Data Organisation and Storage

Data Organisation effective data manipulation involves structuring files, documenting the research process, and metadata.

File structuring requires creating a project folder and individual sub-pages with publication sources, statistical (experimental) data, research results (program code, data analysis tables, work text), and a README file.

Documentation of metadata, data analysis, and transformation methodology provides an understanding of data and research process by all stakeholders.

Metadata information about the original data – describes the data and helps them to classify, organize and characterize. The key elements of metadata are the definition and designation of indicators, units of their measurement, a brief description of the assessment methodology and data sources.

File names should be unique, informative, not very long. It is advisable to use a standardized form for different versions of documents.

Recommended items for file names:

➠ name of the project or the name of the researcher

➠ job type or date of file creation (YYYYMMDD)

➠ a version of the document (example: V1, V1_2, V2)

➠ use of characters from sets A-Z, a-z, 0-9, hyphen, underscore and dot

Examples: MultivariteAnalysis_Part2_20190221.docx, Protsiuk_Thesis_V1.pdf, UkrStat _2000-2019.xlsx

Use the following data formats:

➠ Data Tables – CSV instead XLSX

➠ Text Data – TXT or PDF instead of DOC

➠ Databases – XML or SQLITE instead of MDB, DBF, SQL

➠ Visual – PDF, TIFF, JPEG2000, MPEG-4, WAVE, AIFF

Data Storage

Backing up information is used to store data and play it back in case of damage.

Backup-Rulecopy

Custom programs for project management and file versions: GIT: GitHub, GitLab, BitBucket, Trello.

Platforms for storing and sharing files: Open Science Framework, Google Drive, Dropbox, Box.