eResearch: the open access repository of the research output of Queen Margaret University, Edinburgh This is an author-formatted version of document published as: McGuire, D. , Garavan, T. N. , Saha, S. & O’Donnell, D. (2006): “The impact of individual values on human resource decision-making by line managers”, International Journal of Manpower, Vol. 27, No. 3, p. 251 – 273. Accessed from: http://eresearch. qmu. ac. uk/265/ Repository Use Policy
The full-text may be used and/or reproduced, and given to third parties for personal research or study, educational or not-for-profit purposes providing that: • • • The full-text is not changed in any way A full bibliographic reference is made A hyperlink is given to the original metadata page in eResearch eResearch policies on access and re-use can be viewed on our Policies page: http://eresearch. qmu. ac. uk/policies. html Copyright © and Moral Rights for this article are retained by the individual authors and/or other copyright owners. http://eresearch. qmu. ac. uk
THE IMPACT OF INDIVIDUAL VALUES ON HUMAN RESOURCE DECISIONMAKING BY LINE MANAGERS David McGuire, Thomas N. Garavan, Sudhir K. Saha ; David O’Donnell David McGuire* School of Management, Napier University Business School, Craiglockhart Campus, 219 Colinton Road, Edinburgh, EH14 1DJ, Scotland E-mail: [email protected] edu [email protected] com Thomas N. Garavan Department of Personnel and Employment Relations Kemmy Business School, University of Limerick, Ireland Email: thomas. [email protected] ie Sudhir K. Saha Professor of Human Resource Management and Organisational Behaviour Memorial University of Newfoundland St.
John’s, Canada Email: [email protected] ca David O’Donnell, Intellectual Capital Research Institute of Ireland Ballyagran, Limerick, Ireland Email: david. [email protected] com Reference as: McGuire, D. , Garavan, T. N. , Saha, S. ; O’Donnell, D. (2006): “The impact of individual values on human resource decision-making by line managers”, International Journal of Manpower, Vol. 27, No. 3, p. 251 – 273. Brief autobiographical notes: David McGuire David McGuire is lecturer in HRD at School of Management, Napier University, Edinburgh, Scotland.
His PhD dissertation on the relationship between individual and managerial values and human resource decision-making was completed at the University of Limerick (2004). He has just returned from the U. S. where he was a Fulbright Scholar in HRD at Oakland University, Michigan. His main research interests are in values-based decision-making, boundary-setting in HRD and critical theory. E-mail: d. [email protected] ac. uk Thomas N. Garavan Thomas N. Garavan is senior lecturer in HRD at the Department of Personnel and Employment Relations at the University of Limerick, Ireland.
He is author of six leading textbooks, over sixty academic articles, Editor of the Journal of European Industrial Training for the past seven years and Associate Editor of Human Resource Development International for the past three years. His main research interests are in workplace learning, management development and cross-cultural HRD. E-mail: thomas. [email protected] ie Sudhir K. Saha Sudhir K. Saha is Professor of Human Resource Management and Organisational Behaviour at Memorial University of Newfoundland, Canada.
He has conducted extensive trans-national research on managerial values and employment equity issues and is a regular contributor on international teaching assignments. E-mail: [email protected] ca David O’Donnell David O’Donnell, Intellectual Capital Research Institute of Ireland, is an interdisciplinary researcher in the fields of intellectual capital, organization theory, HR, and critical management studies. He has a particular interest in bringing insights from the Frankfurt school of critical theory, particularly the work of Jurgen Habermas, into discourse on the knowledge economy and society. E-mail: david. [email protected] com The impact of individual values on human resource decisionmaking by line managers Keywords Human resource decision-making, Line managers, Individual values Abstract Research on values has provided significant insights at individual, organizational and societal levels of analysis. One area that remains under-explored is how the individual values of managers influence decision-making on human resource (HR) issues. This article explores this relationship between the individual values of managers and HR decision-making by drawing on data collected from Canadian and Irish line managers.
Questionnaire data was collected from a total of 340 managers. Results provide modest support for the proposed model in that capability values were shown to be a significant positive predictor of the importance of health and safety, and peace values were a significant positive predictor of the importance of employment equity. The findings emphasise the need to simultaneously examine both individual values and organisational factors as predictors of human resource decision-making. Introduction Research on values has provided significant insights at individual, group, organizational and societal levels of analysis.
Such research is primarily premised on the notion, prevalent since Aristotle, that values provide strong explanations of behavior. A vast literature now exists, attesting to the influence of values on individual behaviour in groups, organisations and society (Brief, 1998; Kleindorfer et al. , 1993; Munson and Posner, 1980). At the group level of analysis, Earley (1989, 1993) found that the degree of collectivistic values displayed by a group will affect individual contributions to the group; at the organisational level of analysis, Finegan 2000) found that the affective values of an organisation will significantly influence the affective commitment of individual employees within that organisation, while at the societal level, Hofstede’s (1980) seminal work notes the significance of values in explaining the productivity of nations. There is, however, little published research exploring the role of individual values in explaining the decisions that managers make concerning key human resource (HR) issues (Saha and Fisera 1999).
This article explores this relationship between the individual values of managers and HR decision-making by drawing on data collected from line managers in Canada and Ireland. The structure of the remainder of the paper is as follows; first we present a review of the relevant literature on values; the theoretical model is then proposed; the methodology is then outlined followed by a presentation and discussion of the main findings; the paper concludes by arguing for further research on he individual values of managers and identifies several potentially rewarding avenues for conducting such research. What are values? Values are difficult to define because they share similar characteristics with concepts such as attitudes, preferences and viewpoints. Rokeach (1973), for example, defined an attitude as an organization of several beliefs focused on a specific object or situation whereas a value is an enduring belief in a specific mode of conduct or desirable end state of existence.
A number of research studies demonstrate that attitudes and values are different constructs, not only in the minds of researchers, but also in the minds of practitioners (Hofstede, 1998). Attitudes are most likely to vary in terms of the level of importance attached to the object or situation. They differ from values primarily due to their lifespan within an individual’s cognitive schemata. Values can influence the holding of certain attitudes; however, unlike attitudes, values are considered imperative for action (Bates et al. 2001). This latter point leads to the distinction between values and norms. Norms have an obligatory character; values, on the other hand, have the element of free choice towards teleological or goal directed action (Habermas, 1996; Turner, 1991). Individual values are addressed in decision-making and, therefore, can be expected to influence managerial decision-making on HR issues at deeper more personal levels. The fact that orms and values differ in this way has obvious implications for giving preference to a study of individual values, as distinct from norms, in managerial decision-making on HR issues. The literature on values theories (Schwartz and Bilsky, 1990; Rokeach and BallRokeach, 1989; Sagie and Elizur 1996; Elizur 1996) places emphasis on the stability of values and their structure. Values are acquired and hierarchically organized to become somehow part of a relatively enduring system.
Schwartz (1994) argues that values are structured through a combination of social and psychological conflicts, experienced by the individual expressing these values. Over time, these conflicts and harmonies among value priorities result in a structure or values system—such structures providing an individual with an ordered framework for decision-making. It follows that values must influence decision-making. It is now generally accepted that variations in values structures motivate individual behaviors and impact upon actions in both managerial and non-managerial contexts.
In terms of decision-making, there is increasing consensus that a good decision maker is not just someone who can rationally assess choices, or make the most logical decision given for that moment, but someone who can arrive at a decision through balancing competing values and goals (Hartmann & Patrickson 1998; Cohen, 1993). Traditional approaches view decision-making as rational, resulting from a deliberate and intentional choice taking into account structural constraints that restrict the set of possible action courses to a subset of feasible actions (Zafirovski 1999).
Such approaches however, tended to dissociate the processes of decision making from the people that made decisions (Hambrick and Mason, 1984; Simons and Thomson 1998) Consequently, contemporary perspectives on decision-making increasingly acknowledge a socially constructed reality that emphasizes the contextualised nature of decision-making and the important role played by values in driving decision-making processes (Tacconi 1996). Research evidence (Kleindorfer et al. 993; Brief, 1998) suggests that major decisions are generally referred to an individual’s values for approval. This process of referral leads to the selection of the most favourable decision option or the formulation of a hypothesised solution. An individual’s internalised values function as personal standards of conduct (Kluckholn, 1951; Meglino and Ravlin, 1998) and the strength of such values influences the perceived level of attractiveness or perceived legitimacy of alternative actions.
The link between individual values and managerial decision-making is considered to be dynamic, complex and less well understood. Gamble and Gibson (1999) argue that values develop as individuals are exposed to layers within a social system (family, work, employment) and a range of these external factors must also be taken into account in examining causal relationships. Failure to study values in the context of other variables may be one reason for this dearth of robust research findings (Shrum and McCarthy, 1992).
Some studies have highlighted antecedents such as age, gender, education, and the moderating influence of culture and group orientation in examining the effect of values on various decision outcomes (Bigoness and Blakely, 1996; Erez and Earley 1987; Giacomino and Akers, 1998; Johnson and Elder 2002; Wagner 1995;). Becoming more specific, there is considerable justification for studying values in the context of human resource management (HRM).
The HRM field has increasingly become values-driven with employee-centred HRM practices increasingly advocated. This supposed shift from hierarchical and directive to more people centered and participatory managerial styles suggests a change in managerial mind-sets and the individual values of managers. The transition from rigid bureaucratic structures to more flexible adaptive organizations has been accompanied by some shift in management styles from hierarchic traditions to more human relations oriented expertise (Henderson 1996).
Henderson (1996) highlights four transitions which, we argue, must impact on the individual values of managers: first, managers are departing from resistance and fear of cultural differences towards accepting them; second, there is a noticeable trend away from directive supervisor-subordinate relations towards seeking authentic peer behaviors; third, there is a greater emphasis on collaboration over internal competition; and, finally, there has been a movement away from avoidance or denial of the expression of employee feelings towards facilitating their appropriate and productive expression.
While this may be the case in some organizations, it is certainly not the case in all. In an increasingly individualized and highly competitive global business climate a massive gulf exists between the espoused values sets of managers, not to mention HR policies and practices, of the humanistic ‘mutual gains enterprise’ (Kochan and Osterman, 1994) contrasted with the much more mercenary ‘individualised corporation’ (Ghoshal and Bartlett, 1998). The extent to which such changes are emerging remains an open, and essentially empirical, question.
Ranney and Carlson (1992) argue that an organisation’s HR policies, practices and rules are good indicators of its approach to dealing with people and human behavior. They argue that an organisation’s managerially directed actions affect employee perceptions leading to a strengthening or readjustment of individual values. Similarly, Begley and Boyd (2000) suggest that individual values should be considered in the development of HR policies and strategies leading to greater consistency in organizational decision-making and a more harmonious working environment.
Although there is support for the link between management leadership and values alignment at the organizational level (Bontis and Fitz-enz, 2002), Begley and Boyd (2000), as in too much of the humanist oriented HR discourse (McGuire et al. , 2005), take too one-sided a perspective here by not adequately treating the reality of individualized corporate life where senior managers are as likely to resort to laying off thousands of employees as they are to inquire about their affective states or personal feelings.
Notwithstanding the heterogeneous nature of organizations, very little is really known about how individual values of managers relate to decision-making on HR issues. We do know from extant research that managers operate from a structured set of values that can be described and reasonably measured. We do not yet, however, fully understand how particular individual values impact on particular HR related decisions. This latter issue is of particular concern in the HRM field. Of its nature, HRM is a value laden and value intensive arena of decision-making, simply because it focuses on issues that raise mportant questions concerning equity, equality of treatment, due process, the promotion of individual and organizational outcomes, and the tension between respecting the human in employees and exploiting their human capital and productivity for the cumulative imperative of capitalist organization. HR decisions are, moreover, increasingly devolved to line managers and are no longer the sole preserve of the functionalist personnel or HR department of the industrial era (Gibb, 2002; Bacon et al. 1998).
Many commentators see the increasing responsibility of line managers in all areas including HR as a direct consequence of organizational delayering (White et al. 2004; Torrington et al. 2002). Line managers conduct many operational dimensions of HRM, including selection, promotion, performance management and appraisal, corrective action and discipline, access to training and development opportunities, and compensation. In these decision situations it is likely that the individual values of managers will influence their decision choices.
As front line representatives of the organization, line managers are tasked with the responsibility of setting the agenda, dealing with workplace issues and providing direction to employees. The potential for conflict exists, however, when organizational decision-making conflicts with individual values. In such circumstances, the conflict may end up being resolved through discussion and dialogue, it may be ignored, denied or suppressed, or the manager may simply leave the organization.
We focus in this paper on investigating the influence of individual values on two areas of human resource decision-making, namely (i) employment equity and (ii) health and safety. However, before proceeding to the analysis, we first review the principal studies examining the effects of individual values on various aspects of decision-making. Conceptualising the effect of the individual values of line managers on decision-making In research terms, values may be considered as independent, dependent or moderating variables. In this paper we focus on values as independent predictors of HR decision-making.
There exists some empirical support for this position. A study of individual values and decision styles of expatriates and host country nationals based in the United Arab Emirates revealed a significant correlation between the types of values and decision styles. Using Flowers et al. (1975) values for working instrument, Ali et al. (1995) investigated the relationship between these values and six types of decision styles (autocratic, pseudo-consultative, consultative, participative, pseudo-participative and delegative styles). The study found that socio-centric value systems correlated significantly with consultative (r = 0. 5) and delegative (r = 0. 21) styles. Unsurprisingly perhaps, they found that egocentric values negatively correlated with participative styles (r = -0. 17), but correlated positively with autocratic styles (r = 0. 15). Both conformist and realistic values were found to be significantly correlated with consultative, participative and pseudo-participative styles. The results indicate a strong relationship between individual values and decision styles, but also show that national or societal influences can affect the individual values of managers.
Argrawal and Krishnan (2000) examined the relationship between managerial values and leadership styles. Based upon a sample of fifty manager-subordinates pairs based in South-East India, they found that high task oriented leaders ranked achievement values significantly higher than low-task leaders. High relations leaders gave significantly higher rankings to security and benevolence and lower rankings to self-direction and power as compared to low relations leaders. Task oriented leadership style was significantly related to both achievement and benevolence.
Relations-oriented leadership was significantly related to benevolence, security and achievement. Sharfman et al. , (2000) examined the effects of managerial values on social issues evaluation. A total of 129 managers from the US participated in the study. They found a clear relationship between the issues evaluated as important and the values of the manager conducting the evaluation. Economic values oriented managers emphasised both community and regulatory values less than managers of other values orientations.
The contention here, which is supported by Carroll (1993), is that managers with a strong set of economic values will put economic issues ahead of social issues. Sharfman et al. (2000) also found that managers with a strong ethical values focus emphasised community issues, regulatory issues and political issues equally. Similarly, legal values oriented managers emphasise regulatory and political issues equally with community issues. A series of regressional analyses were conducted which showed that managerial values had a moderate to low effect in explaining the evaluation of community issues (R2 = 0. 1) and regulatory issues (R2 = 0. 10). The third regression model, examining the effect of managerial values on the evaluation of political issues, was insignificant indicating that the evaluation of political issues could not be adequately explained, in this research instance, from the managerial values orientation proposed. Keast (1996) examined the relationship between values and the decision-making of 10 CEOs in public schools in Greenland. Using a qualitative research design, the study showed that values played an important role in the decision-making process.
It found that a degree of similarity appears to exist in the frequency with which the same values reoccurred in the decision-making of all ten CEOs. The study also revealed that some values that showed high occurrences in the decision cases were also found to be highly occurring in the follow-up interviews, indicating their high level of prevalence in the CEOs’ value make-up. Shared decision-making occurred most often in organizational redevelopment decision cases and this value, alongside trust, featured most frequently in the follow-up organizational redevelopment interviews.
Sparrow and Wu (1998) examined the relationship between national culture and HRM preferences of Taiwanese employees. Although the study is based upon employees and not managers and uses cultural values rather than individual values as its unit of measurement, the study is noteworthy for its focus on HRM decision-making. National culture accounted for 5% to 10% of the total individual variance in HRM preferences. A model to predict preferences for training and development was significantly linked to the value orientations of high ‘Relationship to Nature: Harmony’ and low ‘Relationship to Nature: Subjugation’.
The model explained 12% of the variance observed. The study revealed that the relationship between value orientations and compensation preferences was weaker than for other HR functions. In terms of employee performance, this was linked to the value orientation of high ‘Relationships: Collateral’ and the resulting model explained 3. 7% of the variance observed. This limited base of empirical studies indicates that there is a significant relationship between the individual values of managers and decision-making processes.
Individual values explain, to a certain degree, a range of organisational decisions, from leadership and decision styles (Ali et al. 1995; Agrawal and Krishnan, 2000) to decisionmaking and social issue evaluation (Sharfman et al. 2000; Keast 1996). Indeed, the study by Sparrow and Wu (1998), although based upon cultural values, demonstrates good support for the relationship between cultural values and HRM decision-making. Methodology Sample and Response Rates A cross-national sample of Canadian and Irish managers participated in the research study.
In relation to the Canadian sample, 200 managers were contacted in a mail-distributed survey from a cross-section of organizations. The managers were mainly alumni of the Masters of Business Administration Programme at Memorial University of Newfoundland or participants in a Management Development Programme. 150 responses were received, indicating a response rate of 75%. A cross-sectional design was also employed in relation to the Irish sample. A total of 250 managers were contacted.
The managers were participants in management development programmes conducted on behalf of two large multinational organisations (one organisation operating in the technology sector and located in the mid-west region and the other organisation operating in the financial services sector operating in Dublin) or were alumni of the Masters of Business Studies (Human Resource Management) programme run by the University of Limerick or the Diploma in Personnel Management programme run by the Galway-Mayo Institute of Technology. 80 questionnaires of a total of 250 questionnaires were administered in the presence of the researcher, while the remaining 70 questionnaires were distributed by mail. A total of 190 responses were received indicating a response rate of 76%. The research instrument was administered to Irish and Canadian participants in the months of January and February 2002. In both the Irish and Canadian data sets, the research instrument was administered either directly (in the presence of the researcher) or was sent to the manager by post.
In the latter case, a stamp-addressed envelope was enclosed with the research instrument. The overall sample can be categorised as a convenience sample. The overall response rate was calculated at 75. 5%. In a seminal study of response rates in business research, Baruch (1999) reports an average response rate of 55. 6%. He examines response rates of five leading journals in the management and behavioural sciences across three specific points in time – 1975, 1985 and 1995. He notes that the average response rate has dropped from 64. 4% in 1975 to 48. 4% in 1995.
The high response rate in this study can be attributed to the mode of administration of the research instrument and the personal connections between the researchers and respondents. The Research Instrument The research instrument comprises of a ten-page questionnaire, divided into three parts. Part 1 consists of an abridged version of Rokeach’s instrumental and terminal value instrument designed to measure the individual values of respondents. In order to fit the context of the study, it was decided to reduce the number of value items in the survey from thirty-six to thirty.
A further consideration related to the validity of some of the scale items given the nature of the research study. The instrument invited respondents to rate the importance of each of the value items on a 5-point likert scale, ranging from 1 (strongly disagree) to 5 (strongly agree). Whilst a rankings approach is normally adopted in relation to Rokeach’s Value Survey, a ratings approach was adopted as ranking (ordinal) measures severely restrict the type of permissible statistical analyses.
Furthermore, there is an increasing volume of research indicating that ratings techniques can be applied to Rokeach’s Value Survey without adverse consequences. Studies by Moore (1975) and Feather (1973, 1975) have found that average preference orders measured by ratings and rankings are quite similar. Rankin and Grube (1980) showed that ratings have a higher predictive validity than rankings. Indeed, both McDonald (1993) and Finegan (2000) have stated that it is possible to change a rank-order task to a ratings task without hurting the integrity of the scale.
Despite the fact that Schwartz’s Value Survey is now widely acknowledged as the seminal instrument in the field (Burroughs & Rindfleisch 2002; Glazer et al. 2004; Steenkamp et al. 1999), a decision was made to use the Rokeach Value Survey as this study forms part of a larger stream of research being conducted by the third-named author using this instrument (Saha 1995, 1996, 1999, 2001). In addition, a number of research studies continue to use the long established Rokeach Value Survey to investigate the value systems of individuals and managers (Murphy and Gordon 2004; Connor & Becker 2003; Lenartowicz & Johnson 2003).
Part 2 consists of two HR decision scenarios designed to establish both the importance that respondents attached to health and safety and employment equity and the rationale underlying decision-making for each decision scenario. The two decision scenarios of health and safety and employment equity were used to examine the decisionmaking of respondents. These two decision scenarios were selected as it was anticipated that both scenarios would be affected by similar types of values (i. e. concern for the individual).
In both instances, the employee occupies the focal point of the decision scenario and decisions rendered signal the level of concern towards the employee and their welfare. According to Duncan and Wack (1994) decision scenarios present an important opportunity to examine decision-making and in particular the range of possibilities open to managers when making decisions They argue that use of this technique in a developmental context can help to avoid the trap of the corporate one-track mind. The use of decision scenarios has been adopted in studies examining choice preferences (Erb et al. 2002), public policy decision-making (Zahir, 1999), hotelier decision-making (Sheel, 1995) and strategic decision-making (Fredrickson, 1984). The facts of the decision scenario in relation to health and safety were presented as follows: The quarterly health and safety report of company X indicates that accidents are up 10 percent from the last quarter and 20 percent from last year. The severity of accidents is also up. In the past, the company has tried to improve safety by using posters and talks from supervisors.
A government inspector recently visited the company for health and safety improvement. The inspector has advised the company to improve the current safety record. Should Company X invest money to improve safety is the decision that needs to be made now. Your inputs to this decision will be greatly appreciated. An investment equivalent to 2% of X company’s present total overhead needs to be made in safety training, safety contests and purchase of safety equipment and fixtures. According to experts, these measures would reduce accidents and injuries.
Respondents were asked to rate the importance of this issue on a scale ranging from 1 (extremely important) to 7 (not at all important). The question was then posed to respondents as to whether they would recommend no funding, partial funding or full funding for the health and safety plan described in the decision scenario. The decision scenario on employment equity presented respondents with the following set of circumstances: Company X hires employees based upon selection criteria that appear to be in favour of white males.
Company records show that a disproportionate number of white males have the best jobs. Two candidates, a white male and a white female are finalists for a job in the company. The department manager wants to hire the white male because he has better scores. The human resource department is pushing the female candidate to satisfy the requirements of employment equity. Again respondents were asked to rate the importance of this issue on a scale ranging from 1 (extremely important) to 7 (not at all important). The choice was then posed as to whether they would hire the male or female candidate.
In the third section of the research instrument, demographic and human capital information across five dimensions was collected on respondents to the study. Personal information was gathered from respondents regarding their gender, marital status, educational attainment, managerial position and the type of company in which they are employed. Table I shows that gender composition was very similar across the Canadian and Irish sample. The majority of respondents in both Canada and Ireland were married, with a higher proportion of single Irish respondents.
In terms of education, most Canadian respondents were university postgraduates, whereas most Irish respondents had completed a university undergraduate programme. An examination of the positions held by respondents showed that in the case of Canadian respondents, most held middle and upper managerial positions, whereas Irish respondents held middle and lower management positions. Finally, 63. 4% of Canadian respondents worked for local companies, whereas 83. 7% of Irish respondents were employed by multinational companies. Insert Table I about here] Data Analysis The data set was analysed using the Statistical Package for Social Sciences (Version 11. 0).
Univariate analyses (Frequency Distributions), bivariate analyses (Crosstabulations, Analysis of Variance (ANOVAs)) and multivariate analyses (Simple Correlations, Partial Correlations and Regression Analysis) were undertaken to examine the relationship between the independent and dependent variables. A factor analysis was the instrument for data reduction purposes. As there were no prior expectations relating to the factor structure, principal components actor analysis was the type of factor analysis selected. Research suggests that components and common factor analyses produce essentially the same results (Velicier et al. , 1982; Arrindell and Van Der Ende, 1985). Varimax rotation was the choice of rotation selected. The results of this analysis, presented in Table II, indicate the presence of eight distinct factors. All items exhibited a factor loading of greater than 0. 30. The eight factor solution accounted for 57. 7% of the variance and all eight factors have an eigenvalue of greater than one.
The eight-factor solution was examined using the Kaiser-Meyer-Olkin (KMO) measure of sampling adequacy and the Bartlett test of sphericity. The KMO test produced a statistic of 0. 839 indicating that factor analysis could usefully be performed on the data. The Bartlett test of sphericity resulted in a significance level of less than 0. 001, indicating the presence of strong significant relationships among the scale items. The Cronbach alpha for the complete scale was calculated at 0. 781. The Cronbach scores for the eight factors produced range from 0. 735 to 0. 476. [Insert Table II about here]
Results A summary of the means, standard deviations and intercorrelations is provided in Table III. It shows that capability values and relationship values are strongly correlated with the importance of health and safety and that peace values, tolerance values, selfrealisation values and principle values are strongly correlated with the importance of employment equity. There is also a high degree of intercorrelation amongst the values factors. [Insert Table III about here] Table IV presents two regression models and the beta values, R2 values and adjusted R2 statistics for each model.
For the purposes of the regression analysis, several dummy variables were created to permit the inclusion of demographic and human capital variables. As there was no a priori knowledge of the expected outcomes stepwise regression was the form of regression modelling used. In the first model predicting the importance of health and safety, the three variables of education, employer status and capability values accounted for 10. 2% of the overall variance. Employer status was found to be the strongest predictor in the model accounting for over half of the variance explained.
Education was identified as a significant negative predictor of the importance of health and safety, indicating that the higher educated the person, the lower the perceived importance of health and safety issues. The remaining variable, capability values explained 3. 5% of the overall variance. The F statistic of 5. 381 (p < 0. 001) confirmed the presence of significant predictor variables within the model. [Insert Table IV about here. ] In terms of recommending funding for health and safety, 80. 8% (N = 261) of respondents recommend full funding for health and safety.
A further 18. 6% (N = 60) of respondents recommend partial funding for health and safety. Only two respondents (0. 6%) recommend no funding for health and safety. In the second model, three variables were identified as predictors of the importance of employment equity. In all, the independent variables accounted for 12. 7% of the variance observed. Accounting for 5. 4% of the total variance, the nationality of respondents proved to be the strongest predictor of the importance of employment equity.
This points to a significant cultural effect, or perhaps broad societal effect (O’Donnell et al. , 2001), in the perceived importance of employment equity, with Irish respondents placing a higher importance on matters of employment equity. Gender was identified as a significant negative predictor of employment equity, indicating that female respondents place a higher priority on employment equity than male respondents. Peace values accounted for the remaining 4. 6% of the variance observed in the model. The F statistic of 13. 075 (p