6
Research Gaps

The Workshop on Public Response to Alerts and Warnings on Mobile Devices: Current Knowledge and Research Gaps had two principal goals: (1) to present what is known about the public response to alerts and warnings and how what is known about that response relates to the design, operation, and future development of the Commercial Mobile Alert Service (CMAS) program; and (2) to identify gaps in that research. The following sections present research opportunities identified by the committee and drawn from plenary presentations and discussions in breakout sessions of the workshop, along with associated implementation challenges, and a list of new technologies for alerts and warnings that are likely to raise additional research questions.

RESEARCH OPPORTUNITIES

Message Content

The CMAS specification provides for 90-character messages and prohibits the inclusion of uniform resource locators (URLs) that link to additional sources of information. Workshop participants indicated that relatively little is known empirically about how people will respond to such short alerts. Following is a list of research topics in the areas of message content and length identified by workshop participants:

  • How does a 90-character limit for alerts constrain the ability to



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 46
6 Research Gaps The Workshop on Public Response to Alerts and Warnings on Mobile Devices: Current Knowledge and Research Gaps had two principal goals: (1) to present what is known about the public response to alerts and warnings and how what is known about that response relates to the design, operation, and future development of the Commercial Mobile Alert Service (CMAS) program; and (2) to identify gaps in that research. The following sections present research opportunities identified by the committee and drawn from plenary presentations and discussions in breakout sessions of the workshop, along with associated implementation challenges, and a list of new technologies for alerts and warnings that are likely to raise additional research questions. RESEARCH OPPORTuNITIES Message Content The CMAS specification provides for 90­character messages and prohibits the inclusion of uniform resource locators (URLs) that link to additional sources of information. Workshop participants indicated that relatively little is known empirically about how people will respond to such short alerts. Following is a list of research topics in the areas of mes ­ sage content and length identified by workshop participants: • How does a 90­character limit for alerts constrain the ability to 

OCR for page 46
 RESEARCH GAPS provide the public with alerts? What implications does the 90­character limit have for public response? • Can such a short message provide enough information to let indi­ viduals know that a significant event has taken place? Does it provide enough information for individuals to obtain additional information and take appropriate action to protect themselves? • What are the message characteristics that lead to effective instruc ­ tion in crisis situations? • What does the public want the alert or warning message to say? What do they need to hear? • To what extent will CMAS alerts trigger information­seeking behav­ iors, and what forms will such behavior take? Might that information­ seeking behavior end up leading to the network overloads that the Com­ mercial Mobile Service Alert Advisory Committee (CMSAAC) voiced concern about? Message Dissemination Segments of the population are becoming increasingly accustomed to receiving mobile text messages, including alerts and warnings, from other individuals, from businesses, and from government agencies. Indeed, there has been some experience with the use of text alerts in munici ­ pal and countywide systems as well as some research looking at the effectiveness of these systems, but there has been no experience with national­scale systems. Moreover, the user bases for the systems in place today are small, and participation is entirely on an opt­in basis—so these systems only reach users who are most interested in receiving such alert ­ ing information. Where opt­out systems have been established, such as at the Virginia Polytechnic Institute and State University, the populations have been relatively small and centered around a particular institution. In contrast, CMAS will establish a large­scale opt­out system that covers much of the general population. Following is a list of unanswered ques­ tions about how the general public might respond to CMAS messages: • What are the consequences of too many messages (e.g., if the threshold for events which trigger alerts is set too low, if alerts cover too large a geographical area, if messages are repeated too often, or if there are too many false alarms)? Is there a threshold level that would cause people to ignore the messages or opt out from participating? • What are the consequences of too few messages (e.g., if the thresh ­ old for alerts is set too high or messages are repeated too infrequently)? • How does an alerts and warnings system generate the credibility needed to garner attention and guide the public’s response?

OCR for page 46
 PUBLIC RESPONSE TO ALERTS AND WARNINGS ON MOBILE DEVICES • What level of geographical targeting is needed to make messages relevant? Is the targeting by county or equivalent entities sufficiently precise? • Will the distinctive alert tone and other special features be sufficient to distinguish CMAS alerts from other text messages, including spam? Information-Seeking behavior Given the limited information that can be provided in a CMAS alert, gaining an understanding of what people will do in response to alerts is a central question. Past research has shown that people respond to alerts and warnings by seeking additional information to confirm the event, determine their risk, and decide on their next action. Following is a list of research questions in the area of information­seeking behavior: • Can the information­seeking behavior of people who receive CMAS alerts be predicted? What are the mechanisms for obtaining information and the sources of information that various subgroups use, and what information are they likely to seek? • Will a CMAS alert create a demand for cell­phone­delivered infor­ mation that could overwhelm bandwidth­limited communications chan­ nels such as the cellular networks over which people receive the alerts? For example, when they have received an alert, will people place phone calls to friends or relatives, search for Internet information, or browse news and information Web sites? • How do individuals determine what are credible sources of infor­ mation, and how does that determination differ by group? • Would a pointer (URL) to a bandwidth­conserving official source of additional information actually reduce network congestion compared to the bandwidth used by individuals seeking information on their own? (That is, might some people be satisfied with that additional information, which could be specially tailored to reduce the bandwidth required to deliver it?) • How could authoritative secondary sources best be incorporated into the CMAS program? Social Media New and social media such as Facebook and Twitter are being used both by professionals and by citizens to disseminate information in emer­ gency situations. These tools may be used as a second and confirming source during an alert or warning. Additionally, social media tools are often designed and used to provide short pieces of information, which

OCR for page 46
 RESEARCH GAPS may provide insight on the best use and content for CMAS messages. The following research topics with respect to social media tools for alerts and warnings were identified at the workshop: • To what extent can results of research on social media be applied to gaining an understanding of what the public response to CMAS alerts might be? • What roles will social media play in emergency communications? Will the social media be among the important secondary sources that peo­ ple turn to for information? Will these media ever play a primary or major role in initial alerts? How do they relate to, or complement, CMAS? • How might social media factor into CMAS and other official mes­ sage dissemination? • How will public education initiatives need to be designed to help the public understand and evaluate the usefulness of unofficial informa­ tion sources? Demographics and Access Although the use of cellular telephones is widespread, not everyone owns, carries, or uses a cell phone, and cell phone service is not available in some sparsely populated areas. The following questions on these topics need further research: • What is the current demographic profile of use of mobile devices in the United States? • How does this use vary demographically (i.e., by age, income, ethnicity, gender)? • How does this use of mobile devices vary with populations that have sensory or cognitive impairments? • How do cell phones and text alerts fit into the broader set of com­ munications sources (including interactions with other individuals and community institutions) that communities use to convey information about emergencies? • What are the implications of supporting messages using text mes­ sages in only the Roman alphabet, which the initial rollout of CMAS supports? How can multiple languages best be incorporated into the subsequent phases of CMAS? Context The very nature of mobile devices means that people will receive CMAS messages in a wide variety of settings—including classrooms,

OCR for page 46
0 PUBLIC RESPONSE TO ALERTS AND WARNINGS ON MOBILE DEVICES highways, public trains and buses, and innumerable other places. Addi ­ tionally, travelers who are away from home will receive alerts for hazards with which they may not be familiar. Following are questions needing further research that are related to the context in which people will receive alerts and warnings: • How will an individual’s location affect his or her response to an alert or warning? (For example, what might be the response of a per­ son who is driving compared with that of a person traveling on public transportation?) • How will people deal with messages about which they have not been educated? (For example, how might someone respond who lives on the West Coast and who then travels to “Tornado Alley” in the midsection of the United States and is not familiar with the meaning of “tornado warning” or does not know what the appropriate protective action is in the case of a tornado?) IMPLEMENTATION CHALLENgES The following sections describe challenges ahead for CMAS with respect to public education, the incorporation of CMAS efforts into a broader context of other alerting systems, and testing, piloting, and research with respect to CMAS messaging. Public Education Workshop participants emphasized that an effective educational com­ ponent is a key to introducing new alert technologies and methods. They cited past research showing that educating the public in advance about what actions to take under particular circumstances is key to effective public response. The constrained message context of CMAS places an even greater premium on educating the population in advance about the steps to take to protect themselves, the best places to go for additional information, and ways in which they might assist others. The size and diversity of the populations that will receive CMAS alerts and the diversity of the hazards about which alerts may need to be sent indicate the challenges of developing educational programs with sufficient breadth. These programs will need to explain not only CMAS but also the necessary public responses associated with different types of alerts.

OCR for page 46
 RESEARCH GAPS Incorporation into the broader Context Workshop participants observed that CMAS will need to be effec­ tively incorporated into the broader context that includes other alerts and warnings systems, broadcast media, social media, and so forth. CMAS will have the greatest effect if messaging is consistent across these sources (because inconsistent messages will hamper an effective public response) and if their use is coordinated. (For example, if a CMAS message says to tune to a local broadcast channel for further information, it is impor­ tant for local emergency management officials to have forged good rela ­ tionships with those broadcasters so that the information will in fact be available.) Testing, Piloting and Ongoing Research The brevity of CMAS messages and the new contexts in which they will be used point to the importance of testing and research. It will be helpful to test the effectiveness of the wording of particular messages with test subjects. Before the CMAS program is introduced nationally, pilot programs can be used to determine what messaging is most effective, how CMAS messages can best be coordinated with other alerts and warnings, and so forth. Finally, ongoing research that gathers lessons learned from the early use of CMAS can be used to improve future generations of the program and to inform local, state, and federal officials on best practices for using the system. FuTuRE TOOLS FOR ALERTS Advances in information and communications technologies are creat­ ing new opportunities for disseminating alerts and warnings. Several of these were discussed throughout the workshop, including the following: • Ad hoc wireless networks. Most mobile devices have the capability to access wireless hot spots such as those found in coffee shops, bookstores, and public buildings. These semi­public wireless networks have much more capacity than that of cellular wireless networks. Can these networks be quickly and easily accessed during emergencies to disseminate infor­ mation? Could they be useful as a redundant way of reaching mobile device users? • Mobile devices and social media as information sources for emergency managers. Cell phones can be used by the public to report information from a disaster site. Social media and microblogging sites will continue to be used by those affected by a disaster or crisis. How can emergency managers best accumulate and access the information posted to social

OCR for page 46
 PUBLIC RESPONSE TO ALERTS AND WARNINGS ON MOBILE DEVICES media sites to respond to emergencies more efficiently? What are the privacy implications of doing so? • Location-based services for geographical targeting. Geographic codes used in CMAS to localize messages by county or equivalent jurisdiction will provide a fair approximation of the geographic area and population affected by an emergency. However, much more precise alerting could be provided by attaching more precise geographic information to alerts and making use of the location capability built into cell phones. In what scenarios might such precision be useful? What are potential drawbacks, such as concerns about privacy? • Automation. How could automation be used to provide more timely alerts for highly time­sensitive messages, such as those from Early Earth­ quake Warning Systems? Do technical or procedural factors make it dif ­ ficult to deliver alerts sufficiently rapidly? Might automated systems lead to an undesirable level of false alarms? • Distributed sensing using mobile devices. What is the potential for incorporating sensor devices in widely distributed devices such as cell phones to detect and provide more detailed information on events?