Managers' Propensity for Participative Decision-Making with the Components of Theory of Planned Behavior: Evidence from Bangladesh

Abstract

The study aims to explore whether Ajzen's theory of planned behavior is determinant of managers' propensity for participative decision-making in Bangladeshi textile industries. Surveys measuring the theory of planned behavior assumptions and propensity for participative decision-making were administered to 384 full-time managers and supervisors from different textile industries across Bangladesh. A measurement and structural equation model were employed to explain the cause-and-effect relationship between the variables. Results generally support Ajzen's prediction that managers' psychological and psychosomatic predisposition is linked to participative decision-making. Findings suggest that managers' attitude toward participative decision-making, subjective norms, and perceived behavioral control significantly impact their avoidance or acceptance of participative decision-making in prevailing organizational settings. The findings also present a convincing argument for managers to constantly challenge their assumptions about workers and objectively analyze whether their perceptions affect their decision-making practices. The previous studies concerning managers' propensity for participative decision-making are conducted in western developing and developed nations. This study has taken the initiative to advance previous knowledge into a least developed country. This study expands the landscape of participative decision-making literature, adding further evidence that individual-difference variables, in this case, theory of planned behavior assumptions, greatly influence managers' and supervisors' biases about employee participation.

1. Introduction

Participative decision-making (PDM) refers to a collaborative process, joint consultation, or joint decision- making with the employees for organizational decision- making (Kumar & Saha, 2017). Employee voice and participation has been a long-sought desire, but recently has increased in frequency. Some authors argued that participa- tion is best when groups are involved, and the individual is given the freedom to exercise more independent judgment (Ekwoaba & Ufodiama, 2019). Alsughayir (2016, p. 68) has opined that PDM has an inevitable implication for management, highlighting the need for companies to show their commitment to employees in decision-making. Managers’ desire, or penchant for participative decision- making is a prerequisite for the program to succeed in any organizational setting (Parnell, 2010).

Of late, Bangladeshi manufacturing industries are facing multi-dimensional constraints to manage people at the workplace. The national culture determines the organizational culture and influences how the business operates. The foreign aid coming in the country might be affected by the economic growth, which is purely dependent on the industrial peace and harmony (Golder et al., 2020). Collective bargaining agency (CBA) and social dialogue are some of the few essential tools that entrepreneurs have to manage people in the workplace. However, previous researchers opined that these two tools have little impact in Bangladeshi manufacturing industries for managing people at the workplace, notably textile industries (Islam & Kalimuthu, 2020). Few multinational companies tried participative decision-making; however, they could not succeed. The managers’ propensity for participative decision-making (PPDM) varies across culture; the managers operating in the developed nations might not have the same perception as those serving in the least developing nation. There is a minimal study for finding managers’ behavior for PDM in Bangladesh, and the variables are poorly defined. Therefore, it is imperative to understand if the manufacturing industries’ managers possess the required PDM proclivity. While people can use PDM at all levels of an organisation, this study focuses exclusively on managers since their opinion or intention for implementing PDM committee is crucial for success.

The theory of planned behavior (Ajzen, 1991) aims to understand and predict behaviors, which maintains that behavior depends on beliefs and attitude. An intention is the result of three components: motivation, attitude toward the behavior, and perceived self-control. The components of the theory of planned behavior (TPB) have been empirically tested by authors trying to find out individuals intention or behavior, especially in business research (Arora et al., 2017; Jayasinghe, 2020). The TPB is primarily concerned with the prediction of intentions. Behavioral, normative, and control values and behaviors, subjective norms, and behavioral control expectations affect and clarify behavioral intentions. The intensity of the intention–behaviour relationship is moderated by absolute control over the behavior (Ajzen, 2011). Three major components of TPB might help predict managers’ predisposition for PDM. This descriptive study aims to predict if the assumptions of TPB predict the man- agers’ intention to adopt or decline PDM in the Bangladeshi manufacturing industry, notably in the textile industries.

2. Literature Review and Hypothesis Development

PDM is a process where individuals or groups can have a voice in decision-making. Their views are valued, recognized, and decisions are taken together (Ugwu et al., 2019). PDM is essential for organizational efficiency and productivity. Managers should make the necessary effort to balance task characteristics, performance and reward that satisfy staffs. Shaed (2018), in his study, opined that PDM strongly correlates with work satisfaction and organizational commitment. Structured participation refers to formal institutions such as works councils, quality circle, employee empowerment program or social dialogue.

Organizational management theorist such as Elton Mayo, Rensis Likert, Edward Locke, and William Ouchi have propagated the forms, dimensions, characteristics and modus operandi of PDM in the mid-nineties. However, Parnell and Bell (1994) introduced managers’ PPDM while developing a measurement scale to assess managers PPDM in the USA, considering the organizational commitment and power-sharing attitude. Later, the scale was modified by adding two more variables, such as managers’ commitment and organizational culture. Such study lacks individual managers’ cognitive attitude since managers’ willingness for PDM is a micro-level study concerned with behavioral beliefs and social norms. A few years later, Russ (2011) conducted an empirical study to expand the current landscape of PDM research by exploring how McGregors X/Y theory impact managers adoption of PDM. His study revealed that managers’ predisposition for PDM is more observed where managers follow theory Y’s assumptions.

The previous studies on managers’ PPDM were conducted in the developed (USA, Mexico and Latin America) and developing nations such as Turkey and Malaysia. (Parnell, 2010; Parnell et al., 2012; Parnell & Crandall, 2003). Prior research argued that PPDM studies need to be addressed in the least developed countries before generalizing the result since PDM varies across cultures. Islam and Raju (2020) conducted a study to ascertain the managers’ PPDM in Bangladeshi textile industries and found that managers’ PPDM in Bangladesh strongly corre- lates with job satisfaction, motivation, and organizational culture. The researchers have suggested conducting a study considering TPB to expand the existing knowledge on PPDM in the least developed countries.

Bangladeshi business entrepreneurs are faced with numerous challenges while managing their workers at factory premises. Workers are often found in the street making riot, protestation, violence, and breaking the law; the resultant effect is factory shutdown. The Bangladesh Labour Act 2006 promotes participative decision-making committee in all manufacturing industries regardless of size and dimension (Islam & Raju, 2020). According to their findings and speculations, managers in the Bangladeshi textile industry frequently insist on forcing their wishes and will on their employees, making how the people would-be- employees feel and desire. Employers discourage employees from offering criticism by asking them not to do so. The lack of solidarity in Bangladeshi ready-made garments factories results in a great deal of confusion in the industry (Islam, Juhi, & Raju, 2020). It is essential to determine if the Bangladeshi manufacturing industries’ managers show a positive attitude to empower workers to make decisions independently. A study could expand our understanding of how attitude, subjective norms and perceived behavioral control determine or influence managers’ PPDM.

The TPB describes the motivating factors for an individual to perform a particular action intention as the conscious decision to carry out a particular action. The model illustrating that intent comprises three distinct entities: attitudes, subjective norms, and perceived behavioral control. The TPB has been advanced from reasoned action theory (Ajzen & Fishbein, 1980) in the early nineties. Aizen and his colleagues have found that the earlier theory’s role and function (Theory of reasoned action) have changed over the period. Hagger (2015, p. 4), while citing Sniehotta and colleagues (2014), has argued that health psychology researchers will have far-reaching implications for future research; the legacy will impact the field significantly, but not to the degree that it did in the past. This theory is still favorable to management science researchers who have conducted empirical studies to identify consumers, managers, and employees’ intention for purchasing goods, job satisfaction, and commitment.

The conceptual framework (see Figure 1) is developed from existing literature prescribing the factors affecting managers’ PDM; however, the previous literature mostly considered organizational levels such as organizational culture, organizational effectiveness, and macro-level factors such as power-sharing communication apprehension. One’s proclivity or willingness to adapt PDM is influenced by his/ her behavioral belief toward the behavior, particularly PDM, belief in important referent and experience and capability to control the knowledge (PDM).

Figure 1: Conceptual Framework and Working Hypotheses

Attitude toward PDM refers to the degree to which a person shows interest. It consists of considering the results of performing the behavior. However, while one can have several beliefs, only the immediate determinants of those beliefs will ascertain their behavior. For example, when managers believe that PDM has a substantial contribution to productivity, they are likely to adopt PDM. Therefore, one’s belief for specific positive behavior toward PDM is likely to be influenced by their belief in the process or subject and its outcome. Psychological attitude measurement tools can objectively measure such behavioral attitude. Kan, Fabrigar, and Fishbein (2017) have argued that maximizing one’s prediction of an attitude requires that the behavior be measured in terms of the specific motivation, besides numerous studies show that more significant involve- ment in decision-making increases employee engagement (Ngoc & Pham, 2021). The effectiveness of PDM depends on many intrinsic and extrinsic factors; however, micro-level factors such as managers’ intention, willingness, and desire to participate might significantly influence managers’ PPDM (Cotton et al., 1988; Singh et al., 2009).

Managers’ predisposition for PDM significantly depends on their belief toward the benefit of PDM, and the anticipated outcome also depends upon the organizational culture (Westhuizen et al., 2012). For example, when participation is prevalent in an organization, managers are more likely to be participating in decisions. Phan (2021) has opined that there is a strong correlation between managers’ ability and firm performance. For PDM to thrive, the cultural context determines its meaning for its members and their reasons for being there, whom it should include, and which issues should be prioritized. The organisational culture influences ones’ belief about PDM. Therefore, it can be hypothesized that managers’ attitude toward the PDM might significantly affect managers’ PPDM (H1).

The subjective norms refer to the assumption that most people consider behavior to be good or bad based on the person’s beliefs about whether the behavior should be performed by others (Hadadgar et al., 2016). The subjective norms (opinions of others about the behavior) also postulate someone’s perception. Managers are likely to show more PPDM following social pressure where people or social groups have a different opinion about PDM. In order to improve innovation, leaders must increase the morale that workers are able to perform assignments, make them believe their job is important, empower them and motivate them to work independently to engage in decision-making (Siswanti & Muafi, 2020). The social dialogue that is being practiced in the textile industries also required a participative committee (a limited number of companies have found social dialogue effective) that might influence managers’ PPDM. Another aspect of subjective norms can be social and labor law pressure. Bangladesh Labour Act 2006 made it compulsory for entrepreneurs to organize participatory decision-making committee regardless of the size and type of industry. The pressure created by the government and social elites might influence managers’ PPDM. Therefore, it can be hypothesized that subjective norms and motivation to comply with social pressure might significantly impact managers’ PPDM (H2).

Aizen advanced the theory of reasoned action (Ajzen & Fishbein, 1980) by adding the third variable, perceived behavioral control. The theory of planned behavior assumes that managers weigh the risks and benefits of various options and select the policy that provides their expected real benefits. Human psychology is well known to be highly complex, consisting of social, moral, and benevolent behavior and self-interested behavior. Behavior is frequently embedded in collaborative and social decision-making contexts, as well as other contextual factors. Individual preferences are constantly shaped and constrained by these factors. Perceived behavioral control refers to when managers think they do not have complete control over their employees’ behavior. The first factor is self-confidence, which relates to how comfortable one feels about one’s capabilities (Sonja, 2018). The second variable is controllability, representing the availability of PDM resources and workers’ motivation and motivation. The absence of enabling conditions can harm the intention to use PDM as an effective tool for managing people at work. Therefore, it can be hypothesized that perceived behavioral control might significantly impact managers’ PPDM (H3).

3. Research Methodology

The descriptive research study aimed to understand if the theory of planned behavior’s assumptions helps predict managers’ PPDM. Data were collected from serving managers, non-managers, and supervisors from textile industries with a paid research assistant in Bangladesh.

3.1. Population, Sampling and Unit of Analysis

The study population consisted of managers and non-managers operating in the Bangladeshi textile industries. The researchers collected the list of textile industries from Bangladesh Textile Mills Association and randomly selected the industries. Thereafter the list of population was collected from the selected industries and samples were chosen randomly form the list. The researchers opted for stratified simple random sampling technique since the entire study population is stratified in strategic, tactical and operational level. The unit of analysis is the individual managers of those industries whose perception is crucial for the study’s success.

3.2. Data Collection Procedure

The researchers appointed a research assistant in Bangladesh for collecting data on behalf of the researchers, who is an expert paid research assistant. The research assistant personally distributed 692 research instruments among 34 textile industries managers. The respondents were supplied with a close-ended questionnaire accompanied by taka 25 for each respondent, and they were encouraged to courier the questionnaire. Three hundred ninety-four valid questionnaires were received within three months. The valid response rate is 56.53%, which is satisfactory. The respondents were given the option to answer within a five-point Likert scale where 1 denotes ‘strongly disagree’ and 5 denotes ‘strongly agree’.

3.3. Measurement Scales

The measurement scales were adapted from existing literature; for example, the attitude measurement scale was adapted from Krosnick, Judd, and Wittenbrink (1975) and Almutairi (2020), subjective norms and perceived behavioral control measurement scales were adapted from Shah and Bahdur (2017). Managers’ PPDM scale was adapted from Parnell and Crandall (2001). The measurement scales are adapted from previous studies, which demand to ensure reliability and validity. There were six demographic variables besides the interval/ratio scale, such as gender, age, marital status, service experience, appointment level, and educational qualification. The validity was achieved by conducting a content validation index test, and reliability was measured after the pilot test using Cronbach’s Alpha (α).

The researcher has consulted six experts for content validity test (three academicians and three senior managers from the industry). The experts were requested to mark the items (the items corresponding to each latent construct) in a four-point relevant scale such as 1-not relevant, 2-somehow relevant, 3-relevant, and 4-perfectly relevant. There are two forms of validation. Item-wise CVI (I-CVI) was calculated for measuring items validation and scale-wise validation (S-CVI) for measuring the entire construct (Yusoff, 2019). The calculation for I-CVI and S-CVI suggested by Yusoff (2019) was consulted for the calculation. The results are shown in Table 1. Due to space and word limitation, only S-CVI is shown. All four constructs were retained since they have enough S-CVI values (S-CVI values > 0.83). The researcher conducted a pilot test using purposive sampling technique (n-65) from industries to confirm the scale’s reliability. The subscales achieved Cronbach’s Alpha above 0.7, indicating that the constructs measure what is intended to measure.

Table 1: CVI Results in Brief

3.4. Data Analysis Procedure

Data were analyzed using Statistical Package for Social Science (SPSS) for demographic variables and Partial Least Squared Equation Modelling (PLS-SEM) for analyzing the model and hypothesis testing. PLS-SEM is a variance-based analysis best suitable for testing a research model when the researchers need to test the cause-and-effect relationship of exogenous latent constructs with the predicted variable. SmartPLS is a 2^nd generation statistical analysis software used by the researcher for testing the SEM model and hypothetical relations.

4. Results

Acquired data were carefully inserted into the SPSS program; 11 respondents did not disclose their marital status, there was no issue of non-response biases since the response was made within three months and the response rate is above 50%. Mahalanobis distance test was done to check for potential outliers; the critical value (43.650) with 30 degrees of freedom was less than the threshold (43.770), indicating the data is free from outliers. The researcher assured the respondent’s extreme confidentiality and carried out Harmen’s single factor test to ensure that the data set is free from common method bias. A single item variance explained less than 50% variance, ensuring that the data is free from common method bias.

4.1. Respondents Profile

The number of male respondents (n-365) is more than their female counterparts (referring to Table 2); this result is not surprising since previous researchers have shown that female participation in a Bangladeshi business organisation is comparatively much lower than in other Asian countries (Akhter et al., 2019). The majority of respondents are from 30 and 40 years old, have 6 to 10 years of service experience and works at the operational level. Such middle-aged managers’ opinion is crucial for any behavioral study’s success since they remain in the business organization’s driving seat.

Table 2: Demographic Profile

4.2. Partial Least Squared – Structural Equation Model Assessment

Before carrying out the predictive analysis, the data set needs to be tested, primarily when a model is studied empirically (Hair et al., 2019). The SEM model is a reflective model where arrows point toward the items meaning that the indicators of the construct are considered to be caused by that construct. The measurement model is formed by the indicators corresponding to their latent construct, and the structural model is the theoretical underpinning regarding how the constructs are related to each other.

4.2.1. Confirmatory Composite Analysis (CCA)

Confirmatory factor analysis (CFA) has traditionally been used to develop and measure constructs reflecting the domain model; however, researchers conducting empirical causal relationship study using PLS-SEM may conduct CCA before the final model test. Hair, Howard, and Nitzl (2020), while introducing CCA, have proposed to conduct a few analyses using SmartPLS software such as 1) estimate of indicators outer loadings; 2) indicators cross loading; 3) composite reliability as per construct; 4) average variance extracted (AVE); 5) discriminant validity – HTMT; and 6) predictive validity.

All indicators of outer loadings are found above the threshold (0.708) except two items under Attitude for PDM construct with extreme loaded low. These two items were deleted from the model since the researchers have another four items to evaluate (Hair et al., 2019). Table 3 presents the construct reliability and validity test report. The output shows that the data set achieved the minimum recommended threshold.

Table 3: Construct Reliability and Validity

Discriminant validity is achieved by conducting the Hetrotrait-monotrait test (HTMT), where all the values were below the recommended threshold (<0.95). CCA analysis, when meeting all the required thresholds, set out the departure point for further test. Hair et al. (2018) also suggested carrying out a model assessment considering the coefficients of determination (r²). The data set achieved an r² value of 0.682, indicating that the exogenous latent constructs (Attitude for PDM, subjective norms and perceived behavioral control) together explain 68.2% variance on the dependent variable.

4.2.2. Predictive Model Assessment (Using PLS Predict Report)

Shmueli et al. (2019) have suggested that a research model’s predictive ability should be assessed as part of every research project. Researchers’ capacity to make falsifiable predictions about recent findings is used to test hypo- theses and their analyses’ functional validity. This narrow emphasis on metrics for evaluating a model’s explanatory power is problematic since the best predictive model may vary from the best explanatory model. PLS prediction focused on the principles of separate training and holdout samples for estimating model parameters and assessing a model’s predictive capacity. Examine the Q² predict value for each predictor from the PLS-SEM study. A negative Q² predict value means that the model is not very good at predicting the future. The PLS predict report is presented in Table 4.

Table 4: PLS Predict Report

According to Shmueli et al. (2019), when the majority of the PLS Mean Absolute Error (MAE) yields smaller prediction errors compared to the linear regression model (LM), this indicates a medium predictive power. As we can see in Table 4, most items (PLS- SME) have lower error than the LM model, except PPDM1 and PPDM3 (number of folds –20, number of repetitions – 20); therefore, we can say that the three exogenous latent constructs have moderate prediction power.

4.3. Hypothesis Testing

Hypotheses were tested using the PLS bootstrapping procedure (5000 subsamples). The bootstrapping report showing path coefficient, t-statistics and p-value are presented in Table 5. H1 posited that managers’ attitude toward PDM might have a significant effect on his/her PPDM. The bootstrapping result shown in Table 5 indicates that the result failed to reject the null hypothesis since the path coefficient (β) is 0.242, t = –5.103 and p = 0.000 (p < 0.05), which is statistically significant. Therefore, the study finds support for the alternative hypothesis (H1 – Supported). Independent variable subjective norms affect the dependent variable (managers’ PPDM) since β = 0.184, t = 2.651 and p = 0.000. The study rejects the null hypothesis in favor of the alternative hypothesis. Therefore, the study confirms that subjective norms significantly affect managers’ PPDM (H2 – Supported). H₃ was posited that perceived behavioral control might have a significant effect on managers’ PPDM. The result supports the hypothesis since the result shown in Table 5 indicates that β = 0.476, t = 6.091, while the p = 0.000. The study rejects the null hypothesis in favor of the alternative hypothesis. Therefore, the study confirms that perceived behavioral control positively impacts managers’ PPDM (H3 – Supported).

Table 5: PLS Bootstrapping Report for Hypothesis Testing

5. Discussion and Conclusion

This study aimed to test the effect of some defined set of micro-level variables (attitude for PDM, subjective norms and perceived behavioral control) on managers’ PPDM. The study addresses a real-life problem in Bangladeshi manufacturing industries where entrepreneurs are faced with numerous challenges for managing people at work. Manager’s behavioral attitude and perception were measured following the assumption of TPB. The data gathered from 384 respondents were found flawless and met all the assumptions for empirical study. All three hypotheses have a significant positive influence on managers’ propensity.

The result revealed that business entrepreneurs at Bangladeshi textile industries should revisit their relevant policy and SOP for PDM. Managers’ attitude toward PDM was found to have a significant positive impact on their PPDM. They did not deny the benefit of PDM, but do not pose the required proclivity for PDM (Azim et al., 2017; Islam et al., 2018). Employees desire social, political, and official compulsion (together termed as subjective norms) to influence managers’ PPDM. Managers have opined that PDM knowledge is significantly poor; they find communication difficulties with the employees. The research has a significant theoretical contribution. The study deepens our understanding of the forces biasing managers’ PPDM. The PPDM study usually is confined to the western developed and developing nations; the study is the first of its kind in the least developing country that tested managers PPDM with the assumptions of TPB (Parnell, 2010; Parnell et al., 2012). In the practical field, the entrepreneurs of manufacturing industries might use the tested model to understand if their managers at all level possess the required PPDM.

This research is not out of limitation. The researcher felt there is a necessity for a longitudinal study for finding the other factors such as honesty and integrity, locus of control and demographical variables. The study is only concerned with managers’ perceptions on PPDM; workers’ perceptions should have been given more attention in the study. The government regulatory body also influences managers’ PPDM, which was not considered in the study.

Future researchers may advance this study’s knowledge in the service sectors and other manufacturing industries such as cement, cold rod and shipbreaking industry. The decision-making model propagated by Jago and Vroom (1977) may be considered to assess managers’ PPDM. The future researcher might also consider the differences or similarity of PDM, collective bargaining agency and social dialogue in different organisational settings. Demographic variable such as manager’s service experience can be considered moderator, which might influence Bangladeshi managers’ behavioral pattern.