Help CenterData Editor

Data Preparation

The Data Editor is where raw files become analysis-ready datasets — load, inspect, clean, transform, and merge your data, then send it straight to the labs.

Overview

The Data Editor is a spreadsheet-style workspace built for getting data ready to analyze. Open a file, work through it tab by tab, and apply cleaning and transformation steps in place — with undo / redo and keyboard shortcuts throughout. When the dataset is ready, hand it to an analysis lab with Send to Statistical Lab.

New here? Start with the Loading Data guide to get a file in, then come back to follow the workflow below.

What you need

A data file: CSV or Excel (.xlsx / .xls) — or a built-in sample dataset
Nothing else: No setup, no import wizard — data is editable the moment it loads
A goal: Know which lab the data is headed to (e.g. Statistical Lab)

What you get

A clean dataset: Missing values filled, outliers and duplicates handled
Reproducible steps: A Pipeline that records and re-applies every transform
A direct handoff: Prepared data sent straight into the analysis labs

What the Data Editor does

Everything you need to prepare a dataset lives in one place. Here is the full set of capabilities — each is covered in detail on its own page below.

Load

Drag-and-drop or click to upload CSV and Excel files, open several datasets in separate tabs, or start from a built-in sample dataset.

Edit

Edit cells and headers, add or remove rows and columns, build formula-based derived columns, bulk-rename, search, sort, and filter per column.

Clean

Fill missing values with several methods, remove outliers using IQR or Z-score bounds, and drop duplicate rows.

Transform

Apply numeric transforms, convert scales (Z-score, MinMax, Robust), and encode categorical columns.

Merge

Join two tabs together by choosing a join type and a join key.

Reproduce

A transformation Pipeline records every step and re-applies it; a query-based extract pulls a reproducible subset.

Typical workflow

Most preparation follows the same five steps. You do not have to use every step — skip the ones your data does not need — but this is the order that keeps things predictable.

1

Load

Upload a CSV or Excel file, or open a sample dataset. Each dataset gets its own tab.

2

Inspect

Scan column types, search, sort, and filter to understand what you are working with.

3

Clean

Fill missing values, remove outliers, and drop duplicate rows.

4

Transform

Scale numeric columns, encode categories, and add derived columns as needed.

5

Analyze

Send the prepared dataset to a lab — for example, "Send to Statistical Lab".

Tip: turn on the transformation Pipeline before you start cleaning and transforming. It records each step so you can re-apply the exact same preparation to a refreshed file later — or to a query-based extract.

Detailed guides

Each part of the workflow has a dedicated, task-oriented guide. Start with Loading Data, then move through the rest as you need them.

Loading Data

Upload methods, supported formats, sample datasets, and what happens after a file lands.

Editing

Edit cells and headers, manage rows and columns, build derived columns, search, sort, and filter.

Cleaning

Fill missing values, remove outliers with IQR or Z-score bounds, and remove duplicate rows.

Transforming

Numeric transforms, scale conversion, and categorical encoding.

Merging

Join two tabs by selecting a join type and join key.

Common questions

데이터 준비가 끝났다면

Data Editor에서 Send to Statistical Lab 버튼을 누르면 현재 탭의 데이터가 바로 분석 화면으로 이동합니다. 또는 Statistical Lab을 직접 열고 파일을 드래그해서 올려도 됩니다.

Statistical Lab 사용 가이드 보기