The Moderating Role of Social Capital In Influencing Innovation in The Hospitality Industry

Vol


Introduction
The purpose of this study is to look into the relationship between knowledge transfer, social capital, and innovation in the hotel industry in Malaysia. Until recently, this topic has been disregarded in the context of knowledge transfer-innovation literature in both Malaysian and other business contexts (Ashtiani, 2014).
With regards to the above matter, literature has proved that to increase the competitive advantage in a global and growing market, the hotel industry needs to focus on innovation (Kallmuenzer, 2018;Río-Rama et al., 2017). Further, studies by Kandampully et al. (2016) and Melhem, Zeffane, and Albaity (2018) also concluded that innovations are the key factors to achieve competitive advantage.
In Malaysia, study by Ahmad & Scott (2019) suggested a high correlation between innovation and an increase in employees' productivity and efficiency. In addition, a study conducted by Asadi et al (2020) recognize the value of innovation toward the achievement of sustainable development in the hospitality industry.
However, the implementation of innovation strategies is still one of the biggest hurdles in the hotel industry in Malaysia (Balasubramanian and Ragavan, 2019). Balasubramanian and Ragavan (2019) also mentioned in their research that the Malaysian hospitality industry is currently facing some issues and challenges due to this shift in service innovation.
In addition, various studies (Hassi, 2019;Segarra-Ona et al., 2018;Sltten and Mehmetoglu, 2015) have examined the role of knowledge transfer and organisational innovation in the hotel industry; however, these studies have not assessed the role of knowledge transfer and organisational innovation in the hotel industry, which is the motivational factor for this research.
In relation to this, one could say that the effect of knowledge transfer on innovation might be different in the social capital context (Ganguly et al., 2019;Nguyen & Ha, 2020;Ashtiani, 2014). In addition, it appears that social capital is the most appropriate triggering factors of innovation (Doloreux & Parto, 2005). On the other hand, social capital promotes cooperation, as well as exchange of resources and knowledge; contributes to innovation (Bengrich et al., 2020). Nevertheless, Bengrich and colleagues suggested that it is worth mentioning that research on the dynamics of social capital, knowledge transfer and innovation are still scarce and could be better developed. Thus, in response to the calls for developing integrative models that clarify the effects of knowledge transfer on innovation development, the possible moderating role of social capital will be examined in this research (Ashtiani, 2014;Works, 2019;Fatemi et al., 2021). Further, Social Capital Theory suggests that social capital strongly influences the extent of knowledge transfer (Chiu et al., 2006;Ganguly et al., 2019;Swanson et al., 2020). Consistent with this theory, it is acknowledged that social capital provides an underlying environment for behavioural change that leads to greater coordination among individual and business units and eventually to more effective knowledge transfer, which enhances innovation capability (Le & Lei, 2019;Ganguly et al., 2019;Nguyen & Ha, 2020). Moreover, there is some evidence to suggest that social capital moderates the relationship between knowledge transfer and innovation capability (see Ashtiani, 2014;Works, 2019 andFatemi et al., 2021).
In summary, the study highlighted the importance of understanding the relationship between knowledge transfer -innovation. This research offers a holistic view of how these variables interact and influence one another. It could contribute to the areas of knowledge transfer study by linking social capital perspective in the knowledge transfer -innovation domain.

Literature Review
A major part of the economic growth of developed countries comes through innovation (Ghorbani et al., 2012) and innovation has been constantly studied to serve customers better (Truong, Dang-Pham, McClelland & Nkhoma, 2020). Literature has proved that to increase the competitive advantage in a global and growing market, the hotel industry needs to focus on innovation (Kallmuenzer, 2018;Río-Rama et al., 2017). Further, studies by Kandampully et al (2016); Melhem et al (2018) also concluded that innovations are the key factors to achieve competitive advantage.

Social Capital moderates the relationship between Knowledge Transfer and Innovation
Social capital provides an underlying environment for behavioural change that leads to greater coordination among individual and business units and eventually to more effective knowledge transfer which enhances innovation (Nahapiet & Ghoshal, 1998;Lesser & Storck, 2001;Widén-Wulff & Ginman, 2004).
In response to the calls for developing integrative models that clarify the effects of basic knowledge transfer on innovation development, the possible moderating role of social capital will be examined in this research (Ashtiani, 2014 andWorks, 2019). This becomes particularly important since this research does so in the context of the hospitality sector; a knowledgeintensive community characterized by high levels of social interaction. Ashtiani (2014) suggested that social capital moderates the relationship between knowledge management enablers and innovation capability.
Further, Fatemi et al (2021) also highlights the possible moderating role different levels of social capital can alter the impact of the organizational practice on the extent of knowledge transfer. In addition, the moderating role of social capital has been tested in a recent empirical study (see Works, 2019). In summary, the literature indicates that the interaction between social capital and knowledge transfer has a significant influence on innovation.

Sample and Procedure
The subjects for the examination were top, middle and first-level managers in all Marriot International Inc. Hotels in Klang Valley (Ritz-Carlton, Sheraton, Westin, Aloft, Le Meridien and Four Point by Sheraton). The proposed data analysis technique for this examination is structural equation modelling (SEM), which is sensitive to test size and requires a sensible number of tests to accomplish sufficient power to test the proposed hypotheses (MacCallum et al., 1996). Taking into consideration suggestions by (Hair, 2010), the minimum total sample size is running from 5 cases for every parameter (40 questions). Thus, the minimum total sample for this research will be 200 respondents.

Measurement of the variables Innovation Capability
Innovation capability was measured by the scale developed by (Jansen et al., 2006). The scale requires ten items to be answered on a 5-point scale (1=strongly disagree to 5=strongly agree). Examples of the questions are as follows: "We introduce improved but existing products and services for our local market"; and "We invent new products and services".

Knowledge Transfer
Knowledge transfer was assessed using the scale by (Ko et al., 2005;Simonin, 1999). Examples of the questions are as follows: "My interaction with colleagues has increased my understanding on how the knowledge integrates with other knowledge" and "I have greatly reduced my initial knowledge dependence upon my colleagues since the beginning of my work." For each question, respondents answered the question on a 5-point scale (1=strongly disagree to 5=strongly agree).

Social Capital
Social Capital was assessed using the tested scale (see Ashtiani, 2014). Examples of the questions are as follows: "Most members knew each other before they joined this community" and "Members relied on each other for the truthfulness of the information shared" For each question, respondents answered the question on a 5-point scale (1=strongly disagree to 5=strongly agree).

Data Analysis Strategy
The data for this analysis was evaluated in accordance with SEM standards and procedures. In order to conduct SEM, the two-stage approach suggested by Anderson and Gerbing (1982) was adopted in this research. The objective of the two-stage approach is to evaluate the measurement model and then to fix the measurement model at the second stage when the structural model is estimated.

Data Screening: Testing of SEM Assumptions
Before they are further analysed, this section presents the screening and cleaning of data sets. The data sets were analysed based on two types of problems: 1) case-related problems such as accuracy of data entry, missing values, and outliers; and 2) data distribution problems such as normality testing (Tabachnick, 2007;Hair, 2010). A total of 212 surveys were received of which ten cases were eliminated because of the constant responses in the questions due to that they considered as dubious and illogical responses. Additionally, another two cases were founded that missing to respond all the questions; therefore, they considered to have missed the values to count (Sekaran & Bougie, 2010). The assessment of missing values using 5% cut-off criteria (Tabachnick, 2007;Hair, 2010) is discussed below.
Following the missing evaluation of values, the data was subjected to review by outliers. The z-scores for each case were compared using SPSS descriptions. No cases have been found to have z-scores above 3.29 (p<.001). Therefore, no univariate outliers have been identified based on the evaluation of z-score (Tabachnick, 2007). Further, multivariate outliers based on distance from Mahalanobis (D2) were evaluated. Results for D 2 indicate no observations having D²/df value exceeding 3 to 4. The removal of outliers resulted in the final 200 cases being kept for further analysis. After the assessment of outliers, data distribution assessment on all observed variables was undertaken. Based on absolute value of skewness and kurtosis, it appears that all measures were within the range of +/-1.0. As such, it can be assumed that the data set is distributed normally (Bentler, 1987;Schumacker, 1996). Finally, the z-statistic of 59.079 in this research is far above the suggested value of +/-2588 (Hair, 2010).

Descriptive
The study consisted of 200 respondents from top, middle and first-level managers in all Marriot International Inc. Hotels in Klang Valley (Ritz-Carlton, Sheraton, Westin, Aloft, Le Meridien and Four Point by Sheraton). They were between 27 and 55 years of age, with a mean age of 40.66 years. Of these, 105 were male and 95 female. Most of those participants were fluent in English.

The Measurement Model Test
A calculation model using the maximum likelihood estimation method was applied to optimise all measurements for the structural model. Confirmatory factor analysis (CFA) was carried out on the initial items linked to 3 main variables, i.e. knowledge transfer, social capital and innovation capability. The CFA results showed a marginal fit for the remaining items (cmin/df) = 1.62, root mean square approximation error (RMSEA) = 0.07, comparative fit index (CFI) = .70. Consequently, for all further investigations, this measurement model was used.
To determine the internal consistency of multiple measures for each construct, a reliability test was performed. As shown in Table 1, all Average Variance Extracted (AVE) values are higher than 0.5. In addition, for almost all structures, the composite reliability was between 0.71 and 0.77. This shows that in this analysis, many tests are reliable for evaluating each construct (Nunnally, 1978). A construct validity test was conducted using the factor loadings within the constructs and as shown in Table 1, all standardized factor loadings emerged to be fairly high. This showed that the measurement had convergent validity (Anderson & Gerbing, 1988).   Hair et al (2013), indicating strong validity of discriminants. In fact, Table 2 indicates that the coefficients of correlation between the latent constructs were not greater than 0.9, and hence the model is presumed to be free of multicollinearity problems (Tabachnick et al., 2007;Hair et al., 2013). Sufficient evidence was found from the reliability and validity tests to indicate that the constructs meet the criteria for their reliability, convergence and discriminant validity. In predicting innovation, attitude (.56) has the highest direct effects. These findings are consistent with other studies on knowledge transfer and innovation capability. Social capital was found to have a medium effect on innovation capability (.38) and statistically significant (p<.001). The relationship of knowledge transfer to innovation capability was found to be improved and statistically significant (p<0.001) with social capital as a moderator which indicates by the total effect figure (0.75).
Apart from path coefficients, squared multiple correlations (R²) were also used as an indicator showing the integrated effect size for predicted endogenous variables. R² values of .01, .09, and .25 could be used as an evidence of small, medium, and large effects respectively (Cohen, 1988). The R² of innovation capability was .46. This indicates that the structural relationships for knowledge transfer and social capital in the proposed structural model explain 46% of the total variation in innovation capability. Based on the R², it can be deduced that the proposed structural model had a robust statistical ability in explaining the innovation capability in the Malaysian Hospitality Industry.

Conclusion
As hypothesized, a positive and significant outcome was found when testing the possible moderating effect of social capital on the relationship between knowledge transfer and innovation capability which supported findings by Fatemi et al (2021); Ashtiani (2014); Works, (2019). The researcher asserted that innovation increased with the impact of knowledge transfer especially when higher social capital was evident. It is assumed that the more connectivity and interaction between community members, the greater the effect of knowledge transfer on innovation capability.
Finally, in order to evaluate the extent of knowledge transfer and innovation capabilities, this study introduces and develops an integrative model that integrates knowledge-based and social capital concepts. According to the research objectives, the amount of social capital may limit the impact of organisational knowledge transfer on innovation capabilities in the business. The social capital features may influence management's decision to support knowledge sharing and innovation processes.