The international journal of tourism research tập 13, số 01, 2011 01 + 02

INTERNATIONAL JOURNAL OF TOURISM RESEARCH
Int. J. Tourism Res. 13, 1–16 (2011)
Published online 26 March 2010 in Wiley Online Library
(wileyonlinelibrary.com) DOI: 10.1002/jtr.779

Taiwan’s International Tourism: A Time
Series Analysis with Calendar Effects
and Joint Outlier Adjustments
Hui-Lin Lin1,*, Lon-Mu Liu2, Yi-Heng Tseng3 and Yu-Wen Su1
Department of Economics, National Taiwan University, Taipei, Taiwan
2
Public Economics Research Center, College of Social Sciences, National Taiwan University, Taipei, Taiwan
3
Department of International Business, Yuan Ze University, Chung-Li, Taiwan

1

ABSTRACT
In this paper, we examine monthly tourist
arrivals from Japan, Hong Kong and the
USA between January 1971 and December
2008. Our purpose is to find events or
variables that affect Taiwan’s international
tourism. We find that the Chinese New Year
has a positive effect on tourist arrivals from
Hong Kong, but negative effects for other
countries. Through outlier detection, we
obtain a better understanding of the effects
of non-recurring events that have impacted
Taiwan’s international tourism. Using

transfer function model with automatic
outlier detection and adjustment, we find
that the exchange rate influences tourist
arrivals from Japan and Hong Kong.
Copyright © 2010 John Wiley & Sons, Ltd.
Received 3 November 2009; Revised 22 February 2010;
Accepted 23 February 2010

Keywords: international tourism; Taiwan;
time series; calendar effects; outliers.
INTRODUCTION

D

uring the past 40 years, the economic
structure of Taiwan (the Republic of
China) has changed greatly. It has
evolved from a mainly agriculture-based
economy in the 1970s to a technology-based
economy in recent years. The rapid growth of
*Correspondence to: Dr. H.-L. Lin, Economics Department, National Taiwan University, No.21 Hsu-Chow
Road, Taipei 100, Taiwan.
E-mail:

Taiwan’s economy inevitably gives rise to
many new economic and societal issues, such
as a high demand for energy, increases in air
and water pollution, and an M-shaped income
distribution (i.e. the rich gets richer and the
poor gets poorer). To partially address such

societal issues, the Taiwan government has
refocused its attention on international tourism,
which was overshadowed by manufacturing
and technology-based industries in the past.
More specifically, according to the statistical
reports of the World Travel and Tourism
Council, the annual output value of the Taiwan
tourism sector amounted to US$19.7 billion in
2008 or 4.7% of Taiwan’s gross domestic
product (GDP), which totalled US$419 billion.
According to the national statistics released by
the directorate-general of Budget, Accounting
and Statistics of Taiwan in 2007, the manufacturing sector accounted for 23.76% of GDP. In
that same year, the agriculture sector accounted
for only 1.45% of GDP. Relatively speaking,
the tourism sector is about one-fifth of the size
of the manufacturing sector and is thus already
an important component of Taiwan’s economy.
Globally, Taiwan has a moderate ranking in
tourism. According to the World Economic
Forum (2009), the Travel and Tourism Competitiveness Index is composed of three major subindices of the travel and tourism sector, namely,
the regularity framework sub-index, the business environment and infrastructure sub-index,
and the human, cultural and nature sub-index.
Based on the index, Taiwan was ranked 30th in
the Travel and Tourism Competitiveness Index
among the 124 countries reported. In comparison, China was ranked 71st and Korea 42nd,
while Japan was ranked 25th, Hong Kong the
sixth and the USA the fifth based on this index.
Copyright © 2010 John Wiley & Sons, Ltd.

2
In recent decades, there has been keen interest in tourism studies in how tourism demand
is affected by various cultural, economic and
institutional factors, as well as major ‘one-time’
events. In such studies, tourist arrivals have
been the most frequently used dependent variable in quantitative analyses (e.g. Martin and
Witt, 1989; Kulendran and King, 1997; Song and
Witt, 2006). Lim (1997) reviewed 124 tourismrelated studies and concluded that 67 of these
studies used tourist arrivals and 54 used tourism
expenditures as the dependent variable. Lim
(1997) also reviewed several commonly used
explanatory variables, such as income, relative
tourism prices, transportation costs, exchange
rates, the time trend, seasonal factors, economic
activity indicators, lagged dependent variables,
marketing and promotion, as well as various
qualitative factors. Among such explanatory
variables, dummy variables were typically used
to deal with the influence of qualitative factors,
including well-known factors such as seasonal
variation (e.g. Goh and Law, 2002; Hui and
Yuen, 2002) and ‘one-time’ events (e.g. Ryan,
1993; Chen et al., 1999; Goodrich, 2001; Huang
and Min, 2002; Kim et al., 2006; Athanasopoulos
and Hyndman, 2008). Such an approach was
also used by Wang (2008) to study four major
local or international disasters potentially relevant to Taiwan’s international tourism: the
Asian financial crisis in 1997, the major earthquake on 21 September 1999 in Taiwan, the terrorist attacks on 11 September 2001 in the USA
and the outbreak of severe acute respiratory

syndrome (SARS) in 2003.
In most studies, traditional regression models
with dummy variables (e.g. Witt and Witt, 1995;
Wang, 2008) or Autoregressive Integrated
Moving Average-related models (e.g. Goh and
Law, 2002; Chu, 2008) were typically employed.
Recently, a rough sets approach was used to
study tourism (Goh et al., 2008). It has the
advantage of being straightforward and directly
interpretable. It considered various economic
and non-economic factors as well as month in a
year. However, it did not consider effects
because of one-time events or calendar variation as shown in this paper. In this paper, both
ARIMA and transfer function time series models
will be used. Effects because of calendar variation are included in the models, and the onetime events are handled through automatic
Copyright © 2010 John Wiley & Sons, Ltd.

H.-L. Lin et al.
outlier detection and estimation in the context
of time series modelling.
In this research, our primary interest is to
study major factors or events that affect international tourism in Taiwan. Such factors or
events may be classified as recurring or nonrecurring in nature. Both will be studied in this
paper. On recurring factors, besides calendar
variables, we focus on investigating the impact
of exchange rate as previous researches (see
e.g. Crouch et al., 1992; Lim, 1997) demonstrate
that exchange rate has a significant influence
on tourism. However, they did not apply time
series models using joint estimation of model

parameters and outlier effects. With rigorous
time series analysis, these models will allow
Taiwan to obtain information and knowledge
to better allocate its resource for promotion
and expansion of international tourism as well
as providing a better ongoing tourism service.
Before studying international tourism in relation to Taiwan, we first provide an overview of
worldwide international tourism at both the
national and regional levels in Section 2: International Tourism Worldwide and Taiwan. This
is then followed by an introduction to the international tourist arrivals into Taiwan. We have
an extensive collection of monthly tourist arrivals data into Taiwan from various countries and
regions between 1971 and 2008, with each series
having 456 observations. In Section 3: Time
Series Models for The Analysis of Taiwan's
Tourism, Box–Jenkins time series models with
calendar effects are introduced. The parameters
of such models are estimated using a joint estimation method of model parameters and outlier
effects in Section 4: Analysis of Calendar Effects.
The effects of recurring and non-recurring
events are presented and discussed in that
section as well. In Section 5: The Effects of
Exchange Rates on Taiwan's Tourism, we
examine the effects of foreign exchange rates on
international tourist arrivals into Taiwan from
major countries. In Section 6: Discussion and
Conclusion, we provide a discussion as well as
the conclusion to this paper.
INTERNATIONAL TOURISM WORLDWIDE
AND TAIWAN
Even though our primary interest is to

study international tourism in Taiwan, it is
Int. J. Tourism Res. 13, 1–16 (2011)
DOI: 10.1002/jtr


Taiwan’s International Tourism

3

Table 1. International tourist arrivals into various regions and countries
Country

Worldwide

Euro area

USA

ASEAN

China

Japan

Taiwan

1995
1996
1997
1998

1999
2000
2001
2002
2003
2004
2005
2006
2007

538 991 962
573 555 988
599 266 326
614 287 034
639 895 637
688 000 108
683 748 532
703 790 318
691 603 848
763 037 433
803 813 233
850 778 039
908 000 000

201 613 206
207 430 975
220 470 384
233 882 000
245 313 000
260 090 915

258 187 000
262 947 000
259 565 000
260 115 000
268 505 000
284 903 000
315 935 000

43 490 000
46 636 000
47 875 000
46 377 000
48 509 000
51 238 000
46 927 000
43 581 000
41 218 000
46 086 000
49 206 000
50 978 000
55 986 000

23 999 000
24 611 000
23 939 000
22 872 000
26 964 000
30 860 000
33 875 000
35 800 000

30 801 000
40 916 000
43 495 000
48 189 000
54 068 000

20 034 000
22 765 000
23 770 000
25 073 000
27 047 000
31 229 000
33 167 000
36 803 000
32 970 000
41 761 000
46 809 000
49 913 000
54 720 000

3 345 000
3 837 000
4 218 000
4 106 000
4 438 000
4 757 000
4 772 000
5 239 000
5 212 000
6 138 000

6 728 000
7 334 000
8 347 000

2 331 934
2 358 221
2 372 232
2 298 706
2 410 248
2 624 037
2 617 137
2 602 037
2 248 117
2 950 342
3 378 118
3 519 827
3 716 063

ASEAN, Association of Southeast Asian Nations.

important to have a good understanding of
international tourism worldwide. Table 1 lists
the annual tourist arrivals worldwide as well
as the tourist arrivals in several important
regions and countries. Except for Taiwan,
tourist arrivals for the different regions and
counties are obtained from the World Tourism
Organization. The annual tourist arrival data
for Taiwan are provided by the Tourism
Bureau, Ministry of Transportation and Communications, China. To facilitate a better

understanding of tourism growth in each
region/country, annual tourist arrivals are
indexed to 1995 levels (i.e. the numbers in 1995
are set to 100) and displayed in Figure 1 (A,B).
The tourist arrival indices for the world as a
whole and the USA are displayed in Figure 1
(A,B) to facilitate the visual comparison.
From Table 1 and Figure 1, we find that the
Euro area accounts for more than one-third of
worldwide international tourism each year
and that the USA accounts for roughly 6–8%
of worldwide tourism. However, the growth
of international tourist arrivals in the Euro
area, along with the growth in the USA, has
slowed substantially in recent years. International tourism has grown at a significantly
faster rate in Asia, including China, Japan and
the Association of Southeast Asian Nations
(ASEAN) area, despite the 11 September terrorist attacks in 2001 and the SARS epidemic
in 2003. While the growth of international
tourism in Taiwan has been smaller in comCopyright © 2010 John Wiley & Sons, Ltd.

parison with that in other countries or regions,
the pace seems to have picked up following the
SARS epidemic in 2003.
In this study, our primary interest is to study
major factors or events that affect international
tourism in Taiwan. Based on the total tourist
arrivals data for 2008, the international tourists
visiting Taiwan came primarily from the following five regions or countries: Japan (28.3%),
Hong Kong (16.1%, including Macao), the

ASEAN area (16.7%), the USA (10.1%) and
Europe (5.2%). In Figure 2 (A), the total tourist
arrivals in each month between January 1971
and April 2008 are displayed. From this graph,
we find that the total tourist arrivals exhibit a
general upward trend. While this trend was
severely affected by the SARS epidemic in
2003, it resumed with higher growth following
the SARS outbreak.
As the total number of tourist arrivals is an
aggregate of many time series, its properties
are harder to interpret and less meaningful in
their application. To improve our study, tourist
arrivals from major countries are displayed in
Figure 2 (B–D). The solid lines in Figure 2
(B–D) represent monthly tourist arrivals from
Japan, Hong Kong and the USA, the three principal sources of international tourism for
Taiwan, and it is these that are the primary
focus of this study. The dashed lines in Figure
2 (C,D) represent monthly tourist arrivals from
the European and ASEAN areas. As these two
series are also an aggregation of tourist arrivals
Int. J. Tourism Res. 13, 1–16 (2011)
DOI: 10.1002/jtr


4

H.-L. Lin et al.

Tourist Arrivals (Indexed to 1995)

280

(A)
Worldwide
United States
China

240

Japan
Taiwan

200

160

120

80
1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007
280

Tourist Arrivals (Indexed to 1995)

(B)
Worldwide
Euro Area

240

ASEAN
United States

200

160

120

80
1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007

Figure 1. International tourist arrivals into various regions and countries. ASEAN, Association of Southeast
Asian Nations.

from various countries, they are not our focus
in this research study and are provided for
information only.
Tourist arrivals from Japan, Hong Kong and
the USA were obviously impacted by the SARS
epidemic in 2003. However, their historical
temporal patterns are quite different. The
average number of tourist arrivals from Japan
is much higher than that from Hong Kong and
the USA, but its growth rate has been much
smaller than the corresponding growth rates
for the other two areas in recent years. The
numbers of tourist arrivals from the USA are

Copyright © 2010 John Wiley & Sons, Ltd.

much smaller than the corresponding numbers
of arrivals from Japan and Hong Kong, but
they display a persistent upward trend. The
numbers of tourist arrivals from Japan and
Hong Kong have sometimes declined or have
remained the same for extended periods of
time.
As in the cases of many other tourist arrival
time series, the numbers of international tourist
arrivals in Taiwan seem to fluctuate seasonally. We display the average monthly tourist
arrivals into Taiwan from Japan, Hong Kong
and the USA, as well as the total international
Int. J. Tourism Res. 13, 1–16 (2011)
DOI: 10.1002/jtr


Taiwan’s International Tourism
400000

(A)

350000

Tourist Arrivals

5

Worldwide

300000
250000
200000
150000
100000
50000
0
1972

1974

1976

1978

1980

1982

1984

1986

1988

1990

1992

1994

1996

1998

2000

2002

2004

2006

2008

1976

1978

1980

1982

1984

1986

1988

1990

1992

1994

1996

1998

2000

2002

2004

2006

2008

1982

1984

1986

1988

1990

1992

1994

1996

1998

2000

2002

2004

2006

2008

1982

1984

1986

1988

1990

1992

1994

1996

1998

2000

2002

2004

2006

2008

140000

(B)
Tourist Arrivals

120000

Japan

100000
80000
60000
40000
20000

0
1972

1974

70000

(C)
Tourist Arrvials

60000

Hong Kong

Euro Area

50000
40000
30000
20000
10000
0
1972

70000

Tourist Arrivals

60000

1974

1976

1978

1980

(D)
United States

ASEAN

50000
40000
30000
20000
10000
0
1972

1974

1976

1978

1980

Figure 2. Monthly international tourist arrivals into Taiwan (1/1971–12/2008). ASEAN, Association of

Southeast Asian Nations.
Copyright © 2010 John Wiley & Sons, Ltd.

Int. J. Tourism Res. 13, 1–16 (2011)
DOI: 10.1002/jtr


H.-L. Lin et al.
200000

100000

80000
150000
60000
100000
40000
50000
20000

0

Average Monthly Tourist Arrivals into Taiwan

Average Monthly Tourist Arrivals into Taiwan (Worldwide)

6

0
1

2

3

4

5

6

Worldwide
Japan

7

8

9

10

Hong Kong
United States

11

12
Month

Figure 3. Average monthly tourist arrivals into Taiwan (1/1971–12/2008, excluding the year 2003 because
of the severe acute respiratory syndrome epidemic).

tourist arrivals into Taiwan in Figure 3. Note
that the data for 2003 are excluded from the
monthly averages because the SARS epidemic
affected international tourism severely in that
year. Except for October and December, the
seasonal pattern for total Taiwan international
tourist arrivals is very similar to the seasonal
pattern for Japan as Japan is the major source
of Taiwan’s international tourism.
In the case of Japan, the numbers of tourist
arrivals are higher between January and March,
lower between April and October (with July
the lowest in a year), become higher in November and decline to a lower level in December.
This pattern is rather different from the international tourist arrivals in the USA and Europe
where the summer months and Christmas
period tend to have higher numbers of tourist
arrivals. The monthly tourist arrival pattern
for Japan may, to a large degree, be related to
the differences in climate between Taiwan and
Japan. The climate in Taiwan between January
and March is much more temperate than that
in Japan and is thus more appealing to Japanese tourists. The summer months (particularly between June and October) in Taiwan are
much hotter than in Japan and are thus less
appealing to Japanese tourists. The climate in
Taiwan in November and December may be
warmer than in Japan, but Taiwan seems to
Copyright © 2010 John Wiley & Sons, Ltd.

lose Japanese tourists to the USA/Europe in
December. As for Hong Kong and the USA, the
tourist arrival patterns are somewhat different
from that for Japan. In these two areas, the
summer months (June to August) and December continue to have relatively high numbers
of tourist arrivals into Taiwan, and the tourist
arrivals in October are particularly high
because of the most important governmentsponsored national holiday celebration that is
held on 10 October each year.
TIME SERIES MODELS FOR THE
ANALYSIS OF TAIWAN’S TOURISM
International tourist arrivals may be affected
by external factors that can be classified as
recurring variables and non-recurring events.
Non-recurring events, such as the 11 September terrorist attacks and the SARS epidemic,
can only be represented by discrete indicator
variables. Recurring variables such as exchange
rates and other economic variables are data
collected systematically and can be represented
by various forms of time series. As tourist
arrivals in Taiwan are compiled as monthly
data, tourist arrivals may be influenced by calendar variation. Calendar variation is recurring in nature, and it is very important to
account for its effects in the analysis of monthly
Int. J. Tourism Res. 13, 1–16 (2011)
DOI: 10.1002/jtr


Taiwan’s International Tourism

7

time series. We shall discuss two important
kinds of calendar effects below.
In Taiwan, all official monthly economic statistics are compiled according to the Gregorian
calendar, just as in all other countries in the
world. However, the dates of all traditional
Chinese festivals and holidays are set according to the Chinese lunar calendar. The parallel
use of these two calendars gives rise to substantial issues in the analysis of monthly time
series data for these countries. The most problematic issue is that where important traditional festivals or holidays fall during different
Gregorian calendar months from year to year.
For example, even though the Chinese New
Year is always on the first of the first month
each year according to the Chinese lunar calendar, it may fall in either January or February
based on the Gregorian calendar. As tourist
arrivals may be greatly affected by the Chinese
New Year, the observed time series may vary
substantially, depending on whether a particular month (January or February) includes the
Chinese New Year or not. Such effects are
referred to as moving-holiday effects (Liu, 1980,
1986, 2006). In addition to moving holidays,
the number of tourist arrivals may depend on
the days of the week. As the composition of
days of the week varies from month to month
and year to year, the observed series may be
affected by such variation as well. Such effects,
which are by and largely because of the composition of trading days (or work days) in each
month, are referred to as trading-day (or working-day) effects (Hillmer et al., 1981; Hillmer,
1982; Bell and Hillmer, 1983).

A general time series model for
tourism analysis
Assuming that Yt is a time series that may be
subject to the influences of recurring variables
and non-recurring events, a general time series
model for Yt can be written as
Yt = C + f (ω , X t ) + N t , N t =

θ ( B)
at ,
(
D B) φ ( B)

at ∼ i.i.d. N (0, σ a2 ) , t = 1, 2, . . . , n

(1)

where B is the backshift operator (i.e. BYt =
Yt−1), C is a constant term, f (ω , X t ) represents
Copyright © 2010 John Wiley & Sons, Ltd.

the total exogenous effects at time t, D(B) is the
differencing operator, φ(B) is the autoregressive operator, θ(B) is the moving average operator, n is the number of observations, and at’s
are independently and identically distributed
(i.i.d.) in normal distribution with mean 0 and
variance σ a2 . The operators D(B), φ(B) and θ(B)
can be expressed in simple or multiplicative
form as shown in Box and Jenkins (1976). The
function f (ω , X t ) can be either in linear or nonlinear form. In this study, we consider a class
of linear and non-linear dynamic relationship

functions (often referred to as transfer functions) described in Box and Jenkins (1976).
Using the terminology of transfer function
modelling (Box and Jenkins, 1976; Liu, 2006),
Nt is referred to as the disturbance or noise of
the model. In the above model, X t contains
variables X1t, X2t, . . . , Xmt that are used to characterise the effects because of various recurring
variables, and ω is a vector of parameters
reflecting the effects of such variables. Even
though the effects because of non-recurring
events (e.g. the SARS epidemic, 11 September
attacks, etc.) may be included in f (ω , X t ) with
the X it ’s being indicator variables, it is more
flexible to treat such events as outliers (Fox,
1972; Chang et al., 1988). Using an estimation
procedure developed by Chen and Liu (1993),
we can automatically detect outliers (nonrecurring events) and perform joint estimations of the outlier effects and model parameters.
Such an approach allows us to account for the
effects of both known and unknown nonrecurring events more effectively.
Model (1) can also be expressed in the following alternative form
D (B ) Yt = C′ + D (B ) f ( ω , X t ) + N ′t ,
θ (B)
N ′t =
at
φ (B)

(2)

where the differencing operator is applied to
both response and explanatory variables. The
latter form of model is used in the estimation

of the model parameters.
Time series models with calendar effects
We are interested in economic variables or
certain tourism-related events that may affect
tourist arrivals to Taiwan from a prospective
Int. J. Tourism Res. 13, 1–16 (2011)
DOI: 10.1002/jtr


8

H.-L. Lin et al.

country/city. To study the effects of such variables or events, we must include important
calendar effects in the model first. Calendar
effects are of interest in this study themselves.
Furthermore, they also may be viewed as
important ‘nuisance’ effects that must be taken
care of before further econometric modelling is
conducted. In addition to Chinese New Year
(referred to as H1t), there are two other major
moving holidays that potentially may affect
Taiwan international tourism. These are the
Dragon Boat Festival (H2t) and the Autumn
Moon Festival (H3t). The Dragon Boat Festival
is on 5 May of the Chinese lunar calendar and
may vary between May and June of the Gregorian calendar. The Autumn Moon Festival is on
15 August of the Chinese lunar calendar and
may vary between September and October of
the Gregorian calendar. A model with such

moving-holiday effects can be expressed as
f (α 1 , β1 , γ 1 , H1t , H 2t , H 3t ) = α 1H1t + β1H 2t + γ 1H 3t
(3)
if such moving-holiday effects are the same
over years (i.e. staying constant), where the
variable Hit (i = 1,2,3) represents the proportion
of a particular holiday in the t-th month. Here,
we assume that extra tourist arrival changes
(either increases or decreases) because of
Chinese New Year are distributed uniformly
during a 10-day period beginning 3 days prior
to the New Year and 7 days during the festival.
As for the Dragon Boat and Autumn Moon
Festivals, we assume that the tourist changes
are distributed uniformly during a 5-day
period beginning 2 days prior to the festival
and 3 days over the duration of the festival.
The assumptions for the length and effect distribution of the festivals are not crucial as most
of the festivals fall in the same months instead
of splitting across two adjacent months. If
these moving-holiday effects increase (or
decrease) linearly over the years (i.e. having an
upward or downward trend), then the following model may represent the effects more
adequately:
f (α 1, α 2 , β1, β 2 , γ 1, γ 2 , H1t , H 2t , H 3t , K t )
= α 1H1t + α 2 H1t × K t + β1H 2t +
β 2 H 2t × K t + γ 1H 3t + γ 2 H 3t × K t
Copyright © 2010 John Wiley & Sons, Ltd.

where Kt is 1 for all Kt in the first year, 2 for all

Kt in the second year and so on.
As for trading-day effects, the following
model may be considered
7

f ( ξ1 , . . . , ξ7 , W1t , . . . , W7t ) = ∑ ξ i Wit

(5)

i =1

where Wit, i = 1, 2 , . . . , 7 represent the number
of Mondays, Tuesdays, . . . and Sundays in the
t-th month, respectively, and ξi, i = 1, 2, . . . , 7
are the effects because of Monday,
Tuesday, . . . and Sunday. To avoid multicollinearity, it is desirable to restrict trading-day
effects to vary around zero, or equivalently
imposing ξ1 + ξ2 + . . . + ξ7 = 0. Thus the model
in Equation (5) can be written as
6

f (δ1 , . . . , δ 6 , D1t , . . . , D 6t ) = ∑ δ i D it

(6)

i =1

where Dit = Wit − W7t and δi = ξi, i = 1, 2, . . . , 6
are the effects because of Monday, Tuesday, . . . ,
Saturday, and the effect for Sunday is (−ξ1

−ξ2 . . . −ξ6). In addition to D1t, . . . , D6t, Hillmer
(1982) and Bell and Hillmer (1983) include an
additional term δ7D7t in Equation (6), where D7t
= W1t + W2t + . . . + W7t is the length of a month.
The interpretation of δ7 depends on the form
of a model. For a stationary time series, the δ7
parameter represents the average of daily
effects and is used to adjust for the length of a
month. A similar interpretation holds if only
the first-order differencing operator (1–B) is
present in the model. However, when the
model includes the seasonal differencing operator (1–B12), the parameter δ7 reflects the effect
because of leap year that may or may not be
important, and may be omitted from the model
in some situations.
The model in Equation (6) implies that the
trading-day effects are constant over time. If
the trading-day effects increase (or decrease)
linearly from year to year, then the following
model may be more appropriate:
f (δ1 , . . . , δ 7 , λ 1 , . . . , λ 7 , D1t , . . . , D7t , K t )

(4)

7

7

i =1

i =1

= ∑ δ i D it + ∑ λ i ( D it × K t )

(7)

Int. J. Tourism Res. 13, 1–16 (2011)
DOI: 10.1002/jtr


Taiwan’s International Tourism

9

where λi represents the linear trend for the
day of the week. Combining the model in
Equations (4) and (7), a calendar effect model
(abbreviated as f(•)) that includes both
moving-holiday and trading-day effects can be
expressed as
f (•) = α 1H1t + α 2 H1t × K t + β1H 2t +
β 2 H 2t × K t + γ 1H 3t + γ 2 H 3t ×
7

7

i =1

i =1

(8)

K t + ∑ δ i D it + ∑ λ i ( D it × K t )

Using the model identification method
described in Liu (1986, 2006), we find that the
disturbance term (Nt) in Equation (1) can be
represented as
12
ˆ t = (1 − θ1 B ) (1 − θ12 B ) a t
N
(1 − B ) (1 − B12 )

(9)

for the monthly tourist arrivals from Japan,
Hong Kong and the USA as well as worldwide
monthly totals.
ANALYSIS OF CALENDAR EFFECTS
To conduct a rigorous examination of the
potential calendar effects on tourist arrivals
into Taiwan from various countries (Japan,
Hong Kong and the USA) and worldwide, we
employ the calendar effect model with a trend
in Equation (8) and the disturbance model
in Equation (9). Both the original and logtransformed series are examined. The results
for these two scales (original and logtransformed) are largely consistent and will be
presented using one or the other for the
purpose of simplifying the interpretation and
in order to increase clarity.

As the tourist arrival time series in Taiwan
are all subject to outliers (e.g. the SARS epidemic in 2003), the outliers must be identified,
and their effects must be jointly estimated with

model parameters. By using the joint estimation method described in Chen and Liu (1993)
and the Scientific Computing Associates Statistical System (Liu and Hudak, 1992), we find
that the trading-day effects are not significant,
and only the Chinese New Year has a significant impact on international tourism among
the three moving-holidays discussed above.
Furthermore, the Chinese New Year effects can
be simply represented by the trend parameter
(α2) as the intercept parameter α1 is insignificant. Thus, the above calendar effects model
can be simply expressed as
Yt = α 2 H1t × K t + N t

(10)

where Nt is the ARIMA model shown in
Equation (9).
In the table below, we list the model parameter estimates obtained by the joint estimation
method of Chen and Liu (1993), where the
critical value 4.0 is used for outlier detection.
Thus, major outliers such as those because of
the SARS epidemic are automatically detected
and adjusted during the joint estimation of
model parameters and outlier effects. Here, a
larger critical value for outlier detection is used
as the series are long and we are only at the
stage of obtaining appropriate model parameter estimates. The number of outliers detected
using this critical value for each model is

reported at the end of each row in Table 2. A
smaller critical value for outlier detection
will be used later when we try to detect nonrecurring events in the time series. More of the
details are discussed later in this section.
The results in the above table show that,
except for Hong Kong, international tourist
arrivals decrease during the Chinese New Year
period, particularly for tourists arriving from
Japan. Chinese New Year is the most important holiday for families to get together during
the year. Therefore, hotels are primarily booked

Table 2. Parameter estimates of models
α2
Japan
Hong Kong
USA
Worldwide

−479.52 (t = −12.29)
233.90 (t = 8.93)
−6.13 (t = −0.41)
−398.09 (t = −5.31)

Copyright © 2010 John Wiley & Sons, Ltd.

θ1
0.48 (t = 10.93)
0.64 (t = 16.49)
0.70 (t = 19.88)
0.49 (t = 11.58)

θ12

σa

Number ofoutliers

0.56 (t = 13.91)
0.56 (t = 13.04)
0.64 (t = 16.29)
0.67 (t = 17.68)

4834.36
2981.74
1582.69
8686.26

9
10
5
9

Int. J. Tourism Res. 13, 1–16 (2011)
DOI: 10.1002/jtr


10
by domestic customers, leaving little availability for international tourists to book a hotel.
Because of family ties between residents in
Hong Kong and Taiwan, tourist arrivals from

Hong Kong increase, rather than decrease.
Outlier detection and estimation
In outlier detection and estimation in time
series, four basic types of outliers are typically
considered (Chang et al., 1988; Tsay, 1988).
These are additive outliers (AO), level shifts
(LS), temporary changes (TC) and innovational
outliers (IO). Other types of outliers can usually
be expressed as combinations of these four basic
types. Details regarding the mathematical formulation of models for these outliers and their
meanings can be found in Liu and Hudak
(1992), Chen and Liu (1993) and Liu (2006).
The benefits of time series outlier detection
and estimation are not limited to providing
better estimates of model parameters. More
importantly, outlier detection often leads to the
discovery of events that may provide useful
information or knowledge (see, e.g. Liu and
Chen, 1991; Chen and Liu, 1993). Thus, outlier
detection can also be used in time series data
mining (Liu et al., 2001). For such applications,
we retain the same estimates of the model
parameters but choose a smaller critical value
for outlier detection so that more outliers can be
detected. In this study, the critical value 2.5 is
used, and only the AO, TC and LS outliers are
considered in the outlier detection. Fewer outlier
types are employed here to avoid spurious outliers and misspecified outlier types. We detect
29 outliers for tourist arrivals from Japan, 30 for
Hong Kong, 29 for the USA and 40 for all tourist

arrivals into Taiwan. These outliers and their
related estimates are listed in Table 3. As AO
affect only one observation, AO are not shown
in Table 3 if their t-values for all four series are
less than three. This allows us to have a more
concise table and a sharper focus on identifying
events that have occurred around the time that
the major outliers were detected.
In Table 3, brief descriptions of events that
may be relevant to the outliers are also listed.
However, for some outliers, particularly those
that have occurred in the more distant past, it is
difficult to find associated events because of
incomplete documentation or lack of informaCopyright © 2010 John Wiley & Sons, Ltd.

H.-L. Lin et al.
tion. For outliers that had been associated with a
particular event, we find that most events can
explain the effects of the outliers well and reveal
interesting information. For example, we find
that the first major energy crisis had more of an
impact on tourism than the second major energy
crisis, and tourists from countries located farther
from Taiwan were impacted more than tourists
from countries geographically closer to Taiwan
(e.g. US tourist arrivals were more affected than
tourist arrivals from Japan). Recent wars
(February 1991, October 2001 and March 2003),
even though they occurred in the Middle East,
still affected international tourism into Taiwan

(shown as negative TC or AO), as wars threaten
perceived travel safety. The 11 September terrorist attacks in the USA created a momentous
change in the share of international tourism. That
is, the share of international tourism shifted from
Europe and the USA to Asia (as discussed in
Section 2: International Tourism Worldwide and
Taiwan). However, the event still negatively
affected tourist arrivals into Taiwan (except for
tourist arrivals from Hong Kong), but to a much
lesser extent. Undoubtedly, the SARS epidemic
had a dramatic impact on international tourism
in Taiwan, in much the same way that it had on
tourism in many other countries.
Hong Kong tourist arrivals into Taiwan
reached a peak in 1981 and then declined for
the next 10 years until 1991 because of China’s
efforts to entice tourists from Hong Kong and
Macao beginning in the late 1970s. It seems
there were some attempts (e.g. tourism promotions) to attract tourists from Hong Kong into
Taiwan (shown as positive AO or TC), but
such attempts did not reverse the overall
decade-long decline. As for the events surrounding Taiwan’s presidential election, the
latest one in March 2008 was the most tense
and attracted an increasing number of overseas voters (shown as positive AO for the USA
and positive LS for Hong Kong). Even though
the voters were not counted as international
tourists, their family members and friends may
have been counted as such depending on
the nature of their visas. The newly elected
government dramatically changed its policy

towards, and relationship with, China. This
seems to have increased the tourist arrivals
from Hong Kong substantially since the election took place (thus shown as positive LS).
Int. J. Tourism Res. 13, 1–16 (2011)
DOI: 10.1002/jtr


10/1972

12/1973
04/1974
05/1974
12/1974
02/1977
04/1978
12/1978
02/1979
10/1979

01/1980
10/1980
02/1981
10/1981
01/1982
04/1982
01/1983
01/1987
06/1987
02/1989

02/1991
03/1991
06/1991
03/1994

06/1994
05/1995
11/1995
03/1996

04/1997
07/1997
02/1998

06/1998
09/1999
10/1999
12/1999

36
40
41
48
74
88
96
98
106

109

118
122
130
133
136
146
193
198
218

242
243
246
279

282
293
299
303

316
319
326

330
345
346
348

Date

22

Time

Copyright © 2010 John Wiley & Sons, Ltd.
9143.8

TC

Major earthquake in Taiwan (on
9/21/1999)
Millennium New Year’s eve
countdown

Reunification of Hong Kong and
China (on 7/1/1997)
Asian financial crisis (began in
7/1997)

Mainland China conducted
missile tests breaching
Taiwan’s airspace

TC

Taiwan eliminated need for
tourist visas for citizens of
select countries

LS
AO

LS
TC

AO
LS
AO

TC
AO

AO

AO

−24 200.3
−11 269.4

−8428.5
−10 287.2

9145.5
10 356.7
16 777.6

−7.18
−3.49

−2.59
−3.06

2.91
3.18
5.35

3.27

−3.95
−3.16

−13 771.6
−10 272.4
11 038.2

−3.93

3.06

−12 408.1

9692.8

−3.31
−5.46

−10 389.7
−17 795.0

AO
LS
2.67

−6.13

−20 562.3

TC

t value

Japan
Estimate

Type

Persian Gulf War launched
against Iraq (1/1991–3/1991)

The death of Emperor Showa of
Japan

Martial law ended in Taiwan

Second energy crisis (began in
11/1978)

First energy crisis (began in
10/1973)

Taiwan and Japan broke off
diplomatic relations

Event

Table 3. Outlier detection and estimation

TC

LS

AO
TC
AO
AO
AO
AO
AO
AO

TC

Type

−11142.2

−6506.8

7492.2

6903.9
−7362.6
9755.6
9080.2
6886.5
7829.9
6387.2

6883.0

Estimate

−5.25

−3.49

3.40
3.10
−3.34
4.33
4.09
3.14
3.55
2.92

3.12

t value

Hong Kong

TC
LS

TC

AO

LS

−3716.5
−3067.9

−2297.1

2525.8

2512.7

5784.9

3552.3
−2096.7

TC
LS

AO

−2187.4

4098.4

Estimate

LS
TC

Type

USA

−4.20
−3.83

−2.60

2.65

3.34

6.08

3.93
−2.75

−2.92
4.62

t value

TC
TC
AO

−31 242.7
−53 737.8
−21 341.7

−15 164.5

24 605.3

AO

LS

21 058.6

20 016.1

−26 760.2

−16 987.3

−12 564.8

−21 230.7

−20 742.4

Estimate

AO

TC

TC

AO

AO

LS

TC

Type

−5.65
−9.68
−4.23

−2.99

4.90

4.19

3.78

−5.04

−3.36

−2.50

−4.20

−3.92

t value

Total arrivals

Taiwan’s International Tourism
11

Int. J. Tourism Res. 13, 1–16 (2011)
DOI: 10.1002/jtr


Copyright © 2010 John Wiley & Sons, Ltd.

11/2002
02/2003
03/2003

04/2003
05/2003
06/2003

07/2003
02/2004
03/2004

04/2005

383
386
387

388
389
390
391
398
399

412

Global financial crisis occurred

Presidential election in Taiwan
(on 3/22/2008)

Promotion of Taiwan tourism
involving the band F4 in
Japan

Presidential election in Taiwan
(on 3/20/2004)

Anti-Secession Law against
Taiwan passed by China.
Major demonstration held in
Taiwan (on 3/26/2005)

The USA launched war against
Iraq (began on 3/20/2003)
SARS epidemic (3/2003–6/2003)

Major airliner crashes occurred
in both Mainland China and
Taiwan (5/7 and 5/25
respectively)

Terrorist attacks occurred in the
USA (on 9/11/2001)
US anti-terrorist war launched
against Afghanistan
Airliner from Mainland China
crashed in Korea (on
4/15/2002)

Presidential election in Taiwan
(on 3/18/2000)

Event

AO

9404.6

19 764.4
−23 804.0

LS
TC

2.78

5.76
−7.05

−13.56
−8.16

−55 802.9
−31 130.0

LS
TC

−6.38

t value

−2.68

−20 886.6

Estimate

Japan

−9831.1

AO

LS

Type

AO, additive outliers; LS, level shifts; TC, temporary changes.

04/2008
09/2008

05/2002
06/2002

377
378

448
453

04/2002

376

04/2007

12/2007
03/2008

10/2001

370

436
444
447

04/2001
06/2001
09/2001

364
366
369

01/2007
03/2007

01/2000
03/2000

349
351

433
435

Date

Time

Table 3. Continued

AO
TC

LS

5.83
−7.77
−3.21

−20 364.2
−8282.8

3.81
12 014.0

7722.0

−5.81

−13 080.7

AO

LS

−3.72

−8179.9

TC

−4.39

12.07

4.56

−7.04

t value

−15.94
−5.87
−6.65

−10 836.1

23 469.6

10 701.0

−14 077.9

Estimate

−39 734.6
−15 104.8
−14 460.9

TC
AO
LS

AO

LS

AO

LS

Type

Hong Kong

AO
AO

AO

AO
AO
AO

AO

AO
AO

AO
AO
TC

LS
TC

Type

3257.5
3621.4

−3589.5

−20 646.9
−27 948.1
−27 495.5
−12 242.8

−3311.2
−6650.8

3155.7
4468.9
−7327.0

4090.9
2780.1

Estimate

USA

3.26
3.42

−3.62

−20.24
−27.58
−27.11
−12.39

−3.34
−6.59

3.25
4.59
−8.10

4.74
2.91

t value

AO

AO

−23 257.7

27 962.9

−14 301.2
19 018.5

−4.18

5.21

−2.58
3.37

−6.14

−33 236.1
TC

AO
AO

−25.57
−19.21
−13.16

2.73
−4.80
−162 983.1
−113 536.0
−69 334.6

15 928.9
−29 577.7

−3.01

−4.28

−25 833.6

−16 519.4

−4.56

t value

−25 441.3

Estimate

TC
TC
AO

LS

AO

TC

TC

LS

Type

Total arrivals

12
H.-L. Lin et al.

Int. J. Tourism Res. 13, 1–16 (2011)
DOI: 10.1002/jtr


Taiwan’s International Tourism

13

The one month drop immediately after the
election (shown as negative AO) is understandable as certain people may have moved
their trip from April to March in order to participate in the election. The global financial
crisis seems to have had a negative effect on
international tourism, particularly from Hong
Kong (shown as TC). However, this event
occurred near the end of the series, hence, the

type and extent of its effect is more difficult to
evaluate (see, e.g. Chen and Liu, 1993).
THE EFFECTS OF EXCHANGE RATES ON
TAIWAN’S TOURISM
Even though monthly time series data contain
rich information for various studies, it is not
appropriate to use such data directly when we
try to explore the relationships between tourist
arrivals and exchange rates. This difficulty is
partially because of the presence of seasonality
in monthly tourist arrivals, while exchange
rates are non-seasonal. Fortunately, we have
38 years of data that allow us to use yearly time
series to explore such relationships. The yearly
tourist arrivals from Japan, Hong Kong and
the USA, and their respective currency
exchange rates (indexed to 1971, i.e. we set the
average exchange rate per New Taiwan Dollar
(NTD) in 1971 to 100 for each country/city) are
displayed in Figure 4. We shall use transfer
function models (also known as time series
regression models) to study the relationships
between tourist arrivals and exchange rates.
To identify the transfer function models for
tourist arrivals from each country/city, we
employ the linear transfer function (LTF) iden-

tification method (Liu and Hanssens, 1982; Liu,
2006). The LTF method is regarded as the most
rigorous method for the identification of transfer function models, more information can be

found in Pankratz (1991, Chapter 5) and Box
et al. (2008, Chapter 12). To obtain the estimates
of the model parameters and outlier effects, the
joint estimation method developed by Chen
and Liu (1993) is used. Using the critical value
3.0 for outlier detection in the joint estimation,
we obtain the following models that best
describe the relationships between the tourist
arrivals and the exchange rates of each respective country/city. In the model below, we use
Yt to represent yearly total tourist arrivals and
Xt to represent yearly average exchange rates.
Japan
∇ n ( Yt ) = (−0.7249) ∇ n ( X t −1 ) + a t , σ a = 0.0973
(t = − 4.21)
(11)
Hong Kong
∇ n ( Yt ) = 0.0786 + ( −0.5529) ∇ n ( X t − 2 ) +
( t = 5.78) ( t = −2.59)
(−0.4957 )∇ n ( X t − 3 ) + a t , σa = 0.0764
( t = −2.36)
(12)
USA
∇ n ( Yt ) = 0.0331 + (1 − 0.4132B) a t , σ a = 0.0554
( t = 6.10)
( t = 2.58)
(13)

Table 4. Outliers detected and estimates
Observation
time/year

Japan

Hong Kong

Time = 9 (1979)
Time = 32 (2002)
Time = 33 (2003)

TC (value = −0.450,
t = −5.00)

AO (value = −0.273,
t = −3.58)

USA
TC (value = −0.252,
t = −4.92)
AO (value = −0.299,
t = −6.42)

AO, additive outliers; TC, temporary changes.
Copyright © 2010 John Wiley & Sons, Ltd.

Int. J. Tourism Res. 13, 1–16 (2011)
DOI: 10.1002/jtr


H.-L. Lin et al.
1200000

1000000

100
80

600000

60

400000

40
20

200000

800000

1974 1976

1978

(B) Hong Kong
Tourist Arrivals

1980

1982

1984 1986

1988 1990

1992

1994

1996

1998

2000

2002

2004 2006

2008

240

Hong Kong Dollar per NTD

600000

200

400000

160

200000

120

80

0
1972 1974 1976

400000

1988 1990 1992 1994 1996

1998 2000 2002 2004 2006

2008

160

(C) United States
Tourist Arrivals

US Dollar per NTD

140

300000
120
200000

100

100000

80

0
1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008

Exchange Rate (Indexeed to 1971)

500000

1978 1980 1982 1984 1986

Exchange Rate (Indexed to 1971)

Tourist Arrivals into Taiwan

Japanese Yen per NTD

800000

1972

Tourist Arrivals into Taiwan

120

(A) Japan

Tourist Arrivals

Exchange Rate (Indexed to 1971)

Tourist Arrivals into Taiwan

14

Figure 4. Annual international tourist arrivals into Taiwan and exchange rates (indexed to 1971). NTD, New
Taiwan Dollar.

where ∇ is the first-order differencing operator
(1-B). The outliers detected and their estimates
are listed in Table 4 and will be explained later
in this section.
It is important to note that a logged time
series with first-order differencing (e.g. ∇ᐍn(Yt)
and ∇ᐍn(Xt)) is equivalent to a percentage
change for the time series. Thus, the above
transfer function weights (similar to regression
coefficients) can be interpreted as elasticity.
In the above models, we see that tourist
arrivals into Taiwan from Japan and Hong
Copyright © 2010 John Wiley & Sons, Ltd.

Kong are influenced by their respective
exchange rates, but this is not the case for the
USA. For tourist arrivals from Japan, it takes a
whole year to reflect the effect of exchange rate
changes. The effect is that 1% increase in the

NTD will reduce Japanese tourist arrivals into
Taiwan by 0.7249% one year later. For tourist
arrivals from Hong Kong, it takes two to three
years for the exchange rate changes to affect
the tourist arrivals. The total effect is that 1%
increase in the NTD will reduce Hong Kong
tourist arrivals by 1.05% (i.e. 0.5529% +
Int. J. Tourism Res. 13, 1–16 (2011)
DOI: 10.1002/jtr


Taiwan’s International Tourism
0.4957%) over a two- to-three-year period. It is
also important to note that after accounting for
exchange rates, there is no trend (i.e. year-onyear increase) for tourist arrivals from Japan,
while both Hong Kong and the USA exhibit a
definitive upward trend from one year to the
next. In fact, the tourist arrivals from the USA
follow an ARIMA (0,1,1) model with an
upward trend and are not affected by NTD
exchange rates.
In the above joint estimation of the model
parameters and outlier effects, a few major outliers are detected and listed in Table 4. The most
important one is because of the SARS epidemic
in 2003. The SARS epidemic had a negative TC
effect on tourist arrivals from Japan beginning
in 2003, and only had a one-year negative effect
(shown as AO) on tourist arrivals from the USA.
SARS also had a negative impact on tourist
arrivals from Hong Kong. However, the tourist

arrivals from Hong Kong decreased substantially in 2002 for a number of reasons, causing
the drop in 2003 (SARS) to be less prominent so
that it is not shown as a significant outlier. In
addition to the SARS epidemic, there was a
negative TC outlier for the US tourist arrivals in
1979, which was because of the second energy
crisis (also see the discussion in Section 4:
Analysis of Calendar Effects).
DISCUSSION AND CONCLUSION
In this paper, we study major variables or
events that affect international tourism in
Taiwan. In the analysis of the time series for
tourist arrivals, it is important to account for
both known and unknown events in the series
when estimating the models, or else the estimates of the model parameters may be seriously compromised. For known events, both
moving-holiday and trading-day effects are
examined. We find that tourist arrivals into
Taiwan are affected by the Chinese New Year
each year, but not by other moving holidays
such as the Dragon Boat Festival or the MidAutumn Festival. The trading-day effects are
not found to be significant for all tourist arrival
series that we examined. We use an automatic
outlier detection method to detect significant
unknown events (shown as outliers) and
employ a joint estimation method to obtain the
parameter estimates and outlier effects. IncorCopyright © 2010 John Wiley & Sons, Ltd.

15
porating outlier detection and estimation in
the model not only improves the quality of the

model parameter estimates, but also reveals
important events that otherwise may be
ignored. By examining such events, we can
understand their effects better and prepare for
the future if similar events reoccur.
Among the three major sources of tourist
arrivals that we studied, we find that international tourist arrivals into Taiwan are affected
by the exchange rates of the respective country/
city in the case of Japan and Hong Kong, but
not in the case of the USA.
Temporal data aggregation often results in a
loss of information. Higher frequency time
series usually provide more information and
are practically more useful than lower frequency
time series (e.g. monthly time series are often
more informative than yearly time series).
However, in some situations, it is necessary to
use lower frequency time series because of noise
or seasonality in the time series. In the transfer
function modelling of tourist arrivals and
exchange rates, we find it is inappropriate to use
monthly time series, and the use of yearly time
series is more informative.
REFERENCES
Athanasopoulos G, Hyndman RJ. 2008. Modeling
and forecasting Australian domestic tourism.
Tourism Management 29: 19–31.
Bell WR, Hillmer SC. 1983. Modeling time series
with calendar variation. Journal of the American
Statistical Association 78: 526–534.

Box GEP, Jenkins GM. 1976. Time Series Analysis:
Forecasting and Control. 2nd edn. Holden Day: San
Francisco, CA.
Box GEP, Jenkins GM, Reinsel GC. 2008. Time Series
Analysis: Forecasting and Control. 4th edn. Wiley:
New York.
Chang I, Tiao GC, Chen C. 1988. Estimation of time
series parameters in the presence of outliers.
Technometrics 30: 193–204.
Chen C, Liu LM. 1993. Joint estimation of model
parameters and outlier effects in time series.
Journal of the American Statistical Association 88:
284–297.
Chen YM, Hui TK, Yuen E. 1999. Modeling the
impact of sudden environment changes on
visitor arrivals forecasts: the case of the Gulf War.
Journal of Travel Research 37: 391–394.
Chu FL. 2008. A fractionally integrated autoregressive moving average approach to forecasting
Int. J. Tourism Res. 13, 1–16 (2011)
DOI: 10.1002/jtr


16
tourism demand. Tourism Management 29: 79–
88.
Crouch GL, Schultz L, Valerio P. 1992. Marketing
international tourism to Australia. Tourism
Management 13: 196–208.
Fox AJ. 1972. Outliers in time series. Journal of the
Royal Statistical Society, Series B 34: 350–363.

Goh C, Law R. 2002. Modeling and forecasting
tourism demand for arrivals with stochastic nonstationary seasonality and intervention. Tourism
Management 23: 499–510.
Goh C, Law R, Mok HMK. 2008. Analyzing and
forecasting tourism demand: a rough sets
approach. Journal of Travel Research 46: 327–
338.
Goodrich JN. 2001. September 11, 2001 attack on
America: a record of the immediate impacts and
reactions in the USA travel and tourism industry.
Tourism Management 23: 573–580.
Hillmer SC. 1982. Forecasting time series with
trading day variation. Journal of Forecasting 1:
385–395.
Hillmer SC, Bell WR, Tiao GC. 1981. Modeling considerations in the seasonal adjustment of economic time series. In Proceedings of the Conference
on Applied Time Series Analysis of Economic Data,
Zellner A (ed.). US Department of Commerce,
Bureau of the Census: Washington, DC; 74–100.
Huang JH, Min JCH. 2002. Earthquake devastation
and recovery in tourism: the Taiwan case. Tourism
Management 23: 145–154.
Hui T, Yuen CC. 2002. A study in the seasonal variation of Japanese tourist arrivals in Singapore.
Tourism Management 23: 127–131.
Kim HJ, Chen MH, Jang SS. 2006. Tourism expansion and economic development: the case of
Taiwan. Tourism Management 27: 925–933.
Kulendran N, King M. 1997. Forecasting international quarterly tourism flows using error correction and time series models. International Journal
of Forecasting 13: 319–327.
Lim C. 1997. Review of international tourism
demand models. Annals of Tourism Research 24:
835–849.

Liu LM. 1980. Analysis of time series with calendar
effects. Management Science 26: 106–112.

Copyright © 2010 John Wiley & Sons, Ltd.

H.-L. Lin et al.
Liu LM. 1986. Identification of time series models
in the presence of calendar variation. International
Journal of Forecasting 2: 357–372.
Liu LM. 2006. Time Series Analysis and Forecasting.
2nd edn. Scientific Computing Associates Corp.:
Chicago, IL.
Liu LM, Chen C. 1991. Recent developments of time
series analysis in intervention in environmental
impact studies. Journal of Environmental Science
and Health A26: 1217–1252.
Liu LM, Hanssens DM. 1982. Identification of
multiple-input transfer function models. Communications in Statistics A11: 297–314.
Liu LM, Hudak GB. 1992. Forecasting and Time Series
Analysis Using the SCA Statistical System, Vol. 1.
Scientific Computing Associates Corp.: Chicago,
IL.
Liu LM, Bhattacharyya S, Sclove S, Chen R, Lattyak
W. 2001. Data mining on time series: an illustration
using fast-food restaurant franchise data. Computational Statistics and Data Analysis 37: 455–476.
Martin CA, Witt SF. 1989. Forecasting tourism
demand: a comparison of the accuracy of several
quantitative methods. International Journal of
Forecasting 5: 7–19.
Pankratz A. 1991. Forecasting with Dynamic Regression Models. Wiley: New York.

Ryan C. 1993. Crime, violence, terrorism and
tourism: an accidental or intrinsic relationship?
Tourism Management 14: 173–183.
Song H, Witt SF. 2006. Forecasting international
tourist flows to Macau. Tourism Management 27:
214–224.
Tsay RS. 1988. Outliers, level shifts, and variance
changes in time series. Journal of Forecasting 7: 1–20.
Wang YS. 2008. The impact of crisis events and
macroeconomic activity on Taiwan’s international inbound tourism demand. Tourism Management 30: 75–82.
Witt SF, Witt CA. 1995. Forecasting tourism demand:
a review of empirical reseach. International Journal
of Forecasting 11: 447–475.
World Economic Forum. 2009. Travel and tourism
competitiveness report 2009. Third Travel &
Tourism Competitiveness Report in 2009,
Jennifer Blanke: Latin America, Rio de Janeiro.

Int. J. Tourism Res. 13, 1–16 (2011)
DOI: 10.1002/jtr


INTERNATIONAL JOURNAL OF TOURISM RESEARCH
Int. J. Tourism Res. 13, 17–31 (2011)
Published online 9 July 2010 in Wiley Online Library
(wileyonlinelibrary.com) DOI: 10.1002/jtr.794

The Influence of Entrepreneurial Talent
and Website Type on Business
Performance by Rural Tourism

Establishments in Spain
Jannine Nieto, Rosa M. Hernández-Maestro and Pablo A. Muñoz-Gallego
Salamanca University, Salamanca, Spain

ABSTRACT

INTRODUCTION

Entrepreneurs provide the engine of
development, especially in economically
unstable times. In Spain, rural tourism is
undergoing a boom, and the Internet plays a
vital role for tourism. This study therefore
considers: (i) the importance of
entrepreneurial talent for implementing a
website; (ii) the relations among
entrepreneurial talent, website
characteristics and business performance;
and (iii) the moderating effect of
entrepreneurial experience. Using a sample
of 150 rural tourism establishments in
Spain, this study reveals how website
content affects performance and how
experience moderates the relationships
between entrepreneurial talent and
performance, and between entrepreneurial
talent and website design. Copyright © 2010
John Wiley & Sons, Ltd.

The importance of entrepreneurial talent

Received 8 July 2009; Revised 9 March 2010; Accepted 26
May 2010

Keywords: entrepreneur; talent; new
technologies; website; performance; rural
tourism.

*Correspondence to: Jannine Nieto, Departamento de
Administración y Economía de la Empresa, Campus
Miguel de Unamuno, Edificio F.E.S., 37007 Salamanca,
España.
E-mail:

A

lthough investigations have focused on
entrepreneurship, generally referred to
as the creation of new businesses, for
nearly a century, we still lack a consensus
about its definition, which remains an obstacle
to developing a conceptual framework of
entrepreneurship (Shane and Venkataraman,
2000). Various terms exist to refer to entrepreneurship,
including
entrepreneur,
entrepreneurial function and more recently,
entrepreneurial orientation and entrepreneurial talent. The latter refers specifically to a
person’s special ability for entrepreneurship.
Further investigations of entrepreneurial

talent are critical for several reasons. Entrepreneurs represent engines of sustainable development in an economy (Huiyuan and Hua,
2008). Entrepreneurship also enables society to
convert technical information into products
and services (Shane and Venkataraman,
2000). Furthermore, through this mechanism,
societies can discover and address temporal
and spatial inefficiencies in the economy
(Shane and Venkataraman, 2000).
In times of economic crisis, these arguments
become even more pertinent, because entrepreneurship and entrepreneurial talent can
help countries deal with declining incomes
and profits.
The importance of rural tourism
Rural tourism, defined as a tourism that takes
place in rural areas, motivated by tourists’
Copyright © 2010 John Wiley & Sons, Ltd.


18

J. Nieto, R. M. Hernández-Maestro and P. A. Muñoz-Gallego

desire to understand this way of life and come
into contact with nature (Hernández-Maestro
et al., 2007), has reached a worldwide peak
with regard to its revenue-generating abilities.
Many studies address its implications and
issues, and institutions such as the World
Tourism Organization offer seminars to
consider its current situation and future

prospects.
Rural tourism is a widespread activity in
Spain, and in recent years, it has been portrayed as part of the portfolio of leisure and
recreation activities available in virtually every
region (Barke, 2004). Spain ranks among the
top three countries in tourism, generating
turnover of 66.4 million euros in 2004 and
experiencing continued strong growth. For
example, compared with 2001, the number of
rural tourism establishments increased from
5497 to 13 887 in 2009 (National Institute of
Statistics (INE), 2010), and they offered a total
of 126 234 beds. In terms of the number of tourists, 2 708 000 travellers used rural tourism
accommodations in Spain in 2009, 90% of
whom reside within the same country and 10%
from abroad, producing a total of 7 902 000
overnight stays, with an average stay of 2.9
days each.
Rural tourism provides an ideal focus for
research into entrepreneurship and entrepreneurial talent because of its growth, the large
amount of governmental support it receives
through subsidies and the changes the rural
population has undergone in the switch from
agriculture to rural tourism.
The importance of the Internet
The Internet has become a vital tool in people’s
daily activities, whether focused on professional, educational or amusement and leisure
activities. In 2004, the Internet had 215 million
users worldwide (Cyr and Trevor-Smith, 2004);
as of 2009, there were more than 1.6 billion

Internet users (CIA, 2009). In turn, it seems
logical that the numbers of people who use the
Internet to look for, plan and even purchase
tourism products also are increasing.
According to the INE, in 2009, 51.3% of the
Spanish population between the ages of 16 and
74 years was already using the Internet. A
study carried out in the context of rural tourism
Copyright © 2010 John Wiley & Sons, Ltd.

also has revealed that 47% of travellers learned
about the establishment they visited through
the Internet (Hernández-Maestro, 2005).
With regard to the importance that entrepreneurs in this sector attach to the Internet, an
Internet Week survey reported that approximately 60% of tourism companies (e.g. travel
agencies, bars, hotels, motels) regard the Internet as a ‘substantial’ tool for acquiring new
customers (Mullen, 2000; Baloglu and Pekcan,
2006). In Spain, 86.5% of rural tourism establishments have a website, 81.6% advertise on
the Internet and most autonomous regions in
Spain include on their sites an official page for
searching for accommodations, though only
35.7% of accommodations can be contracted
online (INE, 2006).
As Porter and Millar (1985) state, these new
technologies have changed ways to do business, altering the structure of the industry,
such that every company must now understand how to use technologies to compete. In
addition, they provide tools to facilitate the
competitive strength of small- and mediumsized enterprises (SMEs) in the global market
(Quelch and Klein, 1996; Baloglu and Pekcan,
2006). Yet, some studies suggest e-commerce is

not being adopted by SMEs as quickly as might
be expected (Beveren and Thomson, 2002).
Despite the above evidence, we find a dearth
of research studies that combine rural tourism
and new technologies. Thus, this is another
reason that has motivated this study.
Research objectives
We pursue three main objectives with this
study: (i) to analyse the importance of entrepreneurial talent with regard to the introduction of new technologies and specifically a
website on the Internet; (ii) to examine the
impact of both entrepreneurial talent and new
technologies on business performance; and
(iii) to examine the potential moderating role
of an entrepreneur’s experience on the effects
of entrepreneurial talent.
The rest of this paper therefore is structured
as follows: first, we present hypotheses derived
from our literature review regarding the
possible relationship among entrepreneurial
talent, the implementation of websites and
business performance; second, we describe the
Int. J. Tourism Res. 13, 17–31 (2011)
DOI: 10.1002/jtr


Entrepreneurial Talent and Websites on Business Performance
methodology we use, including a specification
of the database and measures; third, we outline
the results; and fourth, we present our conclusions and some implications of our study.
LITERATURE REVIEW

Entrepreneurial talent
A person with talent has the intelligence,
ability and fitness to perform a particular activity. Talent means the person can do something
better than others without talent can (Ingenieros, 1913). In this sense, talent is not ‘supernatural’; some people simply may possess
great intelligence, energy or other generally
valuable traits that enable them to become one
of the best in their chosen occupations, whether
work related or otherwise (Murphy et al., 1991).
The most talented people in a particular area
tend to choose occupations that will earn them
returns on their talent. Similarly, companies
need talented employees to achieve their
business goals (Ricker and Leahy, 2009).
Furthermore, talent implies the person can
finish an activity that others would abandon or
never start. Therefore, entrepreneurial talent
implies that the entrepreneur not only conceives of a good idea but also implements it
fully.
As we have noted though, the concept of
‘entrepreneur’ remains vague, contradictory
and imprecise in its various definitions (Shane
and Venkataraman, 2000; Rauch et al., 2009;
Thompson, 2009).
Gartner (1989), in a meta-analytic review of
the term, showed that throughout history,
researchers have attempted to define what an
entrepreneur is instead of what that person
does. Three main features have marked definitions of entrepreneurship throughout history
though: innovation, the search for information
and making decisions under uncertainty to

earn profits.
Innovation. Some researchers limit the idea of
newness to opening a new business (Mescon
et al., 1981), even if others are already exploiting that line of business. Others argue that
newness requires the creation of a totally new
business (Hornaday and Aboud, 1971) and the
exploitation of an original idea. Innovation
Copyright © 2010 John Wiley & Sons, Ltd.

19

also could mean engaging in activities that fall
outside the ordinary limits of business routines
(Schumpeter, 1934). More recently, innovativeness has emerged as a key dimension of entrepreneurial orientation (Rauch et al., 2009).
In relation to the ability to innovate, an
entrepreneur usually attempts to influence his
or her environment and is not limited by an
existing situation, which is usually referred
to as a proactive personality (Becherer and
Maurer, 1999; Rauch et al., 2009). In this sense,
entrepreneurship has been associated with the
active search for and discovery of opportunities. Not everyone has this ability to discover
opportunities in an environment (Kirzner,
1973; Federici et al., 2008).
Search for information. The phrase ‘knowledge
is power’ is well known, but its practical application to entrepreneurship is unclear. Gathering information for decision making is a critical
activity for an entrepreneur (Cooper et al.,
1995). Entrepreneurs reporting higher levels of
information search intensity will identify more
business opportunities (Ucbasaran et al., 2008).

Uncertainty. Although uncertainty may seem
like the normal state of affairs during economic
crises, it remains prominent for entrepreneurs
even when the economy is stable. Therefore,
the ability to make decisions under uncertainty
is critical for entrepreneurs. These decisions
eventually focus on a specific objective: profits
(Hull et al., 1980; de Klerk and Kruger, 2002).
All decisions contain some risk, but entrepreneurs often face particularly risk with regard
to their financial, physical and/or social status
in their efforts to achieve profit objectives (Hull
et al., 1980; Federici et al., 2008; Rauch et al.,
2009).
Considering these three elements and the
concepts of entrepreneur and talent, we note
that these terms are not substitutive but rather
adjective and noun. That is, entrepreneurial
talent would mean a special ability for entrepreneurship, that is, for embarking on and
exploiting new opportunities, searching for
information and making decisions under
uncertainty in pursuit of profits, while assuming implicit risks.
Age, education and experience also have
been studied as characteristics that influence
Int. J. Tourism Res. 13, 17–31 (2011)
DOI: 10.1002/jtr


20

J. Nieto, R. M. Hernández-Maestro and P. A. Muñoz-Gallego

entrepreneurship. Regarding age, the process
of cognitive development continues throughout life (Baron, 2009); thus, age may affect strategic decisions. In contrast, highly educated
persons are more receptive to new ideas (Kimberly and Evanisko, 1981; Hua et al. 2000) and
tend to identify more business opportunities
(Ucbasaran et al. 2008). Finally, experience
acquired from having started multiple new
ventures in the past offers benefits in terms of
developing contacts (Danson, 1999), gaining
knowledge about obtaining the most appropriate sources of financing (Starr and Bygrave,
1991), learning managerial and technical skills
necessary for leading new ventures and identifying ways to serve new and emerging market
segments (Wright et al., 1998).
Entrepreneurial talent and
website characteristics
Even after an exhaustive review, we found
little research regarding the relationship
between entrepreneurial characteristics and
the characteristics of the company websites,
despite the seemingly relationship between
them.
The research that exists indicates that
talented employees lead to greater commitment to innovation and technology by companies (Murray-Leslie, 2009; WEDC, 2009).
The banking sector even aims to hire talented
people who can help drive technology improvements (Murray-Leslie, 2009). Moreover,
according to the Washington Economic Development Commission (WEDC, 2009), talent
and entrepreneurship are two key drivers of
an innovation economy. On the other hand,
a proactive personality appears to have a
mediating role between strategy and innovation in Internet (Kickul and Walters, 2002).

There may be talent in large and small companies, though according to Ferrante (2005),
there should be more talent in big companies,
which manage more workers. In contrast,
larger companies tend to be more sluggish and
driven by established systems and procedures,
such that talented employees have less room
to contribute than they would in small firms.
Yet, large firms have adopted and noted the
value of websites (Ellinger et al., 2003) far more
than small businesses, as is the case in our
Copyright © 2010 John Wiley & Sons, Ltd.

sample, which tend to use simple websites
(Bernert and Stricker, 2001).
Even with simple websites, talented entrepreneurs should recognise the importance of
their websites for increasing consumer awareness. Therefore, the characteristics of their
websites should differ from those of websites
posted by entrepreneurs with less talent. We
propose the following hypothesis:
Hypothesis 1a: Entrepreneurial talent
determines characteristics of websites.

Entrepreneurial talent and
business performance
Empirical evidence in various contexts demonstrate that human capital is positively associated with benefits such as higher income
(Boylan, 1993; Ucbasaran et al., 2008), productivity (Mincer, 1974; Becker, 1975; Ucbasaran et
al., 2008) and objective quality (HernándezMaestro et al., 2009), all of which should
generate better business performance (Flynn
and Saladin, 2001; Meyer and Collier, 2001).
In addition, in a meta-analytic review, Rauch

et al. (2009) found a moderately large (r =
0.242) correlation between entrepreneurial
orientation and performance.
Objective performance measures, such as
changes in sales and profits (Roper, 1998;
Becherer and Maurer, 1999; Ratchford et al.,
2003; Ferrante, 2005), have also revealed positive relationships with proactiveness, though
in some cases, the relation is not significant
(Becherer and Maurer, 1999). Other performance measures include organisational
strategies (e.g. investments in new products,
management, control), which also tend to
produce positive and significant results in
combination with measures such as education,
experience and proactivity (Roper, 1998; Kickul
and Gundry, 2002). Other studies indicate that
entrepreneurs who engage in more intense
information searches identify more business
opportunities (Ucbasaran et al., 2008). Therefore, we posit:
Hypothesis 1b: Entrepreneurial talent
positively affects business performance.
Int. J. Tourism Res. 13, 17–31 (2011)
DOI: 10.1002/jtr


Entrepreneurial Talent and Websites on Business Performance
Web site characteristics
Classifications of the concept. In accordance with
our research goal, we consider the intensity of
Internet presence of a rural tourism establishment and the design of its website as potential
determinants of its competitiveness.

With regard to website characteristics,
various classifications are available. Bart et al.
(2005) used seven categories: privacy, safety,
navigation, brand power, help in solving problems, purchase orders and customers’ testimonies. Baloglu and Pekcan (2006) instead divided
the characteristics into two big groups, each
with subdivisions: design characteristics and
content characteristics. Yoon (2002) also argues
that customer confidence, which positively
affects behavioural intentions, depends on
three website characteristics: transaction
safety, Web page properties and navigation.
This variety of classifications suggests
merging several concepts to evaluate Web
page characteristics for our research.
Website characteristics and business performance.
Some studies confirm a relationship between
website characteristics and business performance. Ellinger et al. (2003) found a positive
and significant relationship between the interactive characteristics of the website (e.g. selfhelp, online transactions, online purchase
order, delivery) and company earnings. Polo
and Frías (2010) summarised several studies
that show information and communication
technology deployment encourages competitive actions and commercialisation in difficult
periods, especially for rural tourism.
Furthermore, customers’ preference for a
website and behavioural intentions can serve
as measures of performance, because such
intentions should favour added sales. In this
regard, prior studies demonstrated that characteristics of the website (e.g. links to customer
service, privacy policy, customer testimonies)
can have positive and significant effects on

customer preferences for specific sites (Resnick
and Montania, 2003). Other studies similarly
found a positive and significant relationship
between customer confidence, as derived from
navigation ease, purchase orders and problemsolving assistance and behavioural intentions
(Yoon, 2002; Bart et al., 2005).
Copyright © 2010 John Wiley & Sons, Ltd.

21

In contrast, some researchers found no such
positive relation between website characteristics and performance. Rather, they assert that
website characteristics cannot positively influence either purchase intentions (Coyle and
Thorson, 2001) or product price changes
(Ratchford et al., 2003).
In our effort to explore the potential for a
positive influence of website characteristics on
company performance, we propose the following hypothesis:
Hypothesis 2: Website characteristics positively influence business performance.
Entrepreneurial experience
Experience has long been studied as a characteristic of the entrepreneur with a potential
effect on performance or as a moderator of
other relationships (Roper, 1998; Ferrante,
2005; Hmieleski and Baron, 2009). As we noted
previously, experience helps to define the
capabilities of human capital, may be an
antecedent of entrepreneurial talent and
likely affects managerial performance (Van de
Ven et al., 1984; Jo and Lee, 1996; Chandler and
Hanks, 1998).

Some studies demonstrate that the greater
the experience of business professionals in a
particular sector, the higher the income is for
the company, growth rate of assets and growth
rate of employees (Jo and Lee, 1996). An entrepreneur’s related business experience (before
starting the company) should positively affect
productivity (Harada, 2004).
In terms of the potential moderating role of
experience, it should have a positive and significant influence on the relationship between
the frequency of visits to the customer and
sales effectiveness (Martín and Román, 2004).
Prior experience with creating ventures may
also moderate strengthening the negative relationship between entrepreneurs’ optimism
and performance (Hmieleski and Baron, 2009).
In contrast, some authors found no such significant relation (Collins and Moore, 1964;
Sandberg and Holfer, 1987; Jo and Lee, 1996).
In the area of the new technologies, Wetering
and Koster (2007) found no positive effect of
experience on innovative performance. Van de
Ven et al. (1984) even suggested a negative
Int. J. Tourism Res. 13, 17–31 (2011)
DOI: 10.1002/jtr


22

J. Nieto, R. M. Hernández-Maestro and P. A. Muñoz-Gallego

correlation between prior start-up experience
and overall performance.

In our study context, more experienced
entrepreneurs tend to be older. We anticipate
that their experience (longer in the business)
gives them greater knowledge that should
increase their chances of success. However,
their age may make them less likely to use the
Internet, because older people generally are
less familiar with new technologies. Furthermore, educational systems change over time
(Baron, 2009); as Polo and Frías (2010) assert,
there may be a lack of knowledge and training
in new technologies among older entrepreneurs, who may not have been exposed to
formal business education and instead acquired
their knowledge through the ‘school of hard
knocks’. Their experience with strong performance does not involve the Internet, so they
may believe they do not need it. We therefore
propose our final hypothesis:
Hypothesis 3: The experience of entrepreneurs in the sector (i) mitigates the relationship between entrepreneurial talent and
website characteristics and (ii) strengthens
the relationship between entrepreneurial
talent and company performance.
METHODOLOGY
Data collection
We used a database that contained information
about a sample of rural tourism entrepreneurs,
gathered between August and October 2004.
Participating establishments were rural houses
with a maximum of 24 beds that appeared in
Spain’s official 2003 guide to rural tourism.
The database contained data about the characteristics of the entrepreneurs, which enabled
us to derive measures of their entrepreneurial

talent and experience, as well as the characteristics of the rural establishment and business
performance, in 17 autonomous communities
in Spain (survey available on request from the
authors).
We also identified rural tourism establishments that posted information on the Internet,
whether through their own websites or on the
sites of other institutions. A survey was applied
to the resulting sample of establishments
Copyright © 2010 John Wiley & Sons, Ltd.

(available on request from the authors) that
aimed to measure the design and content characteristics of their websites, during March and
April 2007. Of the 219 observations in the original database, we obtained a final sample of 150
participants (95%, p = q = 0.5 error = 7.74%).
We acknowledge that the time gap between
the measurements of entrepreneurial talent
and performance and website characteristics
might imply that some of the information is
not entirely accurate; some of the variables
may have changed over time.
Measures
In accordance with our literature review, we
considered the intensity of the information
search entrepreneurs undertake before making
decisions as an important measure of entrepreneurial talent (Cooper et al., 1995; Ucbasaran
et al., 2008). Specifically, two items assess this
intensity: the number of trade fairs in the
tourism industry that the entrepreneur
attended in the previous 2 years and the publications consulted in the last year. The measure
of entrepreneurial talent therefore was limited

to information in the database. Other variables
also might serve to measure this construct,
though we leave that effort to further research.
Moreover, in our study context, many entrepreneurs rely on relatively simple management practices, because their background is in
agriculture or livestock practices. Few of them
have a business education, so being more
active in information searching should discriminate effectively among entrepreneurs
with more and less talent.
To measure managerial performance, we
consider various items in prior empirical
literature. A common distinction separates
financial and non-financial measures (Rauch
et al., 2009). Financial measures are popular
because they offer objectivity through items
such as return on sales and return on investment (Roper, 1998). In line with our literature
review, (Roper, 1998; Becherer and Maurer,
1999; Ratchford et al., 2003; Ferrante, 2005), we
use two financial measures: the level of annual
income and the annual profits of the company.
Respondents indicated on a seven-point scale
the range of income and profits, in euros,
earned by their business establishment.
Int. J. Tourism Res. 13, 17–31 (2011)
DOI: 10.1002/jtr


Entrepreneurial Talent and Websites on Business Performance
To measure website characteristics, we used
a 16-item questionnaire. In line with our literature review (Bart et al., 2005; Baloglu and
Pekcan, 2006), we define website characteristics according to three classifications:

(1) design characteristics: visual appearance,
attractiveness and ease of navigation;
(2) content characteristics: richness of information offered.; and
(3) confidence characteristics: technical confidence the website inspires and opportunity
for interaction (e.g. space for suggestions,
testimonies).
We used five-point scales with objective
measures for most items. For example, to
explain products and services they offered, the
respondents could choose from a range of
options, from none to all types of information
(e.g. prices, number of rooms, room services,
other hotel services).
Finally and again based on our literature
review (Van de Ven et al., 1984; Martín and
Roman, 2004), to measure the respondents’
experience running this type of establishment,
we asked them to indicate the number of years
of practical experience they had in rural
housing or similar areas.

Data analysis and results
The exploratory phase of our analysis, focused
on the data and relationships, relied on the SPSS
Inc., Chicago, USA (SPSS) 14.00 program. We
continued with the confirmatory phase, using
LISREL 8.54, a structural equation model that
can measure several relationships across variables simultaneously. To estimate the model,
we used the robust maximum likelihood
Exploratory phase. Sample representativeness.

To confirm that the sample was representative
of the broader population, we developed a
comparative chart. We obtained similar results
in both groups, which indicate an absence of
bias and the representativeness of the sample
(Table 1).
Pilot test. Subsequently, to demonstrate the
discriminatory power of the questionnaire that
measured website characteristics and confirm
Copyright © 2010 John Wiley & Sons, Ltd.

23

that the three constructs appear in that questionnaire, we conducted a pilot test in which
we applied the questionnaire to a sample of
15 websites.
Exploratory factor analysis. The 16 questions
represented the 16 variables in the exploratory
factor analysis. Four of them do not discriminate (professional Web, simple language, reliability and consistency of the frame) and five
have very low loadings (ease of contracting,
space to express opinions, information about
owners, attention capture and related links).
Therefore, we eliminated these variables from
the study. Because both variables measuring
confidence (reliability and space to express
opinion) were eliminated, the construct disappeared from our analysis. The seven variables
that remained loaded on two constructs: ‘Web
design’ and ‘Web content’.
Therefore, we conducted an exploratory
factor analysis on four constructs: Web design,

Web content, entrepreneurial talent and performance (Table 2).
The analysis results in loadings greater than
0.7 for all the variables except one (address =
0.685); these excellent results far exceed the
recommended minimum significant factor
loading of 0.45 for a sample of 150 observations
(Hair et al., 2001). The Kaiser–Meyer–Olkin
index is greater than 0.5 in all the cases, which
indicates that the correlation matrix is correct
for every construct. Bartlett’s sphericity test
shows a level of significance within the
accepted range (<0.05), so we can reject the null
hypothesis of equality between the correlation
matrix and the identity matrix, which would
have made the factor model unsuitable. The
measure of sampling adequacy indicates the
relation of every variable with the others; in all
the cases, it is greater than the minimum of 0.3.
The communalities are superior to 0.5, so all
the variables are well explained by the factor.
The total explained variance is greater than
65%, and all the autovalues are greater than 1,
which suggests every factor explains the variance of at least one variable. The Cronbach’s
alpha is also greater than 0.6, so the individual
indicators appear consistent in all the
measures.
The result of the exploratory factor analysis
is a general model, which represents the object
Int. J. Tourism Res. 13, 17–31 (2011)
DOI: 10.1002/jtr

24

J. Nieto, R. M. Hernández-Maestro and P. A. Muñoz-Gallego

Table 1. Rural establishments by Autonomous Community*
Autonomous Community
Andalucia
Aragon
Asturias
Baleares
Canarias
Cantabria
Castilla la Mancha
Castilla y Leon
Cataluña
Extremadura
Galicia
Madrid
Murcia
Navarra
P.Vasco
La Rioja
Valencia
Total rural establishments
Significance
Error margin

Amount of rural

establishments in Spain
25
279
166
172
19
158
118
229
286
100
263
17
0
149
226
61
60
2328
0.05
7.74%

Amount of rural establishments,
sample for study
% Sample

%
1%
12%
7%

7%
1%
7%
5%
10%
12%
4%
11%
1%
0%
6%
10%
3%
3%
100%

0
19
12
9
1
10
8
28
12
7
17
1
0
9

12
3
2
150

0%
13%
8%
6%
1%
7%
5%
19%
8%
5%
11%
1%
0%
6%
8%
2%
1%
100%

Comparative between the total amount of rural establishments and used sample.
* Rural establishments for shared rent in Spain with a maximum of 24 accommodations.

Table 2. Exploratory factor analysis
Variable

Load

Construct: entrepreneurial talent
Trade fairs
Publications

0.849
0.849

Construct: Web design
Easy
Fast
Readability

0.837
0.900
0.770

Construct: Web content
Multimedia
Prod-serv info
Contact
Address

0.850
0.767
0.758
0.685

Construct: performance

Income
Profits

0.916
0.916

of this research, as we depict in Figure 1. The
model consists of four constructs that relate
through five causal relationships: entrepreneurial talent–Web design, entrepreneurial
talent–Web content, entrepreneurial talent–
Copyright © 2010 John Wiley & Sons, Ltd.

Cronbach’s alpha
0.612

0.795

0.771

0.807

performance, Web design–performance and
Web content–performance. The first component of each relationship is the independent
construct and the second is the dependent construct. The potential moderating role of the
Int. J. Tourism Res. 13, 17–31 (2011)
DOI: 10.1002/jtr