Frontiers of Architectural Research (2018) 7, 1–10

Available online at www.sciencedirect.com

Frontiers of Architectural Research

www.keaipublishing.com/foar

RESEARCH ARTICLE

Place attachment assessment of a heritage

place: A case study of the Roman
amphitheater in downtown Amman, Jordan
Jawdat S. Goussous, Nessma A. Al-Hammadin

Department of Architecture, Faculty of Engineering and Technology, University of Jordan, Amman, Jordan

Received 21 July 2016; received in revised form 2 December 2017; accepted 8 December 2017

KEYWORDS
Spiritual value;
Place attachment;
Place identity;
Place identification;
Place dependence;
Social bond
Abstract
This research investigated the place attachment of a heritage place, namely, the Roman
amphitheater, by using a valid model, the Kyle, Graefe, and Manning (2005) model. This model
presents three factors to reach place attachment: place identification, place dependence, and
social bonding. Although the validity of the used model was proved, statistical tests were used
to verify the validity of the collected data because the model was used on a heritage site. In
accordance with the mentioned model, the sample was interviewed using the model
questionnaire to evaluate people's attachment to the heritage place during rush hours. Along
with other statistical tests, the exploratory factor analysis of the sample elaborated that the
Kyle, Graefe, and Manning model is not completely valid for this study, because the results
added a new effective factor, namely, spiritual value. The place attachment estimation was
then examined using the new model. The nature of the place was found to affect the model
used to evaluate its place attachment.
& 2017 Higher Education Press Limited Company. Production and hosting by Elsevier B.V. on
behalf of KeAi. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).

1. Introduction related to people's perception and the meanings that

Contemporary cities are suffering from weakening identity
as manifested through poor city planning and the dismissive
attitude toward community relationships to places (Ujang,
2010). Place attachment is linked to place identity, which is
elaborate the direct correlation between them. Owing to
the consistent immigration to Jordan in past years, the
social demography of the country has been changed.
Specifically, the cultural and intellectual aspects of the
people have changed, which indicate the significant need to

n
Corresponding author.
E-mail address: nessma1990@gmail.com (N.A. Al-Hammadi).
Peer review under responsibility of Southeast University.

https://doi.org/10.1016/j.foar.2017.12.001
2095-2635/& 2017 Higher Education Press Limited Company. Production and hosting by Elsevier B.V. on behalf of KeAi. This is an open access
article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
2 J.S. Goussous, N.A. Al-Hammadi
preserve the people's attachment to heritage places in
order to secure their identity (Ujang, 2010). Accordingly,
the present study used the model of Kyle et al. (2005) to
help identify the place attachment of a heritage place, with
the Roman amphitheater as a case study. The model
provides a valid measurement of place attachment by
examining the social bond, place dependence, and place
identity of a people.
1.1. Issues and objectives
The study area is one of the important heritage places in
Jordan. The Roman amphitheater, located in downtown
Amman (Fig. 1), was built in AD 167–177 into the slope of
a forum constructed in 19 years after the amphitheater. It
accommodates 6000 spectators and has 33 seats (Fig. 2)
(Northedge and Bowsher, 1992)
2. Theoretical background
2.1. Place attachment
Heritage, as the UNESCO defines it, is “our legacy from the
past, what we live with today, and what we pass on to
future generations” (UNESCO World Heritage Centre, 2005).
An individual's identification of heritage places can be
changed significantly over the years, and it addresses the
difficult issue of preserving the identity of a valuable place
(Tonkin, 2012). According to Ujang (2010), “to secure
identity is to ensure continuity in the physical, social
together with meanings and attachment held by the peo-
ple” (p. 64). This idea emphasizes the social role of
preserving place identity (i.e., preserving heritage place
identity). Studies have related place attachment (i.e.,
people's attachment to a place) with place identity in many
Fig. 1 Site of the Roman amphitheater. Source: Northedge and Bowsher (1992).
ways. Place identity is a part of place attachment (Kyle
et al., 2005) and vice versa (Lalli, 1992), but while place
attachment and place identity are connected, they are
different from each other (Hernandez et al., 2007). Fried
(2000) explained that “the operational formulation of
attachment theory, as it applies to place behavior, points
to its origin and meaning in response to the availability of
close, local relationships to people and, by extension, to the
places of relational interaction” (p. 195). Low and Altman
(1992) noted that place attachment contains several hidden
concepts, similar to the symbolic aspect that is related to a
specific environment. They also added that emotional
attributes are attended by awareness and behaviors.
2.2. Dimensions of place attachment
Seamon and Sowers (2008), who relied on the studies of
Edward Relph, stated, “If places are to be more thoroughly
understood, one needs a language whereby we can identify
particular place experiences regarding the intensity of
meaning and intention that a person and place hold for
each other” (p. 4). Some studies also considered the term
“person–place bond” to contain both place identity and
place dependence (e.g., Williams and Roggenbuck, 1989;
Schreyer et al., 1981). In addition, self-dimensions define
the individual's identity in a physical environment through
complicated ideas, viewpoints, predictions, emotions, and
so forth (Proshansky, 1978). This concept explains personal
place attachment. Thus, to differentiate between personal
place attachment and actual place identity, the current
study will refer to the first term as place identification.
However, place dependence focuses on setting efficiency
with regard to serving goal achievement by giving different
alternatives (Jorgensen and Stedman, 2001). Furthermore,
social bonds are emotions of belongingness or people
Place attachment assessment of a heritage place: A case study of the Roman amphitheater in downtown Amman, 3

Fig. 2 Drawings of the Roman amphitheater. Source: Northedge and Bowsher (1992).

Table 1 Interview questions.

The Dimension Interview Questions

PLACE ATTACHEMENT Place Identification Q1 (Meaning): The Roman amphitheater means a lot to me.
Q2 (Identify): I know the historical background of the Roman amphitheater.
Q3 (Attached): I am very attached to the Roman amphitheater.
Q4 (Commitment): I feel commitment to the Roman amphitheater.
Place Dependence Q1 (Activities): The place itself is enjoyable despite the festivals.
Q2 (Best): The amphitheater is the best part of the place.
Q3 (Preference): I wouldn’t prefer any other place.
Q4 (Satisfaction): I feel satisfied on the Roman amphitheater than any other
place.
Social Bond Q1 (Memories): I have a lot of fond memories.
Q2 (Residency): I live near this place.
Q3 (Users): I will (do) bring my children to this place.
Q4 (Special bond): I have a special connection to the Roman amphitheater.

Fig. 3 Research model based on the Kyle, Graefe, and Manning model. Source: Author.
4 J.S. Goussous, N.A. Al-Hammadi

from the interview questions. The questions were derived

Table 2 User classification.
from the study of Kyle et al. (2005), who proved the
Frequency Percent Valid Cumulative efficiency of these questions in measuring place attach-
Percent Percent ment. The interview contained 12 questions, as elaborated
in (Table 1), which were divided as follows:
Family 143 56.3 56.3 56.3
Men 70 27.6 27.6 83.9 – The first four questions measured place identification.
Ladies 41 16.1 16.1 100.0 – The second four questions measured place dependence.
– The last four questions measured social bonding.

connection owing to common characteristics (Kasarda and
Janowitz, 1974). The social relationship with specific set-
tings in the place expresses special meaning and experi-
ences (Kyle et al., 2005). Therefore, a place could be
valued because of its efficiency, its symbolic aspects, and
its relationship with users or even by all three (Moore and
Graefe, 1994). Given that previous literature explained
place attachment in terms of its relation to securing place
identity, the present study will use this concept to measure
the place attachment of the Roman amphitheater as a
heritage place to predict the continuity of its identity.
The study used the questions’ model of the three factors
(variables) that allows individual differentiation between
the three dimensions of place attachment (Fig. 3). The
impact of the nature of the study on the model will also be
examined by using the model on a heritage place. However,
the sample will not be divided into groups because the study
measures the whole sample equivalently and will depend on
the analysis of one model. Consequently, the data were
entered and analyzed through the Statistical Package for
the Social Sciences (SPSS) version 23.0.

3. Method 4. Result

3.1. Study setting
The Roman amphitheater that was built by Emperor Pius is
still used for local activities such as concerts and ceremo-
nies, especially in the summer, spring, and fall (USAID,
2013). This site is considered an essential icon of downtown
Amman, which is near a residential area.
3.2. Design and sample
– Data were collected over the spring (March) of 2016 on
Fridays at noon during the rush hours (3:00 p.m.–5:00 p.
m.).
– The population used in the study consisted of about 510
local users of different ages and genders who were
approached during the rush hours. However, the study
depended on random sampling and covered 254 users,
which is about half of the population.
– The authorities concerned were also interviewed to
confirm their role in emphasizing the social attachment
to the Roman amphitheater.
3.3. Analytical framework
A correlation research design was used to describe the
existing relationship between the variables that resulted
Table 2 elaborates on the nature of the users, who
comprised 143 family members, of whom 70 were individual
men and 41were individual women. These figures indicate
that most of the users of the Roman amphitheater are
families. Despite the lack of historical background for all
users of the Roman amphitheater, they were affected by the
heritage importance of the place. (Fig. 4)
Table 3 describes the means of the variables that are the
most effective on every factor. For instance, place identi-
fication was affected the most by people meanings of the
place. For place dependence, people activities were the
most affected question. Moreover, social bond was affected
by people's opinion on the user type. The table also shows
the standard deviations (SD) of the variables that present
the concentration in the sample answers. As clarified in
Table 3, people answered similarly in the last three ques-
tions for the place identification factor that dealt with
attachment, commitment, and identity feelings to the
place. Therefore, SD in the last three questions was close
to M, which indicated people's focus on a certain answer
(completely agree).
Table 4 presents the reliability of the place attachment
measurement that was evaluated by investigating the
internal consistency (Cronbach's alpha) of the questions
for the three factors of place attachment. The Cronbach's
alpha, as shown in the table, is questionable on two factors,
place dependence and social bond. Thus, a t-test was
necessary to check their validity (Table 5). The test verified

*Factor 1: *Factor 2: *Factor 3: *Factor 4

PI: Attached PD: Activities PI: Meaning PD: Preference
PI: Identification SB: Memories SB: Special bond PD: Best
PI: Commitment Sb: Users SB: Residency
PD: Satisfaction

Fig. 4 Description of factors. Source: Author.

Place attachment assessment of a heritage place: A case study of the Roman amphitheater in downtown Amman, 5

their convergent validity, with all the mean differences in

special
the range of the interval.

bond

2.52
.969
Table 5 reports that the significant t-values Z1.96

SB:
provide evidence of the convergent validity of the data.
Note that this study tested the Chi-square and did not find

1.393
users

2.55
SB:
any significant results due to the small sample so no details
will be addressed. However, this result confirms Kyle et al.’s
(2005) study that the three factors of place attachment are

residency
not correlated with one another. Moreover, a non-significant

1.172
Chi-square indicates a good model fit.

2.50
SB:

Owing to the need to explore the psychometric dimen-

sions (factors) of the research instrument, this study will use
Social bond

memories

exploratory factor analysis (EFA). Fabrigar and Wegener

1.092 (2011) explained that EFA neglects the error variances and
1.82

examines only the common ones in the model. The analysis

SB:

starts with the following steps.

satisfaction

4.1. KMO and Bartlett's test

1.30
.594

The (KMO) measure should be o .70 and is inadequate at

PD:

4.50. The measure shows whether the factor predicted

enough items or not. For the Bartlett test, its significance
preference

value should be 4.05 to indicate the rational base of the

factor that comes from the highly correlated variables
1.000
2.72

(Grimm, 1994). Table 6a shows that the KMO sampling

PD:

adequacy is .579, which is between .70 and .50, which

means it is acceptable. The Bartlett test significance value
Place dependence

is 0.000, which indicates the highly correlated variables.

1.000
best

1.70
PD:

4.2. Commonalities
activities

1.005

It presents the item squared multiple correlations with the

2.90
Pl: meaning Pl: identity Pl: attached Pl: commitment PD:

other items, which should be o .30 to consider a good

value, especially with small samples that usually deform the
results (Grimm, 1994). In the current study, although the
sample considered was small, all the values shown in
Table 6b are above 0.30. In other words, the variables are
correlated with one another.
1.005
1.51

4.3. Total variance explained

Table 6c shows the values of the rotation sums of squared

1.099

loadings, which explain how the four rotated factors could

1.76

elaborate 61.6% of the data variances. The table explains

that the maximum factors that can interpret the data
variances are four. However, if we want to extract only three
Descriptive statistic values.

factors (the three dimensions of the model), it will interpret

Place identification

1.076
1.73

about 50% of the data variances, which is not acceptable.

4.4. Rotated component matrix

As shown in Table 6d, which measures the variable distribu-

2.26
.997

tion for the four extracted factors, the study considered only
the magnitude values of the major values. For further
Std. Deviation

elaboration, the neglected values are indicated in the table.

After the factors were extracted, the study performed
Table 3

the correlation test. First, Table 7a shows the means of the

Mean

factors, which indicate that although factor 4 has only three

variables, it still has the maximum mean. The minimum
mean is for factor 3, which has only two variables. However,
6 J.S. Goussous, N.A. Al-Hammadi

Table 4 Reliability (Cronbach's Alpha).

Cronbach's Cronbach's Alpha Based on Mean (M) Variance Std. Deviation (SD)
Alpha (α) standardized Items

Place Identification .783 .777 7.26 10.549 3.248

Place dependence .023 .124 8.55 3.173 1.781
Social Bond .341 .271 9.39 4.335 2.082

Table 5 One-sample T-test.

t df Sig.(2- tailed) Mean Difference 95% Confidence Interval of the Difference

Lower Upper

Pl: Meaning 36.199 253 .000 2.264 2.14 2.39

Pl: Identity 25.629 253 .000 1.728 1.60 1.86
Pl: Attached 25.567 253 .000 1.764 1.63 1.90
Pl: Commitment 23.978 253 .000 1.512 1.39 1.64
PD: Activities 45.969 253 .000 2.898 2.77 3.02
PD: Best 27.177 253 .000 1.705 1.58 1.83
PD: Preference 41.038 253 .000 2.717 2.59 2.85
PD: Satisfaction 34.860 253 .000 1.299 1.23 1.37
SB: Memories 26.547 253 .000 1.819 1.68 1.95
SB: Residency 33.997 253 .000 2.500 2.36 2.64
SB: Users 29.191 253 .000 2.551 2.38 2.72
SB: Special bond 41.524 253 .000 2.524 2.40 2.64

Table 6a EFA analysis. KMO and Bartlett's test.

Kaiser-Meyer-Olkin Measure of Sampling Adequacy. .579

Bartlett's Test of Sphericity Approx. Chi-Square 1107.599
df 66
Sig. .000

the general factor (summation) is about half of the total

mean for the 12 variables (i.e., 12 questions).
Table 6b EFA analysis. Communalities.
Second, Table 7b presents the correlation between the
Initial Extraction four factors and the general factor. Some inverse correlations
(minus values) in the table relate factor 1 with other factors.
PI: Meaning 1.000 .787 In other words, factor 1 has an inverse relationship with
PI: Identification 1.000 .828 other factors, except for the general one. Moreover, given
PI: Attached 1.000 .840 that the Pearson correlation has a higher value of (F2of4),
PI: Commitment 1.000 .786 factor 2 then correlates greatly with the general factor.
PD: Activities 1.000 .679 The reliability test should be repeated because of the
PD: Best 1.000 .548 change in the study model caused by the extracted factors.
PD: Preference 1.000 .508 This study presented the reliability data in one table, but it
PD: Satisfaction 1.000 .339 tested the factors individually for more accuracy. Table 8a
SB: Memories 1.000 .670 indicates that the variables of factor 1 are the only
SB: Residency 1.000 .399 variables that have internal consistency because the Cron-
SB: Users 1.000 .436 bach's alpha is 0.853 4.70. The Cronbach's alpha values of
SB: Special bond 1.000 .576 the other factors are less than 0.7, which reveals a sample
bias to certain answers.
Place attachment assessment of a heritage place: A case study of the Roman amphitheater in downtown Amman, 7

Table 6c EFA analysis. Total Variance Explained.

Component Initial Eigenvalue Explained Rotation Sums of Squares loading

Total % of Variance Cumulative % Total % of Variance Cumulative %

1 3.127 26.057 26.057 3.088 25.730 25.730

2 1.590 13.252 39.309 1.563 13.027 38.757
3 1.422 11.852 51.161 1.414 11.780 50.537
4 1.256 10.465 61.626 1.331 11.089 61.626
5 .991 8.261 69.888
6 .914 7.617 77.505
7 .786 6.550 84.055
8 .713 5.945 90.000
9 .536 4.467 94.467
10 .326 2.720 97.188
11 .282 2.352 99.539
12 .055 .461 10.000

Moreover, Satisfaction (PD) was the most effective variable

Table 6d EFA analysis. Rotated Component Matrix. that could maximize the variance and the Cronbach's alpha.
However, since Table 8c was used to examine the reliability
Component of variables, the results of the variables varied. As noted,
the Cronbach's alpha of the whole test was less than 0.7,
1 2 3 4
which indicates that the test was invalid.
Pl: Attached .914
Pl: Identification .902
Pl: Commitment .823 5. Discussion
PD: Satisfaction .580
PD: Activities .713 The purpose of this assessment was to examine the place
SB: Memories .666 .350 attachment of a heritage place (i.e., the Roman amphithea-
SB: Users .607 ter) by using a valid model to investigate the impact of
Pl: Meaning .873 place attachment of a heritage place on that model. The
SB: Special bond .405 .602 analysis initially adopted the Kyle et al. (2005) measure-
PD: Preference .687 ment tools, starting with interview questions and ending
PD: Best .421 .556 with an EFA analysis. However, the data were analyzed as
SB: Residency .542 follows.

(1) Frequency of users

The users were elaborated and found to be mostly family
members, which implies that people consider the Roman
Table 7a Correlation test. Descriptive Statistics. amphitheater as a park or an amusement place for
families.
Mean Std. Deviation N (2) Means and Standard Deviation of

F1of4 6.3307 3.22062 254 – Means: The most affected variable on every factor is as
F2of4 6.8504 2.47542 254 follows (Table 2):
F3of4 4.7874 1.56357 254 *Place Identification people meanings
F4of4 6.9213 1.93207 254 *Place Dependence people activities
General F (SUM) 24.8898 4.54878 254 *Social Bond user type
– Standard Deviations (SD): The concentration in the sam-
ple answers is as follows (Table 2):

The study examined reliability in two ways: by factors in *Similarities PI: (attachment, commitment,
Table 8b and by variables in Table 8c. By examining identity)
reliability using the factors in Table 8b, Special bond (SB) *Bias SB: concentrate on a certain answer
was noted to be the most affected variable on the measure- (completely agree)
ment tool variance because it has the minimum scale
means, which indicate the sample tended to answer. (3) Reliability of the three factors (PI, PD, SB)
8 J.S. Goussous, N.A. Al-Hammadi

Table 7b Correlation test. Correlation.

F1of4 F2of4 F3of4 F4of4

F1of4 Pearson Correlation 1 .025 .089 .262

Sig. (2-tailed) .686 .158 .000
N 254 254 254 254
F2of4 Person Correlation .025 1 .101 .153
Sig. (2-tailed) .686 .108 .015
N 254 254 254 254
F3of4 Person Correlation .089 .101 1 .055
Sig. (2-tailed) .158 .108 .386
N 254 254 254 254
F4of4 Person Correlation .262 .153 .055 1
Sig. (2-tailed) .000 .015 .386
N 254 254 254 254
General F (SUM) Person Correlation .552 .626 .359 .341
Sig. (2-tailed) .000 .000 .000 .000
N 254 254 254 254

– Cronbach's alpha: Table 4

Table 8a Cronbach's Alpha test. Reliability Statistics.

Cronbach's Cronbach's Alpha N of PD and SB have questionable values.

Alpha Based on Standar- Items
dized Items (4) T-test
(Table 5): Significant t-values were Z 1.96, which means
Factor 1 .853 .843 4 a convergent validity of the data.
Factor 2 .441 .455 3 (5) Exploratory Factor Analysis (EFA)
Factor 3 .420 .420 2
Factor 4 .099 .079 3 This analysis explores the psychometric factors of the 12
General .380 .386 12 variables.
Factor
– KMO and Bartlett's Test:

(Table 6a):

Table 8b Individual reliability test. Item-Total statistics.

Factor No. Variables Scale Mean if Scale Variance Corrected Item Squared Multiple Cronbach's Alpha
Item Deleted if Item Deleted Total Correlation Correlation if Item Deleted

F1 PI: Identification 4.57 5.099 .843 .873 .744

PI: Attached 4.57 4.926 .868 .887 .730
PI: Commitment 4.82 5.864 .719 .536 .802
PD: Satisfaction 5.03 8.592 .410 .194 .909
F2 PD: Activities 3.9528 4.203 .222 .064 .422
SB: Memories 5.0315 3.509 .349 .123 .214
SB: Users 4.7165 2.741 .261 .081 .394
F3 PI: Meaning 2.52 .938 .266 .071
SB: Special bond 2.26 .993 .266 .71
F4 PD: Best 5.22 2.291 .055 .019 .298
PD: Preference 4.20 2.448 .052 .048 .061
SB: Residency 4.42 1.850 .160 .034 .284
 Place attachment assessment of a heritage place: A case study of the Roman amphitheater in downtown Amman, 9

Table 8c General reliability test. Item-Total Statistics.

Variables Scale Mean if Scale Variance if Corrected Item Total Squared Multiple Cronbach's Alpha if
Item Deleted Item Deleted Correlation Correlation Item Deleted

PD: Best 23.1850 19.740 .005 .258 .405

PD: Preference 22.1732 19.575 .000 .175 .405
SB: Residency 22.3898 20.247 .088 .208 .442
PI: Meaning 22.6260 18.235 .172 .334 .347
PI: Identification 23.1339 17.271 .251 .887 .316
PI: Attached 23.1260 16.743 .304 .900 .294
PI: Commitment 23.3780 16.766 .354 .639 .282
PD: Activities 21.9921 19.352 .037 .327 .391
PD: Satisfaction 23.5906 19.539 .152 .312 .361
SB: Memories 23.0709 16.248 .369 .275 .268
SB: Special bond 22.3661 19.972 .025 .244 .409
SB: Users 22.7559 16.707 .156 .152 .352

Fig. 5 Final model. Source: Author.

*KMO sampling adequacy is .70 4 .579 4 .50 – (Table 7a):

Acceptable * Factor 4 has the maximum mean Three variables
*The significant value of the Bartlett test is .000 Factor 3 has the minimum mean Two variables
Correlated variables – (Table 7b): Inverse correlations for factor (1) with the
other factors.
(6) Communalities
(8) Repeated reliability test
(Table 6b): The communalities are o .30 Vari-
ables are correlated with one another – Cronbach's alpha test
(Table 8a):
– Total Variance Explained: *The Cronbach's alpha of Factor 1 is .853 4 .70
(Table 6c): Four factors elaborate 61.6% of the data Consistency of internal variables
variances. *The Cronbach's alpha of other factors is o .7
– Rotated Component Matrix: (Table 6d): Sample bias to certain answers
– Individual reliability test
The data are approximately and acceptably disrupted by
the four factors through the neglect of small values of the (Table 8b):
repeated ones. *(SB: Special bond), (SB: Memories), and (PI: Meaning) are
the most affected variables on the measurement tool
(7) Correlation test variance.
*(PD: Satisfaction) greatly affects reliability, and Cronba-
ch's alpha can be significantly increased if it is deleted.
10
6. The study model
After the previous statistical tests, this research suggests a
developed model. Given that the study estimated the
attachment of a heritage place, the Roman amphitheater,
the uniqueness of the place affected the measurement
model, namely, the Kyle et al. (2005) model. The results
of the EFA analysis clarified that a heritage place attach-
ment cannot be correctly represented by extracting only
three factors, thus addressing the need for a fourth factor.
As the first model had three factors (place identification,
place dependence, social bond), this study will suggest
spiritual values as the fourth factor.
Fig. 5 shows the new factor variables that contain the
people's special bond and subjective meaning to the place.
Owing to the nature of the variables, the study named the
new factor as “spiritual value.” In the end, the general
factor, which is mentioned in Table 7a, represents the
PLACE ATTACHMENT that has half the mean of the total.
This figure means that people attachment to the Roman
amphitheater is neutral (not high and not low). It also
indicates the need for increasing place attachment to
heritage places. Note that since the 1960s, the authorities
have attempted to help in this matter by organizing
festivals from music ceremonies to boxing matches. Nowa-
days, the Ministry of Tourism and Antiquities is trying to
maintain the same vitality in the place by organizing
festivals for different occasions, especially during the spring
semester. The consistent maintenance and the convenient
security of the place likewise increase the visitors and,
consequently, it will increase people's attachment to the
place.
7. Conclusion
This research aimed to assess the place attachment of the
Roman amphitheater in downtown Amman, a heritage place
that sits near a residential area. Many studies have defined
place attachment and related it to several concepts. In this
research, we relied on Kyle et al. (2005), who developed a
valid model to measure place attachment by place identi-
fication, place dependence, and social bond. Owing to the
specialty of the heritage place, the model used did not
perfectly fit the study, as shown by the statistical results.
Statistical tests suggested a new model that contains four
factors instead of the mentioned three; it adds spiritual
value as an effective factor of place attachment. In other
words, the nature of the place affects the model that
J.S. Goussous, N.A. Al-Hammadi
evaluates its attachment. Eventually, after the study ana-
lysis, place attachment was found to have a neutral
condition that requires greater efforts to raise up.
References
Fabrigar, L.R., Wegener, D.T., 2011. Exploratory factor analysis.
Oxford University Press.
Fried, M., 2000. Continuities and discontinuities of place. J.
Environ. Psychol. 20, 193–205.
Hernandez, B., Hidalgo, M.C., Salazar-Laplace, M.E., Hess, S.,
2007. Place attachment and place identity in natives and non-
natives. J. Environ. Psychol. 27, 310–319.
Jorgensen, B.S., Stedman, R.C., 2001. Sense of place as an
attitude: lakeshore owners' attitudes toward their properties.
J. Environ. Psychol. 21, 233–248.
Kasarda, J., Janowitz, M., 1974. Community attachment in mass
society. Am. Sociol. Rev. 39, 328–339.
Kyle, G., Graefe, A., Manning, R., 2005. Testing the dimensionality
of place attachment in recreational settings. Environ. Behav. 37,
153–177.
Lalli, M., 1992. Urban-related identity: theory, measurement and
empirical findings. J. Environ. Psychol. 12, 285–303.
Low, S.M., Altman, I., 1992. Place attachment: a conceptual
inquiry. In: Altman, I., Low, S.M. (Eds.), Place Attachment.
Plenum, New York, pp. 1–12.
Moore, R.L., Graefe, A.R., 1994. Attachments to recreation set-
tings: the case of rail-trail users. Leis. Sci. 16, 17–31.
Northedge, A., Bowsher, J., 1992. Studies on Roman and Islamic
ʻAmmān: The Excavations of Mrs. C-M Bennett and Other
Investigations. Published for the British Institute in Amman for
Archaeology and History by Oxford University Press, Oxford.
Proshansky, H.M., 1978. The city and self-identity. Environ. Behav.
10, 147–169.
Schreyer, R., Jacob, G., White, R., 1981. Environmental meaning as
a determinant of spatial behavior in recreation. Proc. Appl.
Geogr. Conf. 4, 294–300.
Seamon, D., Sowers, J., 2008. Place and Placelessness, Edward
Relph, Human Geography. David Seamon & Jacob Sowers, 43–51.
Tonkin, S., 2012. WHAT IS HERITAGE?, Australian Heritage Strategy,
Public Consultation Paper.
Ujang, Norisdah, 2010. Place Attachment and Continuity of Urban
Place Identity. Department of Landscape Architecture, Faculty
of Design and Architecture, University Putra Malaysia (UPM).
UNESCO, 2005. UNESCO World Heritage Centre - New Inscribed Proper-
ties. Retrieved November 30, 2017, from 〈www.whc.unesco.org〉.
USAID, 2013. A Walk Around Al-Balad Downtown Amman. Jordan
Tourism Development Project (Siyaha).
Williams, D.R., Roggenbuck, J.W., 1989. Measuring place attach-
ment: Some preliminary results Abstracts of the 1989 Leisure
Research Symposium (p. 32).