ChakataStat
Download

Getting Started

This page takes you from a fresh install to your first finished analysis. It assumes you have read ChakataStat at a glance; if not, skim that first for the big picture.

Installing ChakataStat

ChakataStat is a self-contained desktop application — there is nothing to configure and no separate runtime to install. Download the package for your platform and run it:

  • Windows — run the installer (.exe) or unzip the portable build, then launch ChakataStat.exe.
  • macOS — open the .dmg and drag ChakataStat to Applications.
  • Linux — install the .deb, the Snap, or run the bundled binary.

The statistics engine is bundled inside the app. If it ever fails to load, ChakataStat does not start silently broken — it shows a full-screen diagnostic with a Copy report button so you can see exactly what went wrong.

Building from source instead? The developer guide ships with the source, not with the application.

The first-run welcome

The first time you launch, a welcome overlay offers three ways to begin:

  • New Dataset — start with an empty grid and type or paste data in.
  • Import Data… — load a CSV/TSV, Excel or Parquet file.
  • Open Recent — reopen something you worked on before.

Press Esc or pick any action to dismiss it. It only appears once; you can always get back to these actions from the File menu.

A tour of the workbench

The window is organised around the view tabs on the left. The three you use constantly:

  1. Data View (Ctrl+1) — the spreadsheet. Each row is a case (one observation, e.g. one survey respondent); each column is a variable (e.g. Age). Hover a column header to see that variable's label, type and measurement level.
  2. Variable View (Ctrl+2) — the definition of each variable, one per row: its Name, Type, Label, Value labels, Missing values, Measure, and display width/decimals. Time spent here pays off everywhere downstream.
  3. Output View (Ctrl+3) — your results log. Each analysis or chart appends a titled card; nothing is overwritten.

On the right is the AI assistant panel (toggle with Ctrl+Shift+A); at the bottom, the status bar shows the case count and any active filter, weight or split.

Your first analysis in five minutes

We will use the bundled sample dataset, a small health study (File → Open Sample Dataset — 600 participants, 21 variables: age, sex, study zone, blood pressure, glucose, cholesterol, and more).

1. Open the data

File → Import Data…, choose health_study.csv, and accept the import options (ChakataStat infers each column's type — numbers become numeric variables, text becomes string variables). You now see 600 rows in the Data View. A progress dialog appears while the file loads.

2. Glance at the variables

Switch to Variable View (Ctrl+2). You will see Age, Sex, BMI, SystolicBP, and the rest. Notice that text columns like Sex and StudyZone came in as Nominal and numbers like Age as Scale — that measurement level is what tells the analyses and charts how to treat each variable. You can adjust any of it here; for now, leave it.

3. Run Descriptives

Back in any view, choose Analyze → Descriptive Statistics → Descriptives…. In the dialog:

  • Move Age, BMI, SystolicBP and FastingGlucose_mmolL into the Variables list (double-click, or select and use the arrow).
  • Click Run (or press Enter). The button shows a spinner while the engine works.

4. Read the output

ChakataStat switches to the Output View and shows a Descriptive Statistics table: for each variable, the valid N, minimum, maximum, mean and standard deviation. That is your first result.

5. Try a chart

Choose Graphs → Histogram…, pick BMI, and Run. A histogram card appears in the output, right under your table.

6. Save your work

File → Save (Ctrl+S) and save as health_study.ckd. The .ckd file is ChakataStat's native document: it stores the data and every variable definition together, so reopening it restores exactly this state. Your results log, AI conversation and syntax are remembered with the session, too.

What just happened (and why it matters)

Each thing you did was also recorded as a command. Open the Syntax console (Ctrl+4) and you will see lines like:

descriptives variables=[Age, BMI, SystolicBP, FastingGlucose_mmolL]
histogram variable=BMI

You can edit and re-run these, save them as an .cks script, or replay them on a different dataset — the foundation of reproducible analysis. That is covered in Reproducibility.

Where to go next