Web Usability for Older Users

Document Sample
Web Usability for Older Users Powered By Docstoc
					Improving Web Usability for
Older Users, Low Vision
Users, and Everyone
Does Visual and Audio Help
Improve Performance?

Ann Chadwick-Dias, Marguerite Bergel,
& Tom Tullis
Fidelity Center for Applied Technology
Fidelity Investments
October 18, 2005
    Fidelity Investments
One of the world's largest providers of financial services:
    One of the largest US mutual fund companies
    No. 1 provider of workplace retirement savings plans
    Human resources and benefits outsourcing services
    A leading online brokerage firm
    Customer assets of over 2.1 trillion
A few Website stats (as of January 2005):
    94% of our commissionable trades occur online
    An average of more than 1 million contacts a day
    Over 5.5 million online accounts
    12 million US employees save via a Fidelity-administered plan
    78,000 online trades daily, on average

                          Aging by Design October 2005        Slide #2
Recent Research
   Baby boomers represent an important
    demographic for any ecommerce company
   Prompted research into learning their
   Includes accommodating higher incidences of
    all impairments, focusing on partially sighted
   Then blind JAWS users
   Meanwhile considering financial, Web, and AT

                   Aging by Design October 2005   Slide #3
Help Study Overview
   69 volunteers from 4 age groups (20-34, 35-
    49, 50-64, 65+)
   Completed 16 tasks on a prototype financial
    services site
   Between-subjects design
       Participants were assigned to either the help
        condition (interface with help) or control condition
        (interface without help)
   Collected performance and subjective data

                        Aging by Design October 2005   Slide #4
Help Study Prototype

            Aging by Design October 2005   Slide #5
    Sample Tasks
    What 529-type college savings plans does Fidelity
    What would the commission be if you wanted to buy 100
     shares of any stock? Assume that you are an infrequent
     trader (less than 20 trades per year) and the current
     price of the stock is $25.67.
    How much money do you have invested in FIDELITY
     LOW PR STK fund (symbol is FLPSX)?
    What percentage of money (in all your investments
     combined) do you have invested in domestic stocks?
    You need to make sure that Fidelity has your correct
     date of birth. Check to see that it is correctly
     documented as October 5th 1960.

                       Aging by Design October 2005    Slide #6
Session Description
   Signed consent forms
   Background survey
   Web expertise survey
   Financial expertise survey
   Completed 16 common tasks on prototype
   Completed subjective questionnaire

                Aging by Design October 2005   Slide #7
 Summary of Results
Data was analyzed for <65 and 65+:
 Help (audio, visual, or both) improved performance of
  older users while not affecting the performance of
  younger users.
 Using both audio and visual to complete a task

 Help also improved the performance for those with
  lower financial expertise.
 Financial expertise was strongly correlated to
  performance across all age groups.
 All participants rated help as strongly favorable, even if
  they felt that they would not use it themselves.
 Help redirects user to proper path

                      Aging by Design October 2005     Slide #8
Performance by Age

           Aging by Design October 2005   Slide #9
Performance by Financial Expertise

            Aging by Design October 2005   Slide #10
Text Size

            Aging by Design October 2005   Slide #11
    Facts on Low Vision
   Roughly 12-14 million Americans have some degree of
    vision impairment that cannot be corrected by glasses.
    (National Advisory Eye Council, 1998)
   Age-related eye diseases (AREDs) are leading causes of
    vision impairment and blindness in U.S.
       E.g. Cataracts, glaucoma, macular degeneration (AMD)
   Poised to grow exponentially with ―boomers‖.
   Visual impairments often combined with other
    impairments – motor, hearing.
   Can’t get unrestricted driver’s license in most states with
    <20/40 vision

                            Aging by Design October 2005       Slide #12
What’s Low Vision?
   Catch-all phrase meaning diminished vision
       do not use when recruiting!
   Also more specific term meaning 20/70 vision or
    worse in ―good‖ eye without ability to be corrected via
    surgery, corrective lens, etc.
   Context
       20/20 = normal vision
       20/40 vision in at least one eye = pass driving test
       20/200 vision or worse = legally blind
   Visual acuity may not indicate how well someone can
    see online.

                               Aging by Design October 2005    Slide #13
Diversity Among Low Vision Users
   Trouble with different parts of vision
       Central, peripheral, or both
       General acuity loss
       Isolated visual obstructions (e.g. floaters, diabetic retinopathy)
   Those with same impairment experience different
    degrees of severity and rates of degradation
   Differences in Web experience between those
    who learned Web prior to vision loss and those
    learned Web after
   Familiarity with AT

                            Aging by Design October 2005             Slide #14
Low Vision Study Overview
   13 low vision Web users (+ pilot sessions), mixed
    gender and age, various etiologies of vision loss
   Within-subjects design comparing prototype & live
       Users completed up to 16 tasks, 8 on prototype 8 on live
       Task order fixed, site order counterbalanced
   Collected performance, subjective (SUS), and
    qualitative data
   Goal was to learn about support needs of low vision
    Web users

                            Aging by Design October 2005           Slide #15
Low Vision Study Prototype

           Aging by Design October 2005   Slide #16
Session Description
   Minimized form interaction wherever possible
       emailed preliminary forms to users ahead of time
       had large-print or online versions of any form

   Rated feelings of effectiveness, efficiency, &
    comfort at 3x during interaction with each
    version of site
       Prior to working with site
       After first 4 tasks
       After 8 tasks (before switching to next version)

                           Aging by Design October 2005    Slide #17
Reactions to the Help
   Users reacted positively to Help features
       Appreciated help options whether or not they felt help met
        their personal needs.
       ―These features are just good principle.‖
   Several said features would supplant need for
    assistive software
   Conditions selected by 13 participants:
           4 – Visual Help only
           1 – Audio Help only
           6 – both
           2 – wouldn’t need either

                           Aging by Design October 2005     Slide #18
Contrast & Text Size on Prototype
   5 reversed it
   Among 8 who chose default:
        5 chose largest text option available on interface
        3 further increased text size via browser or by lowering
         resolution to 800x600
   Among 5 who chose reverse:
        All chose 3x or 4x font sizes
        One 4x user also used ZoomText

                         Aging by Design October 2005         Slide #19
Live Site Configuration
   All tried increasing font size via browser but couldn’t
   All but one changed resolution to 800x600
   All leaned in closer to monitor and spoke of eye
   Several said they would not use the live site at home
       Cited tiny font, cluttered layout, bright, white background, prevalence
        of hard-to-read, un-adjustable images
       One tried reversing contrast via Windows OS settings but still could
        not use it
       One user flat out could not use it – too bright

                             Aging by Design October 2005              Slide #20
    Performance & SUS Scores
   No improvement of prototype over live site in either.
Version            Average Task Completion Average Task Duration
Live Site          58%                                1:56 minutes
Prototype Site     57%                                2:04 minutes

   SUS scores out of possible 100
       Prototype = 56.5
       Live Site = 51.8

   Several users said they would not use the live site

                           Aging by Design October 2005              Slide #21
Effectiveness Ratings

            Aging by Design October 2005   Slide #22
Efficiency Ratings

            Aging by Design October 2005   Slide #23
Comfort Ratings

           Aging by Design October 2005   Slide #24
    Quick Lessons Learned: Prototype
   Even larger text sizes
       Most LV users are in process of vision loss so will need increasingly larger levels of
   Magnification will compromise one’s sense of page or site
    structure, or context.
   Magnification and high contrast options should function
   ―Calling out‖ or literally highlighting links, wherever they
    reside on the page is helpful.
       Many reported links often difficult to distinguish without standard treatment and
   Users needed bigger mouse icons.
   Users wanted ability to select visual and/or audio help via
   Accessibility also means accounting for users unfamiliar with
    financial terms & concepts; it’s a literacy all its own.

                                    Aging by Design October 2005                       Slide #25
   Design modifications made to support the needs of
    one user group often help everyone. (young)
   Supplemental, interactive audio and visual help
    improved the user experience of older users, those
    with low financial expertise, and low vision users.
   Even when don’t improve performance, improving
    their subject experience important – trust & credibility
    crucial in our industry.

                       Aging by Design October 2005   Slide #26
   As the Web becomes the increasingly dominant medium for
    information delivery and commerce, the need for universal
    design—one that supports all users with highly varied
    requirements—becomes increasingly compelling.
   Not accessibility as an add-on for those with ―disabilities‖.
   Universal design is preferable to placing burden on users
    requiring specialized technology.
   Do away with notion of ―typical user‖.
       Capabilities often context dependent (noisy environment, car
        browser, broken arm, misplaced glasses etc.)
   Test single design implementation with all vs. just with
    specific demographics.

                            Aging by Design October 2005          Slide #27
Future Research
   Future research should examine whether this
    type of help also improves usability or
    performance for those with visual, auditory,
    motor, or cognitive impairments as well as those
    with low-literacy or minimal language proficiency.
   Video of Fidelity rep in upper right giving
    overview of site e.g. for new investors – signed
    for deaf, captioned, etc.
   Multi-layered interface (Shneiderman 2003)

                     Aging by Design October 2005   Slide #28
Contact Information
Ann Chadwick-Dias

Marguerite Bergel

Tom Tullis

                     Aging by Design October 2005   Slide #29

Shared By: