Effect of Text/Background Color on Reading Comprehension via LCD Display.



 Jon A Kragness

May 2007



Table of Contents


List of Tables. 3

Abstract 4

Acknowledgments. 5

Chapter 1: Introduction. 6

Research Topic: 6

The problem.. 6

Review of Related Literature. 6

Video display terminals. 6

Information consumers. 7

Working memory load. 9

Perception. 10

Color and contrast. 11

Design considerations. 12

Statement of Hypothesis. 13

Operational definition of variables. 13

Chapter 2: Method. 15

Research participants. 15

Instruments. 15

Experimental Design. 16

Procedures. 16

Chapter 3: Results. 17

Discussion. 17

Limitations. 18

Anecdotal Accounts by Participants. 18

Need for Further Research. 19

Summary. 19

References. 20


List of Tables


Table 1: Current institutional student demographics. 15

Table 2: Scores by test order taken and background condition. 17





This study examined what effect, if any, using a yellow colored background with black text for complex reading material may have on reading comprehension compared to a white background with black text.  The yellow background with black text combination was chosen because it is frequently used as a default combination to accommodate low-vision users. There is, however, very little research to support the efficacy of this combination when using a light emitting display such as a computer monitor.  To compare results, 30 college students completed the reading comprehension section of the AccuPlacer test.  Half completed this using a test-retest on a computer displaying a white background with black text first followed by completing this test utilizing a yellow background with black text.  The other half of subjects received the same directions reversed.  Results using a dependent t-test failed to identify a significant difference between conditions.


I would like to thank my adviser and committee chair, Harry Weisenberger, and committee members Dorothy Suomala and Richard Warner for their guidance and support in this process.  I truly value the insights gained by a thoughtful review and subsequent dialogue.


Also, I would like to say a special thanks to John Wasson for offering research guidance and review throughout this process.  It's been a tremendous help to have a source of support that has lasted from project inception through its completion.


Thank you to the MSUM Alumni Foundation for providing the Educational Leadership award to provide financial support necessary to defray the cost of research expenses.  It was because of this award that I was able to secure highly reliable test materials from AccuPlacer.


I would also like to express my sincere appreciation to Minnesota State Community and Technical College - Fergus Falls campus, and particularly to the Interim Academic Dean, Ann Nelson for allowing this study to take place on the MSCTC-FF campus.  Further, thanks to David Seyfried for the patients, persistence, and unending encouragement that have been a vital part of this process.  Lastly, thank you to all of the students that so enthusiastically took part in this study.

Chapter 1: Introduction

Research Topic:

Effect of utilizing high contrast text/background on reading comprehension when information is presented on a computer monitor.


Descriptors for ERIC search: color, cognition, comprehension, VDT, Web, Internet

The problem

The effect of background/text color combinations on the comprehension of college students when reading from a computer monitor.

Review of Related Literature

With the proliferation of personal computers, the Internet, and other personal electronic devices, there is a growing array of options with which information can be shared.  There is no longer a need to rely exclusively on publishers or other forms of printed materials.  Still, there are two central concerns to be addressed when utilizing educational materials and those are content and design.  And of course, the primary goal of any design is to support the content.


Unlike print media and other traditional forms of learning materials, electronic presentation of information can easily take on many forms.  This is because it is created by both experienced and inexperienced designers and it can be viewed on a multitude of devices by the end user.  These variables often confound one another and lead to less than optimal presentation of material.  This is a relatively new phenomena fueled largely by improved technologies with greater display capabilities and the advent of the Internet where virtually anyone can create and display information electronically (Hall & Hanna, 2004).  While this creates an unprecedented level of access for persons to present information, ideas, and commerce it is frequently wrought with design elements lacking support for content.  With careful planning and attention to design during the process of creating learning material intended to be presented electronically, one can avoid the cost of redesigning ill planned material that can require 10 times or more the investment to repair issues caused by lack of planning (Proctor, Vu, Salvendy, Degan, Fang, & Flach, et al., 2002).


Video display terminals.

Looking beyond the issues associated with the design of electronic materials there may be other concerns that become evident when considering the right of modern viewing devices as well.  For example it has long held through research that reading speed suffers if one is reading from a video display rather than printed text, however, differences in reading speed and comprehension between computer monitors and paper-based learning materials may be lessening as the technology behind computer monitors improves (Garland & Noyes, 2004).  As video display technology (VDT) shifts from cathode ray tube (CRT) to liquid crystal display (LCD) both user preference and user performance continues to improve, however, there is evidence that this may be more closely related to monitor size and resolution thus warranting further investigation (Wang, Fang, & Chen, 2003).


In addition, with an ever-increasing variety of video display devices, the viewing of electronic information is no longer confined to a computer monitor.  End users may view this information on a multitude of devices including desktop computers, laptop computers, electronic books, or tablet PCs, among others, and the list continues to grow as technology evolves.  Many of these technologies allow for display devices that are getting more portable than ever before as devices become increasingly smaller and lighter allowing them to be used in environments that were previously inaccessible to electronic information display.


This is in sharp contrast to the long-held notion of a fixed workstation where one can prescribe environmental controls via lighting and ergonomic considerations to optimize health and safety concerns while viewing a VDT.  To address the effects of monitor viewing angle, Turville, Psihogios, Uimer, and Mirka (1998) used objective and subjective data to assess subjects viewing a computer monitor at fixed station 15° below eye level then compared it to subjects viewing a computer monitor at a fixed station 40° below eye level and found no significant difference.  However, it's important to note that Turville et al. did not identify the specific display technology used for this experiment other than that it was a 13” monitor and they used a black 12 point Arial font on a white background and inferred the monitor was a CRT display technology.  Given the limited availability of display technology, and more specifically the display technology apparently used in this study, it would be valuable to replicate this study because of the ever-increasing quality of video displays (Garland & Noyes, 2004) and the trend to shift to other technologies such as LCD, as well as greater monitor size and resolution (Wang et al., 2003).


Information consumers.

When preparing educational materials to be consumed via electronic media, understanding the varying needs and abilities of persons of differing age groups will help guide the development of electronic media appropriate to the intended audience.  While many young children today have never known a world without computers or the Internet, young adults may have varying levels of computer skills to draw upon, and older adults may have little experience and require training or reject this new technology altogether.  Many of these issues may be averted by soliciting ideas from nonexpert end-users providing a superficial level of guidance as a means of solving problems.  This can then be followed up by expert designers skilled at utilizing suggestions in a deeper more forward-thinking manner (Proctor et al., 2002).


This becomes increasingly important as the occurrence of electronically presented information continues to proliferate, and its application can be seen in ways that extend far beyond that of reading material.  As a result, the occurrence of persons utilizing electronic information is becoming ever more commonplace as a primary communication tool.  With the availability of the Internet within reach for most people, electronic information may supplant many of the information and correspondence sources long held by print media.  This becomes readily apparent when considering the frequency with which persons utilizes e-mail and instant messaging for correspondence in addition to gathering information or conducting commerce via the Internet.


As a part of these changes, education has seen a great transition in the way with which electronic information is used.  Many of the informational uses that have long been held by paper and pencil means have gone the way of computer display.  Poynton (2005) compared the results of standardized tests taken by young adults utilizing the traditional paper and pencil method to the results of a similar group taking a computerized version of the same test.  Although both versions of the test have a high level of reliability, computer savvy test takers scored significantly higher on a computerized version than did non computer savvy test-takers.  Another study supported this finding in part when conducting a meta-analytic study of computerized versus paper and pencil tests, and found that there was no medium difference on power tests but there is a significant degradation among scores on computerized speed tests (Mead & Drascgow 2003).  Further, depending on presentation design there is a risk that test takers may fail to perceive subtle changes when navigating between screens or coordinating hand and eye movements to select an answer (Droll, Hayhoe, Triesch, & Sullivan, 2005).  Despite the differences between test takers in computer ability level, when considering gender in terms of preferred navigational methods, button placement, and text/background combinations there are no differences among graduate students (Hsu, 2006).


Not to be overlooked, older adults are becoming an increasingly powerful demographic and may foster significant influence over electronic media both as an independent consumer and as nontraditional students.  As they age, there is an ever greater loss of sensory and motor skills that can make it more difficult to access information readily available to others.  Although they may not comprise a majority among electronic information consumers, at the same time, as a group they have amassed the financial resources to garner a great deal of attention from electronic content providers (Charness & Holley, 2004).


When considering a design for presenting electronic information there are still some age-related differences worth noting.  Phillips and Lesperance (2003) found that although working memory limitations in older adults did not differ significantly from that of younger adults when presented with distracters placed within text, semantic learning in older adults suffered far more than in young adults.  These findings suggest that a straightforward design devoid of unnecessary artifacts will be more useful to adults across the age span.  A well-thought-out presentation that implements a policy of universal design will provide benefits not only to the older population but also to young children in addition to persons with disabilities without sacrificing utility or preference by others.  Therefore, innovations driven by this population may have a highly significant carryover effect in terms of making electronic information more accessible and useful for everyone. 


Working memory load.

After considering the consumer audience and the display devices that may be used, there is the creation of the electronic information itself.  With the dynamic properties of rapidly changing electronic information, by its very nature it brings with it a myriad of issues not typically seen in static printed documents.  This is especially true of information located on the Internet where the control over design often resides at a remote location.  While new technologies continue to drive the addition of more multimedia effects as they become available, designers will often include attention grabbing items that can range from scrolling text to trademarked or themed colors to animations or Macromedia flash files, all of which may distract one's attention from the intended content.  Hong, Thong, and Tam (2004) also cautioned that although some multimedia effects may be useful tools for interacting and demonstrating, they will also often disrupt the critical thinking process and instead encourage learners to become passive viewers.


Among the many problematic design features that plague electronic information via the World Wide Web, one that has become the bane of many Web users is the flashing animation.  These have become a popular way to attract users’ attention when visiting a web page and it has been a technique used particularly by online advertisers as a way to increase click-through rates on banner ads.  Despite the effectiveness of these flashing banner ads in garnering attention, they carry with them a number of significant problematic considerations.


Perhaps most importantly, when a flashing item is placed on a screen it continues to occupy a portion of the limited amount of attention one has available to focus on a given task.  As a result, there is less active attention available to devote to target information, and therefore comprehension of material presented will suffer (Hong et al., 2004).  As a result, information is not as effectively processed where it can be encoded and stored in long-term memory leading to poorer recall.  This issue may be further exacerbated by the disdain for flashing banner ads that was found among persons acquiring information from the Internet, especially among experienced Internet users although not as much so for new Internet users (Hong et al.). 


In spite of this, among the many changes being brought about by electronic presentation of information is the ability to create multidimensional interactive images used to represent abstract data.  This has become increasingly popular as the technology to support it has been developed and refined.  Intuitively, a more robust learning model can be employed, however, because of an overloading of cognitive demands when using three dimensional models there may be a decreased ability to comprehend material, particularly abstract material that has no physical counterpart, compared to using two dimensional images with supporting text (Keller, Gerjets, Scheiter, & Garsofky, 2006).  This may be due in part to the supportive nature of shared cognitive resources where the verbal and visual can lend support to learning without generating cognitive overload (Morey & Cowan, 2005).



Because of the ease with which one can customize the way electronic information is displayed, significant document display changes can be made by simply altering its background color, and there is a great deal of agreement that a high contrast text/background combination is recommended to achieve the greatest degree of readability.  While this rather obvious sounding statement would seem to suggest that simply utilizing maximum contrasts will yield maximum readability, this is not always the case.  If contrast level alone were sufficient to dictate a sound text/background combination, one may simply defer to any color and it’s negative.  However, there is evidence that high contrast alone may not determine ideal readability.  When subjects were asked to read black text on either a gray or white background they performed better when reading black text on gray indicating that so long as contrast levels are sufficient that contrast alone may not be the sole determinant regarding readability (Hall & Hanna, 2004).  Given these results, it's unfortunate that Web browsers have abandoned the gray background that once was a default and have now opted for white in its place.


Despite the general agreement on minimum thresholds, there still does not appear to be a clear consensus on what constitutes a best combination for all users.  Ideally, a user selectable color scheme can be incorporated to allow the end-user the opportunity to self select a text/background color or for a page to utilize stylesheets that allow full customization on the user end.  This may help further relieve problems associated with persons facing light based reading issues (Stone, 2004) as well as accommodating low vision users.


From a practical standpoint, much of the electronic information available today does not use highly customizable text/background combinations.  This may be for a variety of reasons or it may simply not be seen as an important design element when publishing electronic information.  One color combination of particular interest is the light yellow background with black text that is frequently used as the default color scheme in an effort to create the greatest level of visibility for low vision users, yet finding any substantial research to support this proved elusive.  The only formal documentation to support this combination for low vision users was in the Federal Telecommunications Act Accessibility Guidelines (http://www.access-board.gov/telecomm/rule.htm) where there is a rather vague suggestion that one use a cream-colored or yellow non-glossy background with black text.  It's important to note, however, this guideline is a carryover from paper documents and does not suggest that it's a viable application for electronic information.


Further, there is evidence that using a alternative color combination may yield better results when measuring for reading speed and comprehension.  Using transparent overlays, Chase, Ashourzadeh, Kelly, Monfette, and Kinsey, (2003) found that subjects had significantly improved reading speed and comprehension when reading text through a purple overlay compared to reading text through a yellow overlay.  This finding again disagrees with the notion of light yellow providing the ideal level of readability and also disagrees with the findings that older adults have more difficulty with fine details and text that utilize violet or blue colors (Charness & Holley, 2004).


Extending beyond readability alone, Zviran, Te’eni, and Gross, (2006) discovered that by using different background colors that they could elicit more receptive attitudes toward given information.  This may be an especially valuable design tool for presenting information that may be of a controversial nature or otherwise not well received.  When choosing text/background colors intent upon universal design for the purpose of creating the most readable format, there may be some conflict between what is most readable and what is most preferred.  While black text with light yellow background has been seen as very readable, it also ranked lowest in preferred background color being viewed as well as being weak.  This may render the black text with yellow background easily readable hence encouraging a greater degree of recall, but it may fall short in persuasiveness. Some of this, however, may be offset by the level of color saturation (Hall & Hanna, 2004).


These perceptions become increasingly important as information on the Internet becomes more laden with graphics and other content, download times can increase the dramatically, especially for dial-up users.  Although broadband connections are becoming ever more common, this alone may not be sufficient to address the users perception on information gathered electronically.  Gorn, Chattopadhyay, Sengupta, and Tripathy (2004) discovered that background color can have an effect on the perceived time it takes to download information and subsequently influences feelings of anxiety or calm.  Feelings of anxiety decreased the ability to focus on necessary information resulting in poorer recall, and Poynton (2005) cautions that an excess of anxiety may even lead to an aversion of computer use and result in what he calls “information poverty”.


Color and contrast.

To attain maximum readability, a high level of contrast between text and background needs to be maintained while remaining cognizant of the effects of various combinations.  This can be either a positive, dark text on light background or negative, light text on dark background.  The World Wide Web Consortium (W3C) in their Web Accessibility Initiative (WAI) advocates careful use of color combinations to maintain readability and further states that contrast needs to be sufficient when viewed on either a black-and-white or color viewing device (http://www.w3.org/WAI/wcag-curric/chk3-0.htm).  Despite the prevalence of color viewing devices, end users may choose to view electronic material on monochromatic devices that may include electronic books or specialty devices used by persons with disabilities.  Further, developing material that can be easily displayed on black-and-white monitors will generally translate to material that is readily printable on either black-and-white or color printers.


In addition to the immediate text/background area, other factors surrounding the target information may also influence the readability and perception of electronic information.  When viewing information on a VDT that is surrounded by a framing color the apparent contrast of the electronic information will be altered by that frame and further affected by the brightness level of the display device (Carter, Sibert, Templeman, & Ballas, 1999).  Because this information is often nested within navigation buttons, border or frame colors, or additional information the level of readability may vary among users.  While this may not affect the majority of persons for whom this information has been prepared, it may cause significant difficulty for low vision users.


Although there may not be a single specific text/background color combination that is ideal for all end-users, there is support for utilizing colors with short wavelengths when present electronic material for reading.  Chase et al. (2003) found that the color red significantly impedes reading for normal readers.  However, it should be noted that there may be exceptions where the color red may actually enhance reading for persons with certain disabilities.  Chase et al. suggests that persons with dyslexia may be among those to benefit but advocates further research to provide more conclusive evidence.


Additional support for the use of short wavelength colors was found when asking subjects to rate electronically displayed information on readability and aesthetics and a clear preference for high contrast blues resulted.  While these results may apply to most persons, they should perhaps be viewed with some caution because remember that Charness and Holley (2004) found upon further investigation that older adults have greater difficulty resolving fine details on text that involves the use of violets or blue colors because of visual changes associated with aging.  Interestingly, however, subjects still rated a black text on white background highest in preference for educational material.  This may be a result of the level of contrast affording a high degree of readability or from the familiarity most people have for black text on a white background, but in any event black text on a white background was rated as looking most professional and hence most preferred for educational material (Hall & Hanna, 2004).  This may further be a result of state-dependent-learning where one will most easily recall something when conditions are similar to those when it was learned, and in the case of educational material, due to the commonality of black text on a white background.


Design considerations.

When developing information intended to be displayed electronically, an author or designer has a multitude of elements available with which to choose for supporting the content.  The choice of these elements can carry a significant impact on the usability of electronic information for consumers.  Even with the best of intentions, there may be times when this element is overlooked or improperly used that may carry with it unforeseen effects.


In the interest of creating document with universal utility, authors will often use background color or text/background color combinations to support content presented because of the ease with which color can be added or manipulated when displayed electronically.  This leads to a question of what will guide the choice when making a decision on text/background combinations or will this question be bypassed entirely and be left to the default of a viewing device.  An ideal design will allow end-users to select the text/background color of greatest preference or utility to them.  This preference may be driven by age, ability, or viewing device and the capabilities that lie therein.


Information that is intended to be presented and hence useful to the widest audience possible currently is often displayed using a light yellow background with black text.  It has long held that this combination yields the greatest readability to the widest audience and is frequently used on web sites that wish to be accessible to all visitors, although research to support this in the electronic medium is scarce.  In fact, recall that there was some advantage to using a gray background with black text for presenting reading material for many users.  In light of this, is there a significant reason not to use a yellow background with black text?  Will reading comprehension suffer for normal vision users compared to using black text with a white or gray background?  If reading comprehension suffers significantly, then perhaps this design element is not nearly as universal as once believed.


There has been a great deal of discussion in literature above regarding best practice for the display of electronic information via a VDT.  By recognizing the limitations and capabilities of information displayed in electronic form and acknowledging that there are a number of differences between information presented electronically and that of print, one may utilize design principles that will not impede the use by the intended audience.


Both color and contrast are very important for deciding upon text/background combinations and can be easily added yielding significant effect to the support of or to the detriment of the display of electronic material.  Also among the many reasons to use color judiciously when designing information for electronic display is the cultural relevance it may hold (Proctor et al., 2002; Zviran, 2006).  This is true of electronic information even more so than for print material because of the ease with which one can view it globally once it has been placed in a public venue on the Internet.


Careful attention to design and display electronic information also becomes very important because of the multitude of devices with which electronic information can be viewed upon, and materials may not appear as the author had intended.  This risk may be minimized by utilizing universal design principles to ensure equal access for all users and by testing designs in a number of different environments including hardware, software, and printed.  These considerations may disagree with the use of light yellow background and black text in a quest to make a highly readable universal design and instead necessitate further investigation to identify the ramifications of this combination.


Statement of Hypothesis

There'll be no difference in reading comprehension scores between college students reading information on a video display screen utilizing yellow background with black text as compared to white background with black text.


Operational definition of variables

  • Participants received a set of questions from the AccuPlacer test bank presented to them utilizing either of two possible views in a test-retest design.  The difference between these views was limited to text/background color combinations on a computer monitor where each participant received either a screen view that contained a yellow background with black text or white background with black text.


  • Reading comprehension scores acquired from completion of the AccuPlacer test questions were collected and analyzed to identify any significant advantage to a particular text/background color combination.

Chapter 2: Method

Research participants

The participants for this study were selected from the population of students attending a Community College in west central Minnesota. The majority of students come from the surrounding area.  The most current institutional data on student ethnicity reports a largely white population as seen in Table 1.  There are many more female than male students with approximately 62% female students and 38% male students and 75% of all students are under the age of 25.


Table 1


Current institutional student demographics


(N = 938)

American Indian















Spring semester a total of 31 students, 25 female and 6 male, participated in this study and were randomly divided into two groups.  These students were then asked to answer questions to assess reading comprehension using different text/background screen combinations.  All of the participants indicated that English is their first language and none of the participants indicated having a visual impairment on self reporting background questions.



Modern desktop computers utilizing a 15” LCD monitor with a resolution of 1024 x 768 were used to display electronic information in the form of reading questions to be answered in two different text/background color combinations.  These display devices were selected because of the movement in electronic viewing technologies that is increasingly gravitating toward LCD as a preferred display.


To measure reading comprehension, the reading comprehension section of the AccuPlacer college placement exam was used to generate a score under each of two conditions for each subject.  This test has been utilized as a reliable source of student placement for some time at the selected college, and therefore any significant data anomalies could have been readily identifiable had they occurred.  Further, this is an Internet based adaptive test that draws questions from a large question bank thus the number of repeat questions was greatly minimized, however it was not entirely eliminated.  Subjects were asked to complete the reading comprehension section under two different conditions and the group scores between conditions were analyzed.


All tests administered used an Arial (Sans Serif) black (000000) text with either a white (FFFFFF) background or a yellow (FFFF00) background.  The background color changes were made by using the AccuPlacer "Access Wizard" where subjects utilizing the yellow background condition were instructed to open the Access Wizard and select yellow for the: Response Background Color, Question Background Color, and the Passage Background Color.


It's important to note that the yellow background encompassed only material that subjects needed to read, thereby creating a block area on screen, leaving the remaining vacant screen area white.  The black on yellow was chosen because of its reputation for providing preferred contrast for persons with low vision and the black on white background was chosen because of the volume of information using that combination thus its familiarity to students.  The AccuPlacer test was used to assess reading comprehension under both of these conditions.


Experimental Design

The design of this study compared the scores on reading comprehension tests using two different display conditions.  Each participant was required to take the brief reading comprehension test under both conditions using a test-retest design.  The results were analyzed using independent t-test to identify any significant advantage for either condition.



Once selected, participants were randomly divided into two groups whereby one group was instructed to complete the AccuPlacer reading comprehension test using black text on white background followed by again completing the AccuPlacer reading comprehension test with black text on yellow background using the Access Wizard described earlier.  The other group received the same instructions in reverse order.  Although this is a test-retest design, subjects had a low probability of receiving an identical test because of the adaptive nature and large bank of test questions.  These questions were presented on a 15” LCD computer monitor with each condition differing only in background/text color.  Scores for both groups were collected, analyzed, and group scores were compared for reading comprehension level.

Chapter 3: Results


When comparing the results of the white background test condition (M = 81.73, SD = 22.78) to the yellow background test condition (M = 78.91, SD = 23.59), data gathered in this study failed to reject the null hypothesis using a dependent t-test t(29) = 1.49, p = 0.15 (two-tailed) indicating there was no significant difference found in test scores when comparing the yellow to white background condition.


Of the 31 participants that took part in this study, one participant was unable to complete the second test leaving no data for comparative measure and was therefore omitted from the final results yielding a total of 30 participants.  In all, participants completed a total of 60 reading comprehension tests under test-retest conditions.  Upon completing a reading comprehension assessment, scores for both groups were collected. 


By further breaking this data down, it becomes apparent that the group scores earned by participants utilizing the white background condition first followed by utilizing the yellow background second were higher under both conditions as seen in Table 2.  However, because this study was concerned with differences between background conditions, the test-retest design resolved differences seen between groups.


Table 2

Scores by test order taken and background condition

                                                                  Test Order                                                

Background                            First                                         Second                       

                                    M                     SD                   M                     SD

White                          92.40               12.80               67.78               25.77

Yellow                        65.51               24.39               89.54               16.79


Data was also compared to identify differences in time required to complete these tests under each condition.  The time needed to complete these tests was nearly identical under both background color conditions.  When taking a test using a white background it took an average of 15.6 (SD = 4.69) minutes to complete compared to an average of 15.7 (SD = 5.50) minutes to complete when using a yellow background.



Results of this study suggest there is no difference in reading comprehension levels when reading information from an LCD monitor using black text on a yellow background compared to black text on a white background.  It must be noted that this study does not address the efficacy of any particular background for usefulness to low vision users in providing readability via contrast levels, but rather it specifically addresses the effects of background color on reading comprehension.



An important limitation to note in this study is the amount of area that the yellow background covered.  Even when choosing the yellow color, the entire background was not changed, but rather, only the area surrounding the information presented for this test had yellow behind it thus leaving all vacant areas on the page white.  This also meant that the size of the yellow background area would change from one question to the next depending on how lengthy the questions or instructions were for a given question.  This meant that a similar effect to the image framing found by Carter et al. (1999) may have taken place by altering the apparent contrast.  Further, although there does not appear to be a specific standard when using a yellow background to create high contrast, perhaps covering the entire page with a more restful or pastel shade of yellow may indicate a different result.  These lighter shades may be more reflective of commonly used yellow backgrounds, but they were not available to employ within tools available for this study.


Another possible limitation of this study lies in its emphasis on providing relatively short passages and asking subjects to select the most appropriate answer based on that particular passage.  It may be useful to instead to present subjects with a much longer complex reading passage to assess reading comprehension in an environment more reflective of real-world application.


In addition, there are many other color combinations that warrant consideration to assess what, if any, reading comprehension relationship may exist.  This may be especially important where user selectable color schemes are available because its direct view does not filter the display of a light emitting device such as a monitor like the color filters did in the study by Chase et al. (2003).  While they may not be widespread at this time, at least limited user selectable background/text color combinations are nevertheless available in some venues, especially for computer delivered tests.


Anecdotal Accounts by Participants

Although this study was not designed or intended to collect qualitative data for analysis, a number of participant comments at the conclusion of this project warrant consideration.  It's important to note that because these were informal comments and were not recorded in any measurable format, they're not intended to provide quantitative data for the purpose of statistical analysis.  They are instead included in this section to provide insights shared by end-users.


The reports of color preference were somewhat mixed with no clear consensus on a single preferred background color.  Some of the participants preferred the yellow background while some preferred the white background, and at least one other had used the computer mouse to select the reading area and turn the background dark blue under both conditions.


Further, reporting a preference for a particular background color did not always correlate with achieving a higher score on reading comprehension.  Without recorded data to clearly identify the strength of this correlation, it was not possible to determine the significance of this effect within this study.  There was also no way to determine the specific reason for a preference whether it was for ease of reading, level of contrast, or other reason.


Need for Further Research

In light of the results of this study, the limitations that were part of this study, and the anecdotal accounts of participants involved, there is a need for additional research to clarify a number of issues raised.


To successfully complete a sound future study, a greater control over page color and page area covered would make an important improvement to the reliability of results.  This restriction in color and page area covered severely limited the design ability within this study.  This is one of the limitations that were imposed on the current study because the tests resided at a remote location leaving only limited tools available to the researcher with which to manipulate test conditions.


The results of this study also suggest that a greater array of text/background color combinations should be used to identify any advantage to other colors.  These different combinations should assess not only reading comprehension levels, but they should also include a number of user preference questions at the end of each testing condition.  This data, once collected, can be used to identify any correlation between user preferences and effects on reading comprehension level.  This will allow the researcher to make important comparative analysis that was not available in the current study.



In sum, the results of the study did not suggest a significant difference in reading comprehension level when presenting information on a computer monitor using a black text with either a white background or a yellow background, hence failing to reject the null hypothesis.  There are, however, a number of text/background color reading comprehension issues that are not addressed or resolved by this study.  To that end, further research on this topic is needed to provide more conclusive evidence of whether a significant relationship between text/background color combinations and reading comprehension may exist.



Carter, R., Sibert, L., Templeman, J., & Ballas, J. (1999). Luminous backgrounds and frames affect gray scale lightness, threshold, and suprathreshold discriminations. Journal of Experimental Psychology: Applied, 5(2), 190-204. Retrieved May 11, 2006, from PsycARTICLES database.


Charness, N., & Holley, P. (2004). The new media and older adults: Usable and useful? American Behavioral Scientist, 48(4), 416-433. Retrieved May 11, 2006, from PsycINFO database.


Chase, C., Ashourzadeh, A., Kelly, C., Monfette, S. & Kinsey, K. (2003). Can the magnocellular pathway read? Evidence from studies of color. Vision Research, 43, 1211-1222.


Droll, J. A., Hayhoe, M. M., Triesch, J., & Sullivan, B. T. (2005). Task demands control acquisition and storage of visual information. Journal of Experimental Psychology: Human Perception and Performance, 31(6), 1416-1438. Retrieved May 15, 2006, from PsycARTICLES database.


Garland, K. J. & Noyes, J. M. (2004). CRT monitors: Do they interfere with learning. Behaviour & Information Technology, 23, 43-52.


Gorn, G. J., Chattopadhyay, A., Sengupta, J., & Tripathi, S. (2004). Waiting for the web: How screen color affects time perception. Journal of Marketing Research, 41(2), 215-225. Retrieved May 11, 2006, from PsycINFO database.


Hall, R. H. & Hanna, P. (2004). The impact of web page text-background colour combinations on readability, retention, aesthetics and behavioural intention. Behaviour & Information Technology, 23(3), 183–195


Hong, W., Thong, J. Y. L., & Tam, K. Y. (2004). Does animation attract online users' attention? the effects of flash on information search performance and perceptions. Information Systems Research, 15(1), 60-86. Retrieved May 11, 2006, from PsycINFO database.


Hsu, Y. (2006). Better educational website interface design: The implications from gender-specific preferences in graduate students. British Journal of Educational Technology, 37(2), 233-242. Retrieved May 11, 2006, from PsycINFO database.


Keller, T., Gerjets, P., Scheiter, K., & Garsoffky, B. (2006). Information visualizations for knowledge acquisition: The impact of dimensionality and color coding. Computers in Human Behavior, 22(1), 43-65. Retrieved May 11, 2006, from PsycINFO database.


Mead, A. D. & Drasgow, F. (1993). Equivalence of Computerized and Paper-and-Pencil Cognitive Ability Tests: A Meta-Analysis. Psychological Bulletin, 114(3), 449-458.


Morey, C. C., & Cowan, N. (2005). When do visual and verbal memories conflict? the importance of working-memory load and retrieval. Journal of Experimental Psychology: Learning, Memory, and Cognition, 31(4), 703-713. Retrieved May 15, 2006, from PsycARTICLES database.


Phillips, N. A., & Lesperance, D. (2003). Breaking the waves: Age differences in electrical brain activity when reading text with distractors. Psychology and aging, 18(1), 126-139. Retrieved May 15, 2006, from PsycARTICLES database.


Poynton, T. A. (2005). Computer literacy across the lifespan: A review with implications for educators. Computers in Human Behavior, 21(6), 861-872. Retrieved May 11, 2006, from PsycINFO database.


Proctor, R. W., Vu, K. L., Salvendy, G., Degan, H., Fang, X., & Flach, J. M. et al. (2002). Content preparation and management for web design: Eliciting, structuring, searching, and displaying information. International Journal of Human-Computer Interaction, 14(1), 25-92. Retrieved May 11, 2006, from PsycINFO database.


Stone, R. (2004). Light and color count: Light-based reading issues from the point of view of those living with it. Journal of Optometric Vision Development, 35(1), 24-36. Retrieved May 11, 2006, from PsycINFO database.


Turville, K. L., Psihogios, J. P., Uimer, T. R. & Mirka, O. A. (1998). The effects of video display terminal height on the operator: a comparison of the 15° and 40° recommendations. Applied Ergonomics, 29, 239-246.


Wang, A., Fang, J. & Chen, C. (2003). Effects of VDT leading-display design on visual performance of users in handling static and dynamic display information dual-tasks. International Journal of Industrial Ergonomics,32(2), 93-104.


Zviran, M., Te’eni, D. & Gross, Y. (2006). Does Color in Email Make a Difference? Association for Computing Machinery, 49(4), 94-99