Using Pictograms to Make Transit Easier to Navigate for Customers with Communication Barriers (2013)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

TCRP A-33A Final Report 10 CHAPTER 2 Research Approach 2.0 Approach – Literature Review Preliminary research included a literature review to identify new and emerging research or recommendations regarding the use of pictograms for emergency communication in transit. Research topics that guided the literature review process included existing symbols, pictograms, and illustrations in use; processes for developing and testing effective symbols and forms of communication for populations with functional communication needs; pictograms for conveying safety and emergency information; identified communication needs of populations with communication barriers; and barriers to communicating with these populations. Literature sources included industry specific journals, online articles and resources, websites, and Transportation Research Board (TRB) staff-recommended research reports. The literature review primarily benchmarked representative research conducted in the fields of health, transportation, hospitality, and manufacturing as it related to pictograms for the purposes of communicating specific actions or behaviors, rather than wayfinding or locators. The review identified common terminology or symbols already in use. 2.1 Summary From cave drawings to emoticons, people have used picture symbols, such as pictograms, to convey information to one another across time and cultures. Pictograms, specifically, are “pictorial signs used to warn, guide or protect and need to be immediately decipherable. They must get right to the heart of the matter by visually conveying a vital piece of information in such a way that it cannot be misunderstood.” (Abdullah and Hübner 2006) The use of picture-signage at the 2008 Olympics in Beijing gave broad international exposure to contemporary graphic pictograms as a means of communicating around language barriers. As a result, pictograms have become important as a means of communicating in the current global community and are starting to play a role in daily life for all populations, including those with functional communication needs (people who are deaf or hard of hearing, have Limited English Proficiency or no English skills, are illiterate, or have cognitive impairments.) (Ting-Ju 2006) Few pictograms, however, are universally understood and all groups do not always correctly interpret even those that have wide acceptance. It takes years for a pictogram to reach maximum effectiveness – no pictogram is instantly recognized. (CDC/SNS 2006) Research for the development of a pictogram system in the health field confirmed that a well-designed system could assist English-speakers as well as people with communication barriers, but would not solve all the issues involved in delivering vital, actionable information that is understood by all populations. (Hablamos Juntos) Of interesting note to the research team was the extensive practice and creation of universal health symbols. A few of the lessons learned in the symbol design testing process included focusing on a limited number of distinct symbols that could be recognized instead of a large

TCRP A-33A Final Report 11 group of symbols similar in appearance. It was also learned that while some symbols, representing easy to understand destinations, could be read with few problems, others were difficult to comprehend. This is endemic of a lack of understanding of the meaning of certain hospital functions by the general population, and brought to light the need to use symbols for tough-to-comprehend destinations as educational tools. (Hablamos Juntos) Directional signage and picture-based instruction are used extensively in the transportation field. Picture-based identification developed by the U.S. Department of Transportation is seen in airports, rail and transit stations, and ports to convey wayfinding information, such as location of stairs, elevators, restrooms, and gift shops. However, pictograms, which may be designed to shape behavior as well as show direction or location, for the specific purpose of directing transit passengers in an emergency, are few. They are most familiar to users of subways, or light rail and commuter trains. For example, the Bay Area Rapid Transit (BART) system and the Metropolitan Transportation Authority (MTA) New York City Transit (NYCT) system have installed picture-based messages to guide passengers during emergency situations. These usually depict evacuation routes or means of exiting a conveyance and contain a series of actions. Other than signs at bus stops to indicate a hurricane evacuation site, pictograms to relay emergency information and actions to passengers on a bus were not found in the research. Nor did the research uncover transportation-related picture-based instructions for behaviors other than entering or exiting vehicles or spaces (e.g., no expressions for “Stay in your seat”). At the same time, an increasing number of people can now access public transportation systems as a result of as the Americans with Disabilities Act (ADA) and the Individuals with Disabilities Education Act (IDEA). “The ADA recognizes the critical role that public transportation plays in the lives of many people and mandates that public transit systems become accessible to people who have disabilities and that paratransit services are available and accessible to individuals who are unable to use public transportation.” (Groce 1996) Many travel training programs have become available across the nation to help people with disabilities navigate the public transportation system or paratransit services. Because population groups who would be vulnerable as bus passengers during an emergency event comprise a variety of communication characteristics and needs related to physical and mental vulnerabilities, pictograms need to address the following: • Different cultures • Different cognitive abilities to understand the meaning • Different languages • Mental state at time of incident (high stress, etc.) (Galea 2011) In the event of an emergency while riding a bus, all populations need to be able to see pictograms and quickly recognize messages. “Visual perception occupies by far the largest area of the human brain, at 80 percent (followed by hearing at 10 percent), and sight is therefore the most influential of the senses.” (Abudullah and Hubner 2006) This makes the clear, visual representation of important messages vital in disseminating information in emergency situations.

TCRP A-33A Final Report 12 According to many graphic design experts, the best examples of pictograms for emergencies included the following traits: • Black on white; color tends to distract from the message. • Simple. Not a lot of shading or lines. Simple shapes and movement. • Size dependent on the usage. Be sure it is big enough for the surroundings. • File size and type important for distribution to a large audience. • Maximum message in the minimum space. • Look the same/seem related in a series. The viewer will not have to “start over” with each pictogram. For instance, if each pictogram is in a square, with the same width of line used in its creation, the viewer will focus on the message, not on the fact that one may be in a circle, another using thinner lines. • Messages: o Simple. This allows for greater ease in creating a visual (e.g., “Sit Down” vs. “Sit down slowly on the leather couch by the window”). o Concise. o Direct, with a call to action. Clear direction for the viewer to see and act on. Viewers should not have to make their own decision. o Understood without words. o Make sense within its surroundings o Understood immediately or within one-to-two seconds. If the viewer has to think about it, he/she has wasted time that may be in short supply in an emergency. • Public education is most important. If the viewers have never encountered pictograms before, they may be confused or want more information. It is best to explain pictograms several times before an emergency, so viewers have a basic level of familiarity. • Responses are different when pictograms are read independently versus used as a support for a verbal direction. In addition to the graphic elements required, federal regulations and other code and regulating agencies govern regulatory and information signs used in transportation, including the use, placement, and size. (Harding 2011) Many of these regulations may also apply to pictograms, such as ADA-specific requirements. ISO 7001 ("Public Information Symbols") is a standard published by the International Organization for Standardization that defines a set of pictograms and symbols for public information. The latest version, ISO 7001:2007, was published in November 2007. The set is the result of extensive testing in several countries and different cultures and have met the criteria for comprehensibility set up by the ISO. Common examples of public information symbols include those representing toilets, car parking and information, and the International Symbol for access. Research also examined emergency preparedness within the public transportation service sector. In 2003, the U.S. Department of Transportation Federal Transit Administration published “The Public Transportation System Security and Emergency Preparedness Planning Guide,” which focuses primarily on a response to terrorism. “Since the early 1990s, the nation’s 100 largest rail and bus properties, which combine to move approximately 85 percent of all passengers who use public transportation, have been working to address the credible threat from terrorism.” The National Transit Institute offers courses to help improve overall security, such as System Security

TCRP A-33A Final Report 13 Awareness for Transit Employees, which covers skills for observing, determining, and reporting people or things that are suspicious. (Balog, Boyd and Caton 2003) Although the FTA Guide recognizes that major incidents, such as train accidents, fires, floods, violent crime, and terrorist attacks, have been an issue for public transportation for decades, it does not specifically address the communication needs of bus passengers and the responsibilities of drivers and operators to meet those needs during an emergency. (Balog, Boyd and Caton 2003)

TCRP A-33A Final Report 14 2.2 Interviews with Transit Providers Approach The team developed and internally reviewed an interview discussion guide (see Appendix A). The project team worked with panel members and transit providers to assemble a list of potential candidates representing six to seven different geographic regions of the United States. Where possible, the team requested “cluster” interviews with groups of drivers/operators to gain the widest possible perspective. The research team interviewed 21 drivers, operators, and safety professionals in New York City; Houston, Texas; Portland, Oregon; Des Moines, Iowa; and Tuscaloosa, Alabama. Summary Bus transit drivers, operators, and agency representatives take their responsibility for the safety of the traveling public extremely seriously. Interviewers participating in this research effort share complete admiration for the professionalism and capability of the drivers and operators who were willing to give their time to participate in this study. In addition, drivers and operators indicated support for the study and the potential of pictograms as a communication aid in emergency situations. • Current methods for communicating messages to passengers include drivers making announcements via the bus Public Address (PA) system or dispatch announcements via the PA system. Pictograms would be somewhat limited in their use, in that it would require the bus to be stopped. Drivers may use a microphone to make announcements (sometimes using a pedal that operates the microphone while they are driving). When able, drivers will stand and face passengers when making announcements. Eye contact and clear engagement with passengers was emphasized. • Driver behavior and rapport with passengers was identified as very important. Drivers and operators were clear that they are key to keeping control of circumstances and situations on the bus. Drivers emphasized the importance of confidence and control and building relationships with riders. All of this aids in riders’ willingness to accept directions and improves the likelihood riders would come to the driver’s aid in difficult situations. • Passengers with cognitive or other communication challenges represent a different level of responsibility. Drivers report that when they observe that a passenger is confused, they may augment their announcements with hand gestures. It was widely reported that IF ASKED, other passengers are often willing to help in translating information or messages to passengers with communication challenges. People who are deaf or hard-of-hearing will often carry paper with them so that critical information can be written down. For passengers with cognitive challenges, messages must be communicated in a very simplified form (e.g., “the bus won’t move” versus “the bus is out of service”). • If picture-based tools and communication aids were to be made available to the industry, the following suggestions were made by drivers, operators, and safety professionals:

TCRP A-33A Final Report 15 o Public awareness and training on the tools would be very important. o Periodic reminders/refresher training about the picture-based messages, their purpose and meaning would be required. o The best means for informing passengers about pictograms and their meaning would include posting the pictograms on buses (in driver’s area, laminated), at bus shelters, in transfer centers, on screens at bus stops, and on E-cards. Some drivers saw a need to alert people through a universal signal with light and sound that everyone would recognize (similar to Emergency Broadcast signals) prior to other important messaging. o Drivers would prefer a single consistent area in the bus where emergency information would always be posted. * o Driver safety training would be required, as described by drivers, with combined emphasis on how and what to communicate in an emergency and safe bus operation and driving. • Tips/Suggestions for Pictogram Development: o Simple words/pictures o Not too large o Ease of use by the bus drivers o Not too many o Public awareness essential o Permanent diagram of all exits posted on bus* • Potential Pictograms/Messages/Message themes o Look o Listen o Turn off Electronics o Stay Calm or Remain Calm o Follow Me o Stay Seated/Remain Still/Don’t Stand Up o Delay o Change in Route o Danger o Help is Coming o Will Get You to Safe Place o Move/Transfer to Another Bus/Vehicle o Stay together (message for paratransit passengers who disembark the vehicle.) *Exit Strategies/Bus Evacuation Instructions.

TCRP A-33A Final Report 16 2.3 Interviews with Experts Serving Vulnerable Populations Approach The research team relied on long and trusted relationships with a range of advocates and service providers for vulnerable population groups in carrying out this information-gathering task. In addition to specific information sources referred by panel members (and two project panel members), a critical resource for this task was the Kentucky Outreach and Information Network (KOIN), a network of more than 300 different organizations and individuals who work in direct service to a range of vulnerable population groups across the Commonwealth of Kentucky. Members of the project team have long and trusted relationships with various members of this network and were able to gather critical information relevant to this research effort. (See Appendix B.) Summary Advocates and professionals who work as service providers to vulnerable population groups are passionate about their work and dedicated to the needs of the people they serve. Interviewees were enthusiastic about the concept of pictograms for use in emergencies where transit providers were critical to public safety and believed the idea has merit. Interviewees had experience in working with a wide range of vulnerable populations including people who are: 1. Deaf and hard-of-hearing and/or mute 2. Disabled by intellectual and other cognitive limitations 3. Limited English Proficient (LEP) or non-English speaking 4. Illiterate 5. In need of medical care or supervision 6. Socially and geographically isolated 7. Low income 8. Elderly or very young 9. Homeless Interviewees had experience in preparedness and response to a number of different emergency incidents including the World Trade Center bombing in 1993; the 9/11 terror attacks; multiple weather incidents (including hurricane evacuation and sheltering scenarios); rural public transportation; and public health challenges, including the development and implementation of pictograms to overcome communication barriers. The interviewees suggested that all vulnerable population groups need and would respond to pictograms with education and training. Similar to findings from interviews with transit providers, the professionals who work with vulnerable populations believed the use of pictograms as a communication tool will only be successful if ongoing training and education are provided to all users. Additional recommendations from this group of interviewees include: 1. Focus on characteristics of picture-based signage as well as the picture itself. Characteristics to be considered would include the placement height of signage on the bus or in transit stations, vertical versus horizontal design and placement, and signage that is relevant to passenger travel decision points.

TCRP A-33A Final Report 17 2. Test pictograms with very little or no wording combined with the pictures. Some settings where pictograms are in use and showing promise have no words attached, such as traffic signs for “No Parking,” a simple circle with a negation sign through the letter “P.” 3. Use messages of high importance, such as: a. Stay on Bus b. Train/Bus not Working c. Go to…(arrow pointing the direction) d. Stay Calm e. Don’t Worry f. Don’t Move g. Stay Where You Are h. Help is Coming/Help is Not Coming i. Time to Leave/Need to Leave Now j. Move k. Stay in Safe Spot (off transport) 4. Establish ongoing training in the meaning and use of pictograms for multiple target groups. Police, fire, and other public safety officials; transit providers (drivers and operators); and end users will all need awareness and education in order for pictograms to work at an optimum level. Awareness building and educational initiatives could include Public Service Announcements (PSAs); posters on buses, in immigrant or English as Second Language (ESL) service provider facilities; transit, emergency management, and public safety websites, etc. 5. Consider the following: a. Make the final pictogram products available and accessible to transit providers via simple technology and widely available software (e.g., Microsoft Word, PowerPoint, jpeg, or pdf files). b. Include color, white space, lines, borders, logos, and other information that may ultimately be included on a pictogram sign in the project research and testing. c. Incorporate individuals with cognitive impairment in target audiences for pictograms, as there is little research on this vulnerable population group. d. Use multiple methods of communication, including gestures, verbal, and static and dynamic messaging. Make sure people aren’t receiving contradictory information between what’s being said, shown or gestured.

TCRP A-33A Final Report 18 2.4 Select a Scenario for Developing and Testing Messages Approach A scenario useful for studying the potential of pictograms in an emergency affecting bus transit needs several characteristics, if the study is to be broadly applicable in the transportation industry. These characteristics include an emergency circumstance that could occur: • In a wide variety of locations throughout the U.S.; • As an event that could be sudden in some instances, anticipated or forecast in others; • With sufficient time for a bus driver to address the emergency with passengers as it is unfolding, not just in aftermath (e.g., not a collision or explosion); • With fundamental elements that are similar no matter the geographic location; and • With sufficient predictability in some locations that results of this study would potentially be valuable immediately for application to real-time emergencies. Summary The scenario that most meets these criteria is flooding. • Scenario exists in a wide variety of locations throughout the U.S. According to the Federal Emergency Management Agency (FEMA), floods are the most common natural disaster in the United States. Capable of striking almost any river, creek, lake, or coast nationwide, they kill about 200 Americans each year and are often more destructive than the storms that caused them – water flowing at 10 mph exerts the same pressure on a structure as 270 mph wind gusts. Across the country, floods destroy approximately $6 billion worth of property every year. Floods constitute potential emergencies for bus transit in communities of every size. (FEMA 2012) • Scenario is an event that could be sudden in some instances, anticipated or forecast in others. To study the applicability of various pictograms to circumstances that would cause abrupt need for emergency directions to bus passengers and guidance of their behaviors, it is important to have emergencies that could be sudden. Flash flooding is a good example. Overflowing rivers create most flooding situations, both flash and slower-rising floods, but quite a number of circumstances can cause flash flooding. Broken dams, broken pipes, urban run-off, as well as bigger events, such as hurricanes or even tsunamis, can create flash floods. Most deaths and damage from floods in the U.S. are due to flash flooding, which is defined by the National Weather Service as "a rapid and extreme flow of high water into a normally dry area, or a rapid rise in a stream or creek above a predetermined flood level." Flash floods develop suddenly, often in a surprisingly few minutes. Heavy rains are the top cause of flash floods, but urban runoff, "ice jams," dam failures, and other factors may contribute sufficient water for flash floods, which occur in all 50 states. They

TCRP A-33A Final Report 19 are most common in hilly areas with steep valleys, or along small waterways in urban environments. Their speed, depth, and the element of surprise make flash floods very dangerous, giving vehicles in their paths little time to prepare, make a course correction, or evacuate. (NOAA 2012) Other sudden bus transit emergencies can be caused by more infrequent, but not unknown, ground failures. Ground failures occur when the water table rises to the surface and washes away chunks of topsoil. This can cause a variety of ground failures, including "subsidence," or sinking soil, and "liquefaction," a process in which water-soaked sediment loses strength and acts like a liquid. Ground failures can create a range of dangerous transportation circumstances from roadway upheaval to "mudfloods," a liquid flood that carries up to 50 percent solid sediment loads, and "mudflows,” which are solid landslides where the downward flow is viscous enough to support large boulders within a wave of smaller particles. Mudfloods and mudflows are most common in California and other Western states, because they tend to occur on hillsides burned bare by wildfire, but ground failures can occur almost anywhere. Even apparently dry land can be subject to a ground failure when other flooding has occurred. (Pierre and Collins) Pictogram-use scenarios need to include foreseen potential emergencies, too. In such cases, bus transit could be interrupted by flooding, but with enough warning to be able to give passengers a set of instructions that might be different than those used for sudden emergencies. For example, King County Washington Emergency Service Network (ESN) regularly designates 52 core bus routes to address anticipated flooding of the Green River as well as snow or ice. (ESN 2011) Pictograms could be useful in such cases to let regular bus riders know what to do (e.g., exit) on unfamiliar routes, in perhaps unfamiliar areas. Some flooding is anticipated well in advance, giving communities and transportation organizations some opportunities for preparation. By early spring 2011, for instance, the National Weather Service forecast that more than half the U.S. was at risk of spring flooding, with many metropolitan areas of 1 million or more residents facing a 95 percent chance of flooding in the spring rains. (NOAA 2012) Getting bus transit ready to cope with flooding in those metro areas would be a good use of pictogram tools, and such areas provide opportunities to pilot this program. • Scenario anticipates an onset with sufficient time for a bus driver to address the emergency with passengers, as it is unfolding, not just in the aftermath (e.g., not a collision or explosion). Pictograms are well suited for emergencies that are unfolding with time to give instructions and flooding most often affects buses in that way. Many bus and flood incidents are available for view on the web, and even in water that would be overwhelming for smaller vehicles, buses experience the event more stably. Presumably, except for anomaly flash flooding, bus drivers would have time to communicate with passengers as the event was under way. • Scenario needs fundamental elements that are similar no matter the geographic location.

TCRP A-33A Final Report 20 The effect of flooding on buses in most areas is similar, so this scenario has widespread use. Obvious exceptions include mudslides or flash floods on mountain roads, but otherwise flooding on roadways has important elements in common, including most people’s lack of understanding about the dangers of running water that does not appear to them to be deep. Nearly half of all flooding deaths each year occur when people drive through or walk through moving water. (NOAA 2008) It’s possible that a pictogram should depict “Stay on the bus,” as many people might disregard the danger and attempt to disembark. Flooding is also often part of a larger weather event, with ramifications for handling the emergency. Rerouting for floods could cause confusion for some passengers, especially in large urban areas. No matter where it occurs, flooding raises people’s anxiety levels (fears for safety, safety of family and others, property damage, etc.) so that the capability of individuals to remain calm and follow instructions may be affected by their stress that is unrelated to their own immediate circumstances. These kinds of elements affect the usefulness of the pictogram no matter where the event is occurring. (NOAA 2012) • Scenario depicts event with sufficient predictability in some locations that results of this study would potentially be valuable immediately for application to real-time emergencies. Because flooding is predicted to increase, transportation agencies are increasingly making flood readiness a part of business and operational planning. Flooded Bus Barns and Buckled Rails: Transportation and Climate Change Adaptation is a Federal Transit Administration study undertaken to provide “transit professionals with information and analysis relevant to adapting U.S. public transportation assets and services to climate change impacts. (FTA 2011) Climate impacts such as heat waves and flooding will hinder agencies’ ability to achieve goals such as attaining a state of good repair and providing reliability and safety.” Many reports from transportation agencies at state and local levels discuss flooding and its impact. Bus drivers interviewed for this pictogram study cited flooding when asked about emergencies they anticipate. As a result, it is likely that the study will find ready participants to further test the pictograms and pilot initiatives. In coastal and river communities, where flooding events are regular, the pilot pictograms could well be put to use in real emergencies in the first rainy season after a study’s completion.

TCRP A-33A Final Report 21 2.5 Develop and Test Pictograms Approach Develop and test a set of pictograms for the selected scenario and test for effectiveness in communicating critical messages. Conduct field tests with people representing populations that have functional communication needs to discover whether the proposed pictograms communicate the message and elicit the desired behavior. Summary Much of the research for this study explored wayfinding pictograms and environmental graphics in hospitals, airports, transportation systems (including transit), public health settings and other public venues, because these are the pictograms currently in widespread use. In transit systems, the purpose of wayfinding signage is to move people from one place to another, using word messages, symbols, and images. Pictograms that attempt to explain or guide exigencies or occurrences (e.g., changes in schedule, location, or state of being), behavior (e.g. wait, follow, sit, etc.) or response (e.g., stay calm, hurry, listen, etc.) are almost non-existent in public settings, and even in other research efforts. In this pilot project, the purpose of the pictograms is primarily not wayfinding, but rather providing information to riders that would aid their decision-making about actions or behaviors that should or must be taken for safety in an emergency situation on a transit vehicle. The challenge for the design team was to create a series of pictograms that captured messages that had been identified earlier in this study as essential by bus drivers and operators, as well as representatives of agencies that serve populations with communication challenges. The messages aligned in two types: directive (“Look at me”) and circumstantial (“Route Change”). The directive messages were in two categories: messages to influence action (“Turn off”) and messages to influence feelings or understanding (“Stay calm” “Help is coming”). The design team also faced the difficulty of creating universal messages for a variety of people and places. Geographic location, cultures, languages, and prior experiences of the user audience affect the ways people decipher and process information. The Design Process The design team consisted of a graphic designer with experience in symbol design, wayfinding signage system designs, and pictograms and research team members with prior experience in evaluating and developing implementation guidance for transportation and healthcare system signage and wayfinding signage and pictograms. The designer participated in several of the interviews with bus driver/operators to understand firsthand the situations drivers and passengers face in an emergency situation. The design approach was based on best or model practices for symbols or pictograms identified in the literature review and subject matter expert interviews. To design original pictograms, the design team considered the following: • Audience – people who have communication challenges (specifically, people who are deaf or hard of hearing; people with developmental cognitive impairments, including

TCRP A-33A Final Report 22 people who are elderly; people with mobility limitations; people with limited or no English proficiency; people who are distracted, such as adult caregivers of children, elders or disabled companions; people who are tuned in to electronic devices (estimated by drivers as more than 90 percent of bus riders). • Scenario – roadway flooding. • Messages – identified by the bus drivers/operators as critical during an emergency (flood) • Messages – identified by representative organizations that serve populations with communication challenges. • Existing pictogram designs and criteria in transit, transportation, health care facilities, and schools. • Pictogram design requirements – size, content, legibility, readability, use of words, and color, etc. • Location – held by driver on a parked bus. The basic test was to be of images shown by a bus driver, not affixed to the bus interior or shown on dynamic messaging technology. The reasoning for this approach was to use a communication tool (pictograms) selected/controlled by the driver and not dependent on the electrical system of the bus (which could be non-functional if the bus were disabled). The messages that the designer selected for the pilot were ones that bus drivers interviewed for the study agreed were messages they would consider essential if the bus were facing an emergency that stopped the bus, but did not require immediate evacuation. This pilot did not study evacuation for fire, bomb threat or similar no-notice, immediate life-threatening danger events, as transit providers have protocols in place for those. The pilot study messages were: • Look at me • Listen • Turn off (electronics) • Stay calm • Follow me • Stay seated • Delay • Change in route • Danger • Help is coming • Will get you to a safe place The design team worked through many versions of the pictograms to identify ways to better express the message graphically. After the preliminary testing, the team concluded that the last message (Will get you to a safe place) was too abstract to convey in a single pictogram (viewers’ ideas about what constitutes “safe place” are almost entirely personal) and removed it from the set of pictograms to be tested with pilot groups of end users. Preliminary evaluation For a preliminary field evaluation of 11 pictograms, the research team reached out to respected experts with a strong interest and experience in working with populations with communication

TCRP A-33A Final Report 23 barriers. The feedback indicated that the pictograms generally “looked very good,’ but needed refinement. In this preliminary review, the pictograms that needed no changes were: 1. “Turn off” 2. “Delay” (not entirely clear but basically communicated message) 3. “Change in route” 4. “Danger” Those that needed to be revised for clarity – and representative comments –were: • “Look at Me” o Change dotted line to continuous line. o Align the line from the passenger’s eye to the driver’s eye, rather than to the driver’s nose. • “Listen” o Too much black. o Side-by-side with just an ear (minus the head). • “Follow Me” o Draw lines to delineate the walkway movement that coincides with following. o Look at pictograms in other settings for Form Line. Those are the lines that this is missing. • “Help is Coming” o I like this one, but could the cross on the help coming vehicle maybe be widened to look more like the first aid cross? o When the cross is italicized to make it look like it is in motion, it takes away from the cross so it needs to be made wider visually. Those that were least intelligible to viewers were: 1. “Stay Calm” o This one is hard. o Looks like the bus driver is lifting the passenger. o Put bus driver hands on shoulders of passenger. 2. “Stay Seated” a. Something looks weird here. b. The arrows are too small? 3. “Will Go to Safe Place” a. Hardest one yet. b. I’m not sure about this one. c. You could go ahead and test it to see what kinds of responses you get.

TCRP A-33A Final Report 24 The final set of pictograms was printed on 8.5” x 11” card stock in black and white; the actual images comprised slightly more than half the sheet (see Appendix C). Literature review research and previous experience of the team in pictogram testing suggested the value of adding one or two words describing the picture, e.g., “Look,” to aid in comprehension. Some Limited English Proficiency (LEP) viewers may have enough English familiarity to decipher simple words when attached to the picture. The pictograms were numbered in sequence, corresponding to the numbers on the participants’ and observers’ comment cards. The development and testing of this first set of pictograms is early-stage research to determine how graphic messages can be effective for transit systems’ communicating with riders, especially those with communication challenges, especially in an emergency. Ideally, pictograms can also be found to be effective in ordinary communication requirements of transit systems, as well. The objective of this study is to be an early step in long-term research around the possible design and implementation of universal, standardized symbols for transit systems. This study models a process for what could become ongoing pictogram design and testing.

TCRP A-33A Final Report 25 2.6 Conduct Pilot Tests Approach Create a pilot program. Apply the findings from all tasks to develop a pilot program that transit managers can use to sample the effectiveness of pictograms to convey transportation information in routine and emergency situations. Recruit three selected transit entities or agencies representing urban, rural or multi-jurisdiction transportation agencies to use the pilot and report outcomes. Pilot program can be made available to other transit agencies interested in trying it, but project scope limited the research report to reflect only outcomes from three selected organizations. Without addition to project cost, the research team decided to include a fourth provider to expand the diversity of participating transit systems. Summary The accepted protocol for assessing whether a pictogram conveys the intended message is a controlled and impartial evaluation by individuals representing the target audience. To accomplish this, the research team started by recruiting four pilot test sites representing the eastern, middle and Western United States. The research team contacted organizations that had indicated an interest in participating as a pilot site during interviews: New York City Department of Education/New York City Transit (part of the Metropolitan Transportation Authority); Portland, Oregon’s TriMet transportation system; St. Petersburg, Florida’s Pinellas Suncoast Transit Authority (PSTA); and the Kansas City Area Transportation Authority (KCATA), which serves customers in the greater Kansas City metropolitan area, comprising seven counties and several smaller cities. Due to budget limitations, the research team was not able to travel to the pilot sites to set up, help recruit community partners and individual group participants, conduct the discussion group, observe the participants’ reaction to the pictograms, or provide direct reporting. This required the pilot site agencies to bear a majority of the effort without any remuneration. The four transit agencies, however, were eager to participate, saying the research was valuable to them and important in finding ways to better serve all their customers. Partnership with these transit agencies made this research possible inside the scope and schedule of the project, and the input from the agencies themselves added immeasurably to the information gathered in the study. Testing location: The research team recommended the use of a parked bus that was not in service. By conducting the test in a bus rather than a conference room, the team believed that both the driver and riders involved in the testing would be able to give more accurate feedback about how the pictograms would work in the environment where they were intended to be used. A parked bus not in service would prevent the risk of testing on a moving vehicle and allow the necessary time for testing and discussion without the pressure of a route schedule. All four of the participating transit agencies were delighted with that approach and provided necessary buses and drivers, as well as representatives of the agencies to serve as observers and note-takers. This represented a substantial commitment of resources by each of these agencies. Discussion Group Guidance: The research team created a “focus group in a box,” a set of step- by-step directions for the transit agency to follow in preparing for the pilot test; locating and working with a community service partner; conducting the focus group; and reporting the

TCRP A-33A Final Report 26 outcomes. Another set of guidelines was created for the community service provider to use in recruiting participants; working with the transit agency; observing during the test; and reporting outcomes. See Appendix D and Appendix E. In addition, the research team provided materials for the discussion group process, including a set of 10 pictograms; nametags, markers, and pencils; comment cards for the participants, drivers, and observers; and pre-paid return envelopes. The packets of instructions and materials were sent simultaneously to all the transit agencies and service providers participating in the pilot tests. What was required: The following is a synopsis of what the transit agency was asked to do: 1. Review materials to become familiar with the overall project. 2. Provide a parked bus for the testing location. 3. Identify a two-hour block of time when it can provide the bus and spare the driver for the testing. 4. Identify a driver who is interested in the project and is willing to assist in the testing by: a. Committing approximately three hours of time (30 minutes of preparation, one hour for the actual testing and 30 minutes to one hour for notes and wrap-up). The research team recognized that transit agencies would be required to pay their drivers for this time commitment. b. Leading the discussion group by following a script provided by the research team. c. Showing the pictures on a parked bus to discussion group participants. d. Collecting the rider/customer response forms. e. Making observations during the testing about i. How the customers/riders respond to the pictograms and ii. How the driver him- or herself evaluates the ease of use from the driver’s standpoint as well as the usefulness of the picture-based messages. f. Preparing written notes about his/her observations. 5. Coordinate with the identified community service provider about the time, date, and location of the testing. 6. Return the testing paperwork in a pre-paid envelope. The following is a synopsis of what the community service provider was asked to do: 1. Review materials to become familiar with the overall project. 2. Coordinate with transit agency to confirm the date, time, and location of the pictogram testing. 3. Commit necessary resources to: a. Identify six to eight individuals who have some sort of communication challenge (e.g., people who are deaf or hard-of-hearing; people with English as a second language or limited English proficiency; people who don’t read well or don’t read at all; people with cognitive and mobility disabilities). b. Provide candidates with an explanation of the research project and how testing works, including how their personal information will be protected. c. Provide transportation for volunteers to and from the test location at the time established by the transit agency (there is no requirement that testing location be on a bus route or that it occur coincidental to regularly scheduled bus routes).

TCRP A-33A Final Report 27 d. Provide staff to support the transit agency by being present during the testing to assist the bus driver with making observations and taking notes. Pilot Process at Four Sites and Results Portland, Oregon: The first pilot test was conducted in Portland by TriMet, which provides bus, light rail, and commuter rail service in the Portland metro area. According to TriMet’s website, more people ride transit in the Portland area than in larger cities, such as Dallas, Denver, and San Diego. The community is a cultural and ethnic melting pot with scores of languages other than English – the Portland School district, for example, reports 94 non-English languages spoken among their student population. TriMet bus operators participated in the interviews for Task 2 and the agency’s representative indicated an interest in participating in the pilot tests. TriMet enlisted Ride Connection as its community service partner. Ride Connection is a nonprofit agency that provides responsive, accessible transportation service to populations with a range of transportation needs, including people who are elderly and people with physical disabilities. Ride Connection recruited and provided transportation for seven people to participate in the discussion group pilot test. Two were non-English-speakers and writers; the TriMet manager of multicultural programs translated their written comments after the test was over. Four persons were unable to write and an assistant recorded their comments. The seventh person wrote down her own comments in English. New York City, New York: The second pilot test was conducted by New York City Transit, a division of the Metropolitan Transportation Authority, North America's largest transportation network, serving a population of 14.9 million people in the 5,000-square-mile area. Drivers and operators from NYC Transit participated in the Task 2 interviews. The point of contact with NYC Transit indicated at that time a willingness to participate in the pilot tests. Peggy Groce, a Project Panel member and director of the New York City Department of Education Travel Training Office, previously in the study had been interviewed as a representative of populations with communication challenges and had volunteered her office as a community service partner for NYC Transit. Travel Training is comprehensive, specially designed, one-to-one instruction to teach high school age students how to travel safely and independently on public transportation, where appropriate. Independent travel and use of public transit is closely connected to successful post-school results for students in terms of employment, education, community inclusion, and independent living. The Travel Training Office recruited eight people between the ages of 24 to 34, all graduates of the travel training program. All of the participants were daily commuters on the fixed-route bus system in NYC. The group included individuals with autism, Down syndrome, language and communication difficulties, English as a Second Language and significant intellectual disabilities with associated communication problems. “At first impression, it may seem that these individuals might not be representative of transit riders who are considered communication vulnerable. But they are because they may travel more than many other communication

TCRP A-33A Final Report 28 vulnerable individuals. They have learned to cope with the number of changes that do occur when one is a regular commuter,” Ms. Groce wrote in her comments, adding, “They were really pleased to be part of your project.” St. Petersburg, Florida. A prior relationship for studies with a transit agency in Des Moines, Iowa, whose former director relocated to Florida, led the research team to Pinellas Suncoast Transit Authority (PSTA) in St. Petersburg. Situated on the Gulf of Mexico, Pinellas County has recurring weather events that can cause disruption; traffic incidents that can aggravate congestion and travel times; a large population of Spanish speaking individuals; and a presence of transit dependent adults (including seniors). The contact at PSTA immediately recognized the value of the research and expressed willingness to be a research partner. PSTA operates the bus, trolley, and paratransit services that run between Clearwater and St. Petersburg with two express routes that connect with Tampa’s Hartline transit. The PSTA representative volunteered to recruit focus group participants from several organizations with which the transit agency had established relationships. Participants included six people who were Spanish-speakers and had interpreters; a person from Africa who had limited English proficiency and no interpreter; and one other person, a native English-speaker. Kansas City, Missouri. The KCATA has worked in the past with members of the research team on other projects related to transit and transportation within the Greater Kansas City Metropolitan Area, serving seven counties and 10 municipalities. It is the only form of public transit in Kansas City, Missouri, and provides fixed-route, express, paratransit, and demand-responsive service. It also coordinates with other bus services on the Kansas side of the state line. The KCATA representative had initially planned to work with the agency’s new travel training program, but scheduling made that impossible and KCATA asked the research team to help assemble a group of people for the pilot test. KCATA arranged for the driver, parked bus, and observer. The research team reached out to community-based and faith-based direct service providers, as well as individuals known to members of the team. Participants included two people who were hard of hearing; one person who had never been on a public transit bus; two teenagers (with headphones); and one other person with experience in using bus service in other nations. See Table 1 for a summary of respondent characteristics by test cities. Respondent Characteristics Portland, Ore. New York, N.Y. St. Petersburg, Fla. Kansas City, Mo. Totals Native English- Speaker 1 1 4 6 Limited English Proficiency (LEP) 2 7 9 Disability 4 8 0 2 14 Totals: 7 8 8 6 29 Table 1. Summary of respondent characteristics by test cities.

TCRP A-33A Final Report 29 Pilot Tests On a parked bus, drivers stood facing the “passengers” for pictogram pilot tests. After brief introductions and an explanation of how the discussion process would flow, the drivers faced the back of the bus, held up the pictograms one at a time, and followed scripts provided by the research team. For the first three pilot sites, Portland, New York, and St. Petersburg, participants were asked to look at each picture underscored with one or two words of text printed in English and write (or ask an interpreter or assistant to write) what they understood the message in the pictogram to be; whether or not they would do what the pictogram told them to do; and what they liked or did not like about the pictogram. At these first sites, drivers and observers noted that individuals who could read English would read the words beneath the pictogram and say or indicate they understood the picture. Or, participants would hear what others said about the pictogram and repeat the same comment. However, because the point of the study was to discover whether picture-based communication could be effective – not to test whether pictures could be identified – results included any combination of information that caused people to understand the message correctly. Based on the findings at these three sites, the research team decided to cover up the explanatory words beneath the pictures at the fourth pilot site in Kansas City where all participants were English speakers (including two people who were deaf/hard of hearing). This made a difference in respondents’ understanding of several of the pictograms, although this group identified as most understandable the same images other groups did.