THE RELEVANCE AND ROBUSTNESS OF GILBERT'S BEHAVIORAL ENGINEERING MODEL IN AN EMERGING MARKET ENVIRONMENT: EMPIRICAL EVIDENCE FROM CHINA

INTRODUCTION

In the past decade, performance consultants and business professionals in China have adopted the human performance technology (HPT) model and conducted thousands of performance improvement projects across industries. Their efforts have generated significant outcomes, and they have won dozens of awards from ISPI and other domestic and international organizations over the years. Among the theories guiding their practice, Gilbert's behavioral engineering model (BEM) has been a popular instrument. Consultants and researchers in China have developed various tools and approaches utilizing this model to identify important factors affecting job performance and to offer solutions accordingly (Fu et al., 2022).

A strong belief shared by many Chinese professionals is that factors in the environment have more significant influences on human performance than those within the individual (Fu et al., 2013; Tan et al., 2016; Winter, 2018). Applying BEM's terminology, environmental or system factors refer to data, instrument, and incentive, whereas individual factors include knowledge, capacity, and motive (Binder, 1998; Chevalier, 2003; Gilbert, 1996). As shown in Table 1, it is believed that environmental factors, combining data, instruments, and incentives, account for 75% of influences on job performance. By contrast, individual factors of knowledge, capacity, and motives account for only 25% of influences on job performance. These numbers came from an empirical study conducted in the United States (Dean, 1997).

TABLE 1 Gilbert's 1978 Behavior Engineering Model and Dean (1997) Ratio

View Fullscreen

Relying on this theoretical notion, consultants and researchers in China created Jikong Mima, or Tech-Driven Code, an innovative product, to help frontline employees and their managers improve performance and achieve significant outcomes in a variety of industries (Fu et al., 2022). Over the years, this innovation has achieved remarkable outcomes in various industries across China, as organizations and individuals have been able to improve their job performances in relatively short periods of time without significant additional investments of resources. It has also helped frontier employees balance life with work. Because of these achievements, The Jikong Mima won the C. K. Prahalad Award for social innovation in 2022 from the Kaufman Center (http://www.kaufmanawards.org/). As a result, many consultants argue that these practices have generated sufficient evidence that the BEM works in China. However, there is a lack of theory-based empirical studies to demonstrate that this is indeed the case. Considering the importance and significance of the Chinese market to the domain of performance improvement, we designed this study in order to investigate the relevance and robustness of Gilbert's BEM in this emerging economy.

RESEARCH QUESTIONS

Because both the theoretical foundation of the BEM framework and the empirical evidence of its validity were primarily established in the Western context, U.S. business organizations to be exact, it is legitimate to ask whether the BEM is as relevant in a non-Western context, such as China, as in the United States. Specifically, it is worthwhile to understand the extent to which the ranking of the factors' importance will hold in a Chinese organization. This issue is related to both the relevance and robustness of the BEM framework in a non-Western context. In addition, it is interesting and valuable to explore whether the importance of these BEM factors varies across different industries, managerial positions, and demographic variables. Exploring these research questions will advance our understanding of the validity and boundary of the BEM framework and provide guidance for better practice in both Western and Chinese cultures.

The general purpose of this study is to explore the Chinese professionals' assessment of each component of Gilbert's BEM in terms of their perceived impact on job performance. Specifically, we investigate whether the responses from Chinese professionals support the BEM's theoretical predictions, and to what extent they differ from those of their Western counterparts. In addition, we explore the effects of demographic variables, such as age and gender, industries and job functions, and managerial and nonmanagerial positions. We posit the following hypotheses:

• Hypothesis 1: There is a difference in the ranking of the relative importance of the six categories of factors in Gilbert's BEM in terms of their perceived impact on job performance between the sample collected in China and those collected in the United States as evidenced by Gilbert (1996), Dean (1997), and Stolovitch and Keeps (2004).

• Hypothesis 2: There is a difference in the ranking of the relative importance of the six categories of factors in Gilbert's BEM among different industries.

• Hypothesis 3: There is a difference in the ranking of the relative importance of the six categories of factors in Gilbert's BEM between managerial and nonmanagerial positions.

• Hypothesis 4: There is a difference in the ranking of the relative importance of the six categories of factors in Gilbert's BEM among different demographic variables.

LITERATURE REVIEW

While this study was conducted to determine whether using the BEM in China had the same validity as using it outside of China, it is necessary to know the underlying purpose of using the BEM at all as part of a performance improvement initiative. Moreover, it is important to understand why Gilbert developed the BEM and how employing it as intended increases the credibility and influence of performance improvement professionals with senior organization leaders. Gilbert positioned the BEM as one of two variables of a benefit–cost ratio equation that resonates with executives and senior management. Using the BEM outside the context of this formula dilutes its relevance for improving organizational performance using measures that matter to the executives and senior managers. The BEM, used properly, is about the economics of leveraging organization and human capital resources to engineer worthy performance. Once this is understood, the implications of this study and other studies related to the BEM become more significant. This section provides a high-level, general overview of Gilbert's worthy performance work associated with the BEM to provide the theoretical foundation for this study.

Engineering Worthy Performance Behavior Equation

The BEM shifts the focus of engineering worthy performance from accomplishments (A) to behavior (B). Remember, behavior is a cost in Gilbert's worthy performance equation of W = A/B. Gilbert (1996) bisected costly behavior (B) into two equally important tiers: a person's repertory of behavior (P) and a supporting environment (E) that “together form a transaction we call behavior” (p. 80). Gilbert further asserted that both P and E influence, and are influenced by, the management system (M) that decides how to manage both the P and E variables to “alter performance through changes in behavior” (p. 83). The M used to influence P and E is also a cost associated with behaviors. Therefore, for the W = A/B equation, this means focusing on the denominator, B = P × E × M. This means that the BEM refines the original definition of human competence (W = A/B) into a more detailed equation of W = A/(P × E × M). The HPT works with management to use the BEM as a diagnostic tool of six variables that management leverages to influence behavior (B) within the larger worthy performance equation of W = A/B.

Gilbert also recommended a sequence for diagnosing the causes of behavior deficiencies using the BEM. This sequence was based on their influence on behaviors. It is important to note that Gilbert (1996) stated that this sequence “does not necessarily indicate the best strategies for correcting the problem” (p. 92). The original sequential BEM shown in Table 2 identifies the six components of behavior and includes the recommended diagnostic sequence.

TABLE 2 Gilbert's 1978 Behavior Engineering Model

View Fullscreen

Gilbert's original BEM is still effective and commonly used, especially after Dean's (1997) study substantiated Gilbert's influence rankings. However, two other renditions of the original model have since been created to simplify Gilbert's BEM language and make it easier to use. They are shown in Table 3 and Table 4 for comparison. These tables include how the author of each model prioritized the influence of the six factors.

TABLE 3 Binder's 1998 Six Boxes Model of Behavior Influences

View Fullscreen

TABLE 4 Chevalier's 2003 Updated Behavior Engineering Model

View Fullscreen

In 1998, on the 20th anniversary of the BEM, Binder introduced an excellent and useful update called the Six Boxes model of behavior influences (see Table 3). Binder also simplified the terms used for the categories and the six variables that influence behavior to make them easier to explain and understand. The BEM name was changed to Six Boxes because clients did not understand what was meant by behavior engineering and found Six Boxes easier to remember. Binder's model mirrors Gilbert's diagnostic sequence.

Five years after Binder wrote about the Six Boxes model, Chevalier (2003) introduced another BEM update that did not change its name but did integrate terms used by Gilbert and Binder (see Table 4). What was unique about Chevalier's model was the diagnostic sequence that reversed the order in the individual category behavior influences. Chevalier argued that the costs to influence performance results increased along a continuum of information, recourses, incentives, motives, capacity, and knowledge/skills. The impact of these behavior influences occurred in the reverse order.

To continue this BEM simplification trend, Chevalier defined the B = E × P × M equation variable E as environment and P as person, rather than the original environmental support and person's repertory of behavior. M remained management system.

The BEM is not a performance improvement diagnostic tool. The BEM is a behavior cost-focused performance improvement diagnostic tool. The HPT uses it to add value during management's effort to learn how to leverage these P and E variables that influence competent human behaviors. Gilbert (1996) put it this way, “For the performance engineer, the question is: ‘Where can we get the greatest leverage; how can we have the biggest effect at the least cost?'” (p. 85). There are opportunities for improvement at both the P and E tiers. Where is the greatest leverage to improve the behaviors necessary to improve the desired accomplishments at the lowest cost? Research related to the BEM identified in this article indicates that environment has the greatest impact on influencing behaviors as either enablers or barriers.

In sum, Gilbert (1996) stated that the BEM has one purpose:

It helps us to observe behavior in an orderly fashion, and to ask the “obvious” questions (the ones we so often forget to ask) toward the single end of improving human competence. Indeed, its purpose is to put behavior into some perspective, with reference to what we might do to engineer superior performance. (p. 95)

These BEM data establish the critical W = A/B denominator price tag (B = P × E × M) for determining the worthiest performance improvement solution to achieve a valued accomplishment.

Empirical studies have been conducted over the decades to test the validity of the BEM model since Gilbert published his seminal work. The findings overwhelmingly support the ranking of the sequence of factors prescribed in the BEM with slight variations. Many of these studies follow the example set by Dean's experiment (1997) in which researchers asked participants to identify the factors enabling them to do their jobs better. The participants' responses showed that environmental factors, especially information and data, were far more influential than individual factors, such as knowledge. Following Dean's (1997) approach, Stolovitch and Keeps (2004) found almost identical results. Cox et al. (2006) developed a 12-item research instrument based on the BEM, surveyed 119 MBA candidates with full-time job experience, and found results that support the concept of leverage espoused by Gilbert, Rummler, and Deming. We report their results in Table 5.

TABLE 5 Comparison of BEM Factor Ranking with U.S. Empirical Studies

View Fullscreen

Similar results were identified in other empirical studies and also supported the importance of environmental forces in the ranking of factors influencing job performance (e.g., Duman et al., 2011). A meta-analysis summarizing findings from 70 performance technology analysis projects conducted between 1986 and 2012 further affirmed the notion that most job performance problems can be attributed to organizational factors (Hartt et al., 2016). The majority of these empirical studies were conducted in the United States with few exceptions. One of these was a study conducted by Stull and Freer (2019) using data collected from a bank in Afghanistan. We report these results in Table 6.

TABLE 6 Comparison of BEM Factor Ranking with Meta-Analysis and Emerging Economy

View Fullscreen

As discussed in previous sections, no empirical study has ever been conducted in the Chinese context. It is therefore imperative to close this research gap. Considering the prolific applications of Gilbert's BEM in China, we feel obligated to explore the validity and robustness of the model in this important emerging market environment. Empirical results of this effort will shed light on the applications of the BEM across the world and provide sorely needed guidance to practitioners and scholars.

METHODOLOGY AND DATA

We used survey research to explore the relevance and robustness of the BEM in a Chinese context. The study was conducted in 2018 and utilized an online survey format. One coauthor in China sent out 1,000 invitations randomly selected from a consulting firm's client database, and 423 business professionals representing different industries responded and participated in the study and provided their responses to the online questionnaire.

Research Instrumentation

The questionnaire consisted of three parts. Part I introduced the purpose of the study as “to have a better understanding of factors influencing job performance in different industries” and invited participants to share their points of view based on their own experience. Part I made clear that all questions in the questionnaire focused on the participants' own job, instead of other people's jobs, and there were no correct or wrong answers. We then asked participants to answer questions related to their industries, the size and nature of their organizations, their jobs, and whether they were in managerial positions. In Part II of the questionnaire, we asked participants to think about their own jobs and choose one factor from the six following areas that “would enable them to do their job better.” Following the examples in the literature (Dean, 1997), we asked participants to choose one and only one factor. The six options were adapted from Dean (1997) and read as follows:

• Clear performance expectations and relevant feedback about the adequacy of your performance (Information)

• Tools, resources, and materials to achieve your performance goals (Resources)

• Adequate pay and nonpay incentives made contingent upon your performance (Incentives)

• Systematically designed training that matches the requirements of your job (Knowledge)

• A match between your skills and the requirements of your job (Capacity); or

• Assurance of job security and social acceptance (Motivation)

In Part III of the questionnaire, we asked a series of questions related to participants' demographic variables, including work experience, position tenure, gender, age, and education. Their demographic variables are displayed in Table 7. As shown in the table, most participants had undergraduate or graduate degrees, and most of them were between 26 and 40 years old. More than 90% of them had at least 3 years of job experience, and more than half of them had at least 3 years of experience in their current positions.

TABLE 7 Participant Characteristics

View Fullscreen

DISCUSSION AND IMPLICATIONS

The results of this research study indicate that using the BEM as a diagnostic HPT tool works equally as well in China as it does in other industrialized countries. The study results also suggest that in China belief about the influence of environmental factors on job performance is understated by 9.7% (i.e., belief = 75% and study findings = 84.7%). This finding is more in line with the 80/20 rule advocated by economist Pareto (1964) and the 85/15 rule used by quality pioneers Juran (1988) and Deming (1982).

The data indicate that the null hypothesis for the first of the five research questions was not supported. Therefore, the first research question, that the rankings of the relative importance of Gilbert's six job performance influence factors in China would differ from those in the United States, was supported with statistically significant data. This study indicates that for the environmental factors, data was ranked as the top factor by almost half of the respondents. For the individual factors, capacity was rated the top factor in China but at about the same rate as in the U.S. empirical studies. Knowledge and motivation were rated lower than the U.S. empirical studies, which is expected given that the individual category factors accounted for only 15% of the BEM factor rankings. This capacity ranking was 0.7% higher than the ranking given to the incentive factor in the environment category. Most other studies rate this factor second and knowledge as first.

The notable exception for both environmental and individual factor rankings is the study conducted in Afghanistan, which showed that there was not the same level of organizational infrastructure support or a work force with the same education levels as industrialized countries.

The data findings did not support the next two research questions. The null hypothesis for both was not supported. The data indicated that these rankings did not differ among different industries or between managers and nonmanagers in China.

The final research question, whether the rankings of the six categories would differ among different demographic variables, was partially supported. The null hypothesis was true for all demographic variables except one—job experience (F = 2.28; p < .05). In general, the longer the job experience participants have, the more likely they find environmental factors to be influential. For example, 88.6% of participants who had more than 10 years of experience chose the three environmental factors (n = 210). By contrast, 82.2% of participants who had less than 10 years of experience chose the three environmental factors (n = 213). Interestingly, although job experience and job tenure are highly correlated, age itself does not have significant effect (F = 1.61; ns). These results may suggest that people who have longer job experience are more likely to realize the importance of environmental forces in their organizations than those who have less job experience.

The implications of these findings are good news for HPT practitioners in China and elsewhere. This good news is especially true for organizations that are not familiar with HPT or the BEM, where the impact of economic sensitivities and quality on accountability matters to the decision-makers. Why? Because it is human nature to resist change or be skeptical of something that is unfamiliar. To overcome resistance and skepticism, it is helpful to associate the BEM findings of this study with language already accepted and used by managers and consultants. A common persuasive communication strategy is to start with the known and then move to the unknown.

The first implication is that this study provides research evidence that the HPT professional can employ known and accepted ratios used by decision-makers to identify and prioritize actions in order to optimize organizational results economically. Pareto's results/action (80/20) ratio can guide how to use the rankings of this study, which identifies the organizational elements as the likely root causes of 80% of organizational performance inefficiencies or deficiencies.

The same is true for the 85/15 rules used by Juran and Deming, who classified Gilbert's organization/individual dichotomy as the vital few/trivial or useful many and the common cause and a particular cause, respectively. Juran and Deming's common cause includes all factors related to system and management.

Executives, managers, and consultants accept Pareto's economic ratio as true. The same is true for Juran and Deming's quality ratios and categories. This is good news for the known-to-unknown strategy to communicate the implications of this study. Gilbert's environment/individual categories mirror Juran and Deming's quality categories. This study, including its comparative data from more than 70 research studies spanning 25-plus years, provides credible evidence that performance factors associated with Gilbert's BEM environmental category have greater influence (with economic implications) on performance than Gilbert's individual category. Furthermore, the results of this study support the known and accepted 80/20 and 85/15 rules.

The second implication is the focus on this study as defined by the five research questions—the prioritization of the six individual factors within the two categories. This is where the real value of this study can build on the known 80/20 and 85/15 ratios and categorizations. However, Juran and Deming did not include in their 85/15 ratio categories a set of defined variables associated with those categories. Gilbert's (2007) intention for using the BEM was “merely a way to organize empirical data” (p. 93). Gilbert also wrote that while the BEM “is a guide to locating the cause of a performance failure, it does not necessarily indicate the best strategies for correcting the problem” (p. 92).

The results of this study indicate that HPT practitioners should begin with environmental factors because almost 50% of respondents selected data as the most important BEM factor affecting human performance. Resources, another environmental factor, accounted for a little over 25% of the remaining influence. Of the remaining four factors, two were rated under 9% (capacity at 8.6% and incentive at 7.9%) and the other two (motivation and knowledge) under 4% at 3.5% and 3.2%, respectively.

What this means to HPT professionals in China is that this BEM study provides evidence for where it is best to begin the diagnostic process to, as Gilbert put it, locate the cause of performance failures. The order of troubleshooting performance problems using the BEM as a diagnostic tool closely follows the order originally used by Gilbert. For Gilbert, as reflected in Table 2, the first three in sequence are environmental (data, resources, incentives) and the next three in sequence are individual (knowledge, capacity, and motives). For China, the individual capacity factor moves up to number three, where it becomes part of the environment category. In addition, the sequence of the knowledge and motivation factors is reversed. This means, as the results of this study indicate, that HPT practitioners in China use the following diagnostic sequence factors—data, resources, capacity, incentives, motivation, and knowledge.

The implications of this study are significant for managers and consultants using Gilbert's BEM in China. First, the BEM can leverage the accepted language of quality improvement. The research findings help the HPT practitioner use the language of Juran and Deming to communicate evidence justifying how they focus their time and attention on finding the vital few and particular causes over the trivial or useful many and common causes. Second, it is important to remind stakeholders that the BEM represents the costs in Gilbert's worthy performance BCR formula (i.e., W = A/B). The focus is on the economics of reducing or avoiding costs to optimize valued benefits resulting from improved workforce performance. The BEM is part of Gilbert's process to engineer worthy performance. Its purpose is “fundamentally economic” (Gilbert, 1996, p. 11).

RECOMMENDATION

This study collected data in 2018, before the pandemic that dramatically changed the world. It provides specific research evidence that held true in China at the time of the study. Likewise, all other earlier studies referenced in this study were conducted under conditions that no longer exist at the time of this writing. Repeating this research study in the postpandemic world of work in China, the United States, and other countries could reveal new prioritization realities for HPT professionals to leverage and navigate and to remain relevant and performance improvement leaders.