Data science & AI · Reference
What is data analysis?
Data analysis is the process of inspecting, cleaning, transforming, and modelling data in order to discover useful information, draw conclusions, and support decision-making.
What data analysis involves
Data analysis turns raw data into understanding. It begins with preparing the data — checking quality, handling missing values, and correcting errors — because conclusions are only as reliable as the data behind them. Analysts then transform the data into a usable form and apply statistical or computational methods to summarise it, find patterns, or test ideas. Throughout, the aim is to answer a clearly stated question, not simply to manipulate numbers; framing the question well is as important as the techniques applied.
Types of data analysis
Descriptive analysis summarises what the data shows — averages, distributions, and trends — without drawing broader conclusions. Exploratory data analysis (EDA), an approach championed by statistician John Tukey, uses visualisation and summaries to uncover patterns, spot anomalies, and generate hypotheses.
Inferential analysis goes further, using a sample to draw conclusions about a larger population and to quantify uncertainty through estimates and hypothesis tests. Each type answers a different question, and a single project often moves through several.
Data analysis in research
In research, data analysis is the stage where evidence is weighed against questions and hypotheses. Methodological care is essential: distinguishing exploratory findings (which generate hypotheses) from confirmatory ones (which test pre-specified hypotheses) guards against false discoveries from "fishing" through data. Reporting standards call for pre-registering confirmatory analyses where possible, stating assumptions, and reporting effect sizes and uncertainty — not only whether a result is statistically significant.
Analysis versus analytics
Data analysis and data analytics are closely related and often used interchangeably. A common distinction treats analysis as the detailed examination of a specific dataset to understand what happened, and analytics as the broader discipline — including the tools, systems, and forward-looking methods — applied to data more generally. Both sit within the wider field of data science.
Key facts
At a glance
- Definition: inspecting, cleaning, transforming and modelling data
- First stage: data preparation and quality checks
- Descriptive: summarises what the data shows
- Exploratory (EDA): finds patterns and hypotheses (Tukey)
- Inferential: draws conclusions about a population from a sample
- Key safeguard: separate exploratory from confirmatory analysis
Common questions
FAQ
What are the main types of data analysis?+
Descriptive analysis summarises what the data shows; exploratory analysis uses visualisation to find patterns and generate hypotheses; and inferential analysis uses a sample to draw conclusions about a wider population with quantified uncertainty.
What is exploratory data analysis?+
Exploratory data analysis (EDA) is an approach, associated with John Tukey, that uses graphs and summaries to understand a dataset, reveal patterns, and spot anomalies before any formal modelling. It generates hypotheses rather than confirming them.
What is the difference between data analysis and data analytics?+
The terms overlap. Analysis usually means examining a specific dataset to understand what happened, while analytics often refers more broadly to the discipline, tools, and forward-looking methods applied to data, including prediction.
Going deeper
Related on CASRAI
- What is data analytics? →
- What is data science? →
- What is PCA? →
- Research methods →
- Computer science, data science & AI →
Sources
The step most authors miss
Doing CRediT right? Don’t stop at the statement.
A CRediT statement credits you inside one paper. The recognition CRediT was built for happens when those roles are tied to you, persistently. Sign in with your ORCID — free — and claim your CRediT contributions on casrai.org, the home of the standard. They become a verified, portable part of your identity, not a line that disappears into one PDF.
Free: claim your contributions, then export a journal-ready CRediT statement, schema.org structured data, JATS XML, CSV or BibTeX — and preview your public profile. A membership publishes that profile publicly and verifies the journals you serve.
