Relationship Between HRM and Employee Attitude

family relationship Between HRM and Employee billetThe blood amongst HRM praxis and employee hold up- connect attitudesIntroductionThe excogitation of human vision solicitude (HRM) has certain focussed attention for around 20 years, with the catalyst macrocosm that m any(prenominal) US companies establish they were being rivalled and in some instances overinterpreted, in markets they had predominate (Ehrlich, 1994, p. 492). As Lodge (1985, p. 319) observesBy the early 1980s at that place was cool off little disagreement that US corpo compute managers, employees and trade unions would devour to change their ship counseling in say to compete successfully for markets in America and abroad.Harvard university academics introduced a new compulsory component of HRM into their MBA syllabus and fortify this so-called Harvard Model with authoritative books and articles (Beer et al., 1984 Walton, 1985b Walton and Lawrence, 1985). While in that location would surely choose been a genuine desire to help US business, US society, and even US employees, there was in any case a foresightful-term lying-in to ensure that the Harvard Business drill faculty provided leadership in human preference watchfulness (Walton and Lawrence, 1985, p. xx).The Harvard concept stresses that HRM should lead to employee load not solely as a means to employer acc utilises of improved productivity and profits, provided beca custom the fulfilment of many employee unavoidably is taken as a goal rather than merely a means to an wipe expose (Walton, 1985a, p. 49).At ab expose the same time as the Harvard concept was being substantial an different descrypoint was being promulgated by academics who supported a strategical concept of HRM, with the major(ip) work edited by Fombrun et al. (1984). This work emphasises that the tetrad generic human resource activities of all musical arrangements survival/promotion/placement processreward process maturation process and appraisal process (Tichy et al., 1984, p. 26) bespeak to be strategically align with the organisations overall strategic target beas. stiff and sponge deal HRMBritish writers perk up foc charged on the differences in the midst of the Harvard committedness concept of HRM and the strategic HRM concept. Keenoy (1990, p. 368) sees the Harvard concept as philosophically grounded in the light of multiple stakeholders and the belief that the dress and benefits of HRM can be achieved through neo-plura come mechanisms, season the strategic concept is al closely uniformly unitarist in predilection and displays a quite a singular endorsement of managerial values. The former is oft referred to as woolly HRM, while the latter is hard HRM.In theory, fragile HRM fulfils employee necessitate as an end in itself, and the favourable attitudes generated from the use of appropriate HRM implements ( customer, 1997) unitedly with communication, motivation and leadership (Storey, 1987, p. 6), guide in lading to the organisation and improved performance. Hard HRM is more thanover concerned with the impelling utilisation of employees ( knob, 2002) and emphasises the quantitative, calculative and business strategic aspects of managing the head count resource in as natural a office as for any other economic cipher (Storey, 1987, p. 6).If the populace of soft HRM bore was that it produced the benefits referred to in the rhetoric of academic evange diagnoses such as Richard Walton and so it should be highly favoured by both employees and employers. Questions countenance been raised by a numeral of look forers, however, just close(predicate) the force of soft HRM to achieve these benefits. on that point argon two concerns. The basic is that the squ atomic number 18 motive behind its admission is to undermine unions (Sisson, 1994). Indeed, going back to the introduction of the Harvard concept of soft HRM, Lawrence (1985, p. 362) writes slightly a se minal colloquium on HRM among 35 senior executives and Harvard academics. He claims that the question as to whether HRM was intentional to keep the union out was emphatically denied, and he protested that all save third of the firms represented were at least partially unionised, with several(prenominal) having had a prodigious number of units of both kinds, with their older plants unionised and the newer ones not. An mindless interpretation therefore would be that these leaders of commitment HRM bequeath happily dispense with unions given the opportunity. Indeed, Lawrence (1985, p. 362) admits that there was a habitual view that in a well-managed unit with decision-sharing, a union was not compulsory to represent employee interests. lymph gland (1990, p. 389) concludes, the primary(prenominal) impact of HRM in the United States may redeem been to provide a smokescreen behind which solicitude can introduce non-unionism or obtain of import concessions from trade unions .The morsel concern is that meeting the leases of employees has never been an objective in itself, and has simply been the normative view of what Harvard academics would like to see as the usage alliance. Truss et al. (1997, p. 70) in a deliberate of soft and hard models of HRM, reason that even if the rhetoric of HRM is soft, the reality is al virtually always hard, with the interests of the organisation prevailing over those of the individual. Similar conclusions have been r distributivelyed by Keenoy (1990), Poole and Mansfield (1992), node (1995) and Legge (1995a, b, 1998).If the soft model of HRM has validity, then there should be a clear blood mingled with the experience of soft HRM traffic pattern and plus employee attitudes ( conjectureing their needs atomic number 18 being met) and increasing employee commitment as well as improved productivity. Surprisingly, however, most of the research and reporting on HRM has ignored the views of employees. Legge (1998, p. 14) points out when reading bills of HRM example in the UK and North America it is noticeable the consequence to which the selective information argon (literally) the voices of management. Guest (1999, p. 5) agrees, claiming that from its conception human resource management reflected a management agenda to the neglect of workers concerns. Some studies have mattered employee reactions to HRM (Gibb, 2001 Appelbaum and Berg, 2000 Mabey et al., 1998) but Guest (2002, p. 335) is perceptive in his criticism that a feature of both advocates and critics of HRM is their neglect of prep atomic number 18 evidence virtually the post and reactions of workers. Even the exceptions noted which gave employee reactions to HRM did not relate the employee experience of HRM executes to their reactions, and Guest (1997) argues for a research agenda that tackresses this gap.This physical composition aims to contribute to this gap by fulfilling two major objectives. The first is to add to t he limited number of studies which have tested to see if there is a family amidst HRM figure and arrogant employee attitudes. The second objective is to see if there is a significantly salubriouser relationship between HRM answer and employee attitudes if employee perceptions of HRM entrust atomic number 18 taken as the bankers bill of HRM rather than employer perceptions, which has been the approach taken in general by previous researchers.HRM rehearse and employee work- colligate attitudesThe soft model of HRM, as stated previously, arouses a relationship exists between the use of appropriate HRM radiation diagrams and positive employee attitudes, and while theoretically these relationships remain poorly real (Guest, 1997, 2001), a number of attitudes be nonetheless wide loted to be an outcome of soft HRM. For example, levels of job satisfaction, which is the affective perception that results from the doing of desired outcomes (Harber et al., 1997), ar erect to be related to levels of HRM practice (Guest, 2002 Ting, 1997). last levels of employee commitment have also been make to be related to the use of appropriate HRM practice (Guest, 2002), and results from investing in HRM practices which benefit employees. For example, the raising of opportunities for homework and skill breeding benefits the employee by equipping them with the necessary knowledge, skills and attitudes to function autonomously and responsibly (Guest, 2002). Furthermore, it improves retention and enables them to deal with change in the work environment (Guest, 2002).Organisational impartiality is the term utilize to describe the role of impartiality as it directly relates to the oeuvre and is concerned with the ways employees determine if they have been treated fairly in their jobs and the ways in which those determinations yield other work-related variables (Moorman, 1991, p. 845). The HRM systems, policies, and procedures that operate in an organisation hav e been identified as impacting on an individuals perceptions of bias and candidness (Kurland and Egan, 1999 Greenberg, 1990). Studies show that where employees believe they ar treated fairly in the body of work then they hold positive attitudes towards the organisation (Moorman, 1991), whereas HRM practices that are perceived to be unfair have been fix to result in the employee intuitive feeling bewildered and betrayed and thus less committed (Schappe, 1996).As far as the relationship between HRM practice and employee attitudes is concerned, the HRM practice dimension has been broadsided by collecting data from employers in one of the two ways victimisation running(a) amount of moneys of HRM practice or self-reports about the extent to which token HRM practices have been operationalised. A realizable tercet approach would be to collect employee views about the operationalisation of HRM practice and then relate these reactions to their attitudes. These approaches are d epict below.The bilinear approachTo date the most common approach for assessing the relationship between HRM practice and employee work-related attitudes is the analogue approach. This involves employers indicating, apply a yes/no response format, which HRM practices from a predetermined list flowingly operate in their organisations. The yes responses are then added together, and the assumption is that a high sum indicates burst HRM. Researchers using the bilinear approach usually correlate the chalk up number of practices with the HRM outcomes being examined. Research exploring this relationship show a strong correlation between high numbers game of HRM practice and positive employee attitudes (Fiorito, 2002 Guest, 1999 Guest and Conway, 2002 Appelbaum and Berg, 2000).While this approach is frequently used, it has attracted criticism (Fiorito, 2002 Guest, 1997, 2002). The additive approach is very(prenominal) simplistic. If an employer claims a practice is use there is no preeminence on the basis on how well, or to what extent the practice exists. For example, there could be token genteelness and development, or a high level of commitment to a maestro and impelling training and development programme. Both examples are credited as a yes. This perceived washed-outness has led some researchers to use employer self-reports about the strength of HRM practice.Employer self-reports about the strength of HRM practiceGuest and Peccei (1994) suggest that an improvement on the additive approach is to assess the extent or strength of practice using response bands such as those contained in Likert scales. Such an approach sees employers self-reporting, either on the extent they consider particular HRM practices to be effective (Delaney and Huselid, 1996), or alternatively on the extent to which HRM practices have been operationalised (i.e. to what extent practices are practised) in their workplace (Kane et al., 1999).Guest (2001, p. 1099) claims further re search is needed to test the extent that employer and employee views on HRM practices coincide or differ. Specifically he suggestsThere is now a need for more research canvass the responses of managers responsible for developing and overseeing HRM practice and employees to identify levels of agreement about the operation of practices such evidence as it is possible to glean from reports about levels of autonomy and consultation suggest that levels of agreement might be quite low.Support for this assertion has already been comprise. For example, Appelbaum and Bergs (2000) get wind comparing supervisor and employee perceptions about participative initiatives in US organisations embed wide differences. Similarly, Kane et al. (1999) also order evidence of perceptual differences about HRM between different organisational stakeholders.While managers and employers are central stakeholders, so in addition are employees. This has led to a strong call for employee voice to be heard i n HRM research (Guest, 2001).Employee self-reports about the strength of HRM practiceGiven the apparent perceptual differences between employers and employees perceptions of HRM then a third approach that could be used would see employees reports on the strength of operationalised HRM practice analysed in relation to their levels of work-related attitudes. There is now growing support for assessing HRM from the employee perspective emerging in the publications (Fiorito, 2002 Gibb, 2001 Guest, 2001), and given that where the concern is employee attitudes, then it is intuitively logical to relate that to employee rather than employer perceptions of the effectiveness of HRM policies and practices.MethodThe most common approach assiduous in studies assessing the relationship between HRM practice and employee work-related attitudes is the survey (Cully et al., 2000 Guest, 1999), and this is the method of data collection used in this say. Four scene of actions of HRM practice are exam ined trustworthy and safe working conditions, training and development, equal employment opportunities (EEO), and enlisting and extract. These theaters of HRM have been selected because they have previously been identified as those likely to have the great impact on employee behaviour and attitudes (Guest, 2001). Three employee work-related attitudes are examined organisational commitment, job satisfaction and organisational comeliness. These attitudes are more often than not considered to be desirable outcomes that result from the use of soft HRM (Meyer and Smith, 2000 Guest, 1997).MeasurementThree approaches are used to judge HRM practice an additive pass judgment stainless by the employer, and two perceptual measures, one completed by employers and one completed by employees. either 3 measures relate to practices characterised in the literature as soft HRM or best practice HRM (Guest, 1999 Guest and Peccei, 1992 Johnson, 2000 Kane et al., 1999). They are employee-c entred (i.e. they should produce benefits for employees when operationalised), and thus should impact on employee work-related attitudes.The additive measure, as stated, is the one most commonly used in research of this type. The additive measure used in this study asks employers to indicate, using a yes/no response format, which practices, from a comprehensive list of 80 practices (20 for for all(prenominal) one utilitarian area covered), currently operate in their organisation (for example, Do you consult employees on their training needs?).A 20- level scale consisting of statements about HRM practice is used to assess the strength of HRM practices (five items for each of the four functional areas). Respondents are asked to indicate, using a five-point Likert scale ranging from 1= potently disagree to 5= powerfully agree, the extent they consider each practice occurs in their organisation, for example, EEO is promoted at heart this organisation (see Appendix 1 for a full list of statements). The same scale is used for both the employer and the employee groups, with the wording amended to reflect their different perspectives. Employer and employee assessments of HRM using this measure are, therefore, concerned with the judgements these groups make about the extent a particular HRM practice has been operationalised in the workplace. The coefficients for the five item measures relating to each of the four areas of HRM practice examined ranged from 0.83 to 0.88, suggesting high internal consistency exists (Br givee, 2000). The statements about HRM practice on the perceptual measure closely reflect the ones used in the additive measure, but there is not a direct correspondence as that would have resulted in an excessively long measure.Three widely recognized measures of employee attitudes are used. Guest (1997) suggests organisational commitment should be measured using the modular measure developed by Mowday et al. (1979) the organisational commitment qu estionnaire (OCQ) because this scale captures the extent the employee identifies with the organisation, their desire to remain in the organisation and their willingness to exert effort on behalf of the organisation. An adapted and telescoped version of the OCQ is used in this study to measure organisational commitment. Again, respondents are asked to indicate the extent to which they agree with each of the statements using a five-point Likert scale ranging from 1=strongly disagree to 5=strongly agree (=0.8535).To measure job satisfaction, an adapted and shortened form of the manganese satisfaction questionnaire (Weiss et al., 1967) is used. The scale has cardinal items and respondents are asked to indicate how satisfied they are with respect to each of the statements, using a five-point Likert scale ranging from 1=very dissatisfied to 5=very satisfied (=0.8664).While organisational fairness has two dimensions distributive and procedural this study is concerned only with procedu ral fairness and HRM policy and practice. The measure used is an adapted and shortened six-item version of the procedural justice scale used by Moorman (1991). This measures employees perceptions about the way formal procedures are carried out. Respondents are asked to indicate the extent to which they agree with each of the statements using a five-point Likert scale ranging from 1=strongly disagree to 5=strongly agree (=0.9160).Employer and employee respondents were also asked to provide a range of demographic information. analysisData are analysed using SPSS, Version 11, using matched employer and aggregated employee data from 37 organisations. Correlations, using Pearsons r, are completed to research the relationships between HRM practice and employee attitudes. The multiple linear infantile fixation technique is used to explore the differential impacts of the collar approaches being tested.The take inIn total, 234 organisations were originally contacted to go into in this s tudy. A total of 40 organisations agreed to participate. The person most responsible for operationalising HRM practice was call for to complete the employer survey. Employee participation was voluntary, and the union fully endorsed their participation. The employer participant was requested to distribute the surveys to a representative sample of their men, in equipment casualty of occupational classification, ethnicity and gender. The targeted population of employees consisted of a total of 1,075 full- and part-time employees from the 40 participating organisations. A total of 626 employees responded (a response rate of 58 per cent) by completing the survey and returning it in the reply-paid envelope provided. In total, 37 usable matched sets of data were obtained. These 37 matched data sets comprise one completed employer questionnaire (total usable employer responses=37) and on intermediate 15 employee questionnaires for each organisation (total usable employee responses=572) . The number of actual employee responses received for each organisation change with workplace size, and ranged from eight to 31.The demographics for the employer and employee samples are inform in Table I. naked as a jaybird Zealand is a comparatively base country and organisations employing in excess of 100 employees are usually class ad as large, with those having between 50 and 99 employees classified as medium and organisations with under 50 employees as small. Three-fifths of the organisations participating in this study can be classified as large one-fifth are medium, and the remaining five small. Of the 37 participating organisations, 19 are located in the private sector, 18 in the public sector. around 35 per cent of the participating organisations are located in the service of process industry, with this group being evenly represented crossways both sectors. All other organisations are classified in the manufacturing, transport, retail, sales, education and wellnes s industries.Both the employer and the employee samples are reasonably a representative of organisations and the composition of the workforce in wise Zealand, with the only notable exception being the employee sample is skewed in terms of occupation towards professional and semi-professional groups.ResultsThe first analysis examines the relationship between additive measures of HRM practice and attitudes. The number of practices for each functional area change considerably between organisations. Generally it is found that there are relatively high levels of practice for the areas of favourable and safe working conditions (M=15.95, SD=3.22), training and development (M=16.77, SD=2.76), along with enlisting and selection (M=16.52, SD=2.50). EEO is the one area, on the other hand, where a considerably lower mean level of practice exists (M=12.17, SD=5.96). consort to prior research, high numbers of HRM practice are associated with positive employee attitudes (Guest, 1999). The resul ts of the analysis for this relationship are reported in Table II.These findings differ from those previously reported. Using aggregated employee means, when levels of organisational commitment (M=3.72, SD=0.78), job satisfaction (M=3.72, SD=0.79) and organisational fairness (M=3.70, SD=0.86) are correlated with the combine additive measure of HRM practice, the results show that higher index scores in the number of HRM practices are not related to higher levels of organisational commitment (r=0.155, df=35, p=0.05), job satisfaction (r=0.085, df=35, p=0.05) or organisational fairness (r=0.146, df=35, p=0.05). When this relationship is examined by functional area the correlations are also found not to be significant. bordering the relationship between HRM practice and employee attitudes is examined using first employer self-reports, and second employee self-reports. These results are also reported in Table II.Statistically significant correlations are found to exist between the three attitudes measured and employee assessments of HRM practice for all the functional areas examined. However, no significant correlations are found between employer self-reports about the strength of HRM practice and employee attitudes.Data are then analysed using a regression technique, change the predictive qualities of each of the three approaches for assessing HRM practice to be explored. The model used for regressions is shown in Figure 1.All the measures of HRM practices were entered into a multiple regression1 equivalence so that the character of the relationship between different measures of HRM practice and employee attitudes could be explored. It is found that nearly 51 per cent of the variance in organisational commitment (R 2=0.506, pThe results of the regressions for the three measures of HRM practice, along with the three measures of employee attitudes, are presented in Table III.When the results of the regression for the three different measures of HRM practice are a nalysed it becomes evident that employee assessments have the sterling(prenominal) level of predictive power for attitudes.The regression reveals that numbers of HRM practice and employer assessments about the strength of HRM practice account for very little of the variance in attitudes. On the other hand, employee assessments of HRM practice appear to be very influential. Around 47 per cent of the variance in organisational commitment is explained by employee assessments of practice across the four HRM areas examined, with even larger and equally significant levels of variance being explained by this for job satisfaction and organisational fairness.If we close employee assessments related to EEO from the analysis, as it uniformly records the lowest across all three attitudes, the statistic for the remaining three functional areas of HRM practice implies that for every increment in employee assessments, levels of attitudes increase by between 0.17 and 0.34. The most statisticall y significant and consistent increments are in relation to employee assessments of good and safe working conditions, along with training and development, and these are found to apply to all aspects of work-related attitudes measured. Also worth noting is that, not surprisingly, assessments of recruitment and selection practices accounted for the largest variance in levels of organisational fairness.Discussion and concluding remarksConsistent with the work of Appelbaum and Berg (2000), Cully et al. (2000) and Guest (1999), this study also finds that a significant relationship exists between HRM practice and employee work-related attitudes. There is a strong statistically significant relationship when HRM is researched from the employee perspective. However, there was no significant relationship when HRM practice was measured using the additive approach or employer self-reports. These results have implications for current approaches used to assess the HRM practice-employee work-relate d attitudes relationship.This study demonstrates that by far the greatest correlate of employee attitudes are employees own assessments of the strength of HRM practice. Interestingly, higher levels of HRM practice, using an additive measure, are not found to be significantly associated with attitudes across any of the four functional areas of HRM examined. In fact, it is revealed that higher levels of HRM practice are in some cases related to decreased levels of job satisfaction and organisational fairness. For example, looking at the data for each functional HRM area, it is possible that higher levels of EEO practice may actually have no, or even a negative, effect on employee attitudes, with a strong inverse relationship being found for job satisfaction.Similarly, weak relationships are found when the relationship between employer assessments about operationalised practice and attitudes are examined. So while Guest and Peccei (1994) suggest these areas are important aspects of bes t practice HRM, this study finds them to be unrelated to attitudes when data are collected from employers.Put together these results suggest that if managers want to maximise employee attitudes, then implementing lots of practices is not sufficient. Practitioners need to be aware that the way they implement their HRM practices may be a more important determiner of employee attitudes than the number of practices they put in place. This builds on the research findings of other recent studies (Armstrong, 2001) that suggest it is possible that it is the timbre of the practice that counts and not the quantity. It may be that practitioners may need to undertake regular attitudinal surveys to assess employee reactions to current HRM practice and assess these on a longitudinal basis to identify what is working and what is not.It was beyond the parameters of this study to examine the complex relationship between HRM practice, employee attitudes and business performance it has, however, exp lored the narrower relationship between HRM practice (for the areas of good and safe working conditions, training and development, EEO and recruitment and selection) and three employee attitudes (organisational commitment, job satisfaction and organisational fairness). While regressions tell us nothing about the cause-effect constitution of relationships between assessments of HRM practice and employee attitudes, these findings strongly suggest that incorporating the employees experience of HRM into analyses exploring the HRM practice, employee attitudes relationship will contribute further to our misgiving about what aspects of HRM have the potential to yield the greatest performance outcomes. Arguably, more so, it would appear, than exploring this particular relationship simply using additive measures or employer self-reports.This study has some limitations. This study found the standard deviations for assessments of HRM practice and work-related attitudes amongst employee grou ps within each organisation to be reasonably low, showing close agreement. still the aggregated data has been developed from a relatively small sample, thus it may not be representative of the bulk of employee views. The appraising(prenominal) nature of the questions used introduces the possibility of common method variance (i.e. an simulated correlation across questions collectable to mood or other contaminants (Fiorito, 2002)). Also, the HRM practices used in this study are supposedly best practice however, there is still insufficient evidence available to confirm that they are indeed best practice.Finally, Gibb (2001) suggests effective HRM is about implementing best practice. While this may very well be true, it is clear from this study that simply introducing best practice, or even assessing the effectiveness of practice from the employers viewpoint, is no tell that they have been successfully implemented. If the organisations objective is to improve employees workplace att itudes, then possibly a better way to confirm which HRM practices are having the greatest impact, and hence are effective, is to correlate employees assessment of them with employees attitudes. Further studies in this area are now required.ReferencesAppelbaum, E., Berg, P. (2000), High performance work systems giving workers a stake, in Blair, M., Kochan, T. (Eds),The New Relationship Human Capital in the American Corporation, Brookings Institution Press, Washington, DC, .Manual request InfotrieveArmstrong, G. (2001), The Change Agenda Performance through People, The New People Management, CODE, London, .Manual request InfotrieveBeer, M., Spector, B., Lawrence, P.R., Mills, D.Q., Walton, R.E. (1984), Managing Human Assets, degage Press, New York, NY, .Manual request InfotrieveBrowne, J.H. (2000), Benchmarking HRM practices in healthy work organizations, American Business Review, Vol. 18 No.2, pp.54-61.Manual request InfotrieveCully, M., Woodland, S., OReilly, A., Dix, G. (2000), Br itain at Work As Depicted by the 1998 Workplace Employee Relations Survey, Routledge, London, .Manual request InfotrieveDelaney, J.T., Huselid, M.A. (1996), The impact of human resource management practices on perceptions of organizational performance, Academy of Management Journal, Vol. 39 No.4, pp.949-69.Manual request InfotrieveEhrlich, C.J. (1994), Creating an employer-employee relationship for the future, Human Resource Management, Vol. 33 No.3, pp.491-501.Manual request InfotrieveFiorito, J. (2002), Human resource management practices and worker desires for union representation, in Bennett, J., Kaufman, B. (Eds),The Future of Private Sector trade unionism in the United States, M.E. Sharpe, Armonk, NY, .Manual request Infotrieve(1984), in Fombrun, C., Tichy, N., Devanna, M. (Eds),Strategic Human Resource Management, Wiley, New York, NY, .Manual request InfotrieveGibb, S. (2001), The state of human resource management evidence from employees view of HRM systems and staff, Emplo yee Relations, Vol. 23 No.4, pp.318-36.Manual request InfotrieveGreenberg, J. (1990), Organisational justice yesterday, today and tomorrow, Journal of Management, Vol. 16 pp.399-43Relationship Between HRM and Employee AttitudeRelationship Between HRM and Employee AttitudeThe relationship between HRM practice and employee work-related attitudesIntroductionThe concept of human resource management (HRM) has received focussed attention for around 20 years, with the catalyst being that many US companies found they were being rivalled and in some instances overtaken, in markets they had dominated (Ehrlich, 1994, p. 492). As Lodge (1985, p. 319) observesBy the early 1980s there was still little disagreement that US corporate managers, employees and trade unions would have to change their ways in order to compete successfully for markets in America and abroad.Harvard university academics introduced a new compulsory component of HRM into their MBA syllabus and reinforced this so-called Harva rd Model with influential books and articles (Beer et al., 1984 Walton, 1985b Walton and Lawrence, 1985). While there would surely have been a genuine desire to help US business, US society, and even US employees, there was also a long-term effort to ensure that the Harvard Business School faculty provided leadership in human resource management (Walton and Lawrence, 1985, p. xx).The Harvard concept stresses that HRM should lead to employee commitment not simply as a means to employer objectives of improved productivity and profits, but because the fulfilment of many employee needs is taken as a goal rather than merely a means to an end (Walton, 1985a, p. 49).At about the same time as the Harvard concept was being developed another viewpoint was being promulgated by academics who supported a strategic concept of HRM, with the major work edited by Fombrun et al. (1984). This work emphasises that the four generic human resource activities of all organisationsselection/promotion/place ment processreward processdevelopment process andappraisal process (Tichy et al., 1984, p. 26)need to be strategically aligned with the organisations overall strategic objectives.Hard and soft HRMBritish writers have cerebrate on the differences between the Harvard commitment concept of HRM and the strategic HRM concept. Keenoy (1990, p. 368) sees the Harvard concept as philosophically grounded in the recognition of multiple stakeholders and the belief that the practice and benefits of HRM can be achieved through neo-pluralist mechanisms, while the strategic concept is almost uniformly unitarist in orientation and displays a quite singular endorsement of managerial values. The former is frequently referred to as soft HRM, while the latter is hard HRM.In theory, soft HRM fulfils employee needs as an end in itself, and the favourable attitudes generated from the use of appropriate HRM practices (Guest, 1997) together with communication, motivation and leadership (Storey, 1987, p. 6) , result in commitment to the organisation and improved performance. Hard HRM is only concerned with the effective utilisation of employees (Guest, 2002) and emphasises the quantitative, calculative and business strategic aspects of managing the head count resource in as natural a way as for any other economic factor (Storey, 1987, p. 6).If the reality of soft HRM practice was that it produced the benefits referred to in the rhetoric of academic evangelists such as Richard Walton then it should be highly favoured by both employees and employers. Questions have been raised by a number of researchers, however, about the ability of soft HRM to achieve these benefits. There are two concerns. The first is that the real motive behind its introduction is to undermine unions (Sisson, 1994). Indeed, going back to the introduction of the Harvard concept of soft HRM, Lawrence (1985, p. 362) writes about a seminal colloquium on HRM between 35 senior executives and Harvard academics. He claims t hat the question as to whether HRM was designed to keep the union out was emphatically denied, and he protested that all but three of the firms represented were at least partially unionised, with several having had a significant number of units of both kinds, with their older plants unionised and the newer ones not. An unreasonable interpretation therefore would be that these leaders of commitment HRM will happily dispense with unions given the opportunity. Indeed, Lawrence (1985, p. 362) admits that there was a general view that in a well-managed unit with decision-sharing, a union was not needed to represent employee interests. Guest (1990, p. 389) concludes, the main impact of HRM in the United States may have been to provide a smokescreen behind which management can introduce non-unionism or obtain significant concessions from trade unions.The second concern is that meeting the needs of employees has never been an objective in itself, and has simply been the normative view of w hat Harvard academics would like to see as the employment relationship. Truss et al. (1997, p. 70) in a study of soft and hard models of HRM, concluded that even if the rhetoric of HRM is soft, the reality is almost always hard, with the interests of the organisation prevailing over those of the individual. Similar conclusions have been reached by Keenoy (1990), Poole and Mansfield (1992), Guest (1995) and Legge (1995a, b, 1998).If the soft model of HRM has validity, then there should be a clear relationship between the experience of soft HRM practice and positive employee attitudes (reflecting their needs are being met) and increasing employee commitment as well as improved productivity. Surprisingly, however, most of the research and reporting on HRM has ignored the views of employees. Legge (1998, p. 14) points out when reading accounts of HRM practice in the UK and North America it is noticeable the extent to which the data are (literally) the voices of management. Guest (1999, p. 5) agrees, claiming that from its conception human resource management reflected a management agenda to the neglect of workers concerns. Some studies have considered employee reactions to HRM (Gibb, 2001 Appelbaum and Berg, 2000 Mabey et al., 1998) but Guest (2002, p. 335) is perceptive in his criticism that a feature of both advocates and critics of HRM is their neglect of direct evidence about the role and reactions of workers. Even the exceptions noted which gave employee reactions to HRM did not relate the employee experience of HRM practices to their reactions, and Guest (1997) argues for a research agenda that addresses this gap.This paper aims to contribute to this gap by fulfilling two major objectives. The first is to add to the limited number of studies which have tested to see if there is a relationship between HRM practice and positive employee attitudes. The second objective is to see if there is a significantly stronger relationship between HRM practice and employ ee attitudes if employee perceptions of HRM practice are taken as the measure of HRM rather than employer perceptions, which has been the approach taken in general by previous researchers.HRM practice and employee work-related attitudesThe soft model of HRM, as stated previously, suggests a relationship exists between the use of appropriate HRM practices and positive employee attitudes, and while theoretically these relationships remain poorly developed (Guest, 1997, 2001), a number of attitudes are nonetheless widely considered to be an outcome of soft HRM. For example, levels of job satisfaction, which is the affective perception that results from the achievement of desired outcomes (Harber et al., 1997), are found to be related to levels of HRM practice (Guest, 2002 Ting, 1997). High levels of employee commitment have also been found to be related to the use of appropriate HRM practice (Guest, 2002), and results from investing in HRM practices which benefit employees. For example , the provision of opportunities for training and skill development benefits the employee by equipping them with the necessary knowledge, skills and attitudes to function autonomously and responsibly (Guest, 2002). Furthermore, it improves retention and enables them to cope with change in the work environment (Guest, 2002).Organisational fairness is the term used to describe the role of fairness as it directly relates to the workplace and is concerned with the ways employees determine if they have been treated fairly in their jobs and the ways in which those determinations influence other work-related variables (Moorman, 1991, p. 845). The HRM systems, policies, and procedures that operate in an organisation have been identified as impacting on an individuals perceptions of bias and fairness (Kurland and Egan, 1999 Greenberg, 1990). Studies show that where employees believe they are treated fairly in the workplace then they hold positive attitudes towards the organisation (Moorman, 1991), whereas HRM practices that are perceived to be unfair have been found to result in the employee feeling bewildered and betrayed and thus less committed (Schappe, 1996).As far as the relationship between HRM practice and employee attitudes is concerned, the HRM practice dimension has been measured by collecting data from employers in one of the two ways using additive measures of HRM practice or self-reports about the extent to which particular HRM practices have been operationalised. A possible third approach would be to collect employee views about the operationalisation of HRM practice and then relate these reactions to their attitudes. These approaches are described below.The additive approachTo date the most common approach for assessing the relationship between HRM practice and employee work-related attitudes is the additive approach. This involves employers indicating, using a yes/no response format, which HRM practices from a predetermined list currently operate in th eir organisations. The yes responses are then added together, and the assumption is that a higher sum indicates better HRM. Researchers using the additive approach usually correlate the total number of practices with the HRM outcomes being examined. Research exploring this relationship show a strong correlation between high numbers of HRM practice and positive employee attitudes (Fiorito, 2002 Guest, 1999 Guest and Conway, 2002 Appelbaum and Berg, 2000).While this approach is frequently used, it has attracted criticism (Fiorito, 2002 Guest, 1997, 2002). The additive approach is very simplistic. If an employer claims a practice is utilised there is no differentiation on the basis on how well, or to what extent the practice exists. For example, there could be token training and development, or a high level of commitment to a professional and effective training and development programme. Both examples are credited as a yes. This perceived weakness has led some researchers to use employ er self-reports about the strength of HRM practice.Employer self-reports about the strength of HRM practiceGuest and Peccei (1994) suggest that an improvement on the additive approach is to assess the extent or strength of practice using response bands such as those contained in Likert scales. Such an approach sees employers self-reporting, either on the extent they consider particular HRM practices to be effective (Delaney and Huselid, 1996), or alternatively on the extent to which HRM practices have been operationalised (i.e. to what extent practices are practised) in their workplace (Kane et al., 1999).Guest (2001, p. 1099) claims further research is needed to test the extent that employer and employee views on HRM practices coincide or differ. Specifically he suggestsThere is now a need for more research comparing the responses of managers responsible for developing and overseeing HRM practice and employees to identify levels of agreement about the operation of practices such e vidence as it is possible to glean from reports about levels of autonomy and consultation suggest that levels of agreement might be quite low.Support for this assertion has already been found. For example, Appelbaum and Bergs (2000) study comparing supervisor and employee perceptions about participative initiatives in US organisations found wide differences. Similarly, Kane et al. (1999) also found evidence of perceptual differences about HRM between different organisational stakeholders.While managers and employers are important stakeholders, so too are employees. This has led to a strong call for employee voice to be heard in HRM research (Guest, 2001).Employee self-reports about the strength of HRM practiceGiven the apparent perceptual differences between employers and employees perceptions of HRM then a third approach that could be used would see employees reports on the strength of operationalised HRM practice analysed in relation to their levels of work-related attitudes. Ther e is now growing support for assessing HRM from the employee perspective emerging in the literature (Fiorito, 2002 Gibb, 2001 Guest, 2001), and given that where the concern is employee attitudes, then it is intuitively logical to relate that to employee rather than employer perceptions of the effectiveness of HRM policies and practices.MethodThe most common approach employed in studies assessing the relationship between HRM practice and employee work-related attitudes is the survey (Cully et al., 2000 Guest, 1999), and this is the method of data collection used in this study. Four areas of HRM practice are examined good and safe working conditions, training and development, equal employment opportunities (EEO), and recruitment and selection. These areas of HRM have been selected because they have previously been identified as those likely to have the greatest impact on employee behaviour and attitudes (Guest, 2001). Three employee work-related attitudes are examined organisational commitment, job satisfaction and organisational fairness. These attitudes are generally considered to be desirable outcomes that result from the use of soft HRM (Meyer and Smith, 2000 Guest, 1997).MeasurementThree approaches are used to measure HRM practice an additive measure completed by the employer, and two perceptual measures, one completed by employers and one completed by employees. All three measures relate to practices characterised in the literature as soft HRM or best practice HRM (Guest, 1999 Guest and Peccei, 1992 Johnson, 2000 Kane et al., 1999). They are employee-centred (i.e. they should produce benefits for employees when operationalised), and thus should impact on employee work-related attitudes.The additive measure, as stated, is the one most commonly used in research of this type. The additive measure used in this study asks employers to indicate, using a yes/no response format, which practices, from a comprehensive list of 80 practices (20 for each functional area covered), currently operate in their organisation (for example, Do you consult employees on their training needs?).A 20-item scale consisting of statements about HRM practice is used to assess the strength of HRM practices (five items for each of the four functional areas). Respondents are asked to indicate, using a five-point Likert scale ranging from 1=strongly disagree to 5=strongly agree, the extent they consider each practice occurs in their organisation, for example, EEO is promoted within this organisation (see Appendix 1 for a full list of statements). The same scale is used for both the employer and the employee groups, with the wording amended to reflect their different perspectives. Employer and employee assessments of HRM using this measure are, therefore, concerned with the judgements these groups make about the extent a particular HRM practice has been operationalised in the workplace. The coefficients for the five item measures relating to each of the four areas of HRM practice examined ranged from 0.83 to 0.88, suggesting high internal consistency exists (Browne, 2000). The statements about HRM practice on the perceptual measure closely reflect the ones used in the additive measure, but there is not a direct correspondence as that would have resulted in an excessively long measure.Three widely accepted measures of employee attitudes are used. Guest (1997) suggests organisational commitment should be measured using the standard measure developed by Mowday et al. (1979) the organisational commitment questionnaire (OCQ) because this scale captures the extent the employee identifies with the organisation, their desire to remain in the organisation and their willingness to exert effort on behalf of the organisation. An adapted and shortened version of the OCQ is used in this study to measure organisational commitment. Again, respondents are asked to indicate the extent to which they agree with each of the statements using a five-point Likert scale ranging from 1=strongly disagree to 5=strongly agree (=0.8535).To measure job satisfaction, an adapted and shortened form of the Minnesota satisfaction questionnaire (Weiss et al., 1967) is used. The scale has six items and respondents are asked to indicate how satisfied they are with respect to each of the statements, using a five-point Likert scale ranging from 1=very dissatisfied to 5=very satisfied (=0.8664).While organisational fairness has two dimensions distributive and procedural this study is concerned only with procedural fairness and HRM policy and practice. The measure used is an adapted and shortened six-item version of the procedural justice scale used by Moorman (1991). This measures employees perceptions about the way formal procedures are carried out. Respondents are asked to indicate the extent to which they agree with each of the statements using a five-point Likert scale ranging from 1=strongly disagree to 5=strongly agree (=0.9160).Employer and employee respondents were also asked to provide a range of demographic information.AnalysisData are analysed using SPSS, Version 11, using matched employer and aggregated employee data from 37 organisations. Correlations, using Pearsons r, are completed to explore the relationships between HRM practice and employee attitudes. The multiple linear regression technique is used to explore the differential impacts of the three approaches being tested.The sampleIn total, 234 organisations were originally contacted to participate in this study. A total of 40 organisations agreed to participate. The person most responsible for operationalising HRM practice was requested to complete the employer survey. Employee participation was voluntary, and the company fully endorsed their participation. The employer participant was requested to distribute the surveys to a representative sample of their workforce, in terms of occupational classification, ethnicity and gender. The targeted population of employees consisted of a total of 1,075 full- and part-time employees from the 40 participating organisations. A total of 626 employees responded (a response rate of 58 per cent) by completing the survey and returning it in the reply-paid envelope provided. In total, 37 usable matched sets of data were obtained. These 37 matched data sets comprise one completed employer questionnaire (total usable employer responses=37) and on average 15 employee questionnaires for each organisation (total usable employee responses=572). The number of actual employee responses received for each organisation varied with workplace size, and ranged from eight to 31.The demographics for the employer and employee samples are reported in Table I. New Zealand is a relatively small country and organisations employing in excess of 100 employees are usually classified as large, with those having between 50 and 99 employees classified as medium and organisations with under 50 employees as small. Three-fifths of the org anisations participating in this study can be classified as large one-fifth are medium, and the remaining five small. Of the 37 participating organisations, 19 are located in the private sector, 18 in the public sector. Around 35 per cent of the participating organisations are located in the service industry, with this group being evenly represented across both sectors. All other organisations are classified in the manufacturing, transport, retail, sales, education and health industries.Both the employer and the employee samples are reasonably a representative of organisations and the composition of the workforce in New Zealand, with the only notable exception being the employee sample is skewed in terms of occupation towards professional and semi-professional groups.ResultsThe first analysis examines the relationship between additive measures of HRM practice and attitudes. The number of practices for each functional area varied considerably between organisations. Generally it is fo und that there are relatively high levels of practice for the areas of good and safe working conditions (M=15.95, SD=3.22), training and development (M=16.77, SD=2.76), along with recruitment and selection (M=16.52, SD=2.50). EEO is the one area, on the other hand, where a considerably lower mean level of practice exists (M=12.17, SD=5.96).According to prior research, high numbers of HRM practice are associated with positive employee attitudes (Guest, 1999). The results of the analysis for this relationship are reported in Table II.These findings differ from those previously reported. Using aggregated employee means, when levels of organisational commitment (M=3.72, SD=0.78), job satisfaction (M=3.72, SD=0.79) and organisational fairness (M=3.70, SD=0.86) are correlated with the combined additive measure of HRM practice, the results show that higher index scores in the number of HRM practices are not related to higher levels of organisational commitment (r=0.155, df=35, p=0.05), job satisfaction (r=0.085, df=35, p=0.05) or organisational fairness (r=0.146, df=35, p=0.05). When this relationship is examined by functional area the correlations are also found not to be significant.Next the relationship between HRM practice and employee attitudes is examined using first employer self-reports, and second employee self-reports. These results are also reported in Table II.Statistically significant correlations are found to exist between the three attitudes measured and employee assessments of HRM practice for all the functional areas examined. However, no significant correlations are found between employer self-reports about the strength of HRM practice and employee attitudes.Data are then analysed using a regression technique, enabling the predictive qualities of each of the three approaches for assessing HRM practice to be explored. The model used for regressions is shown in Figure 1.All the measures of HRM practices were entered into a multiple regression1 equatio n so that the character of the relationship between different measures of HRM practice and employee attitudes could be explored. It is found that nearly 51 per cent of the variance in organisational commitment (R 2=0.506, pThe results of the regressions for the three measures of HRM practice, along with the three measures of employee attitudes, are presented in Table III.When the results of the regression for the three different measures of HRM practice are analysed it becomes evident that employee assessments have the greatest level of predictive power for attitudes.The regression reveals that numbers of HRM practice and employer assessments about the strength of HRM practice account for very little of the variance in attitudes. On the other hand, employee assessments of HRM practice appear to be very influential. Around 47 per cent of the variance in organisational commitment is explained by employee assessments of practice across the four HRM areas examined, with even larger and equally significant levels of variance being explained by this for job satisfaction and organisational fairness.If we exclude employee assessments related to EEO from the analysis, as it uniformly records the lowest across all three attitudes, the statistic for the remaining three functional areas of HRM practice implies that for every increment in employee assessments, levels of attitudes increase by between 0.17 and 0.34. The most statistically significant and consistent increments are in relation to employee assessments of good and safe working conditions, along with training and development, and these are found to apply to all aspects of work-related attitudes measured. Also worth noting is that, not surprisingly, assessments of recruitment and selection practices accounted for the largest variance in levels of organisational fairness.Discussion and concluding remarksConsistent with the work of Appelbaum and Berg (2000), Cully et al. (2000) and Guest (1999), this study also fi nds that a significant relationship exists between HRM practice and employee work-related attitudes. There is a strong statistically significant relationship when HRM is researched from the employee perspective. However, there was no significant relationship when HRM practice was measured using the additive approach or employer self-reports. These results have implications for current approaches used to assess the HRM practice-employee work-related attitudes relationship.This study demonstrates that by far the greatest correlate of employee attitudes are employees own assessments of the strength of HRM practice. Interestingly, higher levels of HRM practice, using an additive measure, are not found to be significantly associated with attitudes across any of the four functional areas of HRM examined. In fact, it is revealed that higher levels of HRM practice are in some cases related to decreased levels of job satisfaction and organisational fairness. For example, looking at the data for each functional HRM area, it is possible that higher levels of EEO practice may actually have no, or even a negative, effect on employee attitudes, with a strong inverse relationship being found for job satisfaction.Similarly, weak relationships are found when the relationship between employer assessments about operationalised practice and attitudes are examined. So while Guest and Peccei (1994) suggest these areas are important aspects of best practice HRM, this study finds them to be unrelated to attitudes when data are collected from employers.Put together these results suggest that if managers want to maximise employee attitudes, then implementing lots of practices is not sufficient. Practitioners need to be aware that the way they implement their HRM practices may be a more important determinant of employee attitudes than the number of practices they put in place. This builds on the research findings of other recent studies (Armstrong, 2001) that suggest it is possible that it is the quality of the practice that counts and not the quantity. It may be that practitioners may need to undertake regular attitudinal surveys to assess employee reactions to current HRM practice and assess these on a longitudinal basis to identify what is working and what is not.It was beyond the parameters of this study to examine the complex relationship between HRM practice, employee attitudes and business performance it has, however, explored the narrower relationship between HRM practice (for the areas of good and safe working conditions, training and development, EEO and recruitment and selection) and three employee attitudes (organisational commitment, job satisfaction and organisational fairness). While regressions tell us nothing about the cause-effect nature of relationships between assessments of HRM practice and employee attitudes, these findings strongly suggest that incorporating the employees experience of HRM into analyses exploring the HRM practice, employee a ttitudes relationship will contribute further to our understanding about what aspects of HRM have the potential to yield the greatest performance outcomes. Arguably, more so, it would appear, than exploring this particular relationship solely using additive measures or employer self-reports.This study has some limitations. This study found the standard deviations for assessments of HRM practice and work-related attitudes amongst employee groups within each organisation to be reasonably low, showing close agreement. Nonetheless the aggregated data has been developed from a relatively small sample, thus it may not be representative of the bulk of employee views. The evaluative nature of the questions used introduces the possibility of common method variance (i.e. an artificial correlation across questions due to mood or other contaminants (Fiorito, 2002)). Also, the HRM practices used in this study are supposedly best practice however, there is still insufficient evidence available to confirm that they are indeed best practice.Finally, Gibb (2001) suggests effective HRM is about implementing best practice. While this may very well be true, it is clear from this study that simply introducing best practice, or even assessing the effectiveness of practice from the employers viewpoint, is no guarantee that they have been successfully implemented. If the organisations objective is to improve employees workplace attitudes, then possibly a better way to confirm which HRM practices are having the greatest impact, and hence are effective, is to correlate employees assessment of them with employees attitudes. Further studies in this area are now required.ReferencesAppelbaum, E., Berg, P. (2000), High performance work systems giving workers a stake, in Blair, M., Kochan, T. (Eds),The New Relationship Human Capital in the American Corporation, Brookings Institution Press, Washington, DC, .Manual request InfotrieveArmstrong, G. (2001), The Change Agenda Performance through P eople, The New People Management, CODE, London, .Manual request InfotrieveBeer, M., Spector, B., Lawrence, P.R., Mills, D.Q., Walton, R.E. (1984), Managing Human Assets, Free Press, New York, NY, .Manual request InfotrieveBrowne, J.H. (2000), Benchmarking HRM practices in healthy work organizations, American Business Review, Vol. 18 No.2, pp.54-61.Manual request InfotrieveCully, M., Woodland, S., OReilly, A., Dix, G. (2000), Britain at Work As Depicted by the 1998 Workplace Employee Relations Survey, Routledge, London, .Manual request InfotrieveDelaney, J.T., Huselid, M.A. (1996), The impact of human resource management practices on perceptions of organizational performance, Academy of Management Journal, Vol. 39 No.4, pp.949-69.Manual request InfotrieveEhrlich, C.J. (1994), Creating an employer-employee relationship for the future, Human Resource Management, Vol. 33 No.3, pp.491-501.Manual request InfotrieveFiorito, J. (2002), Human resource management practices and worker desires for union representation, in Bennett, J., Kaufman, B. (Eds),The Future of Private Sector Unionism in the United States, M.E. Sharpe, Armonk, NY, .Manual request Infotrieve(1984), in Fombrun, C., Tichy, N., Devanna, M. (Eds),Strategic Human Resource Management, Wiley, New York, NY, .Manual request InfotrieveGibb, S. (2001), The state of human resource management evidence from employees view of HRM systems and staff, Employee Relations, Vol. 23 No.4, pp.318-36.Manual request InfotrieveGreenberg, J. (1990), Organisational justice yesterday, today and tomorrow, Journal of Management, Vol. 16 pp.399-43