Finite Mixture Partial Least Squares Research Articles

Discerning interrelationships among management accounting systems, organizational variables, and managerial performance

In today’s global market, information plays a determinant role in the competitiveness of business organizations. Information systems have become essential resources in daily management, strategy development, and implementation. In this context, management accounting systems (MAS) have undergone substantial changes to adapt to new technological and environmental realities and have become an increasingly valuable instrument to support managers’ decision-making processes. Nevertheless, managers may not effectively leverage the information provided by MAS to enhance the quality of their decisions and, consequently, their overall performance. Discrepancies in research outcomes suggest a potential oversight in analyzing other organizational variables through which MAS might impact managerial performance. Based on an extensive literature review, we identified a scarcity of recent research focusing on three variables that could potentially mediate the relationship between MAS and managerial performance. These variables are intricately linked to people-oriented aspects and include the accountants’ involvement in strategy formulation, implementation, and monitoring, user satisfaction with the information supplied by MAS, and the decentralization of decision-making processes. To address this gap, we used the finite mixture - partial least squares (FIMIX-PLS) to discern the interrelationships among the variables in the research model. The results highlight how the interaction between firm characteristics can enhance the use of MAS in managers’ decision support and consequently improve their performance. This research provides relevant theoretical contributions to the debate about different organizational environments and their influence on MAS effectiveness. It also offers managerial implications for business organizations that need to design, implement, and use management accounting systems.

PROVINCIAL SEGMENTATION IN INDONESIA: EXPLORING FACTORS INFLUENCING EDUCATION WITH SEM-PLS METHOD, INCORPORATING MODERATION EFFECTS AND FIMIX-PLS APPROACH

The significance of education as a developmental metric is underscored by its designation as the 4th goal in the SDGs, which emphasizes ensuring inclusive, equitable, and high-quality education while also expanding lifelong learning opportunities for all. This research relies on two primary sources: secondary data from publications by the Indonesian Central Statistics Agency (BPS RI) in 2023 and the BPS website. The educational variables examined in this study are believed to be influenced by latent variables, including school performance, infrastructure, and poverty levels. Employing the Finite Mixture Partial Least Squares (FIMIX-PLS) approach, the research identified 13 valid and reliable indicators of educational variables. It delineated three regional groups based on the lowest BIC and CAIC values. In this structural equation research, the moderation effect is seen in the significance of the indirect relationship, especially the influence of Regional Poverty on Education with School Outcomes as a moderating construct.

Mapping dust risk under heterogenous vulnerability to dust: The combination of spatial modelling and questionnaire survey

Dust storms are natural events that remove and relocate surface soil, damage vegetation crops, and disrupt many other aspects of the earth's terrestrial ecosystem. Despite the importance of the risk assessment of dust hazards, vulnerability modeling of them is very limited. For this reason, this study provides a conceptual model based on Structural Equation Modeling and the Finite Mixture Partial Least Squares (FIMIX-PLS) approach using interviews and questions for vulnerability modeling of dust in Ahvaz County, Khuzestan province, Iran. Key model drivers included Resilience Actions, Natural-Physical effects, Economic Influence, and Social Influence. The Aerosol Optical Depth (AOD) product of MODIS/Terra was used to develop a dust hazard map. MODIS/Terra performance was evaluated using observed PM10 data from Ahvaz County air pollution monitoring stations. Land use mapping was used for spatial detection of agricultural land affected by the intensity of the AOD map in the previous step. The vulnerability model fitting results showed that the model had acceptable validity (SRMR = 0.013). Results showed that approximately 25 % of agricultural lands are at high and very high dust hazard risk. Based on modeling results, natural-physical variables affect about 89 % and 97 % of social and economic drivers, respectively. Conversely, social influences significantly negatively affect dust storm resilience resulting in agricultural vulnerability. Based on results from the integrated model, strengthening farmers' resilience strategies against dust hazards requires additional research and attention.

Unobserved heterogeneity in firm performance: The alignment of entrepreneurial orientation and organizational error management culture

Most prior research has focused on the positive relationship between a firm’s entrepreneurial orientation (EO) and its performance. However, errors in entrepreneurial strategies are inevitable. We argue that entrepreneurial firms benefit from an organizational error management culture. Drawing on data from 208 German family businesses, we combined partial least squares structural modeling with finite mixture partial least squares (FIMIX-PLS) segmentation to capture unobserved heterogeneity in firm performance. Our results show that firms align organizational error management culture with the behavioral or attitudinal dimensions of EO. The two identified segments indicate that firms align one at the cost of the other. Error communication emerged as a central link in both segments, mediating EO and firm performance. Overall, the results highlight the importance of aligning cultural practices, particularly the dialogue about errors within firms, with entrepreneurial behaviors and attitudes across managerial and firm levels.

Study of intention to use renewable energy technology in Malaysia using TAM and TPB

Despite the increase in the usage of renewable energy in Malaysia, the proportion of renewable energy (RE) within the energy mix lags far behind satisfactory levels, at a mere 23% in 2020 compared to the 31% target by 2025 as per Malaysia's Energy Transition Roadmap. The public acceptance towards RE is one of the many obstacles that hinders the proliferation of RE in Malaysia. It is therefore imperative to determine the key factors which shape an individual's intention to utilize RE. In this study, three models are compared and assessed in terms of their efficacy in predicting such behavioural intention, namely the Technology Acceptance Model (TAM), Theory of Planned Behaviour (TPB) and an extended model that combines the characteristics of the two prior theories. Partial-least-squares structural equation modelling, finite mixture partial least squares (FIMIX-PLS), and important-performance map analysis (IPMA) were performed on the collected survey dataset with 524 responses. Analyses revealed that attitude and perceived behavioural control are important factors of intention to adopt renewable energy technology, with perceived ease of use and perceived usefulness in turn being significant antecedents of attitude, whereas subjective norm did not influence intention. Among the three models, the extended model triumphs in terms of path significance and explanatory power, followed by TPB and then TAM. Results show that costs, ease of use and accessibility of RE remain limiting factors towards RE adoption in Malaysia, hence the need of appropriate policies/programmes, such as targeted incentive campaigns, to alleviate this issue.

Unveiling heterogenous knowledge-oriented leadership and knowledge acquisition based hybrid work agility of knowledge workers

PurposeWith new hybrid working models in place post COVID-19, it is requisite that knowledge workers (KWs) stay agile. Knowledge-oriented leadership (KOL) can help employees with essential knowledge acquisition (KA) facilitating the journey toward hybrid work agility (HWA). This study, thus, aims to explore the impact of KOL and KA on HWA and reveal whether this effect stems uniformly from a single homogenous population or if there is unobserved heterogeneity leading to identifiable segments of agile KWs.Design/methodology/approachData was collected through stratified sampling from 416 employees from 20 information technology enabled services companies involved in knowledge-intensive tasks. Partial least squares (PLS) structural equation modeling approach, using SMART PLS 4.0, has been applied to examine the effect of KOL and KA on HWA. Finite mixture PLS, PLS prediction-oriented segmentation and multigroup analysis have been used to identify segments, test segment-specific path models and analyze the significance of the differences in the path coefficients for unobserved heterogeneity. Predictive relevance of the model has been determined using PLS Predict.FindingsResults indicate that KOL contributes to employees’ KA and HWA. A significant positive relationship is also reported between KA and HWA. The model has medium predictive relevance. A two-segment solution has been delineated, wherein independent agile KWs (who value autonomy and personal agency over leadership for KA) and dependent agile KWs (who depend on leaders for relational and structural support for KA) have been identified. Thus, KOL and KA play a differential role in determining HWA.Research limitations/implicationsThe authors’ major contribution to the knowledge body constitutes the determination of antecedents of HWA and a typology of agile KWs. Future researchers may conduct segment-wise qualitative analysis to delineate other variables that contribute to HWA.Practical implicationsTechnological advances necessitate that knowledge-intensive industries foster agility in employees for strategic agility of the organization. For effecting agile adaption of an organization to the knowledge economy conditions, it is pertinent that the full potential of this human resource be used. By profiling HWA of KWs on the basis of dimensions of KOL and the level of their KA, organizations will be able to help employees adapt better to rapidly changing work conditions.Originality/valueHWA is a novel concept and very germane in a hybrid working environment. To the best of the authors’ knowledge, this is the first study to examine the effects of the dimensions of KOL and KA in relation to HWA, along with an empirical examination of unobserved heterogeneity in the aforementioned relationship.

Investigating speed-safety association: Considering the unobserved heterogeneity and human factors mediation effects.

The relationship between mean speed and crash likelihood is unclear in the literature. The contradictory findings can be attributed to the masking effects of the confounding variables in this association. Moreover, the unobserved heterogeneity has almost been criticized as a reason behind the current inconclusive results. This research provides an effort to develop a model that analyzes the mean speed-crash frequency relationship by crash severity and type. Also, the confounding and mediation effects of the environment, driver, and traffic-related attributes have been considered. To this end, the loop detector and crash data were aggregated daily for rural multilane highways of Tehran province, Iran, covering two years, 2020-2021. The partial least squares path modeling (PLS-PM) was employed for crash causal analysis along with the finite mixture partial least squares (FIMIX-PLS) segmentation to account for potential unobserved heterogeneity between observations. The mean speed was negatively and positively associated with the frequency of property damage-only (PDO) and severe accidents, respectively. Moreover, driver-related variables, including tailgating, distracted driving, and speeding, played key mediation roles in associating traffic and environmental factors with the crash risk. The higher the mean speed and the lower the traffic volume, the higher odds of distracted driving. Distracted driving was, in turn, associated with the higher vulnerable road users (VRU) accidents and single-vehicle accidents, triggering a higher frequency of severe accidents. Moreover, lower mean speed and higher traffic volume were positively correlated with the percentage of tailgating violations, which, in turn, predicted multi-vehicle accidents as the main predictor of PDO crash frequency. In conclusion, the mean speed effects on the crash risk are entirely different for each crash type through distinct crash mechanisms. Hence, the distinct distribution of crash types in different datasets might have led to current inconsistent results in the literature.

Social Media Marketing as a Segmentation Tool

The aim of this study was to determine consumer segments based on the acceptance of shoppable ads from fashion brands on online social media platforms. To achieve this objective, we used the technology acceptance model (TAM) to offer a vision of the perceptions of the shoppable ads, attitudes and behaviors of social network users, using social media marketing activities as a background. Second, we searched for fashion social buyer segments using finite mixture partial least squares (FIMIX-PLS). Third, we sought to characterize these consumer segments. A sample of 486 users of social networks who accessed through mobile devices was obtained. The inclusion of social media marketing variables as antecedents of acceptance allowed us, to a large extent, to understand the intention to buy clothing by these social media users. The a posteriori segmentation technique helps to identify different types of users who use shoppable ads and their relationship with age and concerns about privacy, trust and purchases made on the Internet. The results show that, based on the explained variance and model fit, the proposed variables allow us to explain acceptance, with two groups of consumers within the sample being found.

Identifying unobserved heterogeneity in mobile wallet adoption – A FIMIX-PLS approach for user segmentation

PurposeThe purpose of this study is to segment mobile wallet users using a finite mixture partial least squares (FIMIX-PLS) approach and evaluate the unobserved heterogeneity across segments.Design/methodology/approachPartial least square structural equation modeling (PLS-SEM) using a convenience sample of 744 responses was used to analyze the measurement, structural model and hypotheses testing. To examine unobserved heterogeneity and identify user segments, FIMIX-PLS technique was employed. To generate more precise recommendations, importance-performance map analysis (IPMA) was performed with attitude as the target variable.FindingsA structural equation model revealed that except perceived ease of use (PEOU) all other dimensions, namely perceived usefulness (PU), lifestyle compatibility (LC), facilitating conditions (FC), trust and security significantly influences attitude which, in turn, determines intention. The FIMIX-PLS technique resulted in four segments – The Rationalist, Early Adopters, Late Adopters and The Innovators.Practical implicationsThe paper provides segment specific and between segment differences to derive implications. Identification of relevant predictors and segments will help academicians, marketing researchers and practitioners in gaining further understanding of the mobile wallet adoption. The findings of the paper can guide mobile wallet providers to frame appropriate strategies and offerings pertaining to the obtained segments.Originality/valueThe paper builds upon Technology Acceptance Model (TAM) to propose an integrated model to explain adoption behaviors associated with mobile wallet. To the best of the authors' knowledge, this is one of the first empirical attempts using FIMIX-PLS technique to assess precursors of adoption and substantiates the perceived value-attitude-intention linkage to identify heterogeneity among mobile wallet users.

Cultural influences on future transportation technology usage: The role of personal innovativeness, technology anxiety and desire

AbstractThis study develops and validates a model, based on personal cultural values theory and psychological research, in relation to technology adoption. The model focuses specifically on the future use of on‐demand air mobility (ODAM), which is expected to have significant implications for city commuting and personal well‐being in the years ahead. We use a path modelling approach, in addition to recently advanced analytical methods such as the finite mixture partial least squares (FIMIX‐PLS), measurement invariance of composite models (MICOM) and multi‐group analysis, to validate the model using a data set of 627 young consumers from the Czech Republic. The research model explains 45.2% variation in the future use of ODAM using our global model. This variance explained in the future use of ODAM increases to 62.3% and 64.5% respectively, when we segment our data set into two groups. The results also show that tradition has significant influence on technology anxiety, personal innovativeness and desire to use ODAM. Independence positively affects personal innovativeness but not the desire to use ODAM. We also find that technology anxiety influences the desire to use, which in turn influences the future use of ODAM. However, we find group differences in the influence of ambiguity intolerance on technology anxiety, desire and personal innovativeness. Thus, the study also evaluates the existence of significant differences between two groups in our data set. Overall, the study suggests that individual cultural values play a particularly important role in influencing the future use of ODAM through psychological characteristics. The research implications of the study are discussed in the article.

Heterogeneity in passenger satisfaction with air-rail integration services: Results of a finite mixture partial least squares model

Abstract Air-rail integration services (ARIS) are a promising solution for intermodal travel, but they also offer challenges such as trip complexity and uncertainty that may negatively impact passenger satisfaction. This paper addresses the quality evaluation of ARIS with a twofold purpose, specifically, (1) to explore the causal relationships between the psychological and behavioral variables of passengers and (2) to identify different passenger groups for precise service improvements. The relationships among passenger expectations, perceived quality, perceived value, satisfaction, complaints and loyalty are hypothesized and analyzed by using a passenger satisfaction index (PSI) model based on a representative sample of 1,345 intermodal passengers at Shijiazhuang Zhengding International Airport (SJW). Next, a finite mixture partial least squares (FIMIX-PLS) technique is applied to explore the heterogeneity. The results suggest that perceived quality, perceived value and passenger expectations have a positive effect on passenger satisfaction, which, in turn, influences passenger loyalty and complaints. Moreover, three passenger segments are found, namely, “perceived value and expectation-oriented passengers”, “perceived quality-oriented passengers”, and “satisfied loyal-conscious passengers”. Interestingly, an overall similarity in the perceived quality among the three passenger segments is observed with the importance-performance map analysis (IPMA) method. Nevertheless, some noticeable differences in the psychological-behavioral relationships of the three passenger segments are identified. Finally, managerial implications are discussed.

User's Segmentation on Continued Knowledge Management System Use in the Public Sector

Knowledge management systems (KMS) can help an organization support knowledge management activities and thereby increase organizational performance. This study extends the expectation-confirmation model for predicting mandatory continued KMS use in the public sector. The models are assessed using data from a sample of 627 employees of the Kaohsiung City government in Taiwan and analyzed using the finite mixture partial least squares (FIMIX-PLS) method. The results of this study indicate that (1) data heterogeneity (i.e., educational level) segments two specific groups that show different perceptions toward continued KMS use; (2) the results of aggregate-based data analysis are different from the results of group-specific data analysis; (3) compatibility, relative to confirmation, has larger impact on perceived usefulness regardless of groups; (4) the effect of user satisfaction on continued usage behavior is significant different between the two groups; (5) cognition-driven continued use and emotion-driven continued use are identified in the two groups.

Is Job Satisfaction of Social Sciences Scholars Predicted by Emotions, Job Performance, Work Events, and Workplace Features? A Demonstration of a Data-Driven Policy-Making Approach

This study, using affective events theory (AET) as a framework of reference, focuses on job satisfaction and job performance of academics with social sciences backgrounds working in Malaysian universities and colleges. More specifically, it aims at examining the influence of workplace features such as involvement, workload, and welfare on job satisfaction through role conflict, as a work event, and positive affect. In addition, the mediating role of job performance on the relationship between positive affect and job satisfaction was considered. Data were collected from 1000 academics via an online platform, and partial least squares structural equation modeling (PLS-SEM) method was adopted for data analysis. While all the four hypotheses were empirically supported, the results highlighted the considerable role of workload in increasing role conflict and the negative impact of role conflict on positive affect. We extended the results using finite mixture partial least squares (FIMIX-PLS) segmentation method, as a recommended PLS robustness check, and importance-performance map analysis (IPMA), to identify the major areas of improvement to be addressed by management activities on the grounds of the proposed model. Implications, limitations, and recommendations were discussed too.

Customer engagement with brands in social media platforms

PurposeSocial media brand platforms have become a popular means for engaged customers to share information and experiences with brands and other customers. However, empirical research on how customer engagement (CE) relates to customers’ sharing intentions with the brand is limited. This study aims to investigate causal patterns of four CE dimensions – focused attention, absorption, enthusiasm and interaction – together with two cognitive structure properties in stimulating sharing intentions with the brand.Design/methodology/approachUsing data from 782 Chinese customers of brand pages on the social media platform Weibo, this paper is the first to use both finite mixture partial least squares (FIMIX-PLS) analysis and fuzzy-set qualitative comparative analysis (fsQCA) to empirically assess the impact of CE configurations on sharing intentions.FindingsThe findings imply that not all of the CE dimensions co-occur necessarily and that different configurations of them can produce superior sharing intentions, conditional on the cognitive structure of customers, including their level of brand knowledge and avant-gardism.Research limitations/implicationsAlthough restricted to customers on Weibo, the results inform practice about how social media technology can facilitate different CE configurations and customer sharing intentions.Practical implicationsThe results inform brand managers’ segmentation efforts and CE content marketing initiatives that can induce different CE configurations and customer sharing intentions with customers that possess high avant-gard and brand knowledge characteristics.Originality/valueThis study is the first to substantiate how different CE configurations (as gestalts) affect sharing intentions in social media and to challenge conventional net-effects thinking about CE dimensions. Understanding how such conditional configurations foster sharing via a social media platform is advantageous because it can improve segmentation efforts to strengthen brand relationships.

ANTECEDENTS OF ONLINE PURCHASE INTENTION AND BEHAVIOUR: UNCOVERING UNOBSERVED HETEROGENEITY

The paper aims at exploring the antecedents of customers’ online purchase intention and behaviour, and at uncovering sources of heterogeneity. A sample of customers was surveyed to measure perceived risk and benefits, trust, online purchase intention and behaviour. The study confirmed the causal chain of perceived risks-trust-perceived benefits-online purchase intention-actual purchase. A Finite Mixture Partial Least Squares (FIMIX-PLS) was performed to uncover sources of heterogeneity. It found that the level of security of the payment methods is relevant to understand the relationship between purchase intention and behaviour, while the level of previous experience with the online medium clarifies the relationship between perceived risk and trust. The study contributes to understanding the antecedents of online purchase intention and their relationship with actual purchase behaviour. Additionally, it offers evidence of heterogeneity in the proposed causal relations, particularly, concerning the level of trust in the payment methods and the level of Internet experience.

Determining the Key Capacities of Effective Leaders in Malaysian Public and Private Focused Universities

The study was undertaken to identify the essential leadership capabilities and managerial competencies as the key leadership performance drivers in Malaysian focused universities. To collect data, the previously developed scales of capabilities, competencies, and leadership performance in the context of Malaysian Higher Education (HE) were distributed among the leaders in seven public focused and 12 private focused universities. In total, 172 completed surveys were collected among which 94 had been filled out by the leaders in Malaysian public focused and 78 had been completed by leaders in private focused universities. The data were screened and SmartPLS 3 was employed to analyze the data. Also, Finite Mixture Partial Least Squares (FIMIX-PLS) segmentation and Importance–Performance Map Analysis (IPMA) were run to extend the results. The outcome of FIMIX-PLS didn’t reveal unobserved heterogeneity within the data and, through IPMA, change-oriented capability was identified as the main improvement area to be addressed by management activities. Moreover, the implications of the findings were discussed and future directions were recommended.

Management Control of New Product Development and Perceived Environmental Uncertainty: Exploring Heterogeneity Using a Finite Mixture Approach

ABSTRACTResults regarding the association between perceived environmental uncertainty (PEU) and management control (MC) of new product development (NPD) are ambiguous and sometimes contradictory. Based on survey data for 276 firms, we first contribute to MC research by finding that firms have two completely different responses regarding their MC of NPD when facing high PEU. Whereas the first group of firms increases all MCs when PEU increases, the second group responds with a loosening of MC. The first group consists of slightly larger and older firms than the second group, with relatively emergent innovation strategies, greater innovation capabilities, and weaker perceived complexity of their environment. Second, we reveal these two contrary groups of firms behind an aggregated sample by introducing finite mixture partial least squares (FIMIX-PLS) as a new approach to MC research. Our study may inspire future researchers to control for previously unobserved heterogeneity in structural models by using FIMIX-PLS.JEL Classifications: D81; L60; M11; M40; O32.

Evaluation of employee brand using typological analysis in Indian airline organisation

A favourable employee brand gives service organisations competitive advantage. We analyse the typology of employee branding in an Indian airline company using finite mixture partial least squares (FIMIX-PLS). The employees of an airline organisation, Chennai division, run by the Government of India were surveyed and the types of employee brands prevalent in the organisation were identified. The empirical analysis substantiates the typology by observing the presence of four types in employees namely, all-stars, injured reserves, rookies and strike-out kings. The employees in all-star category have high knowledge of the desired brand (KDB) and perceive that the psychological contract (PC) has been upheld. The KDB of 'injured reserves' employees are high whereas they perceive violation of PC. The 'rookies' employees' KDB is low while their PC is favourable. Both the KDB and PC of 'strike-out kings' are low. The result also uncovers 'department', using chi-squared automatic interaction detection (CHAID), as a categorical variable that explains the segmentation. The management should enhance brand-centric human resource practices to make the organisation an 'all-star'.

Evaluation of employee brand using typological analysis in Indian airline organisation

A favourable employee brand gives service organisations competitive advantage. We analyse the typology of employee branding in an Indian airline company using finite mixture partial least squares (FIMIX-PLS). The employees of an airline organisation, Chennai division, run by the Government of India were surveyed and the types of employee brands prevalent in the organisation were identified. The empirical analysis substantiates the typology by observing the presence of four types in employees namely, all-stars, injured reserves, rookies and strike-out kings. The employees in all-star category have high knowledge of the desired brand (KDB) and perceive that the psychological contract (PC) has been upheld. The KDB of 'injured reserves' employees are high whereas they perceive violation of PC. The 'rookies' employees' KDB is low while their PC is favourable. Both the KDB and PC of 'strike-out kings' are low. The result also uncovers 'department', using chi-squared automatic interaction detection (CHAID), as a categorical variable that explains the segmentation. The management should enhance brand-centric human resource practices to make the organisation an 'all-star'.

Adoption of renewable energy technologies in Mexico

PurposeThis research aims to analyse cognitive factors, innovation attributes and their influence on adoption of solar renewable energy technologies (RETs) for urban households in Mexico. It expands existing cognitive frameworks by including variables from diffusion of innovation theory.Design/methodology/approachOn the basis of the data of 291 urban consumers and through the use of partial least squares (PLS), the proposed model was empirically tested. Finite mixture PLS method helped identify two market segments.FindingsFindings suggest that beliefs about consequences of adopting RETs have significant influence in shaping consumer’s attitudes towards RETs which were found to be an accurate predictor of the behavioural intention to adopt these technologies. Regarding innovation attributes, the results show that for a solar heater to be adopted, it should be compatible with the social values of the consumer. Triability and relative advantage show little influence on attitude formation. Two market segments found differ on the basis of beliefs and attitudes.Research limitations/implicationsThe study was limited to analyse consumer responses to solar energy in residential urban settings.Practical implicationsOrganizations wanting to increase their consumer base need to develop sound technological innovations with high levels of compatibility a low complexity.Originality/valueThe study combines diffusion of innovation theory with cognitive frameworks and finds that innovation attributes become strong predictors of intentions to adopt RETs.