measure captures what it claims to measure.
A measure can be reliable without being valid (consistently wrong), but it cannot be valid without being reliable.
Descriptive statistics summarise and organise data so you can see patterns at a glance.
Frequency distribution: a count of how often each score occurs. Displayed as a histogram or table.
Measures of central tendency: single numbers that represent the centre of the data.
Mean: arithmetic average. Sensitive to outliers.
Median: the middle score when data are ranked. More robust to outliers.
Mode: the most common score. Useful for categorical data.
Standard deviation: measures how spread out scores are around the mean. A small SD means scores are tightly clustered; a large SD means they are spread out.
Normal distribution (bell curve): a symmetrical distribution where most scores fall near the mean, with progressively fewer scores at the extremes. Many psychological variables approximate this shape.
Inferential statistics help researchers decide whether their results are meaningful or simply due to chance.
Null hypothesis: the starting assumption that there is no difference between groups or no relationship between variables. The researcher tries to reject it.
Statistical significance: a result is statistically significant (conventionally p < .05) when there is less than a 5% probability that the result occurred by chance. Important caveat: statistical significance says nothing about the size or practical importance of the effect.
Institutional Review Board (IRB): reviews and approves all research involving human participants before it begins.
Institutional Animal Care and Use Committee (IACUC): reviews and approves all research involving animal subjects.
Voluntary, uncoerced participation: people must choose to take part freely. Incentives (e.g. course credit, small payment) are acceptable as long as they are reasonable and do not pressure people into participating.
Informed consent: before the study, participants must be told what will happen, any risks, and their right to withdraw at any time without penalty.
No irreversible harm: research must not cause permanent physical or psychological damage.
Mild deception is permitted if the study cannot be done any other way and participants receive a full debriefing afterward explaining the true purpose and any deception used.
Privacy and confidentiality: participants' data must be kept confidential and their identity protected.
The research must have a clear scientific purpose that justifies the use of animals.
Animals must receive excellent treatment and housing.
Procedures must minimise pain and suffering as far as possible.
Students often think "correlation does not equal causation" means correlated variables are unrelated. They are related, you just cannot tell which one causes the other (or whether a third variable causes both).
Students sometimes confuse random assignment with random sampling. Random sampling is about how you select participants from the population (for generalizability). Random assignment is about how you place already-selected participants into groups (for causation).
A statistically significant result is not necessarily a large or important result. A tiny effect can be statistically significant if the sample is big enough.
"Theory" in everyday language means a guess or hunch; in science it means a well-supported, broadly explanatory framework. Calling something "just a theory" misunderstands the term.
⚠️ Expect questions asking you to identify the IV and DV in a described experiment.
⚠️ Be prepared to distinguish among descriptive, correlational, and experimental methods and state what conclusions each allows.
⚠️ The difference between correlation and causation is almost certain to appear. Know the directionality problem and the third variable problem by name.
⚠️ Random assignment vs. random sampling: know the purpose of each.
⚠️ Understand when a study can and cannot make causal claims, and why.
⚠️ Know the ethical requirements for human and animal research, including informed consent, debriefing, and the roles of IRB and IACUC.
True or false: A longitudinal study compares different age groups at a single point in time.
Fill in the blank: The ______ variable is manipulated by the experimenter; the ______ variable is measured.
True or false: A correlation coefficient of -0.90 indicates a weak relationship.
Fill in the blank: ______ assignment distributes individual differences evenly across experimental conditions.
True or false: Peer review occurs after a study is published.
Answers: 1. False (that is cross-sectional). 2. Independent; dependent. 3. False (the absolute value is high, so it is a strong negative relationship). 4. Random. 5. False (it occurs before publication).
Q: A researcher finds that students who sleep more tend to have higher GPAs. Can the researcher conclude that sleep causes better grades? Why or why not?
A: No. This is a correlational finding. The directionality problem means better grades could lead to less stress and more sleep, and the third variable problem means some other factor (e.g. conscientiousness) could be driving both more sleep and higher grades.
Q: In an experiment testing a new study technique, what would the independent variable, dependent variable, experimental group, and control group be?
A: The IV is the study technique (new method vs. the existing method). The DV is exam performance. The experimental group uses the new technique; the control group uses the existing method.
Q: A researcher wants to know whether a drug reduces anxiety. She gives the drug to Group A and a sugar pill to Group B. Neither the participants nor the research assistants know who received what. What type of design is this?
A: A double-blind, placebo-controlled experiment.
Q: What is the key advantage of a mixed longitudinal design over a purely cross-sectional one?
A: It can separate genuine age-related changes from cohort effects, because it follows multiple age groups over time rather than comparing them at a single moment.
Q: A measure gives very similar scores when the same person takes it on two different occasions, but experts agree it does not actually measure what it claims to measure. Is this measure reliable, valid, both, or neither?
A: It is reliable (consistent scores) but not valid (does not measure the intended construct).
Research methods underpin everything else on this midterm. When you study behavioural neuroscience, you will see case studies (Phineas Gage), brain-imaging correlations (fMRI studies), and controlled experiments (lesion studies, drug trials). In sensation and perception, psychophysics experiments rely on precise operationalization of thresholds. In consciousness, much of the sleep research uses within-subjects longitudinal designs and EEG measures whose reliability and validity matter for interpreting results.
Related Terms / Search Tags
research methods, scientific method, hypothesis, theory, peer review, replication, case study, naturalistic observation, survey, self-report, sample, population, generalization, correlation, positive correlation, negative correlation, zero correlation, correlation coefficient, scatterplot, directionality problem, third variable problem, experiment, independent variable, dependent variable, IV, DV, experimental group, control group, placebo, placebo effect, single-blind, double-blind, random assignment, confounding variable, operationalization, meta-analysis, publication bias, file drawer problem, cross-sectional study, longitudinal study, cohort effects, mixed longitudinal design, reliability, validity, frequency distribution, mean, median, mode, standard deviation, normal distribution, bell curve, null hypothesis, statistical significance, p-value, IRB, IACUC, informed consent, debriefing, ethics, PSYCH 1100, intro to psychology, OSU