RESEARCH WORKFLOW · 05

Research Data Analysis

Build a reproducible path from data provenance and an analysis plan through cleaning, modelling, and sensitivity checks.

AI can generate code, suggest charts, or explain output, but validity depends on data quality, study design, assumptions, and validation. Treat generated analysis as a draft to be tested, and preserve the lineage from raw observations to reported results.

When this workflow is useful

  • Cleaning and exploring experimental or observational data
  • Building statistical or machine-learning analyses
  • Turning one-off analysis into a reproducible record

A practical sequence

  1. 01

    Record provenance and structure

    Retain read-only raw data, a dictionary, collection conditions, units, coding, and exclusions. Define boundaries before processing personal or restricted data.

  2. 02

    Specify the analysis question

    Distinguish exploratory from confirmatory work and state variables, outcomes, comparisons, confounders, and intended methods. Preserve result-driven changes.

  3. 03

    Make cleaning auditable

    Perform conversions, missing-data handling, anomaly checks, and derived variables in code. Test generated transformations on small cases, boundary values, and known outputs.

  4. 04

    Model, diagnose, and validate

    Examine assumptions, fit, leakage, overfitting, and multiplicity. Run sensitivity or external validation and pin software, seeds, parameters, and logs.

Quality-control questions

  • Can raw, cleaned, and analysis data be traced?
  • Does the method fit design and assumptions?
  • Was generated code tested, reviewed, and versioned?
  • Are uncertainty and sensitivity results reported?
Expected outputThe result should include reproducible code, transformation records, diagnostics, and uncertainty.

RELATED TOOLS

Tools that can support parts of this workflow

Tool inclusion does not replace method selection or validation.

All tools →
Research Data Analysis

JupyterLab

An open-source interactive computing environment that combines code, data, narrative, and visual output.

Open sourceOpen source
notebooksopen source
View tool
Research Data Analysis

MATLAB

A programming and numerical-computing environment used for algorithm development, data analysis, simulation, and engineering models.

Paid
numerical computingengineering
View tool
Research Data Analysis

Wolfram Mathematica

A technical computing system for symbolic mathematics, numerical methods, modeling, and visualization.

Paid
symbolic computationscientific computing
View tool
Research Data Analysis

KNIME Analytics Platform

An analytics platform for assembling data access, transformation, modeling, and machine-learning workflows from visual nodes.

Open sourceOpen source
visual workflowsdata analysis
View tool
Research Data Analysis

Orange Data Mining

An open-source visual programming application for data exploration, machine learning, and teaching.

Open sourceOpen source
visual programmingmachine learning
View tool
Research Data Analysis

ChatGPT Data Analysis

A ChatGPT capability that runs code to inspect uploaded data, perform analyses, and produce tables or charts.

Freemium
code executiondata analysis
View tool