This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
The Changing Information Environment• Increasing volumes of data (overload?)• Greater emphasis on ‘evidence’ to back-up action• More obligations on public sector reporting +
more comprehensive auditing procedures• Information reaching all levels - tools getting
easier to use• Lay users often lack exploratory data skills –
need training to create ‘barefoot statisticians’• Knowledge management and metadata often not
considered high priority – management culture issue
• Data experts often under-valued and few in number
Guideline 1• Different user types have different needs – one size does
not fit all! Get to know your users, their skill levels, their level of data knowledge, what their ‘thinking task’ is likely to be - refine your end products as your understanding improves. Some users will find raw data meaningful, others meaningless. Public nature of Internet makes understanding citizen user needs a BIG challenge > try running consultations with sample user groups.
• Use presentation techniques appropriate to the nature of your data - 'a picture is worth a thousand words (or numbers)' – use comparisons - many prefer textual synopsis (80% people ‘number blind’).
"Bad graphics fail because they omit or manipulate context, deceive by discouraging comparison or obscuring important details, and confuse with visual miscues. Tell the truth. Show the data in its full complexity. Reveal what is hidden. Especially, respect the reader.“
Guideline 3• Use presentational techniques that take
advantage of the web – eg. don't replicate a hardcopy output in PDF if you can bring new insights and understanding through user interaction
• Communicating numeric statistics effectively to non-specialists requires thought, preparation and testing. Generic rule of thumb – 33% effort should go on data capture/research, 33% on analysis, 33% on dissemination. Dissemination often an after-thought!
Guideline 5• Reality is complex - problems are multivariate. Avoid
jargon. Make content engaging and useful. Where appropriate provide multiple routes into presenting the same data to suit different user needs – don’t assume novice users have ‘simple’ needs. ‘Dumbing down’ may not be the best solution.
• Presenting the figures well may not be enough – users may need training and impartial ‘expert’ advice at the end of the phone or in person (“answers on tap”).
Guideline 7• Presenting data ‘on-the-fly’ is a MAJOR challenge to do
well. Often puts the onus on non-expert users to do interpretation. Try to add tools and text to lead users through interpretation process.
• Delivering content dynamically may not be the best solution – use publishing approaches (data often not dynamic) to add more control over content, function and ultimately message. Can also include interpretation.
• Don’t underestimate the demand for embedding dynamic content into static content for publishing (Word, PDF etc) – try to make outputs easy to copy-paste and, if possible, ‘printer-friendly’
• Try to follow agreed web guidelines and standards eg. usability (response times, colours etc), metadata (eGMS, DDE, SDMX), web accessibility (WAI) etc.
Guideline 10
• Make metadata obvious/explicit – it should describe purpose, data sources and data characteristics (dataset -> record). Also metadata about elements of graphical presentation, searchable tags in XML/SVG files etc.
“Making decisions based on evidence requires the appropriate display of that evidence. Good displays of data help to reveal knowledge relevant to understanding mechanism, process and dynamics, cause and effect.”
“When consistent with the substance and in harmony with the content, information displays should be documentary, comparative, causal and explanatory, quantified, multivariate, exploratory, skeptical...it also helps to have an endless commitment to finding, telling and showing the truth.”