axisgroup / evaluation-toolkit Goto Github PK

Replace icons used for best practices

Not everybody shares the same passion for cakes as you @manasvil

Just in general, it'd be nice for some things to be linked, like apps, websites, methodologies, articles, and other authorities that can back up the claims and lend multiple perspectives to the issues covered here.

otherwise I won't know which file(s) to go to next

Update section: https://github.com/axisgroup/evaluation-toolkit/blob/master/3.Plan-the-test/README.md

Use: Writing the right tasks https://projects.invisionapp.com/d/main#/projects/boards/4517347/137764798
http://design.canonical.com/2013/08/usability-testing-how-do-we-design-effective-tasks/
and this summary image https://github.com/axisgroup/evaluation-toolkit/blob/master/Assets/images/task-types.png

Add links to folders such as Setting the stage, Determine research questions as well a description of what these folders encapsulate.

Eg. Effectiveness can be measured as a % of people who completed assigned tasks across multiple usability tests, or Likert scale ratings from a survey

Anyone have any recommendations on how to make it non-redundant?

Include cool icons, references to related articles, make sentences crisper.

Content should be the nav, but have nav as the exit strategy

The challenges document is currently in a bullet form
Needs sewing up+add additional issues

Write down risks of evaluation

How UCD is based on three pillars

In planning the test, lead with an intro to the Research Landscape; How you can have attitudinal vs. behavioral measurements +qual and quant

Then say what are some things people do for data viz and pros and cons of those
Then,final section should be Putting a stake in the ground-

1. Usability Test
2. Expert Review/ Heuristic Evaluation

Clean up based on -
http://www.webdesignerdepot.com/2015/02/how-to-test-the-usability-of-prototypes-like-a-pro/
and Charline Poirier's article https://design.canonical.com/2012/03/about-usability-testing-recruiting/

Too quote-y right now

Questionnaire can be found here: https://www.usability.gov/how-to-and-tools/methods/system-usability-scale.html

Add attribution at the end
This questionnaire is based on the System Usability Scale (SUS), which was developed by John Brooke while working at Digital Equipment Corporation. © Digital Equipment Corporation, 1986.

https://www.nngroup.com/articles/no-validate-in-ux/

Method template should be as such-

Executive Summary
What
When
Why
How
Method in Action
References

For each section, add a visual example of a best practice and a visual example of it opposite for a this vs that effect

https://github.com/axisgroup/evaluation-toolkit/tree/master/1.%20Following%20best%20practices

• a delicate balance, trade-offs
  -beyond lip service
• link to visual design principles
• Include emphasis on Accessibility
• EXPOSURE to an interface before feedback gathering
• Deferring changes vs. Denying
• Cognitive, mechanical burden
• UX Research Landscape
  how your methodology can fall on a gradient
  context matters
  go back to goals of why we are testing: diagnostic, realism
  -Operationalizing measures Eg. Effectiveness can be measured as a % of people who completed assigned tasks across multiple usability tests, or Likert scale ratings from a survey

We can use a custom dashboard we've designed and try the expert review/heuristic evaluation against it, to provide more specifics around how this evaluation would work in context of data viz.

Start with Axis Style Guide

https://github.com/axisgroup/evaluation-toolkit/tree/master/3.%20Planning%20the%20Test#the-complexity-in-our-stars

Add this snippet to the know what to measure section

https://github.com/axisgroup/evaluation-toolkit/blob/master/Assets/images/style-of-data.png

https://measuringu.com/seq10/

Make it easier for folks to scan the list, then dive into the section/best practice that they need to reference

https://github.com/axisgroup/evaluation-toolkit/blob/master/1.%20Following%20best%20practices/README.md

The Information Discrimination section looks very awkward-
https://github.com/axisgroup/evaluation-toolkit/tree/master/1.Following-best-practices

https://github.com/axisgroup/evaluation-toolkit/tree/master/1.%20Following%20best%20practices#the-mantra-we-follow-for-seeking-visual-information

https://github.com/axisgroup/evaluation-toolkit/blob/master/1.Following-best-practices/Design-Reference-Sheet.pdf

The word existent on the data set familiarity is misspelled. Does anyone have the .docx file?

Executive Summary (if applicable)
What
When
Why
How
Method in Action
Resources
References

Such as ambiguous, frustrating

1. Agreement about what counts as a heuristic- In most UX scenarios, Nielsen and Norman's heuristic are used as an accepted usability standard. But in our case, we have a list of info vis best practices, which though generally agreed upon within the team, may not be as comprehensive as it could be. For instance, during the demo test, there was a question about "I cannot find details about my target for this month but how important is it really to know the number?" and the expert did not know how to categorize this issue. One approach is to capture these questions as miscellaneous. Another approach would be to increase the number of best practices and classifying them under the useful, usable, desirable framework.

2. Setting expectations with the expert- The expert felt the burden to find an issue in violation of every best practice or not even though the best practices section is just a guide to be able to categorize issues and also to prime them to think about potential issues. This goes to having a common understanding about the expert about what they are required to do.

3. Task make every test- There is a very important precursor to every testing method and that is to write tasks that allow for the evaluator to explore the interface. FInding issues is an incidental process of performing these tasks and hence the burden of this falls on the facilitator and this is not something the evaluators should worry about.

4. Writing task guideline- Whom you interviewed and whom you are testing with can be an important determinant of writing tasks and establishing usability criteria.

5. Heuristic evaluation - needs scoring method and details on how to conduct the test

Add proper links and citations to Quoted Text
Right now they're in quote blocks

The readme is classically (and on Github) used as a way to get around a project (or repo). Right now, the readme. Overall it reads very well and describes process, but it's written in a narrative style, which is good for a one-sitting read-through, but there isn't an intro or quick section listing that would allow for someone to quickly reference something. There also ought to be a succinct "this is what this repo is about" bit at the beginning and a more robust one with links to each of the folders in the How To Use section.

Folder names have changed so some links may have broken