Physical and Online Card Sorts: A Practical Overview and Case Study

Physical and Online Card Sorts: A Practical Overview and Case Study

Bob ThomasKatelyn ThompsonMay 28, 2008

Outline

• Scope, Goals, and Objectives

• Current Navigational Hierarchy

• Methodology

• Card Sorts

• Analysis of Results

• Recommendations

• Next Steps

Scope, Objectives, and Goals

• Mission• “Our mission is to improve teaching and learning by

providing customized assessment products and educational services.” (http://measuredprogress.org/aboutus/index.html)

• Scope• This project is focused only on Insite, the company’s

intranet web site for employees.

• Objectives• To make it easier for Measured Progress employees to find

information on the intranet, i.e., to “lead with the need.”

• Goals• Redesign Insite, focusing on the navigation and

organization • Make the site user friendly• Make the site attractive to employees

Home HR Policies and Procedures

The goal is to unite the separate Insite and HR Policies and Procedures sites, each created with different software applications.

Participants commented that that they would like to use Search and Site Map features. These were linked only from the home page.

InService

InForm

“You’d think the [Accident and Injury Report] form is under InForm.”InService > HR > Policy Guide > General Administration > Operational Policies > Safety > Accident Reporting and Investigation Plan > [Text… Scroll Down Page] … Accident Report Form (7 clicks)

“A person could bleed to death before the form is even located!”

Current Navigational Hierarchy

• Insite is broad and shallow• HR Policies & Procedures is narrow and deep

Insite Job Request Form:InForm > Job Request Forms > Form (2 clicks)

Policies and Procedures Safety Form:InService > HR > Policy Guide > General Administration > Operational Policies > Safety > Accident Reporting and Investigation Plan > [Text] … Accident Report Form (7 clicks)

Card Sorts: What

• Participants sort “a series of cards, each labeled with a piece of content or functionality, into groups that make sense to” them- Maurer & Warfel (2004), “Card sorting: a definitive guide”

• Closed card sorts• Participants sort cards into predefined bins

• Open card sorts• Participants sort cards and define bins

Card Sorts: Why

• “Card sorting is excellent for situations where you want the users’ mental model to drive the information architecture of the product.”- Courage & Baxter (2005), “Understanding Your Users”

• “Open sorts are used for discovery. Closed sorts are used for validation.”- Rosenfeld & Morville (2006), “Information Architecture for the World Wide Web”

Card Sorts: When

• Beginning of site creation

• Beginning of a site redesign

• Middle of a site redesign/creation

Card Sorts: Who

• Regarding users• Recruit users familiar with the “lingo”• But be cautious in a redesign with users who

are familiar with the website or application

Card Sorts: Where

• Lab• Moderated• Physical or online

• Remote• Moderated or un-moderated• Restricted to online

Card Sorts: How

• Physical• Participants may be apt to put more

thought into sorting

• Online• Cheaper• Higher number of participants• Eliminates data entry

Card Sorts: How

• Tools• EZCalc• MindCanvas• OptimalSort• SynCaps• WebSort

Tools: Mind Canvas

Card Sorts: How

• Analysis• More information to sift through with open

sorts• Dendrograms to see common groupings

Methodology

• Create and run an open card sort with 15 MP employees, including follow-up interviews

• Analyze quantitative and qualitative data

• Make recommendations for a new navigational hierarchy and structure

Open Card Sort# Card name

1 Corporate Graphics2 Floor Plans3 Facilities and Locations4 Contacts5 Employee Directory6 Corporate Travel7 Glossary and Acronyms8 Mission and Philosophy9 Safety

10 Reimbursement11 Payroll12 Purchasing and Requisitions13 Workplace Policies14 Hiring and Employment15 Benefits16 Work Schedules17 Personnel Records18 Shipping and Distribution19 Printing and Copying20 Helpdesk21 Forms22 Computer and Networks23 Corporate Marketing and Branding24 Information Technology25 Phone and Conferencing26 Contracts27 Wellness Programs28 Diversity29 In/Out Board30 News and Announcements31 Photo Gallery32 Corporate Events33 Department Sites34 For New Employees35 Workplace Conduct36 Recognition and Accomplishments

Participant Demographics

P# Gender Department Used Insite?0-5 6-10 IE Firefox Word processing Emailing Web Browsing

1 F Testing Services Y x x x x x2 F Human Resources Y x x x x3 M Client Services Y x x x x x4 F Human Resources Y x x x x x5 F Testing Services Y x x x x x6 F Office of Technology Y x x x x x7 F Finance Y x x x x x8 M Testing Services Y x x x x x9 F Client Services Y x x x x x

10 F Operational Services Y x x x x x11 M Scoring Y x x x x x12 F Process Coordination Y x x x x x13 F Marketing Y x x x x x14 F Client Services Y x x x x x15 M Marketing Y x x x x x

Hours Per Week Web Browser for Insite? What Do You Do on a Computer?

• All 15 participants had experience using Insite.

• We had participants from 9 different departments.

• We had 4 men and 11 women.

Sample Open Card Sort Results: Major Navigation CategoriesWHAT I WANT TO USE• Home Page• Contracts• Forms

WHAT I WANT TO REFER TO• Department Sites• Employee Directory• Working at Measured Progress• Helpdesk

Analysis of Results

• We began to see trends after the first day of testing (5 participants)Table 2. Main categories among first five participants.

Analysis of Results

• After all 15 card sorts were completed, standardized the main categories.

• Gave categories “with similar names or concepts a consistent name” and combined “groups where participants used the same basic concept but a slightly different label”- Maurer (2007), “Instructions for Use: Card Sort Analysis Spreadsheet”

Analysis of Results

• We started with 106 original categories from all 15 participants, and applied consistent naming conventions.

Table 3. Standardized categories derived from original categories.

Analysis of Results

• Second, combined similar standardized categories and arrived at a total of 13 standardized categories.

Table 4. Final list of 13 standardized categories.

0.67

Analysis of Results

Table 5. Mapping the dendrogram groupings to primary and secondary navigation areas.

Analysis of Results

• “Mental image the user forms to understand how software works and how to operate it”- Arnowitz et al (2007), “Effective Prototyping for Software Makers”

• In the case of Insite, many participants in our card study could not express a mental model of the website beyond the use of “In” headings

• But many did express their mental model of the ideal Measured Progress intranet. One participant noted:

• “There’s stuff there that people need to do their jobs. They need a reference library for working at the company… a big bookshelf of benefits and other [information].”

Analysis of Results

• Participants wanted easier access to information using primary, secondary, and third-level navigation

• Participants wanted a consistent navigational structure, i.e., one that didn’t change.

• Participants wanted “HR Policies and Procedures” broken into two separate categories: • HR Forms (such as “Corporate Travel,” “Payroll,”

“Requisitions”)• HR Policies (such as “Diversity,” “Safety,” “Workplace

Conduct”)

Analysis of Results

• Use a navigation hierarchy of medium depth and breadth• Use a primary navigation hierarchy of no more than

8 or 9 categories• Go 3 or 4 levels deep

• Put Search at the top of every page

• Put Site Map link on every page

Recommendations

Recommendations

Next Steps

• Develop wire frames of all eight primary navigational categories.

• Present the wire frames to the design team.

• Develop paper prototypes of the Insite Home page and one other subsection, such as HR.

• Test the usability of the paper prototypes with eight participants, using task-based scenarios. For example: “You are going on a business trip for Measured Progress. Using the web site, how would you complete the pre-approval process for your trip?”

• Develop electronic versions of the Insite Home page and one other subsection, such as HR.

• Test the usability of the electronic versions with eight participants, using task-based scenarios, as discussed above.

• Redesign Insite based on results from the usability test.

Thank You

• Questions?