Who commits? Who engages?

John Sutherland (Scottish Centre for Employment Research, Department of Human Resource Management, Strathclyde Business School, University of Strathclyde, Glasgow, UK)

Employee Relations

ISSN: 0142-5455

Publication date: 2 January 2018

Abstract

Purpose

The purpose of this paper is to address two questions: who commits? And who engages? For example, does an individual’s likelihood of committing/engaging vary with his/her age; or with the level of his/her qualifications; or with his/her occupation? Of what consequences are the characteristics of the workplace at which the individual is employed?

Design/methodology/approach

The investigation uses the Skills and Employment Surveys Series Data set to construct the indicators of commitment and engagement. Using an ordered-logit model and an OLS model, these indicators are analysed to identify their covariates.

Findings

Who commits and who engages depends upon the indicator used to measure the attitude/behaviour in question. Changing these indicators sometimes means that an individual no longer commits/engages. Further, even for the same indicator of commitment/engagement, who commits/engages varies across individuals.

Research limitations/implications

The indicators of commitment and engagement examined are derived from the responses in a pre-existing data set which has its origins in survey instruments which had quite comprehensive terms of reference. Owning to the cross-sectional nature of this data set and the statistical methodology applied, the statistical results are correlations between some possible indicators of commitment and engagement and some variables which denote the personal characteristics of individuals and the characteristics of the organisations with which they are employed. Causation cannot be inferred from these correlations.

Originality/value

Commitment and engagement are central to many models of the management of human resources. However, the likelihood that an individual commits and/or engages differs across the workforce has rarely been examined. This paper addresses this research lacuna using a data set which is rich in detail about an individual’s personal characteristics.

Keywords

Citation

Sutherland, J. (2018), "Who commits? Who engages?", Employee Relations, Vol. 40 No. 1, pp. 23-42. https://doi.org/10.1108/ER-02-2016-0033

Download as .RIS

Publisher

:

Emerald Publishing Limited

Copyright © 2018, Emerald Publishing Limited


1. Introduction

Commitment and engagement are central constructs in many prescriptive models of the management of human resources that detail policies by which management may both enhance worker well-being and improve organisational performance. For example, following the insider econometrics studies of high performance work systems (HPWS) in the USA, commitment was introduced as one possible transmission mechanism to explain why the implementation of such policies resulted in improved outcomes. More recently, if more especially in the UK, engagement has become an important focus for policy makers, subsequent to the publication of a series of case studies demonstrating how the introduction of enablers (or drivers) of engagement resulted in improvements in organisational performance. However, the possibility that the likelihood of committing or engaging may differ across individuals in the workforce, for example according to their personal characteristics and the characteristics of the organisation with which they are employed, has rarely been examined. This paper addresses this research lacuna.

The research investigation uses data extracted from the Skills and Employment Surveys Series Data set to construct the indicators of commitment and engagement. It applies an ordered logit model and an OLS model to identify the covariates of these indicators. The data set has its origins in a series of surveys designed to examine the employed workforce in Great Britain. Consequently, Great Britain is the focus of the empirical component of the paper.

2. A context from some literature of relevance to the empirical study

There are many diverse models of the management of human resources that detail the policies and practices by which management may fully realise labour’s capacity to produce, and thereby simultaneously improve the performance of the organisation and enhance the well-being of the worker. Employee commitment to and employee engagement with organisations are central constructs in many of these models. The specific roles played by these constructs in these models are many, varied and frequently contested. The aim of this section of the paper is to provide an informative context to the empirical investigation which follows. Therefore, it seeks neither to survey nor to evaluate the many issues in question. Often, it cites seminal references in preference to more contemporary ones. It is presented in two sections, one dealing with commitment, the other with engagement.

Commitment

Commitment in the early literature was conceptualised as a psychological contract between employee and employer, characterised by the former’s identification with the values and goals of the latter (Selznick, 1957; Kalleberg and Berg, 1987; Meyer and Allen, 1997). The construct was to become central to those models of the management of human resources that were influenced by writers associated with the neo-human relations school (e.g. Maslow, 1943; McGregor, 1960; Hertzberg, 1966) where it was viewed as a potentially important policy instrument. These models prescribed what Green (2006, p. 7) identifies as a “new ideology of control”, replacing traditional control structures reflected in, for example, compliance, hierarchy and bureaucracy (Legge, 2005). Articulated well in Storey’s (1989) frequently cited “soft version” of human resource management, these models advocated investments over the long term in human resources and the design and implementation of policies and practices that sought to motivate employees, gain their trust and thereby their commitment. The presumption was that “committed” employees were more likely to improve organisational performance.

Consequently, much within the relevant literature was about seeking to understand the conditions under which human resources became strategic assets capable of generating distinctive capabilities – and/or complementing other factors, notably technology to produce this outcome – and then devising and implementing the appropriate strategies. In principle, the exercise was one of building and developing a bundle of human and related technical resources, the latter manifest most especially in terms of work organisation, designed to enhance organisational performance (Boxall and Purcell, 2008). An unashamedly managerialist agenda dominated the mainstream literature that focussed upon identifying then prescribing “best practice” (Delbridge and Keenoy, 2010). As the high performance paradigm literature both in the USA and the UK was to illustrate, however, there was no definitive list or bundle of policies and practices appropriate to this task (e.g. Becker and Gerhard, 1996; Ichniowski et al., 1996; Wood, 1999; Procter, 2008).

The emergence of models reflecting the high performance paradigm was associated with empirical research published in the USA (e.g. Appelbaum and Batt, 1994; Huselid, 1995; MacDuffie, 1995). The principal focus of these studies was the attempt to establish a causal relationship between work practices – reflected in various sets of human resource (hereafter HR) policies and practices and various forms of work organisation – and organisational performance. Always fraught with problems pertaining to theory, methodology and data, this research sometimes did succeed in establishing positive statistical associations between HR policy and practice and organisational outcomes. Nonetheless, it was less successful in explaining how this relationship came about (Boselie et al., 2005).

Latterly, the research agenda was to change to one which sought to identify and examine the nature of the “transmission mechanisms” (Procter, 2008) – or “linkages” (Ramsay et al., 2000) – by which this relationship did come about. The conventional wisdom was that this relationship had its explanations in changes made to employees’ attitudes and behaviours by the policies in question. Two possibilities were mooted: one associated with “commitment” (Walton, 1985) and termed “high commitment management”; and the other associated with “involvement” (Lawler, 1986) and termed “high involvement management”. The former addressed employment practices and required a policy framework designed to commit (or re-commit) the worker with the cultural norms and expectations of the organisation. The latter addressed work practices and emphasised the salience of employee participation (Lansbury and Wailes, 2008). According to this latter perspective, the central assumption was that the implementation of policies and practices such as the creation of semi-autonomous work teams, the adoption of employee profit sharing schemes, etc., designed to create involvement, improved worker effort and, in turn, organisational performance. Irrespective of the transmission mechanism in question, however, and very much in accordance with the assumptions of the unitarist perspective of the organisation, the ultimate outcome was one of “mutual gains”. As well as the organisation improving its performance, employee well-being was enhanced (Kochan and Osterman, 1994).

Engagement

As with commitment, engagement is a contested construct. Bailey et al. (2017, p. 31) report a “bewildering multiplicity of definitions, measures, conceptualisations and theories of engagement”. Further, in some of the literature there is confusion as to whether the concept is an “attitude”, a “behaviour” or an “outcome” (Guest, 2013).

Of the variety of conceptual frameworks that exist, most were drawn from the literature of (organisational/work) psychology (Rich et al., 2010). The first to note was Kahn (1990, p. 702) who maintained that personal engagement occurred when “people bring in or leave out their personal selves during work-role performances”. When employees were engaged, they were assumed to express themselves cognitively, emotionally and physically. Engagement, therefore, was the converse of attitudes and behaviours such as alienation, apathy and detachment (May et al., 2004). Subsequently, there has been a shift away from Kahn’s original construct towards a more multidimensional perspective that seeks to distinguish between job engagement and organisational engagement (Saks, 2006), a change in focus that has consequences for an empirical component of this paper.

Although a job demands-resources framework has tended to dominate theoretical explanations of engagement, more recent developments make use of social exchange theory with its identification of reciprocity as a form of social exchange. This framework has featured in several important empirical studies of engagement (e.g. Saks, 2006; Shuck, 2011; Truss et al., 2013).

To the extent that its roots lie within positive psychology, the concept of employee engagement fits well with the unitarist perspective of the organisation (Watson, 2010). That said, an employee may be “engaged” but not necessarily behave in ways that benefit the organisation. Consequently, Arrowsmith and Parker (2013, p. 2697) maintain that “the pursuit of (employee engagement) is much more complex and dynamic, and the outcome uncertain”.

Engagement has proved to be one of the most significant concepts in management studies in recent years. Further, subsequent to and principally as a consequence of the publication of what is referred to as the “Macleod Report”, there has been a tendency for a consultancy approach to prevail (MacLeod and Clarke, 2009). Consultancy companies devise engagement surveys and HRD professionals are called upon to develop strategies designed to improve employee engagement (Keenoy, 2013; Purcell, 2014). Employees are viewed as passive actors within the system. Engagement, therefore, is assumed to be driven by the organisation and the search is for “drivers” of engagement (Robinson et al., 2004; Emmott, 2010; Francis and Reddington, 2012).

Research demonstrates associations between engagement and both individual and organisation performance. This research uses case study methodology and most of the case studies were undertaken by consultancies (Purcell et al., 2003). Within this research there are many studies that seek to uncover the mechanisms by which HR policy and practice may impact upon individual behaviours. Rich et al. (2010) argue that engagement is the core mechanism. Truss et al. (2013, p. 2661) contend that “Engagement demonstrates the potential to become the ‘new best practice’ HRM approach, with the prospect of ‘high engagement HRM’ becoming the dominant discourse within mainstream HRM”.

Many of the discourses on commitment and engagement – most especially those which employ a micro-econometric methodology – tend to assume that attitudes and behaviours are homogenous across individuals. Little attention is paid to the possibility that the likelihood of committing and/or engaging may differ between individuals, for example according to their personal characteristics and/or the characteristics of the organisations with which they are employed. If commitment and/or engagement are indeed heterogeneous across individuals, therefore, it is quite probable that the impact of a given set of HR policies and practices on individual and organisational outcomes also will be heterogeneous rather than homogeneous.

Accordingly, this paper addresses two questions: who commits? And who engages? For example, does an individual’s likelihood of committing/engaging vary with his/her age; or with the level of his/her qualifications; or with his/her occupation? Of what consequence are the characteristics of the organisation with which the individual is employed?

3. Research methodology

Commitment and engagement are complex social phenomena. Moreover, an individual’s propensity to commit and/or engage both influences and is influenced by equally complex work situations such as the nature of the job, management style and the culture of the workplace. Frequently, therefore, case study methodology – making use of research methods such as interview and observation – has been used to generate the rich qualitative data necessary to understand and explain the complexities of the manifold inter-relationships which exist between actors, processes and contexts (e.g. Jenkins and Delbridge, 2013; Dromey, 2014).

The principal advantage of case study methodology is its potential to address research questions such as “why”? Its principal disadvantage, however, is its limitations when addressing research questions such as “who?” Case study methodology does not make possible an examination of the nature of the statistical relationships between commitment and engagement and, for example, the characteristics of workers and the organisations with which they are employed, the ultimate justification of the construction of the survey instrument designed and applied by Mowday et al. (1979). Further, case study methodology does not lend itself to investigating trends over time in these concepts. To do this, survey methodology is the preferred methodological option. Consequently, there are many instances of “black box”-type models of commitment and engagement in which micro-econometrics is used to examine the survey data to analyse the possibility of relationships between “policy” and “performance” with commitment and engagement operating as the transfer mechanisms (e.g. Alfes et al., 2013; Bakker and Xanthopoulou, 2013). However, a feature of most of these studies is the (often implicit) assumption that commitment and engagement are homogeneous across individuals.

Survey methodology is not without its own potential disadvantages. For example, analysing cross-section data sets only produce statistical associations between the dependent and independent variables. Only the use of less frequently available panel data sets eliminate unobserved heterogeneity and makes the identification of causal relationships between the dependent and independent variables possible.

This research investigation uses survey methodology. The data set analysed has its origin in the Skills and Employment Surveys Series Data set, 1986, 1992, 1997, 2001, 2006 and 2012 (Felstead et al., 2014). The “Skills Surveys” is a series of surveys undertaken in 1997, 2001, 2006 and 2012 designed to examine the employed workforce in Great Britain (from 2006, the UK). Although with a more explicit skills focus, the series builds upon two previous studies, namely the Social Change and Economic Life Initiative Surveys, 1986-1987 and Employment in Britain, 1992. Each survey was administered to nationally representative samples of individuals aged 20-60 in employment as employees. The Skills and Employment Surveys Series Data set, 1986, 1992, 1997, 2001, 2006 and 2012 extracts data from the six original surveys where common questions are asked. The data set is created by retrospectively pooling selected cross-sectional data.

This research investigation, therefore, analyses responses from a pre-existing data set that has its origins in a survey instrument that had quite comprehensive terms of reference. It is data driven, hence atheoretical. It does not attempt to address – indeed cannot address – the manifold theories that relate to either commitment or engagement, nor seek to discriminate between them. The indicators of commitment and engagement that are constructed and examined, therefore, are inevitably constrained by the nature of the questions and statements posed in survey instruments originally devised for other purposes. Although there is precedence for the three indicators of commitment used in the earlier research of van Wanrooy et al. (2013), the same cannot be said for the four indicators of engagement. Indeed, some such as Saks (2006) may argue that the responses made to the quasi-thought experiment-type questions used to measure engagement may be more indicative of measures of commitment. Finally, and perhaps most importantly, there are no variables within what is an employee-based data set which makes possible an examination of how higher levels of commitment or engagement on the part of individuals may translate into higher levels of organisational performance.

4. Data

Questions pertaining to commitment and engagement appear in years 1997, 2001, 2006 and 2012 of the original data set. Observations from these four years, therefore, constitute the working data set.

Commitment is associated with an individual identifying with the goals and values of the organisation. In this investigation commitment is analysed initially using three indicators produced by examining responses to the following three statements, namely:

  1. “I feel little loyalty to this organisation” (where the name of the corresponding dependent variable in the analysis which follows is “loyal”).

  2. “I find that my values and the organisation’s values are very similar” (where the name of the corresponding dependent variable in the analysis is “values”).

  3. “I am proud to be working for this organisation” (where the name of the corresponding dependent variable in the analysis is “proud”).

There were four possible responses to these statements namely. “strongly agree”; “agree”; “disagree”; and “strongly disagree”. The original responses in the data set were re-coded where necessary to produce outcomes whereby the highest (lowest) degree of commitment was associated with the highest (lowest) number.

The percentage frequency distributions of responses to each statement are presented in Panel A-C of Table I. The response rate for voicing a positive commitment is (approximately) 75 per cent for each indicator.

To produce a more holistic perspective of commitment, the (re-coded where applicable) Likert scale responses were treated subsequently as scalar variables and aggregated across the three statements to produce a composite indicator (where the name of the resultant dependent variable in the analysis is “commitment”).

Engagement is associated with particular behaviours on the part of the employee, manifest, for example, in exercising discretionary effort (cf. Rees et al., 2013). However, rather than probing only an individual’s willingness to use her/his “own initiative”, in this research investigation engagement is probed further, if in a hypothetical manner. This is reflected in questions that relate to the willingness of an individual to increase the supply of effort and to make potential monetary sacrifices.

Engagement is investigated initially using four indicators which are produced by examining responses to one question and three statements namely.

  1. “How much effort do you put into your job beyond what is required?” (where the name of the corresponding dependent variable in the analysis which follows is “effort”).

  2. “I am willing to work harder than I have to in order to help this organisation succeed” (where the name of the corresponding dependent variable in the analysis is “help”).

  3. “I would take almost any job to keep working with this organisation” (where the name of the corresponding dependent variable in the analysis is “takeany”).

  4. “I would turn down another job with more pay to stay with this organisation” (where the name of the corresponding dependent variable in the analysis is “turndown”).

There were four possible responses to the question, namely “a lot”; “some”; “only a little”; and “none”. And there were four possible responses to the three statements: “strongly agree”; “agree”; “disagree”; and “strongly disagree”. Again, the original responses in the data set were re-coded where necessary to produce outcomes whereby the highest (lowest) degree of engagement is associated with the highest (lowest) number.

The percentage frequency distributions of responses to the question and each statement are presented in Panels A-D of Table II. Perhaps reflecting their diverse nature, marked differences are discernible. Whereas (approximately) 80 per cent of respondents would “engage” by increasing the supply of effort (i.e. the indicators “effort” and “help”), (approximately), 70 per cent would not do so if this was to entail making personal monetary sacrifices (i.e. the indicators “takeany” and “turndown”).

To produce a more holistic perspective of engagement, subsequently the (again recoded where applicable) Likert scale responses were treated as scalar variables and aggregated across the one question and three statements to produce a composite indicator (where the name of the resultant dependent variable in the analysis is “engagement”).

5. The models estimated

Two models were used to analyse the data set, namely an ordered logit model and an OLS regression model.

An ordered logit model is an econometric model designed to analyse the dependent variables that have ordered multinomial outcomes, a classic example of which is the responses presented in a Likert-scale format. Consequently, it is the most appropriate model to analyse three of the four indicators of commitment (namely “loyal”, “values” and “proud”) and four of the five indicators of engagement (namely. “effort”, “help”, “takeany” and “turndown”).

The model used conforms to convention and is presented as a latent variable model that reflects an individual’s likelihood of committing/engaging (Long and Freese, 2014). Defining y* as this latent variable, that takes the values −∞ to ∞, the corresponding structural model is:

y i * = X i β + ε i
where “y” is the indicator in question, “i” is an observation, “X” a vector of independent variables, “β” a set of coefficients to be estimated and “ε” a random error term. In the corresponding measurement model, y* is divided into J (in this instance four) ordinal categories:
y i = m if Ƭ m 1 y i * < Ƭ m for m = J
where the cutpoints (or thresholds) Ƭ1 through to ƬJ−1 are estimated.

The vector of independent variables contains the variables of two discrete types. The first reflects an individual’s personal characteristics, such as gender, age, highest educational qualifications, occupational classification and length of tenure with the organisation. The second reflects the characteristics of the workplace at which the individual is employed. The surveys upon which the working data set are based are the surveys of employees. Consequently, the number of variables appropriate to capturing fully a workplace’s characteristics is limited. Nonetheless, it is possible to incorporate variables which denote the size of the workplace, the Standard Industrial Classification of the activity undertaken and its sector. Integral to these workplace characteristics are four dummy variables identifying HR policy interventions often considered to be conducive to engendering commitment and engagement. To examine whether commitment and/or engagement has changed over time, a set of year dummies are included. Further details about the independent variables used in the seven ordered logit estimations may be seen in Column 1 of Table III and the footnotes to this table.

The two composite indicators “commitment” and “engagement” are scalar-dependent variables. Consequently, they are analysed best using an OLS regression model (Cameron and Trivedi, 2010). Again the model conforms to convention, namely:

y i = α + X i β + ε i
where “y” is the indicator in question, “i” is an observation, “X” a vector of independent variables, “β” a set of coefficients to be estimated and “ε” a random error term. The vector of independent variables in the OLS model is the same as that used in the ordered logit estimations.

6. Results

The detailed results of the estimations of the four indicators denoting commitment are presented in Table III, with the ordered logit results for the three initial indicators (i.e. “loyal”, “values” and “proud”) appearing in Columns 2, 3 and 4, respectively, and the OLS results for the composite indicator (i.e. “commitment”) appearing in Column 5. The results of the estimations of the five indicators denoting engagement are presented in Table IV, with the ordered logit results for the four initial indicators (i.e. “effort”, “help”, “takeany” and “turndown”) appearing in Columns 2, 3, 4 and 5, respectively, and the OLS results for the composite indicator (i.e. “engagement”) appearing in Column 6.

The coefficients on the explanatory variables in the ordered logit results have a qualitative interpretation (because their average marginal effects have not been computed). A positive coefficient for an independent variable implies that an individual with this characteristic has a higher value of latent commitment/engagement and, hence, is more likely to report a higher value of self-reported commitment/engagement (and conversely when the coefficient of an independent variable is negatively signed). In principle, in the OLS regression results a one unit increase in the value of the independent variable changes the expected value of the dependent variable by the magnitude of the coefficient in question. However, in the application of this principle in this particular context, some qualification must be made for the origin and construction of the dependent variables in question. Nonetheless, the signs of these coefficients and their relative magnitudes will measure, the direction and the strength of the relationships between the variables in question respectively.

The aim of remainder of this section is to report the salient results from the seven ordered logit estimations, with a focus upon those independent variables which denote the characteristics of the individual and the workplace at which he/she is employed. Discussion of these particular results and the other results presented in Tables III and IV is deferred to the subsequent section.

An examination of these identified results indicates the independent variables that, generally, produce consistent outcomes across the three indicators of commitment and the four indicators of engagement.

Managers (relative to administrative and secretarial workers, the reference category) are more likely to commit more highly, irrespective of the indicator used. Similarly, high levels of engagement on the part of managers are observed for three of the four indicators used. The exception is managers’ unwillingness to take any job to keep working with the organisation. In contrast, individuals who are in occupations such as skilled trades; sales; operatives; and elementary are more likely to be much less committed and engaged with the organisation, irrespective of the indicator used.

Individuals who are not on permanent contracts of employment (relative to those who are) are more likely to commit less highly, irrespective of the indicator used. Also, they are more likely not to work beyond the terms of their existing contract or not be prepared to do so to assist the success of the organisation. By way of contrast, individuals who have part-time contracts of employment (relative to those who have full-time contracts of employment) are more likely to both commit and engage more highly across all indicators.

Individuals who are not union members (relative to individuals who are) and individuals who do not work at workplaces at which unions are present (relative to individuals who work at workplaces at which unions are present) are more likely to commit more highly across all three indicators. Similar outcomes are observed for the four indicators of engagement, although not all of these results are statistically significant

Individuals working in small workplaces (relative to those working at workplaces with 25-99 employees, the reference category) are more likely to report higher levels of commitment and engagement, irrespective of the indicators used. In contrast, although not uniformly across all size categories, individuals employed at relatively large workplaces are more likely to report lower levels of commitment and engagement.

Further examination of the results reveals that there are other independent variables that produce outcomes which are not consistent across the three indicators of commitment and the four indicators of engagement.

In terms of commitment, women, relative to men, are more likely to express more highly their loyalty to the organisation, although they are less likely either to share its values or to voice pride in working for it. In terms of engagement, they are more likely to be more willing to work harder than necessary for the organisation and be more likely to be more willing to take any job to keep working with the organisation.

There are some discernible patterns to the results of the age categories variables. While individuals in the older age categories (relative to the reference age category) are more likely to report more highly sharing the values of the organisation, the manifestation of their commitment is restricted to this indicator. In the context of engagement, individuals in these same age categories (again relative to the reference age category) are more likely to turn down outside job offers in favour of remaining with the organisation. Whereas individuals in the youngest age category are less likely to increase their supply of effort beyond their immediate job demands, individuals in the oldest age category are more likely to do so.

Individuals with educational qualifications (relative to those who do not have them) are more likely to be less willing to voice pride in working for their organisation. Further, they are more likely to be less willing to take any job to remain with the organisation.

Individuals working at workplaces in the public sector and in the not-for-profit sector (relative to those working in the private sector, the reference category) are more likely to be more proud of the organisation with which they work. Individuals who work in the public sector (again relative to those working in the private sector) are more likely to turn down alternative job offers in favour of remaining with their present organisation.

Some organisations seek to engender commitment and engagement by implementing particular policies and practices with respect to the management of human resources. It was possible to examine the implications of four such policies and practices, namely that an individual has received training; that an individual has been subject to some form of appraisal; that an individual is employed at a workplace where quality circles operate; and that an individual is employed at a workplace where meetings are held with management. In the context of commitment, where the individual has received training; where the individual is employed at a workplace which uses quality circles; and where the individual is employed at a workplace where meetings are held with management, individuals are relatively more likely to commit more highly across all three indicators. The outcomes are more complex in the context of engagement, however. First, for each of the four policy-related variables, an individual is relatively more likely to work more than his/her immediate job demands necessitate. Second, where an individual is employed at a workplace which uses quality circles and where an individual is employed at a workplace at which meetings are held with management, individuals are relatively more likely to engage more highly whatever indicator is used. No other outcome is statistically significant. In essence, therefore, there are positive correlations between only some of the policies and practices examined and only some of the indicators of commitment and engagement.

From this examination of these results it is evident, therefore, that the likelihood of an individual committing and/or engaging is contingent upon how these constructs are measured. Changing these indicators may mean that an individual may/may not appear to commit and/or engage. Furthermore, for any one indicator of either commitment or engagement, the likelihood that an individual commits/engages varies according to the personal characteristics of the individual and the characteristics of the workplace at which he/she is employed. There is evidence of more heterogeneity than homogeneity in an individual’s likelihood of committing and/or engaging.

7. Discussion

Whereas some of the empirical results reported in Tables III and IV accord with a priori expectations, others do not. Whereas some outcomes may be rationalised, mindful of the fact that it is correlations that are being reported, others are open to explanations that have nothing to do with either commitment or engagement.

That managers (relative to administrative and secretarial workers, the reference category) are more likely to commit more highly, irrespective of the indicator used, is compatible with expectations. To the extent that the managers are “leaders” within their respective organisations, often advocates and/or proselytisers of this attitude/behaviour, this outcome is not unexpected. Equally, high levels of engagement on the part of managers are to be expected for the same reasons. It is for three of the four indicators used. The exception is their apparent unwillingness to take any job to keep working with the organisation, an outcome which may be rationalised by their reluctance, perhaps, to lose their managerial status.

Similarly, that individuals who are not on permanent contracts of employment (relative to those who are) are more likely to commit less highly, irrespective of the indicator used, and are more likely not to work beyond the terms of their existing contract or not be prepared to do so to assist the success of the organisation also accords with expectations. By definition, these individuals are transients, willingly or unwillingly, and neither commitment nor engagement are likely to feature highly in their immediate work orientations.

Unionisation, with its attachment, and often allegiances, to agencies outwith the employing organisation, is incompatible with the unitarist perspective of organisations that dominates the managerially prescriptive literature as it relates to commitment and engagement. Consequently, the outcomes that individuals who are not union members (relative to individuals who are) and individuals who do not work at workplaces at which unions are present (relative to individuals who work at workplaces at which unions are present) are more likely to commit and engage more highly accords with expectations.

Tenure, however, is one variable where the statistical outcomes do not accord with expectations. Given the opportunities to do so exist, an assumption could be made that employee-organisation mismatches would result in an individual quitting voluntarily. Therefore, the expectation is that levels of commitment and engagement would be relatively higher for those with relatively longer spells of tenure. This is never found in the context of commitment and is seldom found in the context of engagement. Admittedly, there is a positive correlation between the two longest spells of tenure and a higher willingness to take any job to remain with the organisation, but this relationship could equally well be explained by an individual’s awareness of the potential costs of job changing.

That individuals who have part-time contracts of employment (relative to those who have full-time contracts of employment) are more likely to commit and engage more highly across all indicators are outcomes that may be attributable to the precarious nature of the employment situation of these workers and, therefore, their perceived need to be seen to demonstrate certain attitudes and behaviours, desired or expected, by management (Or, alternatively, it could be that individuals working on this type of employment contract do indeed possess and demonstrate the attitudes and behaviours that are often keenly sought by some organisations).

Individuals working in small workplaces (relative to those working at workplaces with 25-99 employees, the reference category) are more likely to report higher levels of commitment and engagement, irrespective of the indicators used. In contrast, if generally, individuals employed at relatively large workplaces are more likely to report lower levels of commitment and engagement. One possible explanation for these different outcomes may be that the informality sometimes associated with the management of control in small workplaces is, from the perspective of workers, to be preferred to the formality of the same often found in the bureaucracy in larger workplaces and this has positive consequences for an individual’s commitment and engagement.

Women, relative to men, are more likely to be more willing to work harder for the organisation with which they are employed to help it succeed. However, that women are more likely to be more willing to take any job to keep working with the organisation may be attributable more to their sometimes constrained labour supply rather than engagement with the organisation.

Although individuals in the older age categories (relative to the reference age category) are more likely to report sharing the values of the organisation more highly, their manifestation of commitment is restricted to this indicator. Individuals in these same age categories (again relative to the reference age category) are also more likely to turn down outside job offers in favour of remaining with the organisation. Equally, however, this latter outcome may be attributable to their manifold costs of job changing rather than any particular manifestation of engagement with the organisation. Whereas individuals in the youngest age category are less likely to increase their supply of effort beyond their immediate job demands, individuals in the oldest age category are more likely to do so. However, these outcomes could equally well be rationalised in terms of age differences in tacit knowledge as well as expressions of engagement.

Some of the other results reported in Tables III and IV are also worthy of note and some comment.

By their construction, the composite indictors of commitment and engagement are the aggregation of their respective original (amended where necessary) Likert-scale measures. As a consequence, therefore, the results of the OLS estimations inevitably tend to reflect the results of the seven ordered logit estimations. Nonetheless, the OLS estimates do add value in two ways. First, they identify more clearly those variables where the relationship between the variable in question and commitment and engagement differs. To illustrate, females (relative to males) are statistically significantly (positively) associated with engagement but not commitment: individuals who have qualifications (relative to those who do not) are statistically significantly (negatively) associated with engagement but not commitment: and individuals who are employed in the public sector (relative to those who are employed in the private sector) are statistically significantly (positively) associated with commitment but not engagement.

The second way in which the results from the OLS estimations add value is the manner in which the magnitude of the coefficient of the independent variables measures the strength of the relationship between that variable and commitment and engagement. Again for purposes of illustration, the positive relationship between being a manager (relative to the reference category) and commitment (value of coefficient: 0.4279) and engagement (value of coefficient: 0.3162) is almost as equally strong as the negative relationship between a skilled trades person and commitment (coefficient value: −3,559) and engagement (coefficient value: −3,286): and the extent of the negative relationship between an individual having level 4 and above qualifications (relative to having no qualifications) and engagement (coefficient value: −0.6149) is much greater than it is for an individual who has level 2 qualifications (coefficient value: −0.1878).

Finally, one further advantage of the data set used in the investigation is the scope to examine whether there has been a secular change with respect to an individual’s likelihood of committing and/or engaging. The economic environment prevailing in 2006 and 2012 in Britain contrasted with that of 1997, the reference year. The former was a year when the economy was expanding; and the latter was a year when the economy remained within what van Wanrooy et al. (2013) described as the shadow of the recession. In the context of commitment, there is a suggestion that loyalty has decreased since 1997. Although there is no equivalent trend in the context of engagement, nonetheless, there are some results of note. There is some evidence to suggest that an individual’s willingness to increase his/her supply of labour voluntarily has decreased over time. On the other hand, the willingness to accept monetary consequences to remain with the organisation increased in 2012, relative to 1997. Equally, however, this latter result would also be compatible with a hypothesis that features risk adverse employees being reluctant to accept the unknown of other organisations during a period of economic retrenchment.

8. Conclusions and implications for future research

Commitment and engagement play central roles in many of the prescriptive (or what Guest (1997) refers to as “normative”) models of the management of human resources. Although operating in differing ways, frequently these constructs operate as quasi-conduits in the metaphorical “black box” whereby select bundles of HR policies and practices result in improved performance, however performance is defined and measured (Walton, 1985; Procter, 2008).

Over time, the role played by commitment in these models has changed. With Walton (1985), given the requisite bundle of HR policies and practices, commitment was a necessary and sufficient condition to enhance performance. Although a lack of consensus continues to persist with respect to the meaning, structure and measurement of commitment (Klein et al., 2014; Meyer and Herscovitch, 2001), high commitment management has now become subsumed into the more encompassing model of HPWS (Wood, 1999; Cappelli and Neumark, 2001). Whereas high-involvement management and high commitment management once offered alternative, contrasting routes to improved performance, now they have been integrated into a more generic approach to managing organisations that aspire to improve their performance, with the former addressing work practices and the latter addressing employment practices (Belt and Giles, 2009; Boxall and Macky, 2009; Wood et al., 2013).

Although empirically, the causal relationship between HR policy and practice and performance remains contested (Wall and Wood, 2005), there are two important consequences to be observed from this more comprehensive HPWS model. The first is that, inevitably, the size of the critical bundle of HR policies and practices is increased considerably. The second is that the number of components within the metaphorical black box (such as commitment) – reflecting the processes by which policy inputs (the independent variables in the model) are transformed into performance outcomes (the dependent variables in the model) – are also increased. Moreover, their possible inter-relationships become more complex. However, in the expeditious search to examine the central causal relationship between dependent and independent variables in the HPWS model using micro-econometrics, many heroic assumptions are made about the nature of the transmission mechanisms which operate within the black box. One is that the likelihood of committing is homogenous across individuals.

The first research question addressed in this paper is as follows:

RQ1.

Who commits?

This question sought to examine the empirical basis for this assumption. The results suggested that there were none. The likelihood that an individual commits was seen to vary across the sampled population, an outcome that concurs with similar findings reported by Brown et al. (2011).

Engagement, with its still evolving theoretical base, is the more recent addition to the management lexicon in organisations’ never-ending search for competitive advantage. Case study methodology dominates this research agenda, with its focus upon gathering qualitative data about actors, their experiences and the inter-relationships of both to process as management seek to create the environment necessary to generate engaged employees and translate their newly acquired positive attitudes and beliefs into improved performance. Nonetheless, there are instances of black box-type analysis of engagement in which micro-econometrics is used to examine survey data (e.g. Alfes et al., 2013; Bakker and Xanthopoulou, 2013). Too frequently, however, there is a lack of both agreement and precision about how engagement is incorporated in the empirical studies that use this model. Is it antecedent, mediator or outcome (Bailey et al., 2017)?

In all this, the second research question which is rarely raised is as follows:

RQ2.

Who engages?

Once again the implicit assumption is that the likelihood of engaging is homogeneous across individuals. The empirical results of this paper challenge this assumption.

This empirical investigation has demonstrated that who commits and who engages depends upon the indicator used to measure the construct in question. Changing these indicators may mean that an individual may appear to no longer commit or engage. Furthermore, even for the same indicator, who commits or engages may vary across individuals, according to their personal characteristics and the characteristics of the workplace at which they are employed. There is, therefore, considerable heterogeneity in the likelihoods that an individual commits or engages. Consequently, neither commitment nor engagement can be assumed to facilitate non-problematically the conversion of policy inputs into performance outputs. The implications for future research are considerable.

Percentage frequency distributions for the three statements associated with commitment

Response Per cent
(A) “I feel little loyalty to this organisation” (dependent variable “loyal”)
Strongly agree 4.49
Agree 20.29
Disagree 69.54
Strongly disagree 30.46
Number of observations 14,037
(B) “I find that my values and the organisation’s values are very similar” (dependent variable “values”)
Strongly disagree 3.47
Disagree 22.55
Agree 61.84
Strongly agree 12.14
Number of observations 13,796
(C) “I am proud to be working for this organisation” (dependent variable “proud”)
Strongly disagree 3.51
Disagree 18.08
Agree 59.64
Strongly agree 18.77
Number of observations 13,967

Percentage frequency distributions for the one question and three statements associated with engagement

Response Per cent
(A) “How much effort do you put into your job beyond what is required?” (Dependent variable “effort”)
None 1.89
Only a little 4.67
Some 24.12
A lot 69.32
Number of observations 14,105
(B) “I am willing to work harder than I have to in order to help this organisation” (Dependent variable “help”)
Strongly disagree 2.76
Disagree 16.15
Agree 56.69
Strongly agree 24.40
Number of observations 14,028
(C) “I would take almost any job to keep working for this organisation” (Dependent variable “take any”)
Strongly disagree 23.93
Disagree 51.36
Agree 20.10
Strongly agree 4.61
Number of observations 13,971
(D) “I would turn down another job with more pay in order to stay with this organisation” (Dependent variable “turndown”)
Strongly disagree 24.74
Disagree 45.85
Agree 23.84
Strongly agree 5.57
Number of observations 13,669

Estimation results: commitment

Ordered logit results OLS results
Variable Coefficient (robust SE) “loyal” Coefficient (robust SE) “values” Coefficient (robust SE) “proud” Coefficient (robust SE) “commitment”
Female 0.1324** (0.0537) 0.0090 (0.0575) 0.0121 (0.0546) 0.0777 (0.0480)
Age categories
20-29 0.0532 (0.0624) −0.0825 (0.0658) 0.0019 (0.0657) 0.0082 (0562)
30-39 (the reference category)
40-49 0.0663 (0.0535) 0.1891*** (0.0584) 0.0870 (0.0548) 0.1186** (0.0471)
50-60 0.0049 (0.0585) 0.2142*** (0.0639) 0.0289 (0.0613) 0.0783 (0.0517)
Highest educational qualifications
Categories
None (the reference category)
Level 1 −0.1043 (0.0903) 0.0124 (0.0964) −0.3178*** (0.0968) −0.1316 (0.0799)
Level 2 −0.0256 (0.0747) −0.0202 (0.0796) −0.2107** (0.0828) −0.0914 (0.0675)
Level 3 0.0409 (0.0794) 0.0153 (0.0851) −0.2546*** (0.0876) −0.0787 (0.0715)
Level 4 or above −0.0776 (0.0845) −0.0705 (0.0919) −0.3703*** (0.0933) −0.1908** (0.0774)
Not a union member 0.2187*** (0.0553) 0.1629*** (0.0580) 0.2136*** (0.0568) 0.2292*** (0.0491)
SOC categories
Manager 0.4400*** (0.0880) 0.4978*** (0.0937) 0.4134*** (0.0887) 0.4279*** (0.0769)
Professional −0.0927 (0.0985) 0.1135 (0.1052) 0.0195 (0.0954) 0.0250 (0.0892)
Associate professional and Technical 0.0043 (0.0826) 0.0377 (0.0853) 0.1888** (0.0835) 0.0752 (0.0738)
Administrative and secretarial (the reference category)
Skilled trades −0.4426*** (0.0987) −0.2534** (0.1027) −0.3957*** (0.1033) −0.3559*** (0.0890)
Personal services −0.2704*** (0.1008) 0.1099 (0.1086) 0.1704 (0.1054) −0.0194 (0.0888)
Sales −0.2971*** (0.1023) −0.1527 (0.1045) −0.0946 (0.1050) −0.1911 (0.0891)
Operatives −0.4855*** (0.1029) −0.3154*** (0.1047) −0.4465*** (0.1038) −0.4815*** (0.0927)
Elementary −0.4666*** (0.0945) −0.2835*** (0.0967) −0.3103*** (0.0982) −0.3964*** (0843)
Tenure categories
Less than 1 year −0.0277 (0.0856) 0.1778* (0.0963) 0.0799 (0967) 0.0832 (0.0819)
Between 1 and 2 years (the reference category)
Between 2 and 3 years −0.0130 (0.0941) 0.0697 (0.1057) 0.0437 (0.1051) 0.0293 (0.0868_
Between 3 and 5 years −0.0045 (0.0856) −0.0297 (0.0940) −0.0457 (0.0947) −0.0241 (0.0790)
Between 5 and 10 years 0.0707 (0.0766) 0.0007 (0.0864) 0.0474 (0.0873) 0.0486 (0.0722)
10 years or more 0.1248 (0.0787) −0.0514 (0.0875) −0.0193 (0.0870) 0.0185 (0.0728)
Not a permanent contract −0.2911*** (0.1027) −0.2624** (0.1099) −0.2895*** (0.1033) −0.3263*** (0.0941)
Working part time 0.2253*** (0.0841) 0.2314** (0.0916) 0.2285*** (0.0833) 0.2205*** (0.0732)
Sector categories
Private sector (the reference category)
Public sector 0.0243 (0.0762) 0.1074 (0.0836) 0.2224*** (0.0772) 0.1560** (0.0668)
Not for profit organisation 0.2312 (0.1430) 0.7498*** (0.1557) 0.3935*** (0.1371) 0.4209*** (0.1212)
Workplace size categories Less than 25 employees 0.2316*** (0.0556) 0.3207*** (0.0602) 0.3784*** (0.0587) 0.3123*** (0.0492)
Between 25 and 99 employees (the reference category)
Between 100 and 499 employees −0.1925*** (0.0575) −0.1516** (0.0609) −0.0295 (0.0594) −0.1229** (0.0516)
500 or more employees −0.1678** (0.0672) −0.1556** (0.0678) 0.0498 (0.0681) −0.0966* (0.0580)
No unions present at the workplace 0.2556*** (0.0583) 0.3338*** (0.0597) 0.2007*** (0.0601) 0.2705*** (0.0511)
Received training 0.1754*** (0.0471) 0.1072** (0.0490) 0.2411*** (0.0476) 0.1995*** (0.0413)
Was appraised −0.0447 (0.0554) 0.0494 (0.0575) 0.1304** (0.0589) 0.0597 (0.0493)
“Quality Circles” operate 0.2947*** (0.0454) 0.3068*** (0.0483) 0.3544*** (0.0464) 0.3213*** (0.0392)
Meetings with management held 0.4565*** (0.0571) 0.4691*** (0.0595) 0.5432*** (0.0580) 0.5309*** (0.0504)
1997 (the reference category)
2001 −0.0412 (0.0621) 0.0726 (0.0657) 0.0684 (0.0647) 0.0448 (0.0573)
2006 −0.2612*** (0.0613) −0.0427 (0.0661) −0.0116 (0.0639) −0.1060* (0.0566)
2012 −0.1829** (0.0761) −0.0263 (0.0807) 0.1444 (0.0810) −0.0288 (0.0700)
/cut1 −2.0520 (0.3535) −1.7335 (0.3790) −1.5872 (0.3526)
/cut2 −0.3242 (0.3494) 0.6046 (0.3755) 0.5343 (0.3486)
/cut3 2.0498 (0.3500) 3.8680 (0.3788) 3.5428 (0.3498)
Number of observations 12,514 12,319 12,438 12,204
Wald χ2 (59)=812.85 (59)=790.91 (59)=986.19
Prob>χ2 0.0000 0.0000 0.0000
Psuedo R2 0.0423 0.0467 0.0491
F(59, 12,144) 21.44
Prob>F 0.0000
R2 0.1285

Notes: Additionally, the estimated model controlled for the following: ethnicity (via a dummy variable); marital status (via a dummy variable); the length of time it took the individual to learn to do his/her present job well (via a set of 6 dummy variables); the log of the number of hours usually worked; and the SIC of the activity undertaken (via a set of 13 dummy variables). *,**,***Statistically significant at p<0.1; p<0.05 and p<0.01, respectively

Estimation results: engagement

Ordered logit results OLS results
Variable Coefficient (robust SE) “effort” Coefficient (robust SE) “help” Coefficient (robust SE) “takeany” Coefficient (robust SE) “turndown” Coefficient (robust SE) “engagement”
Female 0.6090*** (0.0622) 0.0073 (0.0556) 0.1229** (0.0517) −0.0806 (0.0527) 0.1767*** (0.0544)
Age categories
20-29 −0.2257*** (0.0717) 0.0199 (0.0655) 0.1002 (0.0632) −0.0292 (0.0622) −0.0356 (0.0648)
30-39 (the reference category)
40-49 0.1002 (0.0623) 0.0299 (0.0558) 0.0106 (0.0526) 0.1106** (0.0517) 0.0965* (0.0530)
50-60 0.1829*** (0.0687) −0.0383 (0.0626) −0.1145* (0.0594) 0.1528*** (0.0581) 0.0741 (0.0588)
Highest educational qualifications
Categories
None (the reference category)
Level 1 −0.0475 (0.1126) −0.1634* (0.0943) −0.3753*** (0.1001) −0.1446 (0.0975) −0.2512** (0.1004)
Level 2 −0.0922 (0.0921) −0.0701 (0.0750) −0.3244*** (0.0806) −0.0480 (0.0790) −0.1878** (0.0808)
Level 3 −0.2971*** (0.0955) −0.0062 (0.0788) −0.4894*** (0.0848) −0.1199 (0.0837) −0.3087*** (0.0836)
Level 4 or above −0.4189*** (0.0993) −0.0624 (0.0846) −0.9464*** (0.0902) −0.2758*** (0.0898) −0.6149*** (0.0885)
Not a union member 0.0113 (0.0637) 0.2204*** (0.0588) 0.0023 (0.0560) 0.1801*** (0.0536) 0.1693*** (0.0573)
SOC categories
Manager 0.4345*** (0.0988) 0.2766*** (0.0887) −0.1004 (0.0815) 0.3415*** (0.0860) 0.3162*** (0.0851)
Professional 0.0075 (0.1089) −0.0867 (0.0935) −0.3885*** (0.0906) −0.0038 (0.0932) −0.1759* (0.0959)
Associate professional and Technical 0.0897 (0.0930) −0.0747 (0.0844) −0.0753 (0.0807) 0.1000 (0.0794) 0.0120 (0.0851)
Administrative and secretarial (the reference category)
Skilled trades −0.1256 (0.1092) −0.4428*** (0.1006) 0.0074 (0.0969) −0.2542*** (0.0979) −0.3286*** (0.1002)
Personal services 0.1509 (0.1231) −0.1847* (0.1045) 0.1785* (0.0970) 0.1033 (0.1008) 0.0978 (0.1036)
Sales −0.1017 (0.1190) −0.3375*** (0.1037) −0.0473 (0.1054) −0.2610** (0.1040) −0.3001*** (0.1098)
Operatives −0.1587 (0.1126) −0.4675*** (0.1009) 0.0754 (0.1048) −0.4100*** (0.1032) −0.3992*** (0.1089)
Elementary −0.0264 (0.1062) −0.3858*** (0.0949) 0.2615*** (0.0954) −0.2784*** (0.0910) −0.1899* (0.0994)
Tenure categories
Less than 1 year 0.1234 (0.1035) 0.0585 (0.0910) 0.0986 (0.0895) 0.0398 (0.0896) 0.1233 (0.0934)
Between 1 and 2 years (the reference category)
Between 2 and 3 years 0.2189* (0.1138) 0.0533 (0.0982) 0.1047 (0.1011) 0.0400 (0.1038) 0.1601 (0.1036)
Between 3 and 5 years 0.2988*** (0.1057) −0.0099 (0.0878) −0.0008 (0.0877) −0.0023 (0.0860) 0.0391 (0.0886)
Between 5 and 10 years 0.0777 (0.0931) 0.0575 (0.0812) 0.1950** (0.0790) 0.0988 (0.0789) 0.1470* (0.0821)
10 years or more −0.1183 (0.0927) −0.0824 (0.0830) 0.3024*** (0.0813) 0.2081*** (0.0794) 0.1477* (0.0839)
Not a permanent contract −0.2370** (0.1143) −0.2978** (0.1148) 0.1036 (0.1073) −0.0595 (0.1105) −0.2076* (0.1218)
Working part time 0.2877*** (0.1024) 0.1943** (0.0871) 0.0747 (0.0825) 0.2215*** (0.0835) 0.2555*** (0.0875)
Sector categories
Private sector (the reference category)
Public sector −0.0384 (0.0896) −0.0504 (0.0781) −0.0379 (0.0743) 0.0731 (0.0760) 0.0275 (0.0778)
Not for profit organisation 0.1192 (0.1600) 0.2078 (0.1552) 0.0439 (0.1239) 0.5946*** (0.1432) 0.3928*** (1325)
Workplace size categories Less than 25 employees 0.1941*** (0.0670) 0.2377*** (0.0578) 0.3302*** (0.0554) 0.1890*** (0.0555) 0.3496*** (0.0570)
Between 25 and 99 employees (the reference category)
Between 100 and 499 employees 0.0111 (0.0683) −0.1407** (0.0606) −0.1464** (0.0570) −0.1230** (0.0565) −0.1656*** (0.0582)
500 or more employees −0.0837 (0.0755) −0.1272* (0.0687) −0.0777 (0.0661) −0.0747 (0.0640) −0.1271* (0.0667)
No unions present at the workplace 0.1621** (0.0689) 0.1938*** (0.0589) 0.1227** (0.0572) 0.2197*** (0.0559) 0.2724*** (0.0579)
Received training 0.1677*** (0.0532) 0.0536 (0.0473) 0.0612 (0.0468) 0.1250*** (0.0463) 0.1475*** (0.0480)
Was appraised 0.1646*** (0.0624) 0.0331 (0.0576) 0.0542 (0.0573) −0.4575 (0.0567) 0.0637 (0.0583)
“Quality Circles” operate 0.3861*** (0.0532) 0.3202*** (0.0464) 0.0925** (0.0445) 0.1919*** (0.0443) 0.3343*** (0447)
Meetings with management held 0.1059* (0.0621) 0.2115*** (0.0569) 0.2956*** (0.0583) 0.3225*** (0.0574) 0.3671*** (0.0586)
1997 (the reference category)
2001 −0.0519 (0727) −0.0130 (0.0652) −0.0166 (0.0612) 0.0922 (0.0615) 0.0254 (0.0642)
2006 −0.1501** (0.0723) −0.4576*** (0.0658) −0.0748 (0.0603) 0.0646 (0.0611) −0.2069*** (0.0638)
2012 −0.0414 (0.0865) 0.1159 (0.0761) 0.3721*** (0.0758) 0.1483*** (0.0760) 0.2354*** (0.0786)
/cut1 0.0660 (0.4110) −2.0490 (0.3693) −1.2117 (0.3614) −0.1494 (0.3619)
/cut2 1.3682 (0.4083) 0.0466 (0.3624) 1.2193 (0.3609) 1.9562 (0.3618)
/cut3 3.3262 (0.4086) 2.8220 (0.3638) 3.1688 (0.3638) 3.9609 (0.3648)
Number of observations 12,547 12,489 12,453 12,198 12,083
Wald χ2 (59)=1,132.90 (59)=740.97 (59)=680.60 (59)=510.25
Prob > χ2 0.0000 0.0000 0.0000 0.0000
Psuedo R2 0.0512 0.0393 0.0330 0.0282
F(59, 1,203) 13.83
Prob>F 0.0000
R2 0.0910

Notes: Additionally, the estimated model controlled for the following: ethnicity (via a dummy variable); marital status (via a dummy variable); the length of time it took the individual to learn to do his/her present job well (via a set of 6 dummy variables); the log of the number of hours usually worked; and the SIC of the activity undertaken (via a set of 13 dummy variables). *,**,***Statistically significant at p<0.1; p<0.05 and p<0.01, respectively

References

Alfes, K., Truss, C., Soane, E., Rees, C. and Gatenby, M. (2013), “The relationship between line manager behaviour, perceived HRM practices and individual performance: examining the mediating role of engagement”, Human Resource Management, Vol. 52 No. 6, pp. 839-859.

Appelbaum, E. and Batt, R. (1994), The New American Workplace: Transforming Work Systems in the United States, ILR Press, New York, NY.

Arrowsmith, J. and Parker, J. (2013), “The meaning of ‘employee engagement’ for the values and roles of the HRM function”, The International Journal of Human Resource Management, Vol. 24 No. 14, pp. 2692-2712.

Bailey, C., Madden, A., Alfes, K. and Fletcher, L. (2017), “The meaning, antecedents, and outcomes of employee engagement: a narrative synthesis”, International Journal of Management Reviews, Vol. 19 No. 1, pp. 31-53.

Bakker, A.B. and Xanthopoulou, D. (2013), “Creativity and charisma and female leaders: the role of resources and work engagement”, The International Journal of Human Resource Management, Vol. 24 No. 14, pp. 2760-2779.

Becker, B. and Gerhard, B. (1996), “The impact of human resource management on organisation performance: progress and prospects”, Academy of Management Journal, Vol. 39 No. 4, pp. 779-801.

Belt, V. and Giles, L. (2009), “High performance working: a synthesis of key literature”, Evidence Report No. 4, UK Commission for Employment and Skills, Wath-upon-Derne.

Boselie, P., Dietz, G. and Boon, C. (2005), “Commonalities and contradictions in HRM and performance research”, Human Resource Management Journal, Vol. 15 No. 1, pp. 67-94.

Boxall, P. and Macky, K. (2009), “Research and theory on high-performance work systems: progressing the high involvement stream”, Human Resource Management Journal, Vol. 19 No. 1, pp. 3-23.

Boxall, P. and Purcell, J. (2008), Strategies and Human Resource Management, 2nd ed., Palgrave Macmillan, Basingstoke.

Brown, S., McHardy, J., McNabb, R. and Taylor, K. (2011), “Workplace performance, worker commitment, and loyalty”, Journal of Economics and Management Strategy, Vol. 20 No. 4, pp. 925-955.

Cameron, A.C. and Trivedi, P.K. (2010), Microeconomics Using Stata, revised ed., Stata Press, College Station, TX.

Cappelli, P. and Neumark, D. (2001), “Do ‘high performance’ work practices improve establishment outcomes?”, Industrial and Labor Relations Review, Vol. 54 No. 4, pp. 737-776.

Delbridge, R. and Keenoy, T. (2010), “Beyond managerialism?”, International Journal of Human Resource Management, Vol. 21 No. 6, pp. 799-817.

Dromey, J. (2014), Meeting the Challenge: Successful Employee Engagement in the NHS, IPA, London.

Emmott, M. (2010), “Employee engagement is a strategic issue”, Strategic HR Review, Vol. 9 No. 1, pp. 40-41.

Felstead, A., Gallie, D. and Green, F. (2014), Skills and Employment Series Dataset, 1986, 1992, 1997, 2001, 2006 and 2012 (Computer File), 2nd ed., UK Data Archive (distributer), Colchester.

Francis, H. and Reddington, M. (2012), “Employer branding and organisational effectiveness”, in Francis, M., Holbeche, L. and Reddington, M. (Eds), People and Organisational Development: A New Agenda for Organisational Effectiveness, CIPD, London, pp. 260-285.

Green, F. (2006), Demanding Work: The Paradox of Job Quality in the Affluent Economy, Princeton University Press, Princeton, NJ.

Guest, D.E. (1997), “Human resource management and performance: a review and research agenda”, The International Journal of Human Resource Management, Vol. 8 No. 3, pp. 263-276.

Guest, D.E. (2013), “Employee engagement: fashionable fad or long-term fixture?”, in Truss, C., Alfes, K., Delbridge, R., Shantz, A. and Soane, E.C. (Eds), Employee Engagement in Theory and Practice, Routledge, London, pp. 221-235.

Hertzberg, F. (1966), Work and the Nature of Man, World Publishing, Cleveland, OH.

Huselid, M. (1995), “The impact of human resource management practices on turnover, productivity and corporate performance”, Academy of Management Journal, Vol. 38 No. 3, pp. 635-672.

Ichniowski, C., Kochan, T., Levine, D., Olson, C. and Strauss, G. (1996), “What works at work: overview and assessment”, Industrial Relations, Vol. 35 No. 3, pp. 299-333.

Jenkins, S. and Delbridge, R. (2013), “Context matters: examining ‘soft’ and ‘hard’ approaches to employee engagement in two workplaces”, The International Journal of Human Resource Management, Vol. 24 No. 14, pp. 2670-2691.

Kahn, W.A. (1990), “Psychological conditions of personal engagement and disengagement at work”, Academy of Management Journal, Vol. 33 No. 4, pp. 692-724.

Kalleberg, A.L. and Berg, I. (1987), Work and Industry: Structures, Markets and Processes, Plenum, New York, NY.

Keenoy, T. (2013), “Engagement: a murmuration of objects”, in Truss, C., Alfes, K., Delbridge, R., Shantz, A. and Soane, E.C. (Eds) op cit, pp. 197-220.

Klein, H.J., Cooper, J.T., Molloy, J.C. and Swanson, J.A. (2014), “The assessment of commitment: advantages of a unidimensional, target-free approach”, Journal of Applied Psychology, Vol. 99 No. 2, pp. 222-238.

Kochan, T.A. and Osterman, P. (1994), The Mutual Gains Enterprise, Harvard Business School Press, Harvard, MA.

Lansbury, R.D. and Wailes, N. (2008), “Employee involvement and direct participation”, in Blyton, P., Bacon, N., Fiorito, J. and Heery, E. (Eds), The Sage Handbook of Industrial Relations, Sage, London, pp. 434-446.

Lawler, E.E. (1986), High-Involvement Management: Participative Strategies for Improving Organizational Performance, Jossey-Bass, San Francisco, CA.

Legge, K. (2005), Human Resource Management: Rhetorics and Realities, Palgrave Macmillan, Basingstoke.

Long, J.S. and Freese, J. (2014), Regression Models for Categorical Dependent Variables Using Stata, 3rd ed., Stata Press, College Station, TX.

MacDuffie, J. (1995), “Human resource bundles and manufacturing performance: organisational logic and flexible production systems in the world auto industry”, Industrial and Labor Relations Review, Vol. 48 No. 1, pp. 197-221.

MacLeod, D. and Clarke, N. (2009), “Engaging for success: enhancing performance through employee engagement”, a report to Government, Department for Business, Innovation and Skills, London.

Maslow, A.H. (1943), “A theory of motivation”, Psychological Development, Vol. 50 No. 4, pp. 370-396.

May, D.R., Gilson, R.L. and Harter, L.M. (2004), “The psychological conditions of meaningfulness, safety, availability and the engagement of the human spirit at work”, Journal of Occupational and Organizational Psychology, Vol. 77 No. 1, pp. 11-37.

McGregor, D.C. (1960), The Human Side of the Enterprise, McGraw-Hill, New York, NY.

Meyer, J.P. and Allen, N.J. (1997), Commitment in the Workplace, Sage Publications, Thousand Oaks, CA.

Meyer, J.P. and Herscovitch, L. (2001), “Commitment in the workplace. Toward a general model”, Human Resource Management Review, Vol. 11 No. 3, pp. 299-326.

Mowday, R.T., Steers, R.M. and Porter, L.W. (1979), “The measurement of organisational commitment”, Journal of Vocational Behaviour, Vol. 14 No. 2, pp. 224-247.

Procter, S. (2008), “New forms of work and the high performance paradigm”, in Blyton, P., Bacon, N., Fiorito, J. and Heery, E. (Eds), op cit, pp. 149-169.

Purcell, J. (2014), “Disengaging from engagement”, Human Resource Management Journal, Vol. 24 No. 3, pp. 241-254.

Purcell, J., Kinnie, N., Hutchinson, S., Rayton, B. and Swart, J. (2003), Understanding the People and Performance Link: Unlocking the Black Box, CIPD, London.

Ramsay, H., Scholarios, D. and Harley, B. (2000), “Employees and high-performance work systems: testing inside the black box”, British Journal of Industrial Relations, Vol. 38 No. 4, pp. 501-531.

Rees, C., Alfes, K. and Gatenby, M. (2013), “Employee voice and engagement: connections and consequences”, The International Journal of Human Resource Management, Vol. 24 No. 14, pp. 2780-2798.

Rich, B.L., LePine, J.A. and Crawford, E.R. (2010), “Job engagement: antecedents and effects on job performance”, The Academy of Management Journal, Vol. 53 No. 3, pp. 617-635.

Robinson, D., Perryman, S. and Hayday, S. (2004), The Drivers of Employee Engagement, Institute for Employment Studies, Brighton.

Saks, A. (2006), “Antecedents and consequences of engagement”, Journal of Managerial Psychology, Vol. 21 No. 7, pp. 600-619.

Selznick, P. (1957), Leadership in Administration: A Sociological Interpretation, Harper & Row, New York, NY.

Shuck, B. (2011), “Four emerging perspectives of employee engagement: an integrative literature review”, Human Resource Development Review, Vol. 10 No. 2, pp. 304-328.

Storey, J. (1989), “Introduction: from personnel management to human resource management”, in Storey, J. (Ed.), New Perspectives on Human Resource Management, Routledge, London, pp. 1-18.

Truss, C., Shantz, A., Soane, E., Alfes, K. and Delbridge, R. (2013), “Employee engagement, organisational performance and individual well-being: exploring the evidence, developing the theory”, The International Journal of Human Resource Management, Vol. 14 No. 14, pp. 2657-2669.

van Wanrooy, B., Bewley, H., Bryson, A., Forth, J., Freeth, S., Stokes, L. and Wood, S. (2013), Employment Relations in the Shadow of Recession, Palgrave Macmillan, Basingstoke.

Wall, T.D. and Wood, S.J. (2005), “The romance of human resource management and business performance, and the case for big science”, Human Relations, Vol. 58 No. 4, pp. 429-462.

Walton, R. (1985), “From control to commitment in the workplace”, Harvard Business Review, March-April, pp. 76-84.

Watson, T.J. (2010), “Critical social science, pragmatism and the realities of HRM”, The International Journal of Human Resource Management, Vol. 21 No. 6, pp. 915-931.

Wood, S. (1999), “Human resource management and performance”, International Journal of Management Reviews, Vol. 1 No. 4, pp. 367-413.

Wood, S., Burridge, M., Green, W., Nolte, S., Ruddolf, D. and Luanaigh, A.N. (2013), “High performance working in the employer skills surveys”, Evidence Report No. 71, the UK Commission for Employment and Skills, Wath-upon-Dearne.

Acknowledgements

The Skills and Employment Survey, 2012 was financially supported by the Economic and Social Research Council (ESRC), the UK Commission for Employment and Skills Strategic Partnership and the Wales Institute for Social and Economic Research, Data and Methods for the Welsh boost. The Skills Survey, 2006 was supported by the Department for Education and Skills, the Department of Trade and Industry, the Learning and Skills Council, the Sector Skills Development Agency, Scottish Enterprise, Futureskills Wales, Highlands and Islands Enterprise and the East Midlands Development Agency. The Skills Survey, 2001 was funded by the Department for Education and Skills. The Skills Survey, 1997 and the Social Change and Economic Life Initiative Surveys, 1986-1987 were supported by the ESRC. Employment in Britain, 1992 was supported by the Leverhume Trust and an industrial consortium of funders. The author acknowledges constructive comments from two referees on earlier drafts of this paper.

Corresponding author

John Sutherland can be contacted at: john.sutherland@strath.ac.uk

About the author

John Sutherland is an Honorary Research Fellow in the Scottish Centre for Employment Research (SCER) in the Department of Human Resource Management, University of Strathclyde, Glasgow.