Psychology terms in Data analysis and interpretation

A graph in which separate bars, with gaps between them, show a value (such as a count or a mean) for each category or condition; the categories have no numerical scale.

A distribution with two distinct peaks, which often shows that the data combine two different groups, such as two age groups or two types of participant.

A graph that summarises a distribution with a box from the lower to the upper quartile, a line at the median, and whiskers to the lowest and highest scores, with outliers often shown as separate points.

In qualitative analysis, the process of attaching short labels (codes) to segments of data that capture a feature relevant to the research question, so that similar segments can be collected and compared.

The square of the correlation coefficient (r²), which gives the proportion of the variation in one variable that is accounted for by its linear relationship with the other.

A range of values around a sample estimate, calculated so that intervals made this way would contain the true population value a stated percentage of the time, usually 95%.

The process of organising, summarising, displaying and interpreting research data, with tables, graphs, statistics or thematic analysis, to find patterns and judge what conclusions the data supports.

The way the scores in a set of data are spread across their possible values, shown by how often each value or interval occurs; described by its shape, centre and spread.

A standardised measure of how large a difference or relationship is, independent of sample size, such as Cohen’s d (the difference between two means in standard deviation units) or the correlation coefficient r.

A line drawn above and below a mean on a graph to show the spread or uncertainty around it, such as one standard deviation, the standard error or a 95% confidence interval.

A table that shows how many times each value or category occurs in a set of data, often with the percentage of the total for each.

A graph of the frequency distribution of continuous data, in which touching bars show how many scores fall into each equal-width interval on the horizontal axis.

A qualitative method in which categories are not decided in advance but are developed from the data themselves, by open coding, grouping codes and abstracting them into broader categories.

Statistical procedures that use data from a sample to judge whether a difference or relationship is likely to exist in the population, by estimating how probable the result would be if it were due to chance alone.

A statistical procedure, such as a t-test or chi-squared test, that turns sample data into a test statistic and a p-value so that a decision can be made about the null hypothesis.

A measure of dispersion: the difference between the upper quartile and the lower quartile, giving the spread of the middle 50% of scores; it is not affected by extreme scores.

Numerical data measured on a scale with equal units between values but no true zero point, such as temperature in degrees Celsius or scores on many standardised tests.

A theme that goes beyond what participants said explicitly to identify the underlying ideas, assumptions or social meanings that shape the content of the data.

A classification of quantitative data by the information its numbers carry: nominal (categories), ordinal (ranks), and interval or ratio (equal units). It decides which graphs and statistics can be used.

A graph in which points joined by lines show how a variable changes over time or across ordered levels of another variable.

A straight line drawn through the points of a scatterplot so that it lies as close as possible to all of them, showing the direction and slope of a linear relationship.

A descriptive statistic that shows how spread out the scores in a data set are, such as the range, the interquartile range or the standard deviation.

The increase in the risk of at least one type I error when many significance tests are run on the same data, because each test at the 0.05 level carries its own chance of a false positive.

Data in the form of separate, unordered categories (such as attachment type or country), summarised by counting how many cases fall into each category.

A symmetrical, bell-shaped distribution in which most scores cluster around the centre, the mean, median and mode are equal, and about 68% of scores lie within one standard deviation of the mean.

A significance test used with a directional hypothesis, which places the whole significance level in one tail so only an effect in the predicted direction can be significant; a two-tailed test splits it between both tails.

Data that can be put in order or ranked (such as ratings on a 1–5 scale), where the gaps between the values are not known to be equal.

A score that is far higher or lower than most of the other scores in a data set; one common rule treats any score more than 1.5 interquartile ranges beyond a quartile as an outlier.

Analysing data in many different ways, such as adding participants, dropping outliers or testing subgroups, and reporting only the analyses that reach statistical significance, which inflates the rate of false positives.

The probability of obtaining a result at least as extreme as the one observed if the null hypothesis were true; a small p-value means the data would be surprising if there were no real effect.

A circular graph divided into slices whose sizes show each category’s share of a whole; useful for a few categories that add up to 100%.

The extent to which an effect is large enough to matter in real life, for example to change behaviour or justify an intervention; it is judged from the effect size and context, not from the p-value.

A measure of dispersion: the difference between the highest and lowest scores in a data set. It is quick to calculate but depends entirely on the two most extreme scores.

Numerical data measured on a scale with equal units and a true zero, such as reaction time, age or number of words recalled, so that one value can be twice another.

The discovery, from about 2011 onwards, that many published findings in psychology and other sciences could not be reproduced when the studies were repeated, which led to reforms such as preregistration.

A graph in which each point shows one participant’s (or case’s) scores on two variables, used to show the direction, strength and form of a correlation.

A theme that stays at the explicit, surface level of what participants said, describing and summarising the content of the data without looking beyond it.

The probability threshold, set before analysis and usually 0.05 in psychology, below which a result is treated as statistically significant; it is the accepted risk of rejecting a true null hypothesis.

The degree to which a distribution is asymmetrical. In a positive skew the tail is longer on the right (high scores) and the mean is above the median; in a negative skew the tail is on the left.

A measure of dispersion that shows how far, on average, scores lie from the mean of their data set. A small standard deviation means scores cluster closely around the mean.

The probability that a significance test will detect an effect that really exists in the population; it equals one minus the probability of a type II error and rises with sample size and effect size.

A qualitative method for identifying, analysing and reporting patterns of meaning (themes) across a data set such as interview transcripts, usually following Braun and Clarke’s six phases from familiarisation to the written report.

A diagram used in thematic analysis to show candidate themes, their subthemes and the codes that belong to them, and how the themes relate to one another.

A pattern of shared meaning across a qualitative data set, built from related codes and organised around a central idea that answers something important about the research question.

A graph axis that starts at a value other than zero, which can make small differences between bars look much larger than they are.

A measure of dispersion equal to the average of the squared deviations of scores from their mean; its square root is the standard deviation.

A score expressed as the number of standard deviations it lies above or below the mean of its distribution, which allows scores from different tests to be compared.