학술논문

Ten simple rules for initial data analysis.
Document Type
Editorial
Source
PLoS Computational Biology. 2/24/2022, p1-7. 7p.
Subject
*DATA analysis
*METADATA
*STATISTICAL decision making
*DOCUMENT markup languages
Language
ISSN
1553-734X
Abstract
For this purpose, IDA provides data ready for analysis including reliable information on the data properties. Having well-documented metadata, making data cleaning code-based and repeatable, greatly enhances other people's understanding of the data properties and data quality - including a future version of yourself who may need to revisit a project later. However, by not prospectively planning to assess data quality and context, the boundaries between answering the research question, model checking, data cleaning, or examining data properties can erode and turn into a cyclical investigation. Data is the new oil [[1]], and analyzing data has been described as the number 1 profession of the 21st century [[2]]. [Extracted from the article]