5 page literature review

Full Terms & Conditions of access and use can be found at https://www.tandfonline.com/action/journalInformation?journalCode=wasr20

Journal of Applied Security Research

ISSN: 1936-1610 (Print) 1936-1629 (Online) Journal homepage: https://www.tandfonline.com/loi/wasr20

Students, Faculty, and Staff's Willingness to Pay for Emergency Texting

Manako Yabe

To cite this article: Manako Yabe (2016) Students, Faculty, and Staff's Willingness to Pay for Emergency Texting, Journal of Applied Security Research, 11:4, 437-449, DOI: 10.1080/19361610.2016.1210486

To link to this article: https://doi.org/10.1080/19361610.2016.1210486

Published online: 08 Sep 2016.

Submit your article to this journal

Article views: 91

View related articles

View Crossmark data

Citing articles: 2 View citing articles

JOURNAL OF APPLIED SECURITY RESEARCH , VOL. , NO. , – http://dx.doi.org/./..

Students, Faculty, and Staff’s Willingness to Pay for Emergency Texting

Manako Yabe

Department of Disability and Human Development, University of Illinois at Chicago, Chicago, Illinois, USA

KEYWORDS Text-to-; emergency texting; contingent valuation method; campus safety; willingness to pay; university

ABSTRACT Although emergency texting benefits callers’ safety during vio- lent situations and campus shooters, the service had not been popularized at a public university. To introduce emergency tex- ting, this study proposed to identify the economic benefits of emergency texting and used the contingent valuation method which estimates students, faculty, and staff’s willingness to pay for emergency texting. Despite the small sampling size, the total val- ues of the three groups’ willingness to pay were $7,736 per year, larger than the lowest implementation costs of text-to-911. The estimation of the overall university population’s total values may be expected to be greater than the results.

Introduction

New generation for text-to-911

The Federal Communications Commission (FCC) created a commission to ensure all individuals access to emergency services with next generation technologies: emergency texting services called text-to-911 (Federal Communications Commis- sion [FCC], 2014a). The Commission has recognized that many people believe that text-to-911 is beneficial for not only individuals with hearing loss or speech disabil- ities, but also for nondisabled public populations when a voice call is not safe due to being overheard by offenders, or is unable to connect due to poor transmissions (FCC, 2014a). However, the change to text-to-911 is still in the beginning stages because the text-to-911 service is available in only a few locations across the United States (FCC, 2015b).

Background

Historical evolution of text-to-911

In September 2011, the Commission released Notice of Proposed Rulemaking which proposed the deployment of New Generation 911 and the introduction of

CONTACT Manako Yabe myabe@uic.edu Department of Disability and Human Development, University of Illinois at Chicago,  West Roosevelt Road, Chicago, IL , USA. ©  Taylor & Francis

438 M. YABE

text-to-911 (FCC, 2014b). In December 2012, AT&T, Sprint Nextel, T-Mobile, and Verizon Wireless made an agreement with the National Emergency Number Asso- ciation (NENA) and APCO International (APCO) to be capable to support text-to- 911 services for emergency call centers as Public Safety Answering Points (PSAPs) by May 15, 2014, which is known as the Carrier-NENA-APCO Agreement (FCC, 2012a).

In December 2012, the Commission ordered that all Commercial Mobile Radio Services (CMRS) providers and covered text providers must support text-to-911 in all areas throughout the country where PSAPs are capable of receiving texts (FCC, 2012b). Furthermore, in May 2013, the Commission required both CMRS providers and covered text providers to set up an automatic bounce-back message when text-to-911 service was unavailable so that many callers who attempted text- to-911 would know immediately when texting was not able to be delivered (FCC, 2013).

In January 2014, the Commission adopted a policy statement that all text message service providers should support text-to-911, enabling sending texts using num- bers from the North American Numbering Plan (FCC, 2014c). Recently, in April 2016, the Commission proposed that real-time text (RTT) technology replace text- telephone (TTY) devices for deaf/hard of hearing people on wireless phone net- works in order to become an efficient transition for text-to-911 services, starting in December 2017 (Jackson, 2016). This helps deaf/hard of hearing callers to con- tact the PSAPs directly, instead of through indirect communication through TTYs by callers and communication assistants (Emergency Access Advisory Committee [EAAC], 2013).

Popularization of text message services

In 2011, 83% of American adults owned cell phones, and of these, 73% commu- nicated via text messages. Within the text message users, 31% of them prefer tex- ting rather than calling, and young adults were the most frequent text users (Smith, 2011). Nearly 97% of smartphone owners, especially young adult owners, used text messages (Smith, 2015). Smartphone owners preferred to use short message services because of three main reasons: speed of delivery, reach, and reliability beyond voice message services (Acision, 2012).

Moreover, the Annual Wireless Industrial Survey reported that of the total num- ber of 129.8 million text message users, 24% were ages 25 to 34, and 49% of text users were male (CTIA, 2015a). The survey reported that monthly annual text messages increased by 163.9 billion in December 2014 from 153.3 billion in December 2013, while monthly annual voice minutes of use decreased by 204.6 billion in December 2014 from 218.2 billion in December 2013 (CTIA, 2015b). The Commission also reported that the average rate of voice use minutes for the major CMRS providers had decreased since 2009 (FCC, 2010). The Commission has supported to expand text-to-911 service due to these statistics and the increasing numbers of text message service users over years.

JOURNAL OF APPLIED SECURITY RESEARCH 439

Emergency texting for people with disabilities

Text-to-911 service provides the emergency service access for people with disabil- ities. Nearly 48 million people in the United States have hearing loss, which is the third most common physical condition (John Hopkins Medicines, 2011), and 7.5 million people have speech disabilities (National Institute on Deafness and Other Communication Disorders [NIDOCD], 2010). In addition, 60% of veterans return- ing from Iraq and Afghanistan came home with hearing loss or tinnitus (Hearing Health Foundation, 2015). Aging is also linked to hearing loss, with 25% of indi- viduals aged 65 to 74 and 50% of individuals aged 75 and older having hearing loss (NIDOCD, 2015).

Since few PSAPs are available to provide text-to-911 service, individuals who have hearing loss or speech impairments would face barriers to calling 911 unless they are with someone who could help to complete a voice call. TTY or Video Relay Services for deaf/hard of hearing also has limitations, because callers would have to call translators first, then the translators would call the PSAPs, causing delays and potential translation errors for emergency situations (EAAC, 2013). Thus, text- to-911 service becomes essential for these populations to obtain direct emergency communication access to PSAPs.

Emergency texting for general populations

Text-to-911 service also benefits general populations without disabilities. For exam- ple, when the shooting occurred at Virginia Tech in 2007, students attempted to send texts to 911 as they hid from the shooters on campus. However, the texts were never delivered to the PSAPs because the PSAPs did not have the capability to receive texts at that time (Kleinman, 2014).

Black Hawk County, Iowa carried out a text-to-911 trial, and they found that text-to-911 was useful for callers in domestic and child abuse situations because the callers were fearful of being overheard by offenders during their calls to 911 (Intrado, 2011). Furthermore, Vermont successfully conducted a text-to-911 trial which demonstrated that text-to-911 was efficient to protect callers who were involved in suicide and domestic violence situations (Ortel, 2014).

Text to 911 service is necessary, not only for these types of situations, but also for emergency situations such as disasters, where landline and mobile voice networks become overloaded and make it difficult to place a 911 voice call (FCC, 2015a). Peo- ple who are in disaster areas could use the option of text-to-911 when they are unable to place a voice call (FCC, 2015a).

Remaining lack of text-to-911 in targeted city’s university

However, text-to-911 service is only available in specific areas where PSAPs accept emergency text messages from the public (FCC, 2015b). Unless callers are aware that PSAPs in their areas support text-to-911, they cannot rely on text-to-911 (FCC,

440 M. YABE

2015b). Unfortunately, the largest city in the United States, where I conducted this research project at a public university, did not have PSAPs that had the capability of text-to-911 at that time. Additionally, the public university did not have their own emergency texting services at that time. The university responded to their own emergency voice calls, and they also provided text and e-mail alerts in various emer- gency and nonemergency situations. Meanwhile, students, faculty, and staff mem- bers had only one choice of contacting the campus police or the city’s PSAPs via voice message, but not through text messages. Therefore, a proposal for this research project was made in order introduce emergency texting services at the public uni- versity.

Purpose of the study

Target population

In this respect, it was proposed to examine the students, faculty, and staff’s values toward emergency texting in order to introduce emergency texting services at the public university where PSAPs did not have the capability to receive text-to-911. If the results demonstrated that the emergency texting had more benefits than its intro- duction costs, the university could make a positive choice to implement emergency texting on campus. There have been no previous studies regarding an analysis of individuals’ values on either emergency texting or text-to-911. Therefore, this study adapted the use of the contingent valuation method and applied it to the sample group’s value toward emergency texting.

Contingent valuation method

The contingent valuation method (CVM) is commonly used as a nonmarket val- uation method for estimating the benefits of environmental goods and services. CVM asks a hypothetical question to individuals and estimates their values for better services or products (Mitchell & Carson, 1989). CVM has different types of ques- tionnaire versions. One example of this is regarding a payment card, which asks individuals to choose bid prices that they would be willing to pay as a maximum. However, this method has a disadvantage in that the respondents may give a will- ingness to pay (WTP) amount that differs from their true WTP amounts (Mitchell & Carson, 1989).

Thus, this study adopted the dichotomous choice CVM (DC-CVM) method. DC-CVM gives the bid prices and asks respondents to choose between only two choices of “Yes” or “No” for the bid prices. This strategic plan provides incentives for the truthful revelation of preferences under certain conditions, and the proce- dure minimizes nonresponse and avoids outliners (Bateman et al., 2002). In addi- tion, responding “Yes” or “No” to the bid prices is similar to marketing behaviors, and reduces the cognitive burden placed on respondents. For its disadvantages, the

JOURNAL OF APPLIED SECURITY RESEARCH 441

method requires a large sampling size to attain a given level of precision (Hanemann, Loomis, & Kanninen, 1991).

Research questions and hypotheses

The literature reviews present data that emergency texting benefits not only users who have hearing loss or speech impairments, but also general users who are hear- ing, but are unable to speak in violent situations or who attempt suicide (FCC, 2014a). If the research proved that not just deaf/hard of hearing users, but also hear- ing users would need emergency texting, it could become beneficial for every pop- ulation.

Furthermore, a majority of students and faculty may tend to stay late in the evening at the university, due to evening classes and final exam schedules. The staff may have a regular schedule and tend to leave earlier. Compared to staff, students and faculty may be more likely to consider their safety and demand emergency tex- ting. Therefore, the research question and two hypotheses were proposed as follows:

Research Question: Which populations at the public university benefit from this service? Hypothesis 1: Students and faculty have a higher WTP for emergency texting than does the staff. Hypothesis 2: Not only deaf/hard of hearing users, but also hearing users have a higher WTP for emergency texting.

If the previous hypotheses were accepted and the all groups’ total values of emer- gency texting were higher than the actual introduction costs of emergency texting, it could emphasize the importance of implementing an emergency texting service at the public university.

Method

Theoretical model

DC-CVM is based on the random utility theory, which assumes that choices are based on utility comparisons between alternatives, and that the alternative showing the highest utility would be the preferred choice (Hanemann, 1984).

The probability of “Yes” answers generally decreases as the bid price increases under a hypothetical market scenario (Lee, 1997). In random utility theory, an increasing rate of utility (�V) presents the bid prices (BID) and exploratory vari- ables (x) described as:

�V = α + βBIDBID + S∑

i=1 βixi

The estimation hypothesizes the relationship between specific distribution func- tions and the probability of “Yes” answers to the bid prices. In this study, the logistic

442 M. YABE

function (1) is formed as follow:

Pr(Yes) = F(�V ) = 1/ [ 1 + exp (−�V )

] = 1/

{ 1 + exp

[ −(α + βBIDBID +

S∑ i=1

βixi)

]}

Pr(Yes) = Pr (Rk = 1) = F (�Vk) Pr(No) = Pr(Rk = 0) = 1 − F(�Vk) (1)

BID is the bid prices that respondents are asked to pay, βBID is the coefficient of BID, and S is the total number of exploratory variable xi, such as respondents’ gender, income, or age. βi is the coefficient of variable xi.

The logistic function (2) is estimated with a maximum likelihood. Rk is an indi- cator variable that represents that “Yes” is equal to 1, and that “No” is equal to 0. k is a number of respondents that represents a product of probability of their choice behaviors for the proposed prices, and N is the total number of observations.

ln L = N∑ k=1

{Rk ln (F (�Vk)) + (1 − Rk) ln (1 − F (�Vk))} (2)

ln L calculates a log-likelihood. The logistic function (2) estimates a maximum log-likelihood by finding a value of parameters: α̂, β̂BID, and β̂ and by maximizing the ln L.

Using the previous logistic function (2), the logistic function (3) estimates a prob- ability of acceptance for the bid prices. Additionally, the mean WTP is generally calculated through the integral calculus of the probability function.

Pr(Yes) = 1

1 + exp [ −(α̂ + β̂BIDBID +

∑ β̂ixi)

] (3)

DC-CVM question

In the survey, DC-CVM question asked respondents to imagine that a university offers emergency texting services in the future:

Imagine that the university is considering adding emergency text messaging. This service would be paid for by a fee charged to staff, students, and faculty at the University. This fee would be automatically deducted from the paychecks of faculty and staff and as part of the fees that students pay with their tuition. Under these circumstances, would you be willing to pay $____ for an emergency text messaging service, or would you not be willing to pay $____ for an emergency text messaging service?

� Yes, I would be willing to pay. � No, I would not be willing to pay. � I don’t know.

The bid prices were $1, $2, $3, $5, and $10. One of five questions was randomly presented in the web survey. The results of the answers are substituted for the logistic

JOURNAL OF APPLIED SECURITY RESEARCH 443

function (2). In that case, “Yes, I would be willing to pay” is placed as YES, and “No, I would not be willing to pay” and “I don’t know” are placed as NO.

To examine whether the respondents’ answers for DC-CVM questions are associ- ated with their social attributes, the respondents were also asked about their campus emergency experiences, interests toward emergency texting services, gender, citizen status, hearing level, age, and total household income in a normal year.

Characteristics of the data and the empirical model

Summary of survey

The study subjects were students, faculty, and staff from a public university in the United States (see Table 1). According to this university population, 27,969 students, 1,917 faculty, and 11,599 staff were registered in the academic year 2014–2015. Web surveys were sent to all students, faculty, and staff via mass e-mail. To increase the response rate, the web surveys were sent three times during the survey period.

Within 1,638 collected samples, 1,505 samples were used for data analysis, exclud- ing uncompleted responses. Specifically, responses from 929 students, 125 faculty, and 451 staff were used for data analysis. The respondent ratio of students to faculty to staff were nearly equal to 6: 1: 3, as almost the same ratio representing the overall university population. The respondent ratio of male to female was about equal to 3: 7, as compared to the approximate ratio of 4: 6 for the overall university population.

It was unknown how many students, faculty, and staff were actually registered on the university mass e-mail list because some of them might have opted out of the mass e-mail list, so an accurate response rate was unable to be calculated. If the estimated ratios of the overall respondents compared to the overall university population, a rough response rate would be assumed to be 5%.

The empirical model

This study describes the definitions of explanatory variables as listed (see Table 2). The explanatory variables of the empirical model were recoded into dummy vari- ables or units.

BID represented a proposed price for emergency texting. The five levels were set at $1, $2, $3, $5, and $10. VICTIM represented a witness or participant who

Table . Characteristic of survey participants.

Characteristic All Student Faculty Staff

Results of Survey Male     Female ,    n ,   

Survey Period // ∼ // University Population: Academic Year – Male , , , , Female , ,  , n , , , ,

444 M. YABE

Table . Explanatory variable and definition of logit model.

Variable Defmition

BID Proposed prices for emergency texting ($, $, $, $, $) VICTIM Witness or participant for emergency on campus ( = Experienced;  = Never Experienced) INTRST Emergency texting (Interested = ;  = Not Interested) FEMALE Gender ( = Female;  = Male) INTL International ( = International;  = American) DHH Deaf or hard of hearing ( = DHH;  = Hearing) AGE Age of the respondent in years INC Total household income for a normal year (thousand $)

experienced emergencies on campus. The two levels were set at 1 = Experienced and 0 = Never experienced. INTRST represented respondents who were interested in using emergency texting. The two levels were set at 1 = Interested and 0 = Not interested. FEMALE represented gender. The two levels were set at 1 = Female and 0 = Male. INTL represented international students, faculty, or staff. The two levels were set at 1 = International and 0 = American. DHH represented deaf or hard of hearing respondents. The two levels were set at 1 = Deaf or Hard of Hearing and 0 = Hearing. AGE represented the ages of the respondents in years. INC represented the total household incomes for a normal year at values of a thousand dollars.

Characteristics of respondents

By following the previous definitions of exploratory variables, the next section describes the characteristics of the respondents (see Table 3).

Respondents were 62% students, 8% faculty, and 30% staff. According to Table 3, 22% of respondents had experienced campus emergencies, and 89% of respondents were interested in emergency texting services. Specifically, 71% of the respondents were females and 29% were males, and 90% of the respondents were American, while 10% were International. Only 5% of the respondents were deaf or hard of hearing.

The mean of the entire group’s age was 25.8 years old. Within the groups, the mean for the students’ age was 19.6 years old, the mean for the faculty’s age was 38.6 years old, and the mean of the staff’s age was 34.9 years old.

The mean of all groups’ total household income was $63,482, while the mean of students’ total household income was $49,357, the mean of the faculty’s total

Table . Mean and standard deviation of explanatory variable.

All Student Faculty Staff

Variable M SD M SD M SD M SD

BID . . . . . . . . VICTIM . . . . . . . . INTRST . . . . . . . . FEMALE . . . . . . . . INTL . . . . . . . . DHH . . . . . . . . AGE . . . . . . . . INC . . . . . . . . n    

Note. M = mean. SD = standard error.

JOURNAL OF APPLIED SECURITY RESEARCH 445

Table . Results of logit model estimation.

All Student Faculty Staff

Variable Coeff. SE Coeff. SE Coeff. SE Coeff. SE

Constant .∗∗∗ . .∗∗∗ . − . . − .∗∗ . BID − .∗∗∗ . − .∗∗∗ . − . ∗∗∗ . —. ∗∗∗ . VICTIM .∗ . .∗ . .∗∗∗ . − . . INTRST .∗∗∗ . .∗∗∗ . .∗∗∗ . . ∗∗∗ . FEMALE . . . . . . .∗ . INTL − .∗ . − .∗ . − . . − . . DHH . . . . . . . . AGE − . . .∗ . − . . − . . INC .∗∗∗ . .∗ . . . .∗∗∗ . n     log likelihood −. −. −. −.

Note. ∗p <.., ∗∗p <.., ∗∗∗p <.. Coeff. = coefficient. SE = standard error.

household income was $113,015, and the mean of the staff’s total household income was $78,850.

Estimated results

Estimated results of the logit model analysis

The logit model was undertaken to see the significant levels for WTP, the mean WTP, and confidence intervals by using NLOGIT 5.0 (2012).

If respondents answered “Yes” to the proposed prices for emergency texting, it indicated that they had a positive WTP for emergency texting to improve campus safety.

Table 4 represents the estimated results of the logit model to see which respon- dents’ exploratory variables affected their WTP for emergency texting for the full group as well as for the three individual groups.

The estimated coefficient of BID is β̂BID. It was negative and significant at p < 0.01 for all groups, and indicated that the amount of bid prices increased as the rate of utility levels decreased. The estimated coefficient of VICTIM was positive and significant at p < 0.10 for all groups and for students. It was also positive and significant at p < 0.01 for faculty. Students and faculty who have experienced campus emergencies had a high WTP for emergency texting. The estimated coefficient of INTRST was positive and significant at p < 0.01 for all groups. Students, faculty, and staff who were interested in emergency texting had a high WTP for emergency texting.

The estimated coefficient of FEMALE was positive and significant at p < 0.10 for staff, especially for female staff members who had a high WTP for emergency texting. The estimated coefficient of INTL was negative and significant at p < 0.10 for all groups and for students. The American groups had a higher WTP for emergency texting than did the international groups. The estimated coefficient of DHH showed no significant levels. The estimated coefficient of AGE was positive

446 M. YABE

Table . Estimate of respondents’willingness to pay for emergency texting.

All Student Faculty Staff

Mean . . . . % CI .∼. .∼. .∼. .∼.

Note. CI = confidential interval.

and significant at p < 0.10 for students. Students who had higher ages in years also had a high WTP for emergency texting. The estimated coefficient of INC was positive and significant at p < 0.05 for all groups and for staff, and at p < 0.10 for students. Students and staff members had a high WTP for emergency texting, although the mean of the students and staff’s incomes were lower than the faculty’s incomes.

Estimation of mean WTP

The mean WTP is generally calculated through the integral calculus of the proba- bility function. To avoid overestimation and to determine the mean WTP and con- fidence intervals, the probability curve was calculated by numerical integrations, ranging from $0 to the maximum bid amount of $10. Confidence intervals for the mean WTP were undertaken using the method proposed by Park, Loomis, & Creel (1991).

Table 5 presents the estimated results of the mean WTP and confidence internals. As a result, the mean of all of the group’s WTP is $5.14. Within the three groups, the mean of the student’s WTP is $5.10; the mean of the faculty’s WTP is $4.84; and the mean of the staff’s WTP is $5.29. The mean WTP of each group was out of the other groups’ 95% confidence intervals. There were no significant levels for the mean WTP between each of the groups.

To avoid overestimation, the total values of emergency texting services were calculated by the number of completed respondents and the mean of all of the group’s WTP. As a result, the total values of emergency texting services were at least $5.14 × 1505 = $7,736 per year.

Conclusion

Discussion

To summarize, the first hypothesis of “Students and faculty have a higher WTP for emergency texting than the staff does” was not accepted. The characteristics of the respondents who have high WTP for texting emergency services were the follow- ing: (a) they were interested in a texting emergency service, (b) they had experi- enced campus emergencies, (c) they were older in age, (d) they had a high per capita income, and (e) they were Americans rather than Internationals. The one common reason that students, faculty, and staff members were interested in emergency tex- ting was in order to secure their own safety.

JOURNAL OF APPLIED SECURITY RESEARCH 447

Furthermore, the second hypothesis of “Not only deaf/hard of hearing users, but also hearing users, have a higher WTP for emergency texting” was partially accepted. The estimated results of logit model analysis showed that the estimated coefficient of DHH had no significant levels within any of the groups. It clearly indi- cated that there was no connection between hearing loss and emergency texting. Additionally, 89% of respondents were interested in emergency texting, and 52% of them preferred to call without letting the offenders know of their emergency calls.

Despite the limited sampling size, within the small numbers of respondents who completed the survey, the total values of texting emergency services were at least $7,736 per year, and approximately $30,000 for 4 years. If the study included the rest of the respondents who did not participate in the survey but had shown inter- est in emergency texting, the overall university populations’ total values of texting emergency would be expected to be higher than the result throughout the years.

Moreover, to collect estimate costs for text-to-911 implementation, a few compa- nies were contacted that provided text-to-911 services for traditional phone compa- nies, wireless carriers, cable operators, public safety and government agencies, but it was impossible to set an exact introduction cost for text-to-911 for a city or pub- lic university, as every state and area had different costs to implement the service. For instance, wireless carriers pay the lowest implementation costs, charging only several cents a month per cell tower. The wireless carriers also pay for circuits to connect the companies through Internet virtual private networks, incurring costs which range from $1,000 to $50,000. One option would be to implement a PSAP where the wireless carriers could use free web browser delivery, at a cost which only ranges from $25 to $300 a month for delivery to the PSAP, depending on the delivery options. There are no fees charged for customers to use text-to-911.

Compared to the estimate costs of text-to-911 implementation, the students, fac- ulty, and staff’s total values for emergency texting were larger than the lowest imple- mentation costs of the wireless carriers’ delivery to a PSAP to provide text-to-911 services. These results could be viewed as emphasizing the importance of introduc- ing an emergency texting service at the public university.

For limitations, this study was unable to calculate a response rate because it was unknown how many students, faculty, and staff were registered on the mass e-mail list. More than 90% of the respondents had smartphones and had interests in emer- gency texting services. Meanwhile, the rest of respondents who did not have smart- phones or had no interests on emergency texting services limited the data analysis. Additionally, as the data were also collected from only one public university, it may include characteristics which differ from other public and private universities. How- ever, this study’s information could benefit other universities who plan to introduce an emergency texting to improve campus safety in the future.

References

Acision. (2012, May 9). US research by Acision shows SMS is still the king of messaging with speed, reach and reliability named as key reasons for usage over alternative services.

448 M. YABE

Retrieved from http://www.marketwired.com/press-release/us-research-acision-shows-sms- is-still-king-messaging-with-speed-reach-reliability-1654679.htm

Bateman, I., Carson, R. T., Day, B., Hanermann, M., Hanley, N., Hett, T., … Elgar, E. (2002). Economic valuation with stated preference techniques: A manual. Cheltenham, UK: Edward Elgar Publishing, Inc.

CTIA. (2015a, March 17). Text message statistics. Retrieved from http://www.statisticbrain. com/text-message-statistics/

CTIA. (2015b, June 1). Annual wireless industry survey. Retrieved from http://www.ctia.org/ your-wireless-life/how-wireless-works/annual-wireless-industry-survey

Emergency Access Advisory Committee. (2013). Proposed procedures for the TTY as a text ter- minal in legacy 9-1-1 PSAPs without IP connection. Retrieved from https://apps.fcc.gov/ edocs_public/attachmatch/DOC-321704A1.pdf

Federal Communications Commission. (2010). Implementation of section 6002(b) of the omnibus budget reconciliation act of 1993. Retrieved from https://apps.fcc.gov/edocs_public/ attachmatch/FCC-10-81A1_Rcd.pdf

Federal Communications Commission. (2012a). Chairman Genachowski announces com- mitments to accelerate text-to-911. Retrieved from https://apps.fcc.gov/edocs_public/ attachmatch/DOC-317786A1.pdf

Federal Communications Commission. (2012b). Text-to-911 further notice of proposed rule- making. Retrieved from https://www.fcc.gov/document/text-911-further-notice-proposed- rulemaking

Federal Communications Commission. (2013). Text-to-911 bounce-back message order. Retrieved from https://apps.fcc.gov/edocs_public/attachmatch/FCC-13-64A1_Rcd.pdf

Federal Communications Commission. (2014a). Second report and order and third fur- ther notice of proposed rulemaking. Retrieved from http://transition.fcc.gov/Daily_ Releases/Daily_Business/2014/db0822/FCC-14-118A1.pdf

Federal Communications Commission. (2014b). Facilitating the deployment of text-to-911 and other next generation 911 applications, framework for next generation 911 deploy- ment. Retrieved from https://apps.fcc.gov/edocs_public/attachmatch/DA-14-638A1_Rcd. pdf

Federal Communications Commission. (2014c). Text-to-911 policy statement and second FNPRM. Retrieved from https://apps.fcc.gov/edocs_public/attachmatch/FCC-14-6A1_Rcd. pdf

Federal Communications Commission. (2015a). Emergency communications. Retrieved from https://www.fcc.gov/consumers/guides/emergency-communications

Federal Communications. Commission. (2015b). Text-to-911: Quick facts & FAQs. Retrieved from https://www.fcc.gov/consumers/guides/text-911-quick-facts-faqs

Hanemann, W. M. (1984). Discrete/continuous models of consumer demand. Econometrica, 52(3), 541–561.

Hanemann, W. M., Loomis, J., & Kanninen, B. (1991). Statistical efficiency of double-bounded dichotomous choice contingent valuation. American Journal of Agricultural Economics, 73(4), 1255–1263.

Hearing Health Foundation. (2015). Hearing loss & tinnitus statistics. Retrieved from http://hearinghealthfoundation.org/85

Intrado. (2011). 9-1-1 text messaging helps save lives in Black Hawk County. Retrieved from https://www.west.com/wp-content/uploads/2016/02/Black-Hawk-County-Case-Study- 1.pdf

Jackson, D. (2016, April 28). FCC takes step to replace TTY with real-time text (RTT) technology, points to text-to-911 adoption. Urgent Communication. Retrieved from http://urgentcomm. com/psap/fcc-takes-step-replace-tty-real-time-text-rtt-technology-points-text-911- adoption

JOURNAL OF APPLIED SECURITY RESEARCH 449

John Hopkins Medicine. (2011, November 14). One in five Americans has hearing loss. Retrieved from http://www.hopkinsmedicine.org/news/media/releases/one_in_five_americans_has_ hearing_loss

Kleinman, A. (2014, May 15). Now you can text 911 in some places. Huffington Post. Retrieved from http://www.huffingtonpost.com/2014/05/15/text-911_n_5329811.html

Lee, C. (1997). Valuation of nature-based tourism resources using dichotomous choice contingent valuation method. Tourism Management, 18(8), 587–591. doi:10.1016/S0261- 5177(97)00076-9

Mitchell, R., & Carson, R. (1989). Using surveys to value public goods: The contingent valuation method. Washington, DC: Resources for the Future.

National Institute on Deafness and Other Communication Disorders. (2010). Statistics on voice, speech, and language. Retrieved from https://www.nidcd.nih.gov/health/statistics/statistics- voice-speech-and-language

National Institute on Deafness and Other Communication Disorders. (2015). Quick statistics about hearing. Retrieved from https://www.nidcd.nih.gov/health/statistics/quick-statistics- hearing

NLOGIT 5. (2012). [Software]. New York, NY: Economic Software, Inc. Ortel, M. (2014, August 8). Vermont leads the nation in text-to-911. Verizon. Retrieved from

http://www.verizonwireless.com/news/article/2014/08/vermont-leads-the-nation-in-text- to-911.html

Park, T., Loomis, J. B., & Creel, M. (1991). Confidence intervals for evaluating benefits estimates from dichotomous choice contingent valuation studies. Land Economics, 67(1), 64–73.

Smith, A. (2011, September 19). Americans and text messaging. Pew Research Center. Retrieved from http://www.pewinternet.org/2011/09/19/americans-and-text-messaging/

Smith, A. (2015, April 1). U.S. smartphone use in 2015. Pew Research Center. Retrieved from http://www.pewinternet.org/2015/04/01/us-smartphone-use-in-2015/