Exploring Items for Measuring Behavioural Intention among Bleisure in Malaysia: An Exploratory Factor Analysis

This study used exploratory factor analysis (EFA) to investigate the reliability of a tool for assessing the behavioural intentions of international bleisure tourists to visit leisure attractions in Malaysia during their business trips. The instrument consists of four constructs. 110 responses were collected to conduct the EFA after distributing the questionnaire. The EFA was performed separately for each construct. The analysis obtained that the first and second constructs have three and four components, respectively, while the third and fourth constructs have one component each. Each item in each construct factor is > 0.5, while all construct is significant as they yielded the P value < 0.05 for Bartlett's Test of Sphericity. All constructs reach higher than 0.6 for the Kaiser-Meyer-Olkin measure of sampling adequacy, indicating that the sample size was adequate. All items in the constructs also reach greater than 0.7 for Cronbach's alpha. It showed that all the instrument to measure the behavioural intentions of international bleisure tourists to visit leisure attractions during their business trip is valid and reliable. The outcome of this study is useful for future research in the field of behavioural intentions.


Introduction
In recent decades, the emergence and remarkable growth of business tourism (Pinho & Marques, 2019) in most developed and developing countries have led to substantial development of economies, destinations, and employability (Anas et al., 2020;Konar & Hussain, 2018). It is an alternative and complementary type of tourism that attracts international tourists by combining commercial, governmental, and educational purposes with leisure activities (Pinho & Marques, 2019). However, as the industry has become increasingly globalised, little attention has been paid to its secondary motivation (Caicedo-Barreth et al., 2020;Lichy & McLeay, 2018;Swarbrooke & Horner, 2007) and the impact of implementing destination trust on visit intention. Visit intention is an essential topic in the tourism discipline (Ng et al., 2007), as the analysis conducted by The Travel and Tourism Economic Research (2006) and World's Leading Outbound Markets (2005) showed that it is important for competing for the number of tourists and rising income. According to Chen et al (2014), visit intention is a tourist's willingness or likelihood to visit a destination. Several researchers (Abubakar et al., 2017;Choi et al., 2016;Liang et al., 2018;Swan et al., 1999) reported that visit intention is derived from overall trust and contributes positively to future behavioural intention. Apart from this, it also plays an important role in the marketing management of tourism organisations (Jailani et al., 2019;Mulec, 2010). Identifying the right factors that promote visitation intention will attract potential tourists, enable proper segmentation of tourist arrivals, and generate more revenue (Jailani et al., 2019). Therefore, it plays an important role in promoting international tourism. Although numerous research studies have found the various antecedents of visit intention, they are still insufficient and need to be continuously researched to ensure the suitability of the sector (Zheng et al., 2021). Destination trust is an important construct (Marinao et al., 2012;Yao et al., 2013) for developing and sustaining the destination (Liu et al., 2019). Destination trust determines the success of destination marketing. It thus contributes to the marketability and popularity of destinations (Choi et al., 2016). In this context, destination trust can be defined as the ability of the tour operator to deliver what was promised and to meet or exceed the tourist's expectations based on knowledge (Osman, 2013). Abubakar and Ilkan (2016) describe trust in a destination as "a visitor's willingness to rely on a destination's ability to deliver the advertised features acquired through the integrity and transparency of a destination's service offerings." However, the desire to travel to a destination stems from people's needs of a specific tourism activity, which is known as motivation (Pizam et al., 1979). Motivation is regarded as the critical factor underlying all tourist behaviour (Bozic et al., 2017) because the rapid expansion of international tourism and the increased competition among attractions has influenced the change of travel motivation and the tourism development of leisure attractions in a specific area (Lee & Crompton, 1992;Phosikham, 2010). As s result, more empirical research is needed to understand how travel motivations differ across travel segments (Yousaf et al., 2018). Push and pull motivation is a widely accepted travel motivation theory (Crompton, 1979;John & Larke, 2016;Michael et al., 2017;Prayag & Hosany, 2014;Zhang & Peng, 2014). Push motivation influences tourists who travel to a specific destination influenced by their intangible, internal or intrinsic desires (Chon, 1986;Lam & Hsu, 2006;Uysal & Jurowski, 1994). Pull motivation, according to Lam and Hsu (2006) and Uysal and Hagan (1993), is related to the characteristics of the specific leisure attraction that lead people to perceive the need for certain travel experiences. Numerous studies have looked at push and pull motivation, but few looked at the secondary purpose of bleisure tourists' behavioural intentions towards leisure attractions while on business trips (Pearce, 2016). Only few researchers (Caicedo-Barreth et al., 2020;Fredriksson & Hakansson, 2018;Lichy & McLeay, 2018) have investigated bleisure tourists' push-pull motivation towards leisure attractions during their business trips, but with different methodological approaches and push-pull motivations have been identified separately. Bleisure tourists' push-pull motivation for leisure attractions is essential because, according to NB (2015), some business travellers have the professional resources and flexibility to incorporate leisure days into a business trip. Several scholars (Fredriksson & Hakansson, 2018;Kerr et al., 2012;Smith & Carmichael, 2007) support the proposition that business travellers tend to extend their trips and engage in more tourism activities while travelling for business reasons. Therefore, this paper explores how destination trust mediates the relationship between push motivation, pull motivation, and behavioural intention.

Materials and Methods
An online questionnaire was adapted from several previous researchers to collect the data (Abubakar & Ilkan, 2016;Boger, 2020;Cengizci et al., 2020;Nthebe, 2016;Pujiastuti et al., 2017;Salsabila & Alversia, 2020). Questionnaire adaptation was based on the topic of this research, which targets international bleisure tourists in Malaysia. The questionnaire consisted of three sections: Section One concerned information on participants' travel behaviour, Section Two related to push and pull motivation (independent variables), destination trust (mediator variable), and visit intention (dependent variable), while Section Three focused on participants' demographic profiles. A 10-point Likert scale was used to measure the constructs. Researchers (Alias et al., 2019;Awang et al., 2018;Mohamad et al., 2018) stated that 1 represents "Strongly Disagree," and 10 represents "Strongly Agree." The instruments were developed based on the previous study and adapted accordingly. Pretest and pilot tests were conducted for these adapted instruments to improve them before they were incorporated into the final questionnaire. During the pretest, a few experts reviewed the questionnaire to ensure that all items were appropriate and amendments were implemented before proceeding with the study (Zikmund et al., 2013). Three experts validated the content validity of this research, and their comments were considered. English language experts performed face validation and a statistics expert checked criterion validity. After completing the pretest validation, the instrument was distributed to 15 respondents to ensure no further errors before proceeding with a pilot study of 110 respondents. A pilot study is a small-scale study used to uncover deficiencies of a proposed survey before conducting the final full-scale study (Viechthaner et al., 2015). Few researchers (Bahkia et al., 2019;Rahlin et al., 2019;Shkeer & Awang, 2019;Yahaya et al., 2018) stated that EFA had to be conducted for individual variables to identify whether the dimensionality of items had changed from previous researchers due to previous population differences in characteristics.

Results and Discussion
To test the dimensionality of items altered by previous researchers, the EFA was applied to all constructs in this study. Previous research instruments were modified and adapted to meet the needs of current research. It includes the mean and standard deviation of each item, as well as the Kaiser-Meiyer-Olkin (KMO) measure of sampling adequacy, which shows the total variance explained for each construct, the factor loading of all items, the dimensionality of the items in their own parts, and Cronbach's alpha, which measures the internal consistency of the construct (Baistamam et al., 2020;Ehido et al., 2020;Rahlin et al., 2019).

The EFA Procedures for the First Construct: Push Motivation
17 items (KN1-EC17) from Table 1 were used to measure this construct. A 10-point Likert scale was used to measure each item, with 1 representing "Strongly Disagree" and 10 representing "Strongly Agree." Table 1 displays the mean, standard deviation, and item statement.  Table 2 displays the outcome of Bartlett's Test of Sphericity, which is significant at P < 0.05. The KMO for the first construct is greater than 0.6, at 0.680, indicating that the sample size is adequate (Bahkia et al., 2019;Hoque et al., 2017Hoque et al., , 2018Noor et al., 2015;Shkeer & Awang, 2019). As a result, the current data are satisfactory. 0.000 Figure 1: The scree plot clearly shows the emergence of three components Figure 1 depicts the Scree plot of three components resulting from the EFA. Later, the EFA procedure was determined for each item within each component (Alias et al., 2019;Rahlin et al., 2019;Yahaya et al., 2018). Table 3 shows that the construct explains 61.53% of the total variance (Component 1 contributes 28.15%, Component 2 contributes 17.92%, and Component 3 contributes 15.47%). It is acceptable because it is higher than 60% (Awang, 2010(Awang, , 20122014;Bahkia et al., 2019;Baistaman et al., 2020;Hoque et al., 2017;Noor et al., 2015;Yahaya et al., 2018).  Table 4 shows the results of components or dimensions for all items representing three components. All item factor loading greater than 0.5 were retained (Al-Khamaiseh et al., 2019;Baistaman et al., 2020;Chan & Idris, 2017;Ehido et al., 2020;Hair et al., 2014;Rahlin et al., 2020;Zulkepli et al., 2017). Thus, of the 17 items, 15 items were retained, while two were deleted because of low factor loading.

The Instrument Measuring Push Motivation's Internal Reliability
Finally, all constructs' internal reliability was calculated. Because the construct is measured by three components, calculating Cronbach' Alpha for each component is required to examine the internal reliability of a specific component. Table 5 confirms the component reliability, as Cronbach's alpha test 0.74 is greater than 0.7.  Table 6 were used to measure this construct. A 10-point Likert scale was used to measure the items, with 1 representing "Strongly Disagree" and 10 representing "Strongly Agree." Table 6 shows each construct's mean, standard deviation, and item statement. Shorter travel distance 7.32 0.98 The principal component analysis was used in the study to extract 18 items measuring the pull motivation construct. Table 2 displays the outcome of the Bartlett's Test of Sphericity, which is significant at P < 0.05. The KMO for the second construct, 0.710, is greater than 0.6, indicating that the sample size is adequate (Bahkia et al., 2019;Hoque et al., 2017Hoque et al., , 2018Noor et al., 2015;Shkeer & Awang, 2019). As a result, the current data are satisfactory.     Table 9 displays the component or dimension results for all items representing four components. Every item factor loading greater than 0.5 was kept (Al-Khamaiseh et al., 2019;Baistaman et al., 2020;Chan & Idris, 2017;Ehido et al., 2020;Hair et al., 2014;Rahlin et al., 2020;Zulkepli et al., 2017). As a result, 17 of the 18 items were retained, while one was deleted due to low factor loading.  Table 11 were used to measure this construct. A 10-point Likert scale was used to measure the items, with 1 representing "Strongly Disagree" and 10 representing "Strongly Agree." Table 11 shows each construct's mean, standard deviation, and item statement. The principal component analysis was used in the study to extract five items measuring the DT construct. Table 12 displays the outcome of the Bartlett's Test of Sphericity, which is significant at P<0.05. The KMO for the third construct is greater than 0.6, at 0.675, indicating that the sample size is adequate (Bahkia et al., 2019;Hoque et al., 2017Hoque et al., , 2018Noor et al., 2015;Shkeer & Awang, 2019). As a result, the current data are satisfactory. 0.000 Figure 3: The scree plot clearly shows the emergence of one component Figure 3 depicts the Scree plot of one of the EFA components. The EFA procedure was later determined for each component item (Alias et al., 2019;Rahlin et al., 2019;Yahaya et al., 2018). Table 13 shows that the construct explains 79.69% of the total variance. It is acceptable because it is higher than 60%. (Awang, 2010(Awang, , 20122014;Bahkia et al., 2019;Baistaman et al., 2020;Hoque et al., 2017;Noor et al., 2015;Yahaya et al., 2018;Hoque et al., 2017;Noor et al., 2015;Yahaya et al., 2018).  Table 14 shows the results of components or dimensions for all items representing one component. All item factor loading greater than 0.5 were retained (Al-Khamaiseh et al., 2019;Baistaman et al., 2020;Chan & Idris, 2017;Ehido et al., 2020;Hair et al., 2014;Rahlin et al., 2020;Zulkepli et al., 2017). Thus, of the five items, three items were retained, while two were deleted because of low factor loading.   Table 16 were used to measure this construct. A 10-point Likert scale was used to measure the items, with 1 representing "Strongly Disagree" and 10 representing "Strongly Agree." Table 16 shows each construct's mean, standard deviation, and item statement. Six items measuring visit intention construct were extracted using principal component analysis in the study. Table 17 displays the result of the Bartlett's Test of Sphericity, which is significant at P 0.05. The KMO for the fourth construct, 0.728, is greater than 0.6, indicating that the sample size is adequate (Bahkia et al., 2019;Hoque et al., 2017Hoque et al., , 2018Noor et al., 2015;Shkeer & Awang, 2019). As a result, the current data are satisfactory. 0.000 Figure 4: The scree plot clearly shows the emergence of one component Figure 4 depicts the Scree plot of one of the EFA components. The EFA procedure was later determined for each component item (Alias et al., 2019;Rahlin et al., 2019;Yahaya et al., 2018). Table 18 shows that the construct explains 77.99% of the total variance. It is acceptable because it is higher than 60%. (Awang, 2010(Awang, , 20122014;Bahkia et al., 2019;Baistaman et al., 2020;Hoque et al., 2017;Noor et al., 2015;Yahaya et al., 2018;Hoque et al., 2017;Noor et al., 2015;Yahaya et al., 2018).  Table 19 shows the results of components or dimensions for all items representing one component. All item factor loading greater than 0.5 were retained (Al-Khamaiseh et al., 2019;Baistaman et al., 2020;Chan & Idris, 2017;Ehido et al., 2020;Hair et al., 2014;Rahlin et al., 2020;Zulkepli et al., 2017). Thus, three items were retained of the six items, while three were deleted because of low factor loading.

Contribution of the Study
The contemporary look affords a few insights that provide implications closer to academicians and practitioners. From an educational perspective, this study contributes solutions to a few critical questions that researchers should consider when considering bleisure tourism. It led the researcher to realize and acknowledge the rising developments in bleisure tourism by examining the bleisure tourists' motivation and further illuminating how trust in a leisure attraction influences their intention to visit. Consequently, this research would widen the understanding of the bleisure tourists' trust and provide further insight and valuable knowledge of tourist behaviour in bleisure tourism contexts for upcoming researchers.
In terms of the practical contribution of this study, it is expected to bring many outstanding benefits to the Malaysian government in terms of understanding the motivations of bleisure tourists to visit leisure attractions during business trips. Other relevant institutions that will benefit from the current research are tourism-related institutions (such as business tourism organizers and tourism marketers) and tourism-related companies (such as travel agencies). This study provides valuable information and appropriate marketing strategies for relevant agencies. It finds appropriate approaches to attract and motivate bleisure tourists to enhance their intention to visit leisure attractions and indirectly reduce the cost of marketing and advertising.

Conclusion
This study showed valid results. The EFA provided considerable support for the meaningful use of push motivation, pull motivation, and destination trust for the behavioural intentions of international bleisure tourists to visit leisure attractions during their business trips. As a result, the dimensions of push motivation, pull motivation, destination trust, and visit intention were used in this study to assess their reliability. The result had a high Cronbach's alpha, met the requirements of the Bartlet's test (significant), a high KMO (>0.6), and factor loadings that exceeded 0.50. All of this points to the data's sufficiency. As a result, in this study, this instrument can assess the efficacy of the push motivation, pull motivation, and destination trust components on the behavioural intentions of international bleisure tourists to visit leisure attractions during their business trip. Finally, this method is suggested for future research.