Image: Ingredients for happiness at work: Arch Woodside

Ingredients for happiness at work

Image: Arch WoodsideArch Woodside

Professor Arch Woodside is the author and co-author of more than 400 journal articles in over 40 different social sciences citation index journals, as well as author and editor of more than 50 books.

He is a Fellow of the American Psychological Association, Association of Psychological Science, Royal Society of Canada, Society of Marketing Advances, International Academy for the Study of Tourism, and the Global Innovation and Knowledge Academy. 

He has a PhD in Business Administration from Pennsylvania State University (1968) and a Doctor Honoris Causa from the University of Montreal (2013).


Articles by Arch Woodside

Case study research methods for theory building

Published in Journal of Business & Industrial Marketing

View the Journal of Business & Industrial Marketing Table of Contents.

See more titles like this one in our Marketing eJournals collection.


Related articles

The impact of corporate social responsibility on employee performance and cost - Review of Accounting and Finance

Measuring the effects of quality certification on labour productivity: An analysis of the hospitality sectorInternational Journal of Contemporary Hospitality Management

Human resource management and organisational productivity: A systems approach based empirical analysisJournal of Organizational Effectiveness: People and Performance


Exclusive blogs

The Importance of Collaboration - Howard Thomas

The importance of customer loyalty, trust and advocacy - John Bowen

Exploring the issues of gender and stereotyping in marketing - Victoria Crittenden

The evolving role of the librarian - David Shumaker

The future of teaching cases in the classroom - Gina Vega

Professor Arch Woodside has conducted illuminating research into employee productivity. He discusses how his findings indicate the need for employers to cherry-pick best practice from different countries if they wish to recruit and retain good workers.


How did you come to be involved with hospitality?

I became involved with hospitality when some of my colleagues in Taiwan, who focus on hospitality studies, asked me to give them training on how to look at recipes – or, in other words, algorithms.

To give an example of recipes, in the US, we currently have a problem in that certain segments of the population only work in low-paid service jobs – at McDonald's, for example – while at the same time living off welfare. The government actually supports the profitability of companies like McDonald's, a part of the system that we need to change. Somehow, we must provide work as well as support or a social network for low-paid individuals, because this is the right thing to do.

We are able to identify recipes that are highly predictive of happy and unhappy employees and high and low performance. It is the recipe, not the individual variables, that is actually highly predictive.

With these recipes, what are the applications of your work?

The applications relate to hiring practices. For instance, should you hire young or old workers? There is not enough information to base these decisions on age, because older workers do well in certain recipes while younger workers do well in others. Instead, it is important to look at the combinations of the recipes themselves. For example, people with low levels of education and who are unmarried with grown children may turn out to be excellent workers by both being happy in their work environment and being high performers on the job. Recipes also show that older males are usually high performers as long as they have a high education level and no children living at home.

Of course, it is not ethically correct for employers to impose too many restrictions on who they choose to hire; moreover, they would soon run out of potential employees if they did. Rather, it is more important to provide certain workers with specific benefits, such as extra leave or flexible schedules.

What is service-dominant logic?

In the US, we used to be a manufacturing society. However, many industries with which we are in contact today are service industries, such as education, banking, retail and hotel administration. The recipes that work for service industries are often very different to recipes that relate to product manufacturing. Indeed, service-dominant logic means that the customer is far more actively involved in designing a delivered service compared to a delivered product.

Is there a finding from your research that you consider particularly significant?

Our most important finding is that happiness alone is not a significant indicator of high worker performance. While it is true that there is an overall positive effect association between happiness and high performance, this is not predictive enough because there are a number of people who are happy but do not do much at work. Conversely, some people are unhappy at work and yet are extremely high performers.

It is important to look at configurations of factors such as happiness and the characteristics of the work environment, as well as the demographics of the person in question. For example, which combinations of happiness and demographic factors – such as age, gender, education level and presence of children – result in high work performance? Similarly, what combinations of unhappiness with the work environment and demographic features result in high performance?

How do you see your research translating into practices?

Because performance is so important, my fear is that employers will use this type of research to develop algorithms that screen people out of the work environment. Indeed, this analysis can equally be used for good or bad practices in human resources management. Yet my hope is that companies would recognize that certain profiles of employees need extra support, which is worthwhile to provide and generates extremely loyal workers. There are certain examples we can learn from, particularly in nations such as Denmark, Finland and Canada.



How to deal with a complex problem

Researchers at the Carroll School of Management at Boston College show how management theory modelling that specifically addresses context and dynamics is more useful than simplistic and outdated survey methods.

Large organizations are inherently complex and difficult to manage, with company executives constantly seeking to boost productivity and implement more streamlined processes. Over the years, management theorists and consultants have introduced numerous techniques for changing company culture, management style and decision making processes.

In the 1980s, structural equation modelling (SEM) became the standard statistical technique for testing theories about complex, unstructured organization problems. SEM, which incorporates multiple regression analysis (MRA), remains the primary method for management research today. It typically involves analysing question-and-answer surveys focused on measuring and modelling latent constructs – namely, abstract psychological concepts – against five- or seven-point scales.

Yet according to Arch Woodside, Professor of Marketing at the Carroll School of Management at Boston College, Massachusetts, the continuing widespread use of SEM/MRA is flawed. Through examining the net effects of individual variables, it fails to account for the combined effects of multiple variables.

An outdated method

In a recent paper, Woodside argues that the fundamental problem of SEM/MRA is abstraction, which introduces distance and removes the key factor for meaningful analysis: context. It also lends itself to researcher bias due to the need to interpret survey results. Additionally, although data fit improves with complexity, the paramount requirement – namely, predictive accuracy – reduces.

Asymmetric testing using complexity theory delivers far more useful results, although there is still a need for more rigour in the development of complexity theory. To this end, Woodside recently conducted several studies in the context of the hospitality industry using fuzzy set qualitative comparative analysis, multiple-variable scenarios and contrarian case analysis. These studies explored the five central complexity theory tenets, covering employee happiness at work, supermarket buying committee decision making processes, and beauty and health spa customer satisfaction.

Testing complexity theory

The first tenet of complexity theory states that although a simple antecedent condition may be necessary, it is rarely sufficient for predicting a high or low score in an outcome condition. For instance, while it is necessary to be male to join an all-male professional football team, simply being male does not guarantee admittance. The second tenet is the "recipe principle"; an antecedent condition comprising two or more simple conditions is sufficient to obtain a consistently high score in an outcome condition. This is the main principle that guided Woodside's study of employee happiness.

Third, the equifinality principle states that even if an antecedent condition is not met, a high score outcome can occur; the model does not have to be necessary. Fourth, the causal asymmetry principle states that recipes which indicate a certain outcome are unique and should not be seen as polar opposites of recipes with another outcome. For example, studying the causes of success in obtaining a job offer reveals little about what causes failure, thus separate asymmetric models are necessary to explore the gap between the two outcomes. Finally, the fifth tenet is that an individual ingredient – a feature, an attribute or action in a recipe – can contribute positively or negatively to a specific outcome, depending on the presence or absence of other ingredients. Thus a variable that results in a positive effect can similarly result in a negative effect.

Importantly, the causal asymmetry and recipe principles point to the inadequacy of using the SEM approach to explain and describe success. As the fifth principle demonstrates, a specific attribute can contribute positively or negatively to an outcome, depending on how it is configured with other attributes.

Modelling multiple realities

Woodside operates from the premise that the research proposition should be expressed parsimoniously, with simple descriptions, explanations and predictions regarding patterns. Identifying the patterns guides the removal of redundancy – and the networks of interactions between constructs are variable, while processes are iterative and change occurs frequently.

For instance, in studying customer evaluations of visits to beauty parlours and health clubs, Woodside and his team derived examples of models with equifinal positive results. Using complex recipes to model the attributes of gender, age, education and occupation, certain models obtained consistently high scores in the outcomes of perceptions of pleasure, service quality, treatment effectiveness and value for money. The study showed that the main effect of a lone attribute is insufficient for determining the impact of that specific attribute on an outcome. However, it also showed that one attribute, high education, was a necessary condition for the perception that treatment was effective and that service quality was high.

Contrarian case analysis

Unfortunately, most studies that aim to improve organization management ignore the most unpredictable and dynamic organizational element of all: the employee. Prevailing wisdom maintains that the happier the employee is with their work, the more productive they are. Woodside explored the truth of this in a recent study, whereby he investigated hospitality industry employees' perceptions of their work environments and overall happiness with their jobs. Following quintile analysis of the employees' evaluations of their jobs and their supervisors' assessments of their in-role work performance (IRP), he then examined the relationships between two or more constructs in each quintile.

More than 25 per cent of the cases exhibited two relationships, showing that happy employees are not necessarily productive and unhappy employees are not necessarily unproductive; revealingly, an employee's intrinsic state of happiness alone was insufficient or unnecessary for low or high IRP. However, high IRP was sufficient for predicting a strong probability that the employee would go beyond the call of duty to ensure high customer satisfaction: "There are actually four types of people: happy people who are highly productive; happy people who are unproductive; unhappy people with high productivity; and unhappy people with low productivity. We have to recognize and understand the recipes involved with all four types of people," observes Woodside. He also points out that large datasets will almost always include contrarian cases – and the reasons for these should be examined and explained.

Woodside recommends that employers use his research simply as a guide: "To enable people in different recipes to perform highly and enjoy some degree of happiness at work, it may be necessary to build systems that complement and support the home environment," he concludes.


Tourism and marketing


Objectives

  • To decisively solve the confusion in the literature on worker happiness and on-the-job performances.
  • To answer the question "when does high happiness as well as low happiness associate with high performance?" by looking at recipes (profiles) of workers rather than happiness as a variable.


Key collaborators

  • James Po-Hsun Hsiao, National Taiwan University of Physical Education and Sport, Taiwan
  • Chyi Jaw, National Yunlin University of Science and Technology, Taiwan
  • Tzung-Cheng Huan, National Chiayi University, Taiwan