A fuller explanation is being developed. Verify this concept with cited sources before relying on it.
Introduction to Psychological Statistics
Psychological statistics provides methods for organizing, describing, analysing and interpreting data about mental processes and behaviour. It helps researchers summarize observations, estimate population characteristics, evaluate hypotheses and communicate uncertainty.
Explanation
Example or application
A worked example has not yet been added.
Conceptual Coverage
Statistics in psychology includes descriptive and inferential methods. Descriptive statistics summarize observed data using frequencies, proportions, measures of central tendency and variability. Inferential statistics use sample data to estimate population parameters or evaluate specified hypotheses. Common methods include confidence intervals, t-tests, analysis of variance, correlation, regression, chi-square tests and non-parametric procedures. Bayesian methods provide another framework for updating probability distributions in light of evidence. Statistical choices depend on the research question, measurement scale, sampling design, distribution, independence assumptions and study structure. Statistical significance, effect size and practical importance are distinct concepts.
Scientific Foundation
Statistical methods entered psychology through research on individual differences, psychophysics and experimental measurement. Francis Galton and Karl Pearson contributed to correlation and regression, while Ronald Fisher developed influential methods for experimental design and significance testing. Neyman and Pearson developed a decision-theoretic framework for hypothesis testing. Later developments include generalized linear models, multilevel modelling, Bayesian statistics and meta-analysis. Statistical methods continue to evolve in response to reproducibility concerns and increasingly complex research designs.
Deep Understanding
A sample is used to learn about a population, but sample statistics vary from sample to sample. Statistical inference quantifies this uncertainty under a model and specified assumptions. A p-value describes how unusual data at least as extreme as those observed would be under a null model; it is not the probability that the null hypothesis is true. A confidence interval is produced by a procedure that has a stated long-run coverage rate under its assumptions. Effect size describes the magnitude of a difference or relationship. Researchers should interpret all of these alongside study design, measurement quality and practical context.
Mathematical Coverage
The mean is x̄ = Σxᵢ/n, and the sample variance is s² = Σ(xᵢ − x̄)²/(n − 1). The standard error of a sample mean is SE = s/√n under common assumptions. A standardized mean difference can be expressed as Cohen's d = (x̄₁ − x̄₂)/sₚ, where sₚ is the pooled standard deviation for an appropriate independent-groups design. Pearson's correlation coefficient ranges from −1 to +1. A simple 95% confidence interval for a mean is approximately x̄ ± t*SE, where t* depends on the degrees of freedom and desired confidence level. For a chi-square test, χ² = Σ(O − E)²/E, where O and E are observed and expected frequencies. Each method has assumptions that must be checked.
Practical Learning
Consider the scores 60, 70, 70, 80 and 90. The mean is 74, the median is 70 and the mode is 70. These summaries describe different aspects of the distribution. In a study comparing two groups, report each group's sample size, mean, standard deviation, estimated difference, confidence interval and an appropriate effect size. Use a graph to inspect distributions and outliers before choosing a test. Practise interpreting a statistical result in plain language without overstating its meaning.
Critical Thinking
A small p-value does not guarantee a large or important effect, and a large p-value does not prove that no effect exists. Multiple testing increases the chance of false-positive findings if not addressed. Statistical assumptions may be violated, and inappropriate methods can produce misleading results. Averages can hide subgroups, while correlations can be distorted by outliers or confounding. Researchers should report estimates and uncertainty rather than relying only on a threshold such as p < 0.05.
Knowledge Assessment
1. Distinguish descriptive and inferential statistics. Answer: Descriptive statistics summarize observed data; inferential statistics estimate or test claims about a wider population or model. 2. Calculate the mean of 60, 70, 70, 80 and 90. Answer: 370/5 = 74. 3. What does a standard deviation describe? Answer: The spread of observations around the mean. 4. Does p = 0.01 mean there is a 99% probability that a research hypothesis is true? Answer: No. 5. If two groups differ by 10 points but the confidence interval is wide, what should the researcher emphasize? Answer: The estimated difference and its substantial uncertainty, alongside the design and effect-size context.
Further Learning
Study probability, distributions, sampling theory, estimation, hypothesis testing, effect sizes, regression, analysis of variance, non-parametric statistics and Bayesian inference. Later explore multilevel models, factor analysis, structural equation modelling and meta-analysis.
Sources and verification
Sources are shown as provided in the project data. Not every source has yet been independently verified. Please report errors or missing citations.
- Source verification pending; check an authoritative textbook, university, standards body, or peer-reviewed source.