EgressIII results

Statistical methods
Data for travel of emergency egress routes based on the five accessible formats. Remote infrared audible signs (RIAS), push-button verbal directions (PB), tactile map (TM), Braille (Bra), and raised print (RP) were analyzed for time for Familiarization, Time to Begin traveling the route, Travel Time, Total Time, number of times Braille, raised print and push-button verbal directions were read or listened to, and Restarts. Familiarization time was analyzed descriptively. The first analysis for each of the other dependent variables was always a one-way within subjects ANOVA, n=16, with four levels: RIAS, PB, TM, and RV (raised verbal, which aggregated data on Braille and raised print) to determine whether there were any significant differences between any of the four formats.

*NOTE: Two participants took so much time to begin traveling the route (that is, to acquire the necessary information and prepare to travel), that they did not actually travel routes in one input condition. One participant did not travel a route with tactile map information, and another did not travel a route with raised print information.

Then planned contrasts were conducted to determine which pairs of formats resulted in significantly different or marginally different means. Planned contrasts are not typically conducted unless significant differences appear in the overall ANOVA. We have chosen to conduct them, however, because this is the first research comparing different formats for emergency egress information and the results will be used to frame the questions for future research. It is more important in the beginning investigation of any topic to identify any differences that may be significant than to rule out any which may not be significant. That is, where there may be real differences which do not reach statistical significance in an initial investigation of a topic, it is important to investigate further. If these possibly significant differences are not maximally explored in initial research, future research may wrongly assume that there are no differences. Thus, our approach minimizes the likelihood of a type II error, that is, claiming there is no difference when, in fact there is a difference, while there is some increased likelihood of a type I error, that is, claiming there is a difference when there is in fact, no difference.

To explore differences between Brl and RP, a one-way between subjects ANOVA, n=8, was conducted with two levels: Brl and RP. Finally, paired samples t-tests were conducted, where warranted, comparing Braille with TS, PB and TM and comparing raised print with RIAS, PB and TM.

Results of trials
Time to Familiarize
Familiarization consisted of learning how to use each format, reading or listening to the instructions for a practice travel route, and repeating the travel route instruction back to the experimenter. This familiarization required approximately 1/2 hour per participant. It was noted during familiarization of the first five participants that the different formats required very different amounts of familiarization time; familiarization times were subsequently recorded. Mean familiarization times and ranges based on the 11 participants for which familiarization time was recorded are shown in Fig. 9.

Figure 9. Mean and range of times required for subjects’ familiarization with various formats. Mean times in seconds are: TS=705*, PB=56, TM=273, Brl=170, RP=599

*Time included the time needed for traveling a short familiarization route using the remote infrared audible signs, locating and identifying the closed fire doors to the exit stairway, and locating the Tactile exit sign, tasks not included in familiar experimental conditions. These additional tasks were necessary to be sure participants had an adequate understanding of the use of the remote infrared audible sign system.

Results for familiarization are presented descriptively, only. Inferential analysis was not conducted for familiarization because there was too much missing data in the raised print group.

Several participants had seen remote infrared audible signs demonstrated but no participants had previous experience traveling a route using the remote infrared audible signs system. Thus the time for familiarization (mean=705 sec., range=482 - 965) really reflects the time needed for individual instruction in traveling a short route using information provided by the Remote infrared audible sign system. However, unlike familiarization with other formats, additional time was consumed in actually traveling a short route. Also included in the remote infrared audible signs familiarization was experience finding and identifying the closed fire doors to the exit stairway, and locating the Braille and raised print exit sign which they would be used to confirm arrival at the proper exit door in all the experimental conditions. These additional tasks were necessary to be sure participants had an adequate understanding of the use of the remote infrared audible sign system.
Familiarization using the push-button, until the participant was able to indicate the intended route in words and gestures, required the least time (mean = 56.8 seconds, range = 17 - 91). Participants typically listened to the route multiple times; only one participant listened only once before describing the route to the experimenter.

Mean time for familiarization using Braille (mean = 170 sec., range = 49 - 575) was longer than for the push-button. However, if data from one “outlier” is removed from the computation, the mean for the Braille group is 103 seconds. Most participants read the route instructions more than one time. Two participants were unfamiliar with the use of the Braille capital sign.

Familiarization time using the tactile map varied greatly, but along a fairly normal distribution (mean = 273 sec., range = 106 - 630). Many participants said they had had no or little experience with tactile maps. As no participant was able to independently explore the map and its accompanying legend, in either raised print or Braille (depending on their group assignment), and then successfully describe the route to the experimenter, verbal as well as physical assistance was given during the familiarization to assist in understanding the map. Common problems included:

not understanding the word “Legend,” or not understanding the concept of a Legend
not understanding that once they knew the symbol for “you are here” then they must move from the Legend to the map and begin at the “you are here” symbol
not able to determine whether the turns depicted would be left or right turns along the route to the exit
Once participants understood the Legend and found the “you are here” on the map, most were then able to independently understand the nature of the route.

Neither the mean time or range for familiarization using raised print (mean=599 sec., range=126 - 989), based on recorded times from only four participants, reflect very well the observations made by the experimenters of the total 8 subjects. The distribution of recorded and observed (non-recorded) times indicates that three of the eight participants who used raised print for this project were skilled users, requiring about two minutes to read the route and be able to describe it to the experimenter. The other five were very unskilled, requiring more than nine minutes for this task. Some participants did not know that the letter “I” could be written as a straight vertical line (“I”), as it was in the particular sans serif font used for the raised print signs.

Time to Begin Travel (Reading/Listening Time)
A one way within-subjects ANOVA, with data from the Braille and raised print combined to form the measure ‘raised verbal’ (RV) which includes all 16 subjects, (four levels: RIAS, mean=11; PB, mean=63; TM, mean=156; RV, mean=222) revealed a significant difference in time to begin travel, that is, the time required for participants to acquire route information in each format i.e., read the Braille, press the button on the infrared signs receiver, or listen to the audible push-button message) and to begin travel toward the destination (F(3,15) =12.97, p<.001). (see Fig. 10). Planned contrasts revealed significant differences between all conditions except the tactile map and raised verbal formats (p<.005).

Figure 10. Time to begin travel. Mean times in seconds are: RIAS=11, PB=63, TM=156, Brl=108, RP=336, (RV=222)

A one way between-subjects ANOVA (Brl and RP with 8 participants each), also revealed that participants began traveling sooner using Braille (mean=108 sec.) than using raised print (mean=336 sec.) (F(1,14) = 9.31, p<.009). Planned contrasts to determine whether Braille was significantly different from remote infrared audible signs, push-button and tactile map revealed that remote infrared audible signs resulted in significantly faster time to begin travel than Braille (p<.009), but no other pairs were significantly different. Planned contrasts to determine whether raised print was significantly different from remote infrared audible signs, push-button and tactile map revealed that raised print resulted in significantly longer time to begin travel than remote infrared audible signs, push-button, and tactile map (p<.01).

Travel Time (excluding Reading/Listening Time)
A one way within-subjects ANOVA which includes all 16 subjects (four levels: RIAS, mean=345; PB, mean=222; TM, mean=257; RV, mean=272), did not reveal any significant difference in time to travel, that is, the time required for participants to travel to the destination (until they touched the tactile exit sign to confirm that they had located the double doors to the exit). However, planned contrasts revealed that push-buttons required significantly less travel time than remote infrared audible signs (p<.012). (See Figure 11.)

Figure 12. Total time to travel routes for each format used. Includes time to read/listen to egress instructions. Mean times in seconds are: RIAS=355, PB=285, TM=413, Brl=329, RP=660, (RV=495)

To explore the significant within-subjects effect, planned comparisons revealed that the push-button required significantly less travel time than the tactile map (ps<.025), and raised verbal (ps<.022).

A one way between-subjects ANOVA with 8 participants each also revealed a marginal effect showing an advantage of Braille (mean=329 sec.) over raised print (mean = 626 sec.) (F(1,14) = 3.65, p<.077). Planned comparisons to determine whether Braille was significantly different from Talking Sign, push-button and tactile map revealed no significant differences. However, planned comparisons to determine whether raised print was significantly different from remote infrared audible signs, push-button and tactile map revealed that raised print resulted in significantly longer travel time than push-buttons (p<.013), and marginally longer than tactile maps (p<.069).

Number of Times Reading or Listening to Route information
There were no significant differences between the number of times participants fully read or listened to route directions (Braille, raised print and push-button only).

Restarts
Despite the apparent differences shown in Fig 13, one way within-subjects ANOVA which includes all 16 subjects (four levels: RIAS, mean=.2; PB, mean=.6; TM, mean=.6; RV, mean=.4) revealed no significant difference in restarts, that is, the number of times participants asked to be taken back to the beginning of a route so they could try again. There was also no difference in restarts between Braille and raised print.

Figure 13. Mean number of restarts for each format: RIAS=.2, PB=.6, TM=.6, Brl=.5, RP=.3, (RV=.4)

Results of Survey
Fourteen of the 16 participants completed a short, orally administered survey following the route travel evaluation; the other two, both of whom read Braille, left the experiment early because of time constraints. Each participants was first asked to rank order all of the four formats they experienced in terms of how useful they would be if participants were trying to familiarize themselves with how to get out of a building in a non-emergency situation, that is, if they had recently arrived at a building and wanted to know how to get out in case an emergency occurred.

Looking at the number of times each format was ranked 1st, 2nd, 3rd or 4th, the data reveal a pattern which may be useful when viewed in the context of results of other measures (See Table 2). A ranking of 1 indicates the most useful format and a ranking of 4 indicates the least useful format. Remote infrared audible signs was the first choice of more participants (6) than any other format. Push-buttons were ranked in the top half (1st or 2nd) by 9 participants and 4th by none. The mean rankings ranged from 2.4 to 3.0. Mean rankings were computed by multiplying the number of respondents at each rank by the number of the rank, summing these products, and dividing by the number of participants.

Table 2. Numbers of participants who rank ordered formats 1st, 2nd, 3rd or 4th, on the basis of preference for use in familiarizing themselves with emergency egress routes in a non-emergency situation. A ranking of 1 indicates the most useful format and a ranking of 4 indicates the least useful format.

Text description of Table 2:
Numbers of participants who rank ordered formats 1st, 2nd, 3rd or 4th, on the basis of preference for use in familiarizing themselves with emergency egress routes in a non-emergency situation. A ranking of 1 indicates the most useful format and a ranking of 4 indicates the least useful format. Remote infrared audible signs was the 1st ranked (most useful) for more participants (6) than any other format. Push-button were the 2nd ranked of more participants (7) than any other format. Push-button were the 3rd ranked of more participants (7) than any other format. Tactile maps were the 4th rank (least useful) for more participants (5) than any other format. For mean rankings, the lowest mean indicates the highest ranked format. Mean ranking values are: RIAS=2.4, PB=2.5, TM=2.8, Brl=2.4, RP=3.0.

A one way within-subjects ANOVA includes all 16 subjects (four levels: RIAS, mean=2.4; PB, mean=2.5; TM, mean=2.8; RV, mean=2.6) on mean rankings revealed no significant differences in mean rankings. In addition, a one way between-subjects ANOVA with 8 participants each comparing Braille and raised print was also not significant. A ranking of 1 indicates the most useful format and a ranking of 4 indicates the least useful format.

Participants were then asked to rank order the four formats they experienced in terms of how useful they would be if participants did not know how to get out of a building and an emergency occurred. Responses to this condition showed a greater range in mean rankings of 2.2 to 3.3. Mean rankings were computed by multiplying the number of respondents at each rank by the number of the rank, summing these products, and dividing by the number of participants. Looking at the number of times each format was ranked at each level, a change in ranking between non-emergency and emergency situations can be seen. Remote infrared audible signs and the push-button became more preferred (TS top half=9; PB top half=10), while maps became less preferred (top half=3). Braille was still highly preferred by those who used it, while raised print became even less preferred. (see Table 3)

Table 3. Numbers of participants who rank ordered formats 1st, 2nd, 3rd or 4th, on the basis of preference for use in helping them get out of an unfamiliar building in an emergency situation. The lowest mean indicates the highest ranked format.

Text description of Table 3:
Numbers of participants who rank ordered formats 1st, 2nd, 3rd or 4th, on the basis of preference for use in helping them get out of an unfamiliar building in an emergency situation. Remote infrared audible signs (5) tied with push-button (5) for the 1st ranked (more useful) of most participants than other formats. Push-buttons were the 2nd ranked of more participants (5) than any other format. Tactile map (4) tied with push-button (4) for the 3rd rank of more participants than any other participants (7) than any other format. For mean rankings, the lowest mean indicates the highest ranked format. Mean ranking values are: RIAS=2.2, PB=1.6, TM=3.1, Brl=2.3, RP=3.3.

A one way within-subjects ANOVA which includes all 16 subjects (four levels: RIAS, mean=2.2; PB, mean=2.2; TM, mean=3.1; RV, mean=2.7), revealed a marginally significant difference in mean rankings (F(3,39) =2.73, p<.057). Planned contrasts to determine which pairs of rankings were significantly different revealed that push-button information was ranked significantly more useful than tactile maps (p<.023), and marginally more useful than raised verbal (p<.059).

A between subjects ANOVA with 8 participants each comparing Braille and raised print revealed that Braille (mean=2.3) was ranked marginally higher than raised print (mean=3.3) (F(1,12)=.056). Planned comparisons of Braille with the other formats revealed no significant differences. However, planned comparisons of raised print with the other formats revealed that push-button and remote infrared audible signs were ranked significantly higher than raised print (ps<.05).

To obtain some suggestion of their likelihood of using accessible information, participants were asked to rate how likely they would be to actually use their preferred system to familiarize themselves with exit routes during the first day they were in a building (scale, 1=very unlikely—5=very likely). The mean rating was 3.5. However, when asked how they would prefer to learn exit routes in a non-emergency situation, 8 of 13 respondents said they would prefer to have a person assist them. (One participant seemed unable to understand this question and did not answer.)

Participants were also asked to rate how likely they would be to actually use their preferred system to find out how to get out of an unfamiliar building in an emergency situation (scale, 1=very unlikely—5=very likely). The mean rating was 3.7. Ten of 13 respondents said that if other people were around and an emergency occurred in an unfamiliar building they would prefer to get assistance from another person. (One participant seemed unable to understand this question and did not answer.)

When asked for other suggestions or comments, most comments were made by no more than one person. However several participants mentioned that they would be more comfortable with the remote infrared audible signs system if they were always able to hear a message. Several also suggested that having remote infrared audible signs more precisely indicate the location of the exit would be helpful. Suggestions were also made with regard to Braille, the tactile map, and the push-button that it would be helpful if users had individual copies or recordings they could carry along as they traveled.

Results of Focus Group
The 16 persons who participated in the emergency egress route travel evaluation all returned to Boston College to participate in one of two focus groups convened to obtain additional subjective evaluation and suggestions on providing emergency egress information in accessible formats. Each focus group contained persons who had used Braille and persons who had used raised print during the route travel evaluation. Both were conducted using the same protocol. Question topics were raised by the Focus Group leader and discussed by the group before each question was read and rated by each participant. Participants were asked to rate their responses using a 5 point scale by raising a number of fingers corresponding to their rating. No participant was able to see another participant. Results will be presented for both groups together. One participant arrived late and did not respond to the first four questions. Several questions were responded to only by participants who lingered after the official conclusion of each focus group. The total number of participants responding to each question are noted.

Question 1. Although the accessible information provided to participants in the route travel evaluation consisted only of a route to the exit, the California fire code provides that the emergency procedures information shall include, but not be limited to the following information:

Location of Exits
Location of fire alarm initiating stations
What fire alarm sounds and looks like
Fire department emergency telephone number (911)
Prohibition of elevator use during emergencies

Focus group participants were asked whether it was important to provide all the information in accessible format. Participants noted that providing all information in tactile formats would require considerably more space and that it might therefore become difficult to find the needed unique route information to the nearest exit. Two participants suggested that the redundant information could be in a packet which persons who are blind could pick up as they enter unfamiliar buildings.

Participants were asked to rate the importance of providing all information in an accessible format using a 5 point scale (1 = very unimportant—5 = very important), by raising a number of fingers corresponding to their rating. (See Table 4). The mean response was 2.5 indication that inclusion of information was considered relatively unimportant.

Table 7. Criterion for choosing an accessible format.

Two participants said they would be likely to have difficulty remembering route information in an actual emergency; that it would be helpful if, in emergency, accessible route information such as Braille, a tactile map or recorded directions could be carried along. Another participant suggested placing tactile maps on all fire doors rather than on doors from rooms. Another participant suggested that providing information about the type and location of emergency egress information, and general information about the locations of fire doors, in Braille or tape recorded form when a person checks into a facility such as a hotel, and providing route information in accessible format on or beside the doors of rooms would serve both needs well. All participants were concerned that loud alarms might make it difficult or impossible to hear audible directions or speech messages in case of an emergency. The person who was blind and hearing impaired said that she wouldn’t use remote infrared audible signs in an emergency because she wouldn’t be able to hear the messages. She suggested that remote infrared audible signs incorporate a vibratory feature—the receiver would vibrate if aimed in the direction of a sign.

Question 5. The focus group leader (B. Bentzen) suggested that one way of identifying or choosing an accessible format for emergency egress information was to think about what properties were important. A selection of five properties or characteristics was rated by each of 16 participants on a 5 point scale (1 = strongly disagree; 5 = strongly agree). One property was rated by only one group of seven participants because of insufficient time. (See Table 8)

Table 8. Mean ratings of properties of accessible formats for emergency egress information. (1 = strongly disagree; 5 = strongly agree).

Considerations for particular media
Remote infrared audible signs. The message “To the exit,” (the equivalent of an EXIT route sign with an arrow pointing straight ahead) confused many participants even though during training they had heard this message, learned that it indicated that they were going the right way and should keep going that way until given further directions, and practiced continuing past such a sign until coming to another message. Many participants started looking along the walls for an exit as soon as they no longer heard the message.

Participants were asked to suggest a message that would have worked better to keep them traveling straight ahead after they passed under the message and could no longer hear it. The following suggestions were made:

Continue toward exit.
Moving toward exit.
Headed toward exit.
Keep going.

After considerable discussion, each group reached a consensus that the wording was not critical; more experience with remote infrared audible signs would have led them to keep going.

Raised print. During the focus groups participants had the opportunity to read samples of the raised print of the type used in the route evaluation, but having various character spacing (see Fig. 8). They were asked to choose which spacing was most legible to them. A somewhat wider spacing than used in the experiment was preferred by a majority of participants. (See Table 9)

Table 10. Preference for line spacing for tactile map.

Verbal route descriptions (Braille, raised print and push-button). A majority of participants held the opinion that the directions were too verbose. For example, they recommended shortening “Take the next hall going left.” to “Next hall on left.” Use of “stair” was recommended instead of “stairway.”

Summary of results
Results by task (measure):

Familiarization
*NOTE: Familiarization times were not subjected to inferential analysis as they were not systematically obtained for the duration of the experiment, and there was too much missing data for the raised verbal format.

For remote infrared audible signs, as was the case for PB, participants readily understood the spoken messages. However, familiarization with the unfamiliar technology which presents a series of messages along a route instead of describing a route at the onset, required time to travel a short route. All the other formats presented the entire route at the outset. When being familiarized with all other formats, participants were asked only to describe the route, not to travel it, to demonstrate their understanding. Familiarization using remote infrared audible signs required the most time.

Auditory push-button was the easiest to use as reflected by the shortest familiarization time.

In order to read the legends of a tactile map, the user must know Braille or raised print. Participants found the tactile maps easy to use once they understood it. All participants required at least a minimal amount of individual instruction in order to acquire route information from tactile maps.

Braille required minimal time for familiarization in the context of this research. However, reading Braille requires extensive prior training in the Braille code.

Familiarization using raised print required more than nine minutes for five of the eight participants. Although these participants were able to recognize raised characters, they do not normally do so except occasionally reading of elevator panels and room numbers.
Time to begin travel
Both auditory formats allowed the subjects to start traveling in the shortest amount of time. When choosing which accessible formats will be used for emergency information, from the standpoint of initiating travel most quickly, remote infrared audible signs was objectively the fastest, significantly faster than push-button, tactile map, Braille, and raised print. This is because participants did not need to pause to read or listen to a list of travel instructions to be either memorized or converted into a cognitive map.

The push-button was second fastest, significantly faster than TM and RP, but significantly slower than remote infrared audible signs.

Travel time
When considering the speed at which subjects traveled based upon different formats used, push-button and Braille required significantly less time than remote infrared audible signs.

Total time to travel (Reading/Listening time + travel time)
When considering total time to accomplish the task of locating the emergency exit door, (including Reading/Listening time and travel time), push-button was significantly faster than the tactile map or the raised print. Braille was also significantly faster than raised print.

Number of times reading or listening to route directions
There were no significant differences between the number of times participants fully read or listened to route directions (Braille, raised print and push-button only).

Restarts
There were no significant differences in number of restarts.

Non-emergency vs. Emergency situations
There was no significant difference in mean ranking of formats for use in learning the exit route in a non-emergency situation. However, inspection of the ranking data shows that RIAS was the first choice of more participants than other formats and that push-buttons was ranked in the top half (1st and 2nd) by 9 (of 13) participants.

Mean ranking of the same formats for usefulness in an emergency situation revealed marginally significant differences where PB was preferred over the next most useful format, RIAS. Inspection of the ranking data shows RIAS and PB to both be ranked in the top half (1st and 2nd rank) by 9 and 10 (of 13) participants, respectively.

Results by format type:

Based on results of this study, assuming that the visually impaired person has found the accessible sign, it appears that they can get to exits using all formats tested. However, the data show that participants’ efficiency in getting to exits does vary according to the format, and some formats are preferable to others.

Remote infrared audible signs resulted in the shortest times to begin to travel. Remote infrared audible signs did require longer travel times than the push-button and Braille. However total travel time did not differ significantly between remote infrared audible signs and the push-button. Preference for an accessible format in non-emergency and emergency conditions also did not differ between remote infrared audible signs and the push-button.

Push-button speaker route information resulted in the fastest time to travel and fastest total time. It was preferred for use in non-emergency and emergency situations.

Tactile maps were one of the slower means of familiarization and travel and one of the least preferred formats for prior familiarization and in emergency. Tactile maps were wall mounted for the study.

Braille resulted in faster Times to Begin and Total Time than raised print, but never as fast as the push-button. Braille required considerably less time for familiarization than raised print, and it was also preferred above raised print for use in both non-emergency and emergency situations.

Raised print required rather long times for familiarization and longer total travel times. It was preferred by only one participant for familiarization prior to an emergency and the preference of no participant for use in actual emergencies.

Summary statistic: “Goodness”

In attempting to provide an objective performance measure for each format, a statistic called “Goodness” was developed to collapse the ranking of various measures (i.e., Time to Begin, Travel Time, Total Time, preferences in non-emergency and in emergency situations) into a single measure. This statistic was generated by summing the number of planned comparisons “wins” of each format over other formats (For example, in the measure of Time to Begin, PB “won” a significant planned contrasts difference over TM, Brl and RP (3 contrasts) and so the PB score for this measure is three. Remote infrared audible signs won Planned contrasts over PB, TM, Brl, and RP (4 contrasts) and so the RIAS score is four. On the measure of preference in non-emergency situations, when formats were compared against each, there were no “winning” formats and so all comparisons equal zero.) It is left to the reader to determine how much weighting should be assigned to each of the measures in computing the statistic. The weighting factor shown in Table 11 is one, meaning that the ranking of each measure is treated as being equally important.

Scale	# Participants
1 = no problem	1
2	2
3	1
4	4
5 = a very big problem	7
	n = 15

Scale	# Participants
1 = not at all important	0
2	0
3	1
4	2
5 = very important	12
	n = 15

Criterion	# Participants
Ideal format for use in non-emergency situations	9
Equal consideration for use in non-emergency and emergency situations	4
Ideal format for use in emergency situations	2
	n = 15

Properties	Mean rating
It should be very easy to find the information.	4.9 (n=16)
I should be able to acquire the information and get moving quickly.	4.4 (n=16)
I should be able to start anywhere. That is, I should be able to get the information as I go along and not have to find it at some starting place.	3.9 (n=16)
The system should provide the whole route at the beginning.	3.4 (n=16)
The system should be self-correcting, that is, if I get off track something will tell me how to continue. I don’t have to try to find my way back to a beginning.	3.2 (n=9)

Spacing (kerning factor)	# Participants
Narrowest (.8)	2
Somewhat narrow (.9) [Used in route evaluation]	3
Somewhat wide (1.0)	7
Widest (1.1)	3
	n = 15

	# Subjects	1st	2nd	3rd	4th	Mean Rank Order
Remote IR Audible audible signs	n = 14	6	1	3	4	2.4
Push-button	n = 14	2	7	5	0	2.5
Map	n = 14	3	2	4	5	2.8
Braille	n = 8	2	3	1	2	2.4
Raised Print	n = 6	1	1	1	3	3

	# Subjects	1st	2nd	3rd	4th	Mean Rank Order
Remote IR Audible signs	n = 14	5	4	2	3	2.2
Push-button	n = 14	5	5	4	0	1.6
Map	n = 14	2	1	4	7	3.1
Braille	n = 8	2	3	2	1	2.3
Raised Print	n = 6	0	1	2	3	3.3

Scale	# Participants
1 = very unimportant	5
2	4
3	2
4	2
5 = very important	2
	n=15

Spacing between lines for walls (in.)	# Participants
Narrowest (.250”) Middle (.375“) Widest (.500“) [Used in route evaluation]	1 7 5
	n=13

	Weighting	RIAS	PB	TM	Brl	RP
Time to Begin	1	4	3	1	1	0
Travel Time	1	0	2	1	1	0
Total Time	1	0	2	1	1	0
In Non-Emergency	1	0	0	0	0	0
In Emergency	1	1	3	0	1	0
“Goodness” (Total)	1	5	10	3	4	0