VCE Applied Computing Data Analytics key terms

The accuracy, consistency and completeness of stored data over its whole life. It is maintained through validation rules, access controls, backups and careful editing so that decisions rest on trustworthy information.

A check applied when data is entered or processed that rejects values failing a condition, such as a range limit, a required field or a valid date format. It prevents obviously wrong data entering a system.

Confirming that data already captured matches its original source, for example by double entry or sighting a document. Unlike validation, it checks whether the data is correct rather than merely plausible.

The stages data passes through from creation to disposal: collection, storage, processing, use, sharing, archiving and secure destruction. Each stage raises different accuracy, privacy and security obligations for an organisation.

A documented policy setting how long each type of data is kept and when it is securely destroyed. Keeping data longer than needed increases privacy risk, while destroying it too early can breach legal or reporting duties.

Data that describes other data, such as field names, data types, units, capture dates, sources and version history. Good metadata lets another analyst understand and correctly reuse a dataset without guessing.

A reference document listing each field in a dataset with its name, data type, permitted values, units and meaning. It supports consistent entry, correct interpretation and easier maintenance across a project team.

The category of values a field may hold, such as integer, decimal, text, date, boolean or currency. Correct types allow valid calculations and sorting, and prevent errors such as treating a number as text.

Data collected first-hand by the analyst for the specific purpose at hand, through surveys, sensors, interviews or experiments. It is tailored and current but usually slower and costlier to gather.

Data gathered by someone else for another purpose and reused, such as government statistics, company reports or open datasets. It is quick and cheap but may be outdated, differently defined or unverifiable.

Selecting a subset of a population to represent the whole. Random selection reduces bias, while sample size and method determine how confidently findings can be generalised to the wider population.

A systematic distortion that pushes results consistently in one direction, arising from unrepresentative samples, leading survey questions, missing groups or measurement instruments. It does not shrink with a larger sample.

The documented origin and history of a dataset, including who collected it, when, by what method and what changes were made since. It supports trust, reproducibility and audit of analytical findings.

Detecting and correcting errors such as duplicates, impossible values, inconsistent formats and missing entries. It improves accuracy before analysis so that results are not distorted by faulty records.

Reshaping raw data into an analysable form, for example splitting columns, merging tables, pivoting, standardising categories and converting types. It often consumes more project time than the analysis itself.

Changing data values or structure into a more useful form, such as scaling, aggregating, normalising units or deriving new calculated fields. Each change should be documented so results remain traceable.

Absent values in a dataset, caused by non-response, sensor failure or entry errors. Analysts must decide whether to remove affected records, impute values or report the gap, since each choice affects conclusions.

A value lying far from the rest of the distribution. It may be a genuine extreme case or a recording error, so it must be investigated rather than automatically deleted, as removal can distort the analysis.

Representing data graphically, through charts, maps or dashboards, to reveal patterns and support interpretation. The chosen form must suit the data type and the message, or it will mislead viewers.

A chart that distorts interpretation through truncated axes, inconsistent scales, 3D effects, unlabelled units or selective data. It can create false impressions even when the underlying figures are accurate.

Combining visuals, narrative and context to communicate findings to a specific audience and prompt action. It links evidence to a clear recommendation rather than simply presenting charts.

A single screen presenting key indicators and trends, often updating automatically from a live source. Good dashboards prioritise a few decision-relevant measures over exhaustive detail.

A measurable value chosen to track progress towards a specific objective, such as average wait time or conversion rate. It should be defined precisely, with known formula and data source.

A statistical relationship in which two variables tend to change together, measured on a scale from minus one to plus one. Correlation alone does not establish that one variable causes the other.

A relationship in which a change in one variable actually produces a change in another. Establishing it requires controlled comparison or strong evidence, not merely an observed association between the two.

The general direction of a series over time, after short-term fluctuations are smoothed or set aside. Identifying trends supports forecasting, but projections assume past patterns continue.

Gaining a person's clear, voluntary agreement to collect and use their data, after explaining purpose, scope and recipients. It underpins lawful and ethical data collection and must be recorded.

A person's interest in controlling how their personal information is collected, used, stored and shared. It is protected by law and by practices such as data minimisation and restricted access.

Removing or altering identifying details so individuals cannot reasonably be re-identified from a dataset. It reduces privacy risk but is weakened when combined fields allow re-identification.

Technical and procedural safeguards protecting data from unauthorised access, alteration or loss, including encryption, access permissions, backups and audit logs. Security supports both privacy and integrity obligations.

Legal rights over creations such as datasets, text, images and software, covering copyright, licensing and ownership. Reusing secondary data requires checking licence terms and attribution requirements.

A written statement of what a data analytics project will deliver, including objectives, boundaries, deliverables and exclusions. Clear scope prevents uncontrolled expansion and keeps effort aligned with the client's need.

A significant checkpoint in a project plan marking completion of a deliverable or phase, with a target date. Milestones help a team track progress and communicate status to stakeholders.

A horizontal bar chart showing project tasks against a timeline, including durations, dependencies and overlaps. It helps a team schedule work, spot bottlenecks and monitor progress against deadlines.

A system that records changes to files over time so earlier versions can be compared or restored. In collaborative data work it prevents lost edits and shows who changed what and when.