HAFEM Matrix: Accessibility Features for Disabled eLearners: Microsoft Windows vs. Mac OS X
Author: John W. Brown, Joseph B. Hildreth, Dr. Timur Mirzoev
Contact : Clayton State University, Clayton State University. Georgia Southern University
Published: 2014-04-22 - (Updated: 2015-01-28)
Recent studies indicate that the rate of distance learning is on the rise and it changes the options for disabled eLearners.
It can be assumed that having the necessary technology associated with distance learning and its access is a fundamental requirement for success in distance learning programs. As operating systems evolve, so do the features that are associated with them. This study addresses three accessibility features that are shared between the latest editions of the two platforms that hold the highest market shares of users: speech-to-text, text-to-speech and visual cues for sound. A comparative analysis is provided to rate the important features of accessibility software for disabled eLearners. This analysis is intended to enhance the disabled eLearner's (TopTenReviews.com, 2014) ability to make the most informed decision when choosing between Windows 8 and OS X Mavericks based on the three stated accessibility features.
ACCESSiBILITY FEATURES: win 8 VS OS X MAVERICKS
The act of leveraging the functionality of the Internet to create virtual classrooms is an emergent property of post-secondary education that has been labeled with the term e-Learning. (FitzPatrick, 2012). Throughout this manuscript, the terms distance learning and eLearning will be used interchangeably. A study of postsecondary education from 2000 - 2008 determined that 20% of students enrolled in at least one distance learning class. This study also determined that persons with mobile disabilities enrolled in at least one distance learning class at a rate of 24%. Additionally, students enrolled in a distance learning degree program rose from 2% to 4% during the course of the study. (Radford, 2011) Students with disabilities that participate in e-Learning must rely on accessibility features to help interact with computers and the online courses. Computer operating systems offer a number of accessibility features that provide students with disabilities unique options to succeed in e-Learning. Therefore, we can assume that the effectiveness of the accessibility options available to students may potentially equate to the effectiveness of the e-Learning medium. The purpose of this study is to compare the accessibility features of the latest editions of the two platforms that hold the highest market shares of users per the w3schools.com "OS Platform Statistics": 1) Microsoft Windows 8, 2) Mac OS X Mavericks. (w3schools.com, 2014)
The specific accessibility options of each operating system that provide speech-to-text, text-to-speech, and visual feedback will be considered. Each of these three features will be assessed for their respective efficacy in both operating systems. These specific accessibility features are chosen because they address the majority of the needs of persons with physical disabilities. This research will not include the needs or assessment of accessibility features for those with psychological or cognitive disabilities. The features will be rated based on a detailed system defined within by study named the Hildreth Accessibility Feature Experience Matrix (HAFEM). This detailed assessment and rating of accessibility features may offer people with disabilities an in-depth analysis of the rated features, and an easy to understand model that may allow people to make an educated choice on the operating system that may be most suitable. HAFEM model is important to persons with the associated disabilities because it offers an assessment that holds the possibility of allowing the person to increase the effectiveness of their online education.
When faced with a choice of Windows 8 or OS X Mavericks, persons with disabilities must carefully consider the accessibility features available, and the quality those features hold. This study will compare and rate three accessibility features: 1) speech-to-text, 2) text-to-speech, and 3) visual cues for sound, which are common to both Windows 8 and OS X Mavericks. The accessibility features will be given specific values of quality that will be defined by the authors. This study is based on the three feature equivalents offered by the compared operating systems, and will not study the effects these features on disabled eLearners.
The goal of this text is to provide the reader with a definitive comparison between the efficacy of select accessibility features of the two popular operating system platforms Microsoft Windows OS and Apple Mac OS X. The accessibility features that will be compared between the two platforms are: Speech-to-text, Text-to-speech and Visual cues for sounds.
Accessibility is a component contained within the broader domain of "universal usability" that has been described as, "a focus on designing products so that they are usable by the widest range of people, operating in the widest range of situations, as is commercially practical." (Vanderheiden, 2000) Providing access to disabled eLearners, without question, falls within the realm of Vanderheiden's (2000) usability definition, and both Windows 8 and OS X Mavericks have provided the underlying infrastructure to fulfill what has been described as accessibility. (P. Brunet, 2005)
The three features: speech-to-text, text-to-speech and visual cues for sound, which are the focus of this study have developed dramatically in the decade and one half since the ACM Conference on Universal Usability. Although it is very challenging to find research that compares the specified accessibility features of Windows 8 vs OS X Mavericks, a significant amount of data exists on the subjects independently.
Microsoft, the producer of Windows 8, has released an abundance of information on how to implement visual cues for sound, text-to-speech, and speech-to-text.1 In addition to these features, Microsoft Windows 8 Professional allows for the use of assistive technology from third party vendors (Microsoft, 2014) However, a comparison of these "Ease of Access" (Microsoft, 2014) tools against products from another vendor are not available from this source. Additional research shows a variety of resources that briefly describe different aspects of these features, and occasionally give a rating. A review of the accessibility features designed for persons with blindness, which includes speech-to-text, text-to-speech and visual cues for sound, receives a grade of "Not Good Enough Yet", from betanews. (Erbland, 2012)
A resource exploring the narrator feature of Windows 8 simply states it is, "said to be more responsive." (Webster, 2012) There are examples of informative articles, like one titled "Accessibility features for Windows 8" (Weiss, 2012), that give very good information on the accessibility features, but with no rating or comparison. Unfortunately, the majority of resources encountered offer only generic ratings, if any rating is offered. For this reason, it is important to include into one resource not only an explanation of the features: speech-to-text, text-to-speech and visual cues for sound, but a rating that will allow disabled eLearners the ability to compare Windows 8 accessibility features with those of OS X Mavericks.
A similar experience is gained when seeking existing sources that compare Mac OS X accessibility features to those of other vendors. Many of the sources found reviewed the individual efficacy of the Mac OS X features, but did little to compare them to the accessibility options of other vendors, and none to those of the Microsoft Windows Operating System family.
The experts at TopTenReviews.com reviewed nine (9) commercial software packages for text-to-speech software (TopTenReviews.com, 2014). The site compared all nine (9) products in a matrix of features and attributes. The site appeals to the disabled user with information on what features to consider when purchasing such software. Their analysis helps one to better understand the needs for text-to-speech software packages, and to generalize these needs into metrics for comparisons between the text-to-speech software features of Mac OS X Mavericks and those of the Microsoft Windows OS (TopTenReviews.com, 2014).
In the September 2005 Issue of the American Foundation for the Blind (AFB) AFB AccessWorld Magazine, Jay Leventhal (2005) posted a product evaluation of Apple's text-to-speech software; VoiceOver. In the body of the article, Leventhal examines the feature as it first made its debut into the Mac OS X operating system with version 10.4 Tiger on April 30, 2005 (Leventhal, 2005). The article highlights some of the missing attributes of the software that Leventhal perceives to be oversights. One such attribute is the online help manual that is virtually inaccessible to a "beginner or someone who is not familiar with the Apple keyboard" (Leventhal, 2005). Leventhal is also critical of how the application integrates with Safari, the native web browser of the Mac OS X operating system. Leventhal concedes that the initial version of VoiceOver was "disappointed and frustrating" (Leventhal, 2005).
Media Access Australia published a review of the VoiceOver application on their website. This article highlighted several advanced features that the application offers. Among these features are braille display support, multi-touch gesture support, and the ability to save a user's preferences to a USB storage device and later accessed from a different machine. The insights of the extra features highlighted in this article helps the authors of this manuscript to identify some potential metrics to use when comparing the OS X MAVERICKS screen reader software to the Microsoft Windows 8 equivalent (Media Access Australia, 2014).
Christopher Breen of the popular online magazine for Max OS X news, Macworld, wrote up a brief review of the Mac OS X system preference option "Dictation and Speech". This single feature performs the functions of speech-to-text and text-to-speech. Breen supplied insights on the options available to the user for each function. He also added that Apple uses its remote servers to process and store the user's speech data as part of Dictation's normal operation in earlier versions of the operating system (Breen, 2013). However, Lex Friedman (2014) of Macworld wrote a recent article in the printed edition of the Magazine on quick hints for the Mac OS X platform. In one of those tips he exposes that the Dictation application of Mac OS X, in its Mavericks 10.9 version, is capable of performing the speech-to-text translations locally on the user's machine after a pattern recognition data download (Friedman, 2014). This shows a maturity in the Dictation application that respects the privacy of the user more, and frees the user from the requirement of a reliable broadband internet connection in order to use the application.
The goal of this study is to provide the readers with a definitive comparison between the efficacies of select accessibility features of the most resent versions of two popular operating system platforms: Microsoft Windows OS and Apple Mac OS X. The accessibility features that will be compared between the two platforms are: Speech-to-text, Text-to-speech and Visual cues for sounds
The efficacy of each operating system's accessibility options will be determined by objectively applying a weighted-scoring rubric and aggregating the resulting values. The assignment of the weighted points can either be tertiary or binary in nature. Tertiary weights will be assigned as a result of five (5) trials using the accessibility option under controlled conditions that will be explained. Binary values of the rubric will represent answers to polar (yes/no) questions. It is assumed for this study that the average student with disabilities has an intermediate to advanced level of proficiency in their chosen operating system.
Weighted ratings will be assigned a value of 1 -3. These values are based on the following criteria:
- Poor Performance
- Experienced significant lag or choppiness
- Experienced errors making incoherent or non-usable results
- Experienced a system or feature crash
- Feature failed to launch
- Fair Performance
- Experienced some lag or choppiness, but not enough to interfere with results
- Experiences some errors, but not any that make the results incoherent or non-usable
- System and feature run successfully from beginning to end
- Feature launches successfully, but with some difficulty
- Excellent Performance
- Experienced no lag or choppiness
- Experienced no errors
- System and feature run successfully from beginning to end
- Feature launches successfully and with ease
Windows 8 based accessibility features will be tested using a DELL Inspiron model 7520 using Microsoft Windows 8.1 (64-bit). The system hardware is as follows: CPU: Intel®Core i7-3632QM CPU @ 2.20GHz, Memory: 8.00 GB, HDD: Western Digital 1 TB (WDC WD10JPVT-75A1YT0). The Windows Operating System has all current updates as of 1 March 2014.
OS X MAVERICKS based accessibility features will be tested using an Apple MAC Book PRO. The system is as follows: CPU: Intel®Core i7 CPU @ 2.4 GHz, Memory: 8.00 GB, HDD: 1TB. The MAVERICKS OS will be version 10.9.
Text to speech features were tested using five (5) pre-selected paragraphs, which increased in complexity and length. Additionally, this feature was tested reading on screen text with no running applications, with one application open, within the system's default web browser and with Firefox 27.0.1. The intent of these testing procedures was to reproduce the basic requirements of a user to interact with an operating system.
Speech-to-text features were tested using five (5) pre-selected paragraphs, which increased in complexity and length. Additionally, this feature was tested accessing system components with no running applications, with one application open, within the system's default web-browser and with Firefox 27.0.1. The intent of these testing procedures were reproduce the basic requirements of a user to interact with an operating system.
Visual Cues for Audio
Visual Cues for Audio features were tested by enabling the feature in both operating systems and deliberately triggering three (3) distinct system sound events. The responsiveness of the operating system to each event was assigned a tertiary rating based upon the tertiary rating rubric explained earlier in this text. The cumulative ratings for both operating systems were attained by aggregating the scores for each triggered event.
In addition to the performed testing, each application was measured on the ease with which it was implemented, or set-up, within the operating system. This test was measured by the number of steps required to make the feature active. The intent of this test was to help in determining the feature's overall ease-of-use.
Data Collection and Analysis
All data was collected in a matrix designed in Microsoft Excel as shown in Appendix A-1 of this text. Results from each test were recorded in the appropriate block. After testing was complete, all tests with multiple iterations were averaged for a final score. Testing methods that only required one iteration had the one score as final. Next, the final scores for each application, not including the implementation score, were averaged to determine a single score for each application. The final results included an average score for: Windows 8 (speech-to-text, text-to-speech, visual cues for audio) and OS X MAVERICKS (speech-to-text, text-to-speech, visual cues for audio). The final results were analyzed by comparing the average scores of each feature, and presented in a table, along with a written interpretation of the final results.
Microsoft Windows 8: Text-to-Speech
The text-to-speech feature of Microsoft Windows was tested on five (5) different articles. Each test was performed five (5) times and the mean average of the tertiary scores received was totaled for each article. In addition to these tests, the text-to-speech feature was tested on a web page in Internet Explorer 11, the native web browser of Microsoft Windows, and then tested again on the same web page using the Mozilla Firefox v.27 web browser. The article was retrieved in the two web browsers five (5) times each. The total of the tertiary scores received from each trial was summarized in a mean average. The responsiveness of the feature's being initiated was graded as well as one overall tertiary score. The overall score for the usage of this feature in Microsoft Windows received an overall score of 2.6 on the HAFEM. The feature ran smoothly throughout the entire process. It did have difficulty pronouncing some materials in articles 3 and 4. The roll-over activation with mouse cursor feature was a little too much.
Apple Mac OS X Mavericks 10.9: Text-to-Speech
The text-to-speech feature of Apple Mac OS X Mavericks was tested on the same five (5) different articles that we used to test Microsoft Windows. Each test was performed five (5) times and the mean average of the tertiary scores received was totaled for each article. In addition to these tests, the text-to-speech feature was tested on a web page in Safari, the native web browser of Mac OSX, and then tested again on the same web page using the Mozilla Firefox v.27 web browser. The article was retrieved in the two web browsers five (5) times each. The total of the tertiary scores received from each trial was summarized in a mean average. The responsiveness of the feature's being initiated was graded as well as one overall tertiary score. The overall score for the usage of this feature in OS X MAVERICKS received an overall score of 2.7 on the HAFEM. This feature was not too difficult to get up and running and ran flawlessly with OS components. However, the feature was unable to read content inside of MS Word for Mac files, but was able to read .pdf files perfectly. The feature ran smoothly through the entire process. However, it did require a lot of key strokes to run efficiently, possibly negating the entire reason for rating the feature.
Figure 1- Text-to-Speech Results
Figure 1 - Text-to-Speech Results
Microsoft Windows 8: Visual Cues for Audio
The feature that produces visual cues in lieu of audio on Microsoft Windows 8 was tested using three (3) triggers for sound notifications. The three triggers that were used were the critical error, the new hardware connection notification, and the generic system notification sounds. Each trigger was tested five (5) times and assigned tertiary values of the Hildreth Accessibility Feature Experience Matrix. In addition to the testing of these triggers, the feature was given a single value for its responsiveness to being invoked. The Microsoft Windows 8 operating system received a 2.9 as an overall score. This feature was a difficult to find. Searching the Windows Help Files was required to activate it. Once it was up and running the feature produced flawless results.
Figure 2- Visual Cues for Audio Results
Figure 2 - Visual Cues for Audio Results
Apple Mac OSX Mavericks 10.9: Visual Cues of Audio
The feature that produces visual cues in lieu of audio on OS X MAVERICKS was tested using the same three (3) triggers for sound notifications as used for Microsoft Windows. Each trigger was tested five (5) times and assigned tertiary values of the HAFEM. In addition to the testing of these triggers, the feature was given a single value for its responsiveness to being invoked. The OS X MAVERICKS operating system received a 2.6 as an overall score on the HAFEM scale. This feature was fairly easy to set up. However, there was no way to specify which OS actions would trigger a response. Therefore, the add hardware trigger never produced a result.
Microsoft Windows 8: Speech-to-Text
The accessibility feature of Microsoft Windows 8 that allows the user to have their speech translated into text was tested by voicing five (5) different paragraphs of text. Each paragraph was spoken into the flagship word processor of Microsoft, Microsoft Word, five (5) times. The tertiary value for each trial was recorded in the HAFEM and summarized with a mean average. A single value was recorded for the feature's response to being initiated. The speech-to-text Microsoft Windows 8 accessibility feature scored a 2.4 on the HAFEM scale. This feature was easy to set up. A 6 minute video allowed the ability to both navigate and dictate nearly flawlessly into MS Word. The web browsers proved a little difficult to run by voice command. All OS features were very easy to access.
Figure 3- Speech-to-Text Results
Figure 3 - Speech-to-Text Results
Apple Mac OSX Mavericks 10.9: Speech-to-Text
The accessibility feature of OS X MAVERICKS that allows the user to have their speech translated into text was also tested by voicing five (5) different paragraphs of text. The performance of the OS X MAVERICKS feature was very modest. With each paragraph dictated to the feature five (5) times, each series of trials were averaged to bring the operating system to a total score of 1.7 on the Hildreth Accessibility Feature Experience Matrix. This feature was very difficult to set up. The training included with the feature took over 30 minutes, and did not produce user proficiency. Accessing OS features were somewhat successful, and the performance of the hardware was great. However, dictation results produced no perfect paragraphs, and a majority were completely incoherent.
By designing the HAFEM and recording our test results, we feel that we have given considerable evidence for a fair comparison of the speech-to-text, text-to-speech and visual cues for sound features of MS Windows 8 and OS X MAVERICKS. This comparison shows that MS Windows 8 has the higher average of HAFEM scores, and we feel that this system offers the most user-friendly experience. However, it is important to note that the OS X MAVERICKS HAFEM average of scores was very close in comparison. We feel that additional face to face training with OS X MAVERICKS has the possibility of increasing the usability of its features.
Both platforms scored very close to one another with their respective text-to-speech options. Microsoft Windows 8 out-performed OS X MAVERICKS in both remaining categories. It was observed throughout the testing that having multiple application windows running in the background had no substantial bearing on the performance of the tested features. OS X MAVERICKS had limited capabilities with Video Cues for Audio, but performed well when the feature was triggered. Additionally, person to person training or instruction would probably be needed to become proficient with the Speech-to-Text feature in OS X MAVERICKS.
Possibilities for future research include: Testing in multiple languages. The testing did not account for how each platform handled various languages. Testing of third party accessibility hardware platforms. A comparison of the remaining accessibility features included in both operating systems.
Breen, C. (2013, March). Using Mountain Lion's Dictation And Text-To-Speech Features. Retrieved February 20, 2014, from www.macworld.com/article/2030774/using-mountain-lions-dictation-and-text-to-speech-features.html
Erbland, M. (2012). How good are Windows 8 accessibility features for the blind Retrieved February 16, 2014, from www.betanews.com: betanews.com/2012/03/02/how-good-are-windows-8-accessibility-features-for-the-blind/
FitzPatrick, T. (2012). Key Success Factors of eLearning in Education: A Professional Development Model to Evaluate and Support eLearning. US-China Education Review, A (9), 789-795. Retrieved February 6, 2014, from www.eric.ed.gov/contentdelivery/servlet/ERICServletaccno=ED537174
Friedman, L. (2014, February). Mac OS X Hints. Macworld, 31 (2), 86-87. Retrieved from Macworld.com.
Harvey Hyman, P. (2014). Systems Acquisition, Integration and Implementation for Engineers and IT Professionals.
Jones, S. (n.d.). Internship Section Reports.pdf. Retrieved from https://go.view.usg.edu: https://go.view.usg.edu/d2l/le/content/662276/viewContent/10023687/View
Kaplan, N. C. (2013). Investment Arbitration's Influence on Practice and Procedure in Commercial Arbitration. ARBITRATION & ADR IN ASIA (pp. 122-125). Asian Dispute Review.
Leventhal, J. (2005, September). Not What The Doctor Ordered: A Review Of Apple's Voiceover Screen Reader. AFB AccessWorld Magazine, 6 (5). American Foundation for the Blind.
Media Access Australia. (2014). Access Review of Voiceover. Retrieved February 14, 2014, from Media Access Australia : www.mediaaccess.org.au/digital-technology/assistive-tech/voiceover
Microsoft. (2012, September 10). Accessibility in Windows 8 Fact-sheet. Retrieved February 15, 2014, from www.microsoft.com: www.microsoft.com/enable/products/windows8/default.aspx
Microsoft. (2014). Accessibility in Windows 8 . Retrieved February 15, 2014, from www.microsoft.com: www.microsoft.com/enable/products/windows8/default.aspx
Microsoft. (2014). Making your PC easier to use . Retrieved February 15, 2014, from windows.microsoft.com: windows.microsoft.com/en-gb/windows-8/make-pc-easier-use
Mirzoev, T. (2014, January 12). Course Information - IT Strategy, Design and Devel Section A Spring 2014 CO . Retrieved from https://go.view.usg.edu: https://go.view.usg.edu/d2l/le/content/653148/viewContent/10024075/Viewou=653148
P. Brunet, B. A. (2005). Accessibility requirements for systems design to accommodate users with vision impairments. IBM Systems Journal, 44 (3), 445.
Radford, A. W. (2011, October). Learning at a Distance: Undergraduate Enrollment in Distance Education Courses and Degree Programs. STATS IN BRIEF, NCES 2012-154 , 3. U.S. Department of Education. Retrieved February 6, 2014, from nces.ed.gov/pubs2012/2012154.pdf
TopTenReviews.com. (2014). Screen Reader Software Review . Retrieved February 14, 2014, from TopTenReviews.com: screen-reader-software-review.toptenreviews.com/
Vanderheiden, G. (2000). Fundamental Principles and Priority Setting for Universal Usability. Proceedings of the ACM 2000 Conference on Universal Usability (pp. 32-37). Arlington, VA: ACM Press. Retrieved February 20, 2014
w3schools.com. (2014, February 6). OS Platform Statistics and Trends . (w3schools.com) Retrieved from w3schools.com: www.w3schools.com/browsers/browsers_os.asp
Webster, A. (2012, February 14). New Windows 8 accessibility features detailed. Retrieved February 15, 2014, from www.theverge.com: www.theverge.com/2012/2/14/2798197/windows-8-accessibility-features
Weiss, T. C. (2012, 12 02). Accessibility Features in Windows 8 . Retrieved February 15, 2014, from www.disabled-world.com: /assistivedevices/computer/windows-8.php
APPENDIX A-1: Testing Materials
Test Paragraph 1
This is a very simple paragraph. Words in this paragraph are not difficult to read or say. The intent of this paragraph is to be a very simple test. Joseph B Hildreth
Test Paragraph 2
These criteria reflect what management thinks is relevant in his/her decision. These might include criteria such as price, product model, and efficiency of the existing system, current hardware, software, applications, technology, client/end T user knowledge, and equipment. (Jones)
Test Paragraph 3
Paragraphs are the building blocks of papers. Many students define paragraphs in terms of length: a paragraph is a group of at least five sentences, a paragraph is half a page long, etc. In reality, though, the unity and coherence of ideas among sentences is what constitutes a paragraph. A paragraph is defined as "a group of sentences or a single sentence that forms a unit" (by BRENT) (Mirzoev, 2014)
Test Paragraph 4
The growth of investment arbitration over the last two decades has been monumental on a number of levels. Under public international law, individuals traditionally lacked standing to bring international claims against foreign States. Through investment treaties, private investors may now not only have recourse to arbitration directly against States, but are also protected by substantive rights expressed in such treaties. Moreover, should the ICSID Convention be chosen to resolve an investment treaty dispute, this dispute resolution mechanism provides one of the most de-localized international arbitration systems ever to be implemented. (Kaplan, 2013)
Test Paragraph 5
A significant activity of the SSR is the functional overview. A functional overview of the system will give a brief description of the system requirements in terms of performance requirements, interface requirements with other systems and internal requirements of the System. The supporting documentation of SSR must clearly and specifically define "what is happening here" so that we can state in unambiguous terms, exactly what the constraints and parameters are for the specifications.
Performance Requirements include: Number of cores, throughput, response time, execution time, storage, memory allocation, load, redundancy, fail-over, and OS version.
Interface Requirements with other systems include: delivery method, communication links, speed, portability and compatibility.
Internal Requirements of the System include: data types handled, input, processes, hand-off between processes, storage, and output. There should be traceability from start state to end state. (Harvey Hyman, 2014)
1 Some examples of Microsoft materials documenting the use of accessibility features include: Making your PC easier to use (Microsoft, 2014) and Accessibility in Windows 8 Fact-sheet (Microsoft, 2012).
- 1 - Change TV Channel Using Your Eyes with Xfinity X1 Eye Control : Comcast (2019/06/19)
- 2 - List of Computer Screen Readers for Visually Impaired : Disabled World (2014/08/06)
- 3 - Brain Waves Write Directly to Computer : Mayo Clinic (2009/12/07)
- 4 - Connecting iPad iPhone or iPod to TV or Projector : Disabled World (2012/02/02)
- 5 - Essential Accessibility: Tools for Website Accessibility : Wendy Taormina-Weiss (2012/06/17)
- 6 - List of Hotkeys Keyboard Shortcuts for Computers : Disabled World (2009/02/17)
- 7 - Using Sticky Keys for Accessibility On PC and Mac Computers : Disabled World (2009/02/17)