read the report - John Trent

jumentousmanlyInternet and Web Development

Oct 21, 2013 (3 years and 9 months ago)

84 views

Hook’eMap: Effectiveness of Mobile Campus Tour
Applications


Charlie Dunn, Elizabeth Nitsch, Joseph Stern, John Trent

The University of Texas at Austin School of
Information

1616 Guadalupe Suite #5.202

Austin, TX 78701
-
1213

+1 (0) 512
-
471
-
3821

{char610, betsy.nitsch, josephsternwork, john.e.trent}@gmail.com



ABSTRACT


Adaptive hypermedia
,

as
both a
mode to build context awareness
and
a
s a

personalized
mobile
-
application
,

is rare
among tour
applications
.

Tours
designed
with

a
single user architecture

and a
location
-
based approach

are much more common
.

Hook’eMap, a
mobile campus tour application for the University of Texas at
Austin
,

was

de
veloped to offer a better

campus tour

experience

to
users than the self
-
guided paper brochure
currently
available
through the Visitor Center.

By defining
three measures of
engagement: personalization, relevance and
navigation
, we
developed a field ready

mobile
-
application

to compare
against

the
standard paper brochure.

We recruited campus tour participants
and methodologically captured their
field
experiences using both
tour methods.

We investigate
d findings from field studies and
consider
ed

recommendations to advance the working prototype
that was created. Our study

shows
that users did prefer
Hook’e
Map
, finding it to be more personalized
and relevant

to
their interests

than the paper brochure.

General Terms

Design, Human Factors

Keywords

Adaptive Hypermedia, Context Awareness, Location
-
Based
Services, Campus Tour Application, Mobile Development


1.

INTRODUCTION

There are currently a limited number of guided campus tours
offered by the University of Texas at Austin. A self
-
guided
campus tour,
made available via a paper brochure, does exist for
those visitors unable to take part in the guided tour. However,
unlike
many other u
niversity campuses
, the University of Texas at
Austin does not have a comprehensive mobile campus tour
application.
Given

the advancement of location
-
based services
and the proliferation of smartphone usage, our team
developed a
mobile
-
application

to improve the experience

of the self
-
guided
tour
,
Hook

e
Map.

Hook’e
Map utilizes an adaptive
hypermedia

approach and context
awar
e design to offer
users
a more personalized, relevant, and
navigable

campus tour. In order to test this theory, we measured
the experiences of study participants as they used both the
mobile
-
application

and the paper brochure during a self
-
guided campus
to
ur. Analysis of the traditional tour methodology helped set
a
benchmark against which the
mobile
-
application

could

be judged.

Ultimately
our research looked at whether
Hook
’eMap
provide
d a
better
experience
than the paper brochure for tour participants on

the University of Texas
at Austin campus by answering the
following questions:

a. Does the
mobile
-
application

offer a more personalized campus
tour experience?

b. Does the
mobile
-
application

offer a more relevant campus tour
experience?

c. Does the
mobile
-
application

offer better
navigation of campus?

Extensive
exploratory

interviews with staff at the University of
Texas at Austin Vis
itor Center revealed three main
users of the
self
-
guided tour: prospective students, visitors, and alumni.
These
insights pus
hed development of

Hook’eMap
into an
adaptive
hypermedia content

approach
, whereby content was categorized
by the aforementioned stereotypes: prospective student, visitor, or
alumn
i
.

This allowed individual users to choose a role that
matched their current

status in order to generate relevant content
crafted towards their specific interests and provided awareness of
the

user’s personal context
.

Early user experience testing surveys were pooled from two
sources to streamline the
mobile
-
application
’s layout,
user flow,
navigation, and mapping functionality. Field tests were designed
and run to eliminate user bias by splitting the campus into east and
west sections to run four separate tours, where twenty participants
took both sides of the tour switchin
g evenl
y between the paper
brochure

and
mobile
-
application
. The results showed that users
did prefer the
mobile
-
application
, finding it to be more
personalized
and relevant

to their interests
than the paper
brochure.

However, there were development and technical
issues
with the mapping and navigational features that lead to negative
user experiences for some participants.

The following sections
will go into further detail regarding the research study.

2.

RELATED WORK

2.1

Context Awareness in Mobile Tours

As the percentag
e of smartphone users has increased
exponentially, significant research has been conducted regarding
their use for both navigation and tourism [13] [11].
Hook
’eMap
leveraged findings from
the tourism industry applications and
research therein to identify i
ts mapping and navigational
approach.

Location
-
based services (LBS) access spatial
coordinates in order to integrate a user’s geographic location
within the larger notion of service [9]. LBS exist inside the larger
realm of context awareness, a concept defined by Abowd et al. [1]
as a system’s

use of context “to provide relevant information
and/or services to the user, where relevancy depends on the user’s
task.”
Mobile
-
application
s are some of the most common context
-
aware programs since, “in handheld and ubiquitous computing, a
user's context

is very dynamic.”

Context is commonly taken to mean location, however Schmidt et
al. argue

that “location is only one aspect of the physical
environment” [10]. Context is therefore more accurately described
as “any information that can be used to characte
rize the situation
of an entity, where an entity can be a person, place, or object” or
“anything relevant to the interaction between the user and
a
pplication, including the user
and the application” [1].

Cheverest et al. [4] defined two classes of context,

personal and
environmental, when they attempted to develop a mobile city tour.
Their understanding of personal context encompassed everything
from the visitor’s current location to his interests and refreshment
preferences. Environmental context included
current location in
addition to conditions such as time of day, weather, and
infrastructure information. Their experience testing the mobile
tour showed that, although users reported a high quality of visitor
experience when operating the program, "the fle
xibility provided
by the system” was superfluous to some, illustrating “the need to
enable visitors to choose the level of functionality" required.

2.2

Adaptive Hyper
media

as Context
Awareness

Although, previous research has shown that there are many
different

contexts to focus on when creating a mobile tour
application, one of the most interesting takes into account aspects
of the user’s personal context: interests and self
-
defined role.
Adaptive hypermedia, described by Brusilovsky [3] as “an
alternative to t
he traditional ‘one
-
size
-
fits
-
all’” system, tailors
information to the user, giving the system “the ability to
‘understand’ the user, to customise information and presentation
and to dynamically support navigation” [6].

Many forms of user modeling for adap
tive hypermedia exist. One
of these is the concept of a stereotype user. Rich [8] defined a
stereotype as “a set of traits that often occur together” and his
model, which represents individual us
ers via sets of attributes,

“as
simply a way of capturing som
e of the structure that exists in the
world around us.”

Testing of this model in the context of mobile tours has found it to
be advantageous. Van Hage et al. [12] explored stereotypes when
they created an adaptive art museum guide, tailored to the user’s
p
osition and interests. The guide defined four stereotypes:
children, students, novices, and art experts in order to construct
individualized narratives for the users. Testing of the system
showed users found that the recommendations accurately reflected
ty
pical interests for the stated groups and provided a sufficient
degree of personalization.


The adaptive hypermedia appr
oach was selected for
Hook
’e
Map
to specifically tailor information to the user stereotypes.

Content
was collected from sources on the Un
iversity of Texas at Austin’s
officially authorized department websites
as well as the
university’s visitor center
to broaden the scope of information for
users beyond the paper brochure.

In general, visitors
sought
information

about historical

and cultura
l

aspects of campus,
prospective students learned about campus life
,

and alumni were
interested in

new developments
and historical knowledge
.

Again,
these content choices were
collected

from insights drawn th
rough
interviews with the

University of Texas at

Austin
’s

Visitor Center.

This department supervises all tours given at the university.


2.3

User Interaction Preferences

Further investigations into

various user
interaction preferences
gave us
clarity on

interaction design for the mobile tour
application
.
Cheverst et al. [5] again explored their mobile tour
while grappling with the question of "how to design around the
apparently conflicting goals of adapting to changes in context
while at the same time adhering to the principle of predictability."
They con
sidered two design approaches: information push vs.
information pull. Information push is when the presentation of
information is triggered by contextual events, while with
information pull the impetus is on the user to decide when
information is presented
. They found users were accepting of both
approaches, stressing that "designers need to carefully balance the
way in which the system reacts to environmental triggers (e.g.
location) with a desire for the system to...adhere to the principle of
least astoni
shment."

Barkhuus and Dey [2] further researched this topic when they
inquired: Is context
-
aware computing taking control away from
the user? To answer this question they first defined three levels of
interactivity: personalization, passive context
-
awarene
ss, and
active context
-
awareness. Personalization lets the user specify
his/her own settings for how the application should behave in a
given situation. Passive context
-
awareness (similar to push)
presents updated context or sensor information to the user
but lets
the user decide how to change the application behavior. Active
context
-
awareness (similar to pull) autonomously changes the
application behavior according to the sensed information. After
conducting an experimental case study, Barkhuus and Dey wer
e
able to determine that the "more autonomous the service is, the
less users felt in control." They were surprised, however, to
discover that users preferred context
-
aware applications (either
passive or active) over more personalized ones. Ultimately, the
y
concluded that users were "willing to give up partial control if the
reward in usefulness is great enough."

Lanir, et al. [7] found similar attitudes towards proactiveness
when they considered “proactive vs. passive context
-
awareness
and choice” in a mus
eum setting. Asking the question, “should the
system keep the user in control all the time and only respond to
user requests, or should the system take initiative and propose its
services when needed?” they, too, found that “users’ feelings of
control decr
eased as the proactiveness and the autonomy of the
system increased.” Taking their study further, they were able to
conclude that “decreased feeling of control, however did not
affect visitors’ satisfaction, intention to use the system, or other
subjective

opinions of the system on which reported ratings did
not differ between interfaces.”

Hook
’eMap utilized an active context
-
awareness design with a
pull strategy allowing for proactive user interactions.

User control
of the application was built into the de
sign to enable participants
to select
options along the tour such as
choos
in
g their next
location,
what to read about locat
ions and the ability to
navigate
with

location
-
based services within

the mapping functionality.

3.

MOBILE
-
APPLICATION

3.1

Technical
Development

Hook
’eMap
mobile
-
application

that w
as developed for and used
in this

study was built using the jQuery Mobile framework along
with the Google Maps JavaScript API. jQuery Mobile is a touch
-
optimized web framework which uses HTML, JavaScript,

and

Permission to make digital or hard copies of all or part of this work for
personal or classroom use is granted without fee provided that copies are
not made or distributed for profit or commercial advantage and that
copies bear this notice and

the full citation on the first page. To copy
otherwise, or republish, to post on servers or to redistribute to lists,
requires prior specific permission and/or a fee.

Conference’
10
, Month 1

2, 2010
, City, State, Country.

Copyright 2
010

ACM 1
-
58113
-
000
-
0/0
0/0010

…$15
.00.


CSS and is designed to be compatible with a wide variety of
smartphones and tablet computers. The Google Maps JavaScript
API is an open
-
source JavaScript library, which allows the
embedding of Google Maps and related services within a web or
mobile
-
ap
plication
.

The use of jQuery Mobile allowed for an HTML5
-
driven
configuration for the laying out the pages of the application. This,
along with an Ajax
-
powered navigation for animating page
transitions, allowed for the application to provide a user
experie
nce similar to that of a native
mobile
-
application

while
being accessed from a phone’s mobile browser.

By incorporating the Google Maps JavaScript API, the
mobile
-
application

allowed for users to get walking directions to all of the
locations on the tour,

with both an interactive Google Map and a
corresponding list of directions. By retrieving the user’s current
location via a call to the Geolocation API, the application would
route the user by pulling in the latitude and longitude coordinates
correspondin
g to their current destination. A DirectionRenderer
would then render both the Google Map and the list of directions.

Once the initial prototype was developed, a series of usability
tests were conducted and a number of design revisions were made
to
integra
te

the findings from our
user feedback

surveys
.

The content of the tour was based upon the current self
-
guided,
campus tour
and
consists of ten different spots spread throughout
campus with 36 points of interest

and includes images and text
relevant to the points of interest. The tour content of the
mobile
-
application

was modeled closely after the paper tour and used the
same images; however, the text was modified with the goal of
providing a more personalized tou
r experience

for

prospective
students, alumni and visitors
.

3.2

Design

On the first screen of the application, the user was prompted to
choose a role

(see Figure
1
)
. The application was able to
personalize the tour content once the user selected a role of

prospective student, alumni, or v
isitor. Based on the user’s
selection, the application would apply a filter using JavaScript and
would display
text and graphic
information relevant to that role.

During testing of early versions of the app, it was found t
hat it
was unclear as to how many total spots there were on the tour. To
remedy this, a tour overview screen was added which displayed
all 10 spots on the map and included instructions on how to
navigate between spots within the app
(see Figure
1
)
. The use
r
could then choose to begin the tour by clicking on the “Start the
Tour” button located at the bottom of the tour overview screen.


Figure
1
.
First and Second Application Screens


Once the user had elected to begin the tour, they were given text
directi
ons to head to spot one of the tour from their current
location and also given a map overlay with points A and B, their
current location and destination. This is seen in the Google Map
generated
(see
Figure

2
)

on the left and right. During initial
testing,

users had indicated that it was unclear from using the map
alone as to whether or not they had successfully arrived at their
destination. Therefore, a picture of the user’s current destination
was added to the screen,

(see Figure
2
)
. The user could toggle

back and forth between a view of the map or the image of their
destination by clicking on the icon at the bottom right of the
frame.


At the top of the screen
(see Figure
2
)
, a menu displayed the
number of the spot of which the user was currently at. The
user
was able to navigate between various screens of the application by
clicking on the appropriate navigation buttons at the top of the
screen displayed in orange. As a result of user testing, these
navigation buttons were made orange and positioned in a
single
location at the top of the screen throughout the application. This
was done to both make them more prominent and more intuitive
to the user.

Figure 2.

Google Map Directions and Map



By clicking the Next button at the top of the screen

(see

F
igure

3
)
,
the user was taken to a screen with all of the points of interest at
the current spot. On the Google Map, markers were plotted to
indicate the locations of each point of interest.

In user testing, it was found that should a user become lost, it was

un
clear as to how to find their

next desired location
. Therefore, a
button labeled “Lost? Click here to get directions to this spot.”
was added to the center of the
screen

(see

Figure 3
)
.

By clicking
this button, the user was taken back to the directions screen and
redirected back to the spot.

Figure 3.
Main Locations and Lost Feature


By clicking on a point of interest, the user was brought to a screen
displaying a larger image of the lo
cation as well as some text

(see

Figure 4
)
. As stated before, this text was customized based on the
user’s selected role as indicated at the start of the tour. The user
could also choose to get directions to that spot.

Once the user had finished touring a

particular location, they
could click the Back button and head back to the Spot Information
screen. Then once, finished with a spot, they could head to the
next spot on the tour by clicking on the Next Spot button.

Once the user had progressed through al
l of the spots on the tour,
they were taken to a screen featuring our beloved University of
Texas at Austin mascot and namesake of the application,
Hook’em, which thanked them for taking the tour and presented
them with the option of returning to Spot 1.













Figure
4
.

Point of Interest



3.3

Methodology

The study sought to discern the difference
in experience between
two m
ethods for taking a self
-
guided

campus tour: paper brochure
or
mobile
-
application
. The paper brochure includes ten spots and
36 points of interest
(see Figure
5
).

Hook
'eMap used as

the same
spots and points of interest as the paper brochure. The test phone
was a Samsung Galaxy SIII borrowed from the
University of
Texas at Austin
School

of Information IT Lab.

Figure
5. Campus Tour Route



3.4

Procedure

Participants were recruited via email

to
give

a mix of prospective
students, alumni and visitors

and invited to take part in the study.
In order to compare the different tour experiences we d
ivided
twenty participants into four groups of five (Group A, Group B,
Group C, and Group D). The group designations represented four
distinct combinations of the same campus tour. For example,
participants designated as Group A toured the eastern half of
campus (spots 6
-
10) using the paper brochure. Following this,
Group A toured the western half of campus (spots 1
-
5) using the
mobile
-
application
. The various combinations
(see
Table

1
)

were
created in order to mitigate any preference a participant might
have for a certain half of campus and to

combat fatigue bias.

Participants toured the campus in groups of one or two, with
groups of two sharing one test phone. Prior to undertaking the half
of the tour that used the mobile
-
application, participants were
g
iven a short explanation on how to manipulate the
Samsung
Galaxy SIII
phone, but not the mobile
-
application. For both the
paper and mobile
-
application tours, participants were encouraged
to visit the
points of interest

at their own pace and to
use the
guid
es as they wished.

Table
1
.

Testing
G
roups

Group A

Tour

Method

1st Half

East

Paper brochure

2nd Half

West

Mobile
-
application


Group B

Tour

Method

1st Half

East

Mobile
-
application

2nd Half

West

Paper brochure


Group C

Tour

Method

1st Half

West

Paper

brochure

2nd Half

East

Mobile
-
application


Group D

Tour

Method

1st Half

West

Mobile
-
application

2nd Half

East

Paper brochure


3.5

Measures

In order to evaluate the difference in experience between the
methods for taking a self
-
guided tour, we used two
complementary tools: a survey that participants filled out
before
and after

each section of the tour
and a semi
-
structured interview
conducted
as part of participant debriefing.

The survey was divided into four parts: The first part of the survey
asked for general demographic information and was completed
prior to starting the tour. The second part of the survey was
completed after finishing the
first half of the tour. It asked
participants to rate their agreement with a number of statements
regarding enjoyment, interest, relevance, and ease of use on a 5
-
point Likert scale where values ranged from strongly agree (1) to
strongly disagree (5). The
third part of the survey was a duplicate
of the second, but was completed after finishing the second half of
the tour. The fourth part of the survey immediately followed and
asked users to rate a preference between the two different touring
methods.

The se
mi
-
structured interview was conducted upon completion of
the survey. Here, users were asked open
-
ended questions and
invited to give thoughts and feedback on the mobile
-
application
while the researcher made notes.

Questions focused on rating the
mobile
-
app
lication

experience, identifying likes and dislike
s

regarding features
and
of
fering an opportunity
to make
recommendations for improvements

to
Hook
’e
Map
.

4.

RESULTS

After twenty participants underwent the campus tour testing
process we analyzed and assessed each participant’s survey
responses to gauge the reception of
Hook
’eMap
mobile
-
application
.

The male to female ratio was even in our pool of
participants.

A
55%
ma
jority of participants indicated that they
had never taken a
campus
tour of the Univ
ersity of Texas at
Austin.
A significa
nt portion of our participants 90%
, however,
had visited the University of Texas at Austin campus previously.

Our distribution of part
icipant types was even among those that
identified as alumni of the University of Texas at Austin and those
that identified as tourists. The pool of participants also utilized
current students of the

University of Texas at Austin
(
35%
)
. One
limiting factor

in the recruitment of participants was the access to
prospective students.

Only 5% of our participant pool identified
themselves as a prospective student, leaving us without an even
distribution of tour participant
stereo
types.

The abundance of
University

of Texas affiliated recruits also led to a high level of
campus familiarity among tour participants with
(
95%
)

indicating
that they had at least some familiarity with the University of
Texas campus.

Based on
the self
-
selected user
stereo
type

and the respo
nses given
in the survey
before
the start of

the tour about a participant’s
primary motivation
for

touring the campus, trends began to
emerge
about participant
stereo
type’s motivations.

Alumni
generally wanted to learn about the history of the campus (56%)

or see the new features of the campus (44%).

Prospective
students
, on the other hand,
wanted to familiarize themselves with
campus (75%).

Finally, visitors
exhibited a diverse range of
interests.

29% of visitors indicated they wanted to see the new
featur
es of campus, but still others wrote in responses such as,
“just curious” or “to get some exercise.”


When analyzing these various motivations for touring campus and
cross
-
tabulating it with how users responded to their level of
interest and expectations o
f the tour, the results paint an
interesting picture of a user’s motivatio
n for

use
of
the
mobile
-
application
.
Of those that wanted to learn more about the history
of the campus, 100% agreed or strongly agreed that the
mobile
-
application

was interesting

(S
ee Table 2
)
.

Only 50% of this same
group seeking historical knowledge of campus, however, agreed
that the
mobile
-
application

tour exceeded expectations

(See Table
2
)
.

The other half was undecided in whether or not the
mobile
-
application

tour exceeded expectations

(See Table 2
)
.

















Table
2
.

Primary Interest: Learn about the history of the
campus (mobile)


I found the
majority of
locations

visited on this
tour
interesting

Strongly Agree

17%

Agree

83%

Undecided

0%

Disagree

0%

Strongly
Disagree

0%

Total

100%

p
-
value

0.01


This tour of
the University
of Texas at
Austin
exceeded my
expectations

Strongly Agree

0%

Agree

50%

Undecided

50%

Disagree

0%

Strongly
Disagree

0%

Total

100%

p
-
value

0.02


This
navigation
method was
easy to
manipulate

Strongly Agree

17%

Agree

50%

Undecided

17%

Disagree

17%

Strongly
Disagree

0%

Total

100%

p
-
value

0.05



Along the same lines, participants whose primary motivation was
to see the new features of the University of Texas campus also
agreed that the
mobile
-
application

tour was interesting.

In fact,
71% who wanted to see new features agreed or strongly agreed
th
at the
mobile
-
application

was interesting

(See Table 3
)
.
Additionally, this group who sought new campus features also
found that the
mobile
-
application

tour exceeded their
expectations.

83% of participants agreed or strongly agreed with
this sentiment

(See Table 3
)
.










Table
3.

Primary Interest: See the new features of

the campus (mobile)

I found the
majority of
locations
visited on this
tour
interesting

Strongly Agree

43%

Agree

29%

Undecided

29%

Disagree

0%

Strongly
Disagree

0%

Total

100%

p
-
value

0.01


This tour of
the University
of Texas at
Austin
exceeded my
expectations

Strongly Agree

33%

Agree

50%

Undecided

17%

Disagree

0%

Strongly
Disagree

0%

Total

100%

p
-
value

0.02



This
navigation
method was
easy to
manipulate

Strongly Agree

17%

Agree

67%

Undecided

0%

Disagree

17%

Strongly
Disagree

0%

Total

100%

p
-
value

0.05



Where results differed was with the group whose primary
motivation was to become familiar with the University of Texas
campus.

Their responses were split as to whether they found the
mobile
-
application

tour interesting with 50% indicating they
agreed or strongly agreed and the other 50% undecided as to the
mobile
-
application
’s level of interesting location content

(See
Table 4
)
.
T
hen when measuring to what extent the
mobile
-
application

exceeded expectations, 50% of respondents whose
primary motivation was to familiarize themselves with campus
relayed that they disagreed that the
mobile
-
application

exceeded
expectations

(See Table 4
)
. Another 25% were undecided on the
matter leaving only 25% that strongly agreed that the
mobile
-
application

exceeded expectations

(See Table 4
).








Table
4.
Primary Interest:
Familiarize myself with the campus
and its atmosphere

(mobile)

I found the
majority of
locations
visited on this
tour
interesting

Strongly Agree

25%

Agree

25%

Undecided

50%

Disagree

0%

Strongly
Disagree

0%

Total

100%

p
-
value

0.01


This tour of
the University
of Texas at
Austin
exceeded my
expectations

Strongly Agree

25%

Agree

0%

Undecided

25%

Disagree

50%

Strongly
Disagree

0%

Total

100%

p
-
value

0.02


This
navigation
method was
easy to
manipulate

Strongly Agree

25%

Agree

0%

Undecided

0%

Disagree

50%

Strongly
Disagree

25%

Total

100%

p
-
value

0.05



Another area of significance occurred among the ratings users
gave the ability to manipulate the
mobile
-
application

compared
with user’s primary motivations for taking a tour.

Again, those
who wanted to learn the history of the campus or see the new
features of the campus responded positively to using the
mobile
-
application
.

66% of participants seeking to learn about the history
of the campus agreed or strongly agreed that the
m
obile
-
application

was easy to manipulate (See Table
2
).

Only 17% of
respondents disagreed that the
mobile
-
application

was easy to
manipulate (See Table
2
).

Those whose primary interest involved
seeing the campus’s new features echoed the same sentiment.

83
% of those seeking new campus features agreed that the
mobile
-
application

was easy to manipulate (See Table
3
). In
addition, only 17% of respondents disagreed with that the
mobile
-
application

was easy to manipulate (See Table
3
).

The group
whose primary motivation for taking a tour was to become more
familiar with campus, however, overwhelmingly disagreed that
the
mobile
-
application

was easy to manipulate. 75% disagreed or
strongly disagreed that the
mobile
-
application

was easy to
manipulate (See Table
4
).

These findings all seem to indicate that those seeking to
familiarize themselves with campus via the
mobile
-
application

found a less than satisfactory experience in terms of exceeding
expectations and ease of manipulation. They we
re also of mixed
opinion when determining whether the locations were of interest.

Meanwhile the groups seeking new features of the campus or an
increased knowledge of the history of campus seem to find the
mobile
-
application

particularly interesting, easy
to manipulate,
and found it exceeded expectations.


Although not significant, comparing the
mobile
-
application

responses to the paper brochure version of the tour reveals that the
participants
hoping

to see new campus features and those wanting
to learn th
e history of the campus viewed the paper brochure more
negatively than they viewed the
mobile
-
application
.

Of those
whose primary motivation for touring the campus was to see new
features, only 57% agreed that it was interesting, 33% agreed it
exceeded exp
ectations, and 50% agreed the paper brochure was
easy to manipulate (See Table
5
). This is a decline from the
ratings of the
mobile
-
application

where 71% found it interesting
and 83% found it exceeded expectations and easy to manipulate
(See Table
3
).

Tabl
e
5
.

Primary Interest: See the new features of

the campus (paper)

I found the
majority of
locations
visited on this
tour
interesting

Strongly Agree

14%

Agree

43%

Undecided

29%

Disagree

14%

Strongly
Disagree

0%

Total

100%

p
-
value

0.88


This
tour of
the University
of Texas at
Austin
exceeded my
expectations

Strongly Agree

0%

Agree

33%

Undecided

67%

Disagree

0%

Strongly
Disagree

0%

Total

100%

p
-
value

0.97


This
navigation
method was
easy to
manipulate

Strongly Agree

0%

Agree

50%

Undecided

0%

Disagree

50%

Strongly
Disagree

0%

Total

100%

p
-
value

0.12



Those who indicated their primary motivation as learning more
about the history of campus displayed a similar pattern of
behavior.

Only 50% agreed that the paper section of the tour was
interesting (See Table
6
). 17% indicated that the paper tour
exceeded their expectations (See Table
6
). Finally, 60% disagreed
that the paper tour was easy to manipulate (See Table
6
).

This too
was in
stark contrast to the sentiment displayed in the use of the
mobile
-
application by the same group (See Table
2
).

Those
seeking historical data about the campus seemed to have preferred
the
mobile
-
application
.

Again, the results from the paper tour
were found to have a p
-
value above .05 and are therefore not
significant, but the results do give an interesting look into how the
mobile
-
application

fared in comparison to the paper tour among
those seeking historica
l information about the campus.

Table
6
.

Primary Interest: Learn about the history of the
campus (paper)

I found the
majority of
locations
visited on this
tour
interesting

Strongly Agree

0%

Agree

50%

Undecided

17%

Disagree

33%

Strongly
Disagree

0%

Total

100%

p
-
value

0.88


This tour of
the University
of Texas at
Austin
exceeded my
expectations

Strongly Agree

0%

Agree

17%

Undecided

50%

Disagree

33%

Strongly
Disagree

0%

Total

100%

p
-
value

0.97


This
navigation
method was
easy to
manipulate

Strongly Agree

0%

Agree

20%

Undecided

20%

Disagree

60%

Strongly
Disagree

0%

Total

100%

p
-
value

0.12



Conversely, the tour participants who sought to familiarize
themselves with campus had a much different perspective on
which method

of touring campus they preferred.

While they did
not rate the
mobile
-
application

particularly high, they did enjoy
the paper section of the tour. 75% of those seeking to familiarize
themselves with campus indicated that the tour was interesting
(See Table

7
).

They were still inconclusive on whether the paper
-
based tour exceeded expectations, with 50% choosing that they
were undecided and the other 50% indicating they disagreed that
the paper
-
based tour exceeded expectations (See Table
7
).

They,
however, di
d agree that the paper
-
based section of the tour was
easy to manipulate with 67% agreeing.

These findings for the
paper
-
based section of the tour did not meet statistical
significance, however.






Table
7
. Primary Interest
: Familiarize myself with the ca
mpus
and its atmosphere (paper)

I found the
majority of
locations
visited on this
tour
interesting

Strongly Agree

25%

Agree

50%

Undecided

25%

Disagree

0%

Strongly
Disagree

0%

Total

100%

p
-
value

0.88


This tour of
the University
of Texas at
Austin
exceeded my
expectations

Strongly Agree

0%

Agree

0%

Undecided

50%

Disagree

50%

Strongly
Disagree

0%

Total

100%

p
-
value

0.97


This
navigation
method was
easy to
manipulate

Strongly Agree

33%

Agree

33%

Undecided

33%

Disagree

0%

Strongly
Disagree

0%

Total

100%

p
-
value

0.12



These findings parallel the results found later in the survey. Upon
completion of the tour participants chose which method of touring
the University of Texas campus they preferred.

Several follow
-
up
questions helped to determine the strengths and weaknesses of
each method.

Overall, participants indicated preference for the
mobile
-
application

in touring the University of Texas at Austin
campus.

55% of respondents indicated that they preferred the
mobil
e
-
application

tour to the paper
-
based tour (See Table
5
).

Additionally, 20% of respondents indicated that they did not have
a preference between the two types of tours (See Table
5
).

Interestingly, however, among those that liked both methods
equally,

100% agreed that they enjoyed the
mobile
-
application

tour (See Table
11
).

Furthermore, participants found that the
mobile
-
application was able to offer more relevant information
and a more personalized experience to them while taking a tour of
the campus.

When asked which method provided the most
relevant information 55% of participants chose the
mobile
-
application

(See Table
9
).

Additionally, when asked to choose
which method provided the most personalization 75% indicated
the mobile
-
application (See Table

10
).

This is congruent with the
results found previously.

The mobile
-
application garnered a
positive reaction from the tour participants in meeting their
primary interests in taking a tour.

Meeting this interest may have
created preference for the mobile
-
application among tour
participants.


Table
8
. Which method of touring campus did you prefer?

Answer

%

I preferred the mobile
-
application tour

55%

I preferred the paper
-
based tour

25%

I liked both methods equally

20%

I disliked both methods

0%

Total

100%


Table
9
. Which tour

option

offered the most relevant
information to you?

Answer

%

The mobile
-
application tour offered the
most relevant information to me

55%

The paper
-
based tour offered the most
relevant information to me

0%

I found the
information in both tours to
be equally relevant to me

35%

I did not find the information in either
tour to be relevant to me

10%

Total

100%


Table
10
. Which tour option offered the most personalization

Answer

%

The mobile
-
application tour offered the
most personalization

75%

The paper
-
based tour offered the most
personalization

0%

I found the information in both tours to
be equally personalized to me

10%

I did not find the information in either
tour to be personalized to me

15%

Total

100%


Table
11
. Which method of touring campus did you prefer vs.
mobile
-
application level of enjoyment



I preferred
the mobile
-
application
tour

I preferred
the paper
-
based tour

I liked both
methods
equally

I disliked
both
methods


I enjoyed
this tour of
the
University
of Texas at
Austin
campus

(mobile)

Strongly
Agree

64%

0%

0%

0%

Agree

36%

20%

100%

0%

Undecided

0%

20%

0%

0%

Disagree

0%

60%

0%

0%

Strongly
Disagree

0%

0%

0%

0%

% of Total

55%

25%

20%

0%

p
-
value

0.05






5.

DISCUSSION

In
rationalizing the results of the survey, one is struck by the fact
that two main groups seemed to prefer the mobile
-
application for
touring the university campus and one group did not particularly
care for the mobile
-
application.

The two grou
ps that rated the
mobile
-
application highly were interested in the historical aspects
of campus and in seeing the new features.

In tying these
motivations to user
-
types it seems as though these motivations for
taking a tour are most closely associated with

alumni, and to a
certain extent visitors.

Both of these user
-
types indicated a fairly
high level of familiarity with campus.

Taking this into
consideration, these two types of users would not be as concerned
with the navigational aspects of the applicatio
n.

Based on their
previous experience navigating the campus, they have some
inclination as to their location when walking around the
University of Texas at Austin.

The group that consistently gave
lower ratings to the mobile
-
application was interested in
b
ecoming familiar with the campus and its atmosphere.

This
motivation was primarily found among the prospective student
user
-
type.

It would make sense for this user
-
type to be less
familiar with campus than the others.

A prospective student, or
someone seek
ing to become more familiar with campus, is not
able to draw on past experience to aid in locating their position.
Furthermore, someone whose primary goal is to become more
familiar with campus has indicated that they are presently
unsatisfied with their c
urrent level of familiarity with the campus
and are seeking to improve that by taking a tour.


When taking these considerations into account and comparing it
to in
-
the
-
field knowledge of the mobile
-
application and its
functionality, this discrepancy betwee
n people with some
background navigating the campus and those unfamiliar with
navigating the campus becomes somewhat understandable.

Forgoing the navigational aspects of the application alleviated a
significant amount of application usability frustration.

In order to
quickly and accurately update the user’s present location a
connection to the Internet was required.

In testing the application
tour participants were limited to the campus
WiFi
.

While fairly
substantial in coverage, there are some areas with l
imited
WiFi

on
the campus.

This led to some frustration on the part of the user
when trying to navigate
to

a specific location.

A user unfamiliar
with the campus, or a user whose goal is to become more familiar
with campus would notice these hiccups in con
nectivity much
more than someone that is not entirely reliant upon the
application’s navigation.

A person familiar with campus is much
more apt to recognize the location the application is attempting to
direct them towards and be satisfied to move on to re
ad about
each location within the application.

A user unfamiliar with
campus is entirely reliant upon the application to direct them and
a connectivity error would stall their tour and make them focus on
basic application functionality, rather than enjoyin
g their tour and
the atmosphere of the campus.

This would further explain the
particularly low
-
ranking that ease of
mobile
-
application

manipulation level earned among those seeking to gain a higher
level of familiarity with the campus (See Table 3).

The ne
cessity
of frequently updating their current location in order to guide
them to their destination was hampered in particular instances.


WiFi

connectivity, however, is not the only means of interpreting
the results of the
mobile
-
application

tour testing.

F
rom the
standpoint of the application’s use of adaptive hypermedia, a user
seeking knowledge and information, i.e., historical information or
new additions to the campus, will be more appreciative of
adaptive hypermedia’s use.

The information they seek is
found
within the content sections of each location.

This section has been
specifically adapted for their user
-
type, as selected at the start of
the tour.

In contrast, someone wanting to gain familiarity with
campus is much more inclined to focus on the nav
igational
aspects of the application.

These have not been specifically
tailored for a particular user
-
type.

The path through campus was
kept consistent among all user
-
types for the sake of testability.

Overall the current iteration of the
mobile
-
applicatio
n

is a better
source of information than it is a tool for navigation. A user
focused on following the correct directions and assimilating a
good idea of on
-
campus direction may forgo reading the content
for each location, thus not gaining the full benefit
that the use of
adaptive hypermedia offers within the
mobile
-
application
.


This sentiment of meeting specific interests with the
mobile
-
application
’s use of hypermedia is echoed in the results
showcasing participant’s preference for the
mobile
-
application

in
providing the most relevant information (See Table 8).

Additionally, it could be indicative of the results of whether or not
the
mobile
-
application

exceeded expectations. Users motivated by
gaining historical knowledge or seeing new features of the
camp
us generally indicated that the use of the
mobile
-
application

exceeded expectations (See Table 1 and Table 2).

User’s
motivated by increasing familiarization of campus did not believe
the mobile tour application exceeded expectations (See Table 3).

To some
one not as concerned with navigation, a method of
touring campus
with more customized information
is
preferred,
especially when it can be obtained conveniently via a personal
phone.

When the focus is placed on navigation
Hook
’eMap

does
not live up to expectations.

As of this version,
Hook
’eMap is
much more suited for alumni tourism.

Following
the testing
, many
partipants

commented that with a strong connection the
mobile
-
application

would be superior.

With a more robust navigation
feature and a consistent Internet connection,
Hook
’eMap would
appeal to a wider range of tour interests.



6.

CONCLUSIONS

Smartphone usage has increased dramatically over the last several
years, with nearly half of the United States population owning a
smartp
hone [11]. Additionally, among smartphone users, 74% use
location
-
based services

[13]. Many previous studies have focused
on interaction within mobile tour applications without making
comparisons against more traditional touring methods. This study
allowed

us to compare a
mobile
-
application

developed for touring
the University of Texas at Austin campus, Hook’eMap, with the
paper brochure offered to those interested in taking a self
-
guided
tour. Revisiting our research questions, results indicate that a) The

majority of participants felt that the
mobile
-
application

offered a
more personalized campus tour experience. b) While a number of
participants found the information in both tours to be equally
relevant, the majority of participants still felt that the
mo
bile
-
application

offered a more relevant campus tour experience. c)
Results did not indicate that the
mobile
-
application

offered better
navigation of campus. Participants already familiar with the
campus found the
mobile
-
application

to be an adequate navig
ation
tool. However, those unfamiliar with campus were less satisfied.
This suggests that visitors al
ready familiar with the campus

are
interested in using mobile tour applications for gleaning more
detailed information about the university without need of

navigational information. Meanwhile, first
-
time visitors, who are
naturally unfamiliar with the campus, need better tools for
navigation than the
mobile
-
application

provides.

7.

FUTURE WORK

Now that a
benchmark

has been established, future research
should be concentrated within the
mobile
-
application
. This
research may focus on identifying different types of campus
visitors along with their interests and needs, determining user
preference for fixed vs. custom ro
utes, or verifying the stated
preference for more multimedia content such as audio and video.
Further refinement of the
mobile
-
application

to eliminate
connectivity issues and aid the user in location awareness through
the use of real
-
time GPS and arrival
alerts is also recommended.

8.

REFERENCES

[1]

Abowd, G., Dey, A., Brown, P., Davies, N., Smith, M., and
Steggles, P. 1999. Towards a better understanding of context
and context
-
awareness. In
Proceedings of the 1st
international symposium on Handheld and Ubiquitous
Computing
(HUC '9
9),
304
-
307.

[2]

Barkhuus, L. and Dey, A. 2003. Is Context
-
Aware
Computing Taking Control away from the User? Three
Levels of Interactivity Examined.
In Proceedings of
U
bicomp 2003
, 146
-
156.

[3]

Brusilovsky, P. 2001. Adaptive Hypermedia. In
User
Modeling and User
-
Adapted Interaction

11, 1
-
2 (March
2001), 87
-
110.

[4]

K. Cheverst, N. Davies, K. Mitchell, A. Friday and C.
Efstratiou. 2000. Developing a context
-
aware electronic
touri
st guide: some issues and experiences. In
Proceedings of
the SIGCHI conference on Human Factors in Computing
Systems

(CHI '00)
, 17
-
24.

[5]

Cheverst, K. Mitchell, and N. Davies. 2001. Investigating
context
-
aware information push vs. information pull to
tourists
. In
Proceedings of Mobile HCI 01
.

[6]

Eklund, J., and Zeiliger, R. 1996. Navigating the web:
Possibilities and practicalities for adaptive navigational
support. In
Proceedings of AUSWEB'96
.

[7]

Lanir, J., Kuflik, T., Wecker, A.J., Stock, O., and Zancanaro,
M. 201
1. Examining proactiveness and choice in a location
-
aware mobile museum guide. In
Interact
.
Comput
. 23, 5
(September 2011), 513
-
524.

[8]

Rich, E. 1999. Users are individuals: individualizing user
models.
In International Journal of Human
-
Computer
Studies
,

51,
2 (August 1999), 323
-
338.

[9]

Schiller, J.H. and Voisard, A. 2004.
Location
-
based Services
.
Morgan Kaufmann, San Francisco, CA.

[10]

Schmidt, A., Beigl, M., and Gellersen, H. 1999. There is
more to context than location. In
Computers & Graphics
Journal
, Elsevi
er, V
olume 23, No.6, Dec 99,
893
-
902.

[11]

Smith, A. 2012. Nearly half of American adults are
smartphone owners.
Pew Internet
Research,
1
-
9.

[12]

Van Hage, W.R., Stash, N., Wang, Y., and Aroyo, L. 2010.
Finding your way through the rijksmuseum with an adaptive
mobile mus
eum guide. In
Proceedings of the 7th
international conference on The Semantic Web: research and
Applications
-

Volume Part I

(ESWC'10), 46
-
59.

[13]

Zickuhr, K. 2012. Three
-
quarters of smartphone owners use
location
-
based services.

Pew
Internet Re
search
, 1
-
2.