ICDL Data Analytics (ICDL) key terms

The practice of examining raw information to find patterns, trends and useful insights that support decisions. It combines clear questions with tools and techniques for organising, analysing and presenting data.

Data sets so large, fast-moving or varied that ordinary spreadsheet tools struggle to handle them. Organisations use specialist storage and processing methods to capture and analyse such data.

Answers the question 'what happened?' by summarising past data, for example totals, averages and trends over time. It gives a factual picture of current or historical performance.

Looks at why something happened by comparing data, drilling into detail and searching for causes behind a pattern. It often combines several data sources to explain an observed result.

Uses historical data and statistical models to estimate what is likely to happen next, such as future sales or customer churn. Results are probabilities rather than certainties.

Goes beyond prediction by recommending actions, often weighing options against constraints. It may use optimisation or simulation to suggest the best course of action for a stated goal.

A measurable value chosen to show how well a person, team or organisation is achieving an important goal. Good indicators are specific, relevant and tracked consistently over time.

Information organised in a fixed pattern, such as rows and columns in a table or fields in a database, so software can easily search, sort and analyse it.

Information with no fixed, predefined format, such as free text, images, audio or video. It usually needs extra processing before it can be analysed alongside structured data.

Principles for collecting and using data fairly and responsibly, including respecting privacy, obtaining consent, avoiding bias and being open about how data is used.

Finding and fixing errors, duplicates, gaps and inconsistencies in a data set so that later analysis is reliable. It is often the most time-consuming stage of a data project.

Checking that data meets agreed rules, such as allowed values, correct format or sensible ranges, before it is used. Validation can be applied when data is entered or imported.

A value that sits far away from the rest of the observations in a data set. Outliers may be genuine extremes, data-entry mistakes, or signs of a faulty measurement process.

A blank or absent entry where a value was expected in a data set. Analysts must decide whether to leave it, remove the record, or replace it with a reasonable estimate.

The category defining what kind of values a field can hold, such as whole numbers, decimals, dates, text or true/false. Correct types are essential for sorting, calculating and charting.

The process of identifying and removing repeated records from a data set so that each real-world item appears once. It prevents totals and counts from being overstated.

Changing data into a more useful shape or format, for example converting units, splitting columns, standardising codes or grouping values, so it can be compared and analysed consistently.

Extract, Transform, Load: a three-step process that pulls data from sources, cleans and reshapes it, then writes it into a target store such as a data warehouse for analysis.

Measures that protect data from loss, theft or unauthorised access, including passwords, access permissions, encryption and backups. Security supports both legal compliance and user trust.

The arithmetic average, found by adding all values and dividing by how many there are. It uses every value but can be pulled strongly by a few extreme figures.

The middle value when data is arranged in order, or the average of the two middle values when the count is even. It is less affected by extreme values than the mean.

The value that appears most often in a data set. It is useful for categorical data such as shoe size or product choice, where an average would be meaningless.

A number showing how spread out values are around the mean. A small value means most values sit close to the average; a large one means they are widely scattered.

A measure of how strongly two variables move together. It ranges roughly from minus one to plus one, and a strong relationship does not by itself prove that one causes the other.

Combining many individual records into summary figures, such as totals, counts or averages per group. It makes large data sets easier to compare, chart and report.

Temporarily hiding records that do not meet specified conditions so that only relevant rows are shown or analysed. The underlying data is unchanged and the filter can be removed.

An interactive summary tool that groups and cross-tabulates data by chosen fields, letting you quickly total, count or average values and reorganise the layout without rewriting formulas.

The general direction in which a series of values moves over time, ignoring short-term ups and downs. Trends are often shown with a line chart or a moving average.

A statistical method that models the relationship between one outcome variable and one or more input variables, often used to estimate effects or make predictions from new values.

Testing how results change when input assumptions are varied, for example changing prices or costs, so that decision makers can see the range of possible outcomes.

Presenting data as charts, graphs, maps or other visual forms so that patterns, comparisons and exceptions can be understood quickly. Good visuals match the chart type to the question asked.

A single screen or page that gathers key charts and figures, usually linked to live data, so a user can monitor performance at a glance and spot problems early.

A chart that uses rectangular bars of varying length to compare values across categories. Bars may be vertical or horizontal and suit data with many or long category labels.

A chart that joins points with a line, mainly to show how a value changes over time or another continuous sequence. Plotting several lines allows different series to be compared.

A chart plotting one value against another for each record, revealing whether two variables are related. It also helps to spot clusters, outliers and unusual patterns.

A circle divided into slices to show each category's share of a whole. It works best with few categories and when the message concerns proportions rather than exact values.

Combining data, clear visuals and a short narrative so that an audience understands what the numbers mean and what action is being recommended.

A structured document or file that presents findings, visuals and commentary to a defined audience, usually including a purpose, a method and conclusions or recommendations.