Data File Management

This guide discusses the best practices of data file naming, file formats, and file versioning and control

Data Services

For more information or for assistance, please contact Data Services at datalib@uark.edu.

 

Codebooks

Codebooks

A codebook acts as a place to store and organize information related to your research and data and make take a variety of forms such as .pdf, html, or xml.  If creating an xml mark-up codebook, consider following the Data Document Initiative (DDI) standards.  If you have data in SPSS, SAS, or STATA formats, there are tools available through the software to assist in creating DDI codebook metadata.

You may want to track the following information and include it in your codebook:

  • Dataset creators – full names and/or organizations
  • Funding sources, awards granted, and award amount and reference number
  • Abstract of the research, including the purpose and output
  • Versioning information
  • Products and supplies used in research, where they were obtained, quantity of each
  • Dates of research activities
  • File names and formats
  • Explanations of variables, codes, and abbreviations
  • Sources for data you did not collect

Examples of codebooks in an original format and in DDI xml markup are available here .

Lab Notebooks

Laboratory notebooks are the places where researchers document the experimental process. They are organizational tools and memory aids and can even assist in protecting institutional property rights.  Lab notebooks may be electronic; however, paper is still the preferred method in many labs.

While the format and construction of the laboratory notebook may determined by an institution, department, instruction or researcher, the following are the basic elements of experimental entries:

  • Date
  • Title
  • Hypothesis (goal / purpose)
  • Background
  • Protocols, reagents, equipment, calculations, etc.
  • Observations
  • All that happened  (planned and unplanned)
  • Raw experimental data
  • Reference to data location
  • Data analysis
  • Processing of raw data, graphs, interpretations
  • Future experiments

Ethics are an important part of the scientific process

  • All data should go in notebook - even" bad" data points, outliers and failed experiments are included
  • Do not remove pages from a notebook
  • Do not skip pages. Cross out unused parts of page
  • Do not remove mistakes. Cross them out with a single line
  • Corrections which are signed and data may be pasted in as long as the original is visible.

For more information on lab notebooks see Keeping a Lab Notebook  by the National Institutes of Health, Office of Intramural Education and Training.