Lessons from Sri Lanka
Anil Chandrakumara, Business Administration Department, Faculty of Management Studies and Commerce, University of Sri Jayewardenepura, Colombo, Sri Lanka
Paul Sparrow, Professor of International HRM, Manchester Business School, Manchester, UK
The authors thank David Collings and Michael Morley, HRM Research Group at Limerick University for their encouragement and all the support extended throughout and for anonymous reviews for their valuable comments on the first draft of this paper.
This study extends the contention that national culture affects human resource management (HRM) policies and practices and explores meaning and values of work orientation (MVWO) as an element of national culture in predicting HRM policy-practice design choices. The data were obtained in a sample of 487 employees in domestic and foreign-invested firms (FIF) in Sri Lanka. Eight distinct MVWO patterns emerged from the sample. Twenty-six HRM design choices were clustered into four components: planned and open career and empowering system, qualifications and performance based reward system, generic functional perspective of job-person fit, and job-related competence and rewards. All the four HRM preference practices are influenced by MVWO. The evidence suggests MVWO relativity of HRM design choices in Sri Lankan context. The question of transferability of empowering and performance management to developing countries becomes evident. Moreover, MVWO relativity of HRM design choices is relatively high in FIF, reflecting that the “type of ownership” can have an impact not only on actual HRM practices but also on preferred HRM practices in FIF. The existence of business in the long-run and host government expectations also seem to be important factors in understanding HRM preferences in FIF. Theoretical and practical implications for international HR managers are discussed.
National cultures; Developing countries; Sri Lanka; Human resource management.
International Journal of Manpower
Emerald Group Publishing Limited
World economies are increasingly exposed to international trade, global competition and capital flows, and marketisations. At the same time, many scholars and practitioners today regard human resource management (HRM) as the source of sustained competitive advantage for organisations operating in a global economy (Aycan et al., 2000; Schuler, 2000; Sparrow and WU, 1998). In line with these developments, there has been a great deal of recent research interest in the field of international and comparative HRM. National culture has generated more interest in these fields of HRM because of its impact on HRM practices (Aycan et al., 2000) and the design of HRM policy practices preferences (Nyambegera et al., 2000; Sparrow and Wu, 1998). However, the majority of research has concentrated on differences across developed economies or between newly industrialized countries. The circumstances confronting developing countries have generally been ignored (Budhwar and Debrah, 2001; Nyambegera et al., 2000). To develop the area of comparative HRM, it is, therefore, necessary to understand HRM in the context of developing countries. The significance of examining cultures and HRM practices in these countries is understood by the fact that employees in developing countries hold values completely different from those in developed countries (Mendonca and Kanungo, 1990; Nyambegera et al., 2000; Sparrow and Budhwar, 1997). This has led to the question of transferability of HRM paradigms adopted mainly in developed countries to developing countries. Therefore, some recent researchers have paid a particular attention to HRM issues and socio-cultural characteristics in developing countries (McCourt, 2001; Nyambegera et al., 2000; Wright et al., 2000). Theoretically, one of the major issues associated with the concept of culture is that the dilemma regarding what factors to be included under the broad concept of national culture needs to be resolved (Budhwar and Sparrow, 2002; Cray and Mallory, 1998). Broadly, the aim of this paper is to address these issues in relation to a developing country.
Sri Lanka is a developing country in the South Asia. It is one of the first developing countries to understand the importance of investing in human resources and promoting gender equality. As a result, Sri Lanka has achieved human development outcomes (e.g. literacy rate around 92 per cent) more consistent with those of high-income countries (The World bank, 2000). It liberalised its economy in the late 1970s ahead of other developing nations. Sri Lanka today is South Asia's most opened economy and therefore known as “the Gate Way to South Asia” by many investors. Its per capita income (US$ 900), almost double than that of India remains the highest in the region, after Maldives. The country has adopted the strategy of export-led model of economic development and foreign investments. More than 500 foreign investors from different part of the world, including investment from Fortune-500 companies, have invested in Sri Lanka for past 10 years. Multinational and global companies have been seen as contributing to take home cultures to foreign locations and the existence of such companies within national cultures creates cultural diversities within nations (Punnett, 1998). On the other hand, the entry of foreign investment into developing countries involves tension related to HR utilisation due to divergent contextual factors (Zaffane, 1994). The question arises as to what HRM issues are of most relevance in this country. There is a very little empirical research in examining culture's impact on HRM practices has been carried out in Sri Lanka. The recent renewed interest in these developing countries as potential markets for both international trade and foreign investment makes this type of study worthwhile. Researchers have call for increased efforts directed toward studying such markets (Nyambegera et al., 2000; Sparrow and Budhwar, 1997; Thomas and Philip, 1994). As Frost and Cyr (1990) note, in future, particularly where change is rapid and cultures interact (or even collide), it will be companies and countries that value and ensure effective HRM that will likely gain a strategic edge over those that do not do so. Sri Lanka is a country having such a dynamic culture in which rapid transformation of economic and business activities taking place for last two decades in an opened and liberalised market environment which was introduced in 1978.
Sri Lankas population of 19 million consists of three major ethnic groups: Singhalese (74 per cent), Tamils (17 per cent) and Muslims (8 per cent). Its main religions are Buddhist (69 per cent), Hindu (15 per cent), Christian (8 per cent), and Muslim (8 per cents). It is therefore rich in its diversity of culture, race, language and religion, with a recorded history spanning over 2,500 years. In addition to the influence of a traditional Indian civilisation and cultural phenomenon, Sri Lankan society and culture has been influenced by its colonial heritage and development policies, and the strategies of post-independence governments. The cultural patterns of the society are, therefore, intricately made up of components, which intermesh and influence each other. Interestingly, Sri Lankan exhibit many Asian traits in family and other social interaction, but they tend to mix Asian traits with Western management philosophies and practices in business management. For example, Nanayakkara (1993) notes that many Sri Lankan managers maintain power distance and an individualistic cultural traits. They, however, tend to exhibit collective traits in family and other social interactions. It has also been observed that Sri Lankan management practices evolved from the British system prior to its political independence and has adopted mostly similar Western management practices since independence (Nanayakkara, 1992). Joiner (2001) also notes that progressive firms in the developing countries tend to mimic the practices of successful organisations from more industrialised nations, without reference to surrounding societal values. According to Joiner, the implementation of such cultural changes may jeopardise the success of that change. On the other hand, there is enormous pressure exerted on many organisations to effect changes to enable them to compete successfully in a borderless world. Reflecting these realities, some researchers and national organisations in Sri Lanka have stressed the importance of exploring organisationally/industrially important cultural values and modifying HR practices and work ethics (WE) to overcome some of the HR outcomes (such as low commitment, low productivity, poor relationship between employees and management, etc.) that prevail in many private and public sector organisations (The Central Bank of Sri Lanka, 1999; Vishnath, 1997). There is abundant evidence to indicate that the activity of working and related ethics and values are of major significance to individuals in industrial societies (Buchholz, 1978; MOW International Research Team, 1987; England and Harpaz, 1990; Ruiz-Quintanilla and England, 1996). These work and value orientations represent industrially or organisationally important values. We conceptualised meaning and values of work orientation (MVWO) to represent these variables.
MVWO as an element of national culture
This study uses MVWO as an element of national culture value orientations because of the theoretical reasons explained in this section and the contextual reasons cited above. Literature review revealed that the concept of MVWO could represent definition of working, work centrality (WC), work norms, WE, and work values. There is substantial theoretical and empirical evidence to indicate the fact that the concept of MVWO as an element of national culture (Budhwar and Sparrow, 2002; Claes and Ruiz-Quintanilla, 1995; Dickson and Buchholz, 1977; MOW International Research Team, 1987; Schwartz, 1999). For example, MOW International Research Team (1987) study reports that meaning of working is representative of national cultures as it found national differences in working definition patterns, WC and work norms. Claes and Ruiz-Quintanilla (1995) also found that respondent's country and age have an impact on work meaning patterns. Dickson and Buchholz (1977), in their study in Scotland and the USA, considered work related beliefs as a representative of national culture. Schwartz (1999) suggests that WC, societal norms about working and work values can be used as a way towards utilizing what is known about national differences in cultural values for the study of national differences in work-related values. Budhwar and Sparrow (2002) also conceptualised meaning and values of work as an element of national culture in their model of comparative and international HRM. Thus, in this study we take national culture perspective of MVWO to explore its impact on HRM policy-practice preferences. Individual preference approach to examining the relationship between cultural values and HRM preferences seems to have a great significance for countries or cultures where vast majority of management practices have been uncritically adopted (Aycan et al., 2000; Sparrow and Wu, 1998).
Impact of value orientations on HRM preference practices
We have already explored national culture values relativity of MVWO. People use values, consciously or unconsciously, as standards or criteria for choosing among alternative courses of goals, interest or actions (Dose, 1997). We use the term “preferences” to represent these alternative choices among which people may select according to their values, standards or criteria. According to Locke (1976), individuals differ what they value and this determines their choices. Thus, values influence preferences or choices. The existing individual difference literature under the label of “work-related preferences” has attempted to glean relevant messages to the field of international HRM from the study of national differences based on the socio-cultural characteristics of individual employees (McGaughey et al., 1997). More importantly, the cultural values represent unique measurements, and could be separated from attitudinal measures (Nyambegera et al., 2000; Sparrow and Wu, 1998). Therefore, a more meaningful examination of psychological work-related preferences and their impact on HRM policy practice can be carried out by tapping into the influence of the MVWO items.
The review of literature revealed that the concept of MVWO was associated with three major concepts:
- meaning/definition of working, WC, and societal norms about working;
- WE and beliefs; and
- work values.
Definition of working represents the degree to which people define working in terms of a burden, a constraint, a responsibility, or a social contribution. WC refers to the degree of importance that working has in the life of the individual and the extent to which people put work above other aspects of their life (MOW International Research Team, 1987). Societal norms about working represent “normative evaluation about work in terms of standard of one's duty to and right from the society” (MOW International Research Team, 1987).
The concept of work-related ethics and beliefs in this study refers to the underlying concepts or beliefs people holds about working. This includes WE, organisational beliefs (OB), humanistic beliefs (HB), Marxist related beliefs MRB), and leisure ethics (LE) (Buchholz, 1978). WE represent the beliefs that work is good in itself, offers dignity to a person and that success is a result of personal effort. OB refer to the view that work takes on meaning only as it affects the organisation and contribution to one's position at work. The view that individual growth and development in the job is more important than output represents HB. LE regards work as a means of providing provisions to pursue leisure goals. MRB represent the opinion that work is fundamental to human fulfilment but as currently organised it represents exploitation of the worker and consequent alienation. Work locus of control (WLOC) is defined as a generalised belief and expectancy that rewards reinforcement and outcomes in life are controlled either by oneself or by factors outside oneself, such as other persons (Spector, 1982). WLOC is considered as a culture related work values because of its conceptual relationship to national culture value of “relationship to nature-subjugation” adopted by Maznevski and DiStefano (1995) based on Kluckhohn and Strodbeck (1961) and its contextual relevance.
Work values are defined as desirables or what employees seek from the activity of working. They refer to what a person wants out of work in general, rather than to narrowly defined outcomes of a particular job (Wang, 1996). Work values in this study represent intrinsic and extrinsic values such as attitudes towards earnings, social status, upward striving, achievement and growth orientations, and security orientation.
Then, the three research questions to be addressed in this study are:
- Which MVWO are held by the Sri Lankan sample?
- Whether MVWO influence HRM policy-practice preferences and to what extent?
- Whether there is any difference between domestic firms (DF) and foreign invested firms (FIF) with regard to:
- employees’ preferences for HRM policies and practices; and
- the impact of MVWO on preferences for HRM practices.
Data were obtained from eight Sri Lankan DF and FIF. Sixteen firms were approached that had a labour force of not less than 200, were in the manufacturing sector, had been in the operation for more than five years, and had a human resource/personnel department. Eight companies were participated. Table I shows the background profiles of these companies.
Depending on the size of the firm, between 80 and 140 questionnaires were distributed among employees who belong to middle managers to operational and manual level. Respondents were selected by adopting disproportionate stratified random sampling. Stratified random sampling was used to ensure that different categories of employees were included in the sample and was disproportionately adopted because some strata are too small or too large (Sekaran, 2000). Questionnaires were distributed through heads of departments or divisions. Respondents were provided with stamped envelope and asked to return the anonymous questionnaires directly to the researchers or to put them in a special box placed in the reception area in each company. A total of 487 out of 850 questionnaires returned found usable, representing a response rate of 58 per cent. This is a very acceptable rate in the context of a developing country (Nyambegera et al., 2000). The demographic characteristics of the sample are presented in Table II. Most respondents (75 per cent) were aged 35 years or below and 85 per cent possessed at least G.C.E Ordinary or Advanced level education. While managerial and supervisory levels represent 24 per cent (24 per cent) of respondents, the majority (68 per cent) were from operator, manual, and clerical categories of employment.
A questionnaire that consists of 61 items was used to assess MVWO by individual employees. The scale was developed by the researchers to represent a wider scope of the concept of MVWO. The scale, therefore, was based on a number of measurement scales developed by previous researchers (Buchholz, 1977, 1978; England and Harpaz, 1990; Furnham, 1997; MOW International Research Team, 1987; Ros et al., 1999; Ruiz-Quintanilla and England, 1996; Spector, 1982; Wollack et al., 1971). Sub scale items were selected on the basis of being well-written, cultural relevancy, having clarity of meaning, values of factor loading where it was available, and avoiding overlaps with the measurement of other sub-scales. The concepts of definition of working, WC, and societal norms about working were assessed by using standard questions adopted in MOW International Research Team (1987) as the structures of them have been empirically found stable in many cultures (Ruiz-Quintanilla and England, 1996). Thus, the questionnaire consisted of measures relating to both interval and ordinal scales. It was initially prepared in the English language and then translated into local language (Sinhalese). In order to ensure consistency of meanings, it was back translated.
HRM policy-practice preference scale
A list of HRM practice measures (54 items) was obtained from the measures developed by Sparrow and WU (1998) based on the initial typology developed by Schuler and Jackson (1987). After effecting modifications to some individual items, the list was produced to HR managers or CEOs of participating companies to select practices relevant to their respective organisations. Their review was supplemented by informal discussions with the researchers. This resulted in preparing a questionnaire that contained 42-item bipolar statements. These statements represent practices relating to HR planning, job designing, recruitment and selection, appraising, compensating, and training and career development. Although the six categories may not capture all HRM policies and practices, they represent some of the major and most common policy-practice categories applicable to or used in many cultures. Thus, it provides a basis for any comparative understanding as well. Commensurate measures were developed by modifying the relevant statements to assess HRM preference practices by individual employees. Statements were presented as a pair of alternative HRM practices using a six point bipolar rating scale. Thus, each of the choices runs along a continuum providing respondents with reasonably adequate opportunity to select different viewpoints and to mark their intensity of choices. The participants were asked to decide their individual preferences for how the company should conduct specific HRM policies and practices by circling a number ranging from one to six between the pairs.
The MVWO items that represented interval scale were subjected to principle component analysis. The decision criterion of factor loading was decided as 0.3 because Steven (1992) recommends that for a sample size of 300, a loading of 0.298. The overall results indicted a five-factor solution representing 37 per cent of total variance on an average (Table AI). Table III shows the labelled factor solution along with relevant reliability indicators. The second major MVWO sub-scale, which consisted of ordinal-scale measures relating to the definition of working and WC, was analysed separately. Table III indicates two resulting working definition patterns and a variable relating to WC. The reliability of those variables was assessed by using mean inter-item correlation (MMIC) as Cronbach's α coefficient is not a meaningful indicator of ordinal scale's reliability and for short sub-scales (Briggs and Cheek, 1986; Pallant, 2001). The three sub-scales were then transformed into interval scale (so that the distance between the scale points represent reasonably equal magnitude) by using similar procedures suggested by MOW International Research Team (1987) in their original study for the purpose of multivariate analysis.
The alpha reliability index of 0.60 s is consistent with many other empirical studies that used these and similar sub scales (Furnhan, 1996; Pelled and Hill, 1997; Taris and Feij, 2001; Wollack et al., 1971). For example, Furnhan (1996) study reports an α coefficient of 0.65 and 0.66 for work ethics and organisational belief system, respectively. Given the relatively small number of items on most scales, the coefficient of α in 0.60 s can be regarded as reliable (Taris and Feij, 2001) and acceptable for scales measuring broad constructs such as values (Singelis et al., 1995).
Analysis of HRM policy-practice preferences
The 42 HRM preference measures were cluster analysed. Cluster analysis was performed as an alternative analytical method after a series of factor analysis results indicated no meaningful and reliable component structure. The diverse nature of factor solution seems to reflect at least two major elements:
- the respondents in the sample may not see HRM measures as a coherent set of policies and practices; and
- the existence of dynamic and complex patterns of cultural work values as explained in the research context.
Cluster analysis was performed by using Ward's method and the stability of initial solution was examined by using different clustering methods – “between group average linkage” – (Weiss, 1976) and “split sampling clustering” – based on managerial and non-managerial division (Aldenderfer and Blashfield, 1984). For all the three clustering methods, 32 items were repeatedly clustered into the same category. An indices based on correlation was adopted (e.g. MIIC) to examine the reliability of components derived from adopting non-correlation method (e.g. cluster analysis). The use of completely different methods seems to be one of the best indicators of reliability test of cluster membership (Hamer and Cunningham, 1981; Weiss, 1976). The result indicated four HR preference clusters (Table IV). One of the major implications of adopting the above stability and reliability measures is that the total number of items was first reduced from 42 to 32 and then to 26.
The validity examination and interpretation of the cluster solution were made in relation to the existing HRM theories. Aldenderfer and Blashfield (1984) note that variables to be used in cluster analysis should be chosen within the context of explicitly stated theory. Such a basis supports not only the examination of the validity of clusters but also their interpretation. The HRM variables selected for this study are well supported by the theory. They are based on HRM policies and practices typology first developed by Schuler and Jackson, (1987) and subsequently modified and empirically tested by Nyambegera et al., (2000) and Sparrow and Wu (1998). In addition, these variables can be classified and identified in relation to HRM practice components proposed and used by Aycan et al., (2000); Kanungo and Jaeger (1990) and Kanungo and Mendonca (1994). Hence, the validity of clusters could be examined and established by comparing and agreeing with these conceptual and theoretical structures (see discussion section).
Correlation between MVWO and HRM preferences
The correlation results are shown in Table V. They are taken against a more conservative significance level of p<0.01 to minimise the possibility of type 1 error. Table V shows that all the eight MVWO are significantly related to at least one category of HRM preference practices. On the other hand, all the four HRM preference practices are related to at least one MVWO variable. The MVWO concerned with work norms, work beliefs, extrinsic work values, and working defined as a burden and constraint WE (WDABC) are correlated with three out of the four HRM preference practices.
Impact of MVWO on HRM policy-practice preferences
Table VI shows the results of multiple regression analysis, which reflect the impact of the eight MVWO variables on HRM preferences. It shows that many MVWO-HRM preference relationships suggested by zero-order correlation have become significant (p<0.05 or better), indicating the predictive ability of seven out of eight MVWO variables.
Table VI shows that the MVWO accounts for 17 per cent variation of “planned and open career and empowering practices”. In particular, there is a significant impact of work, WE and work beliefs, and working defined as a social contribution and responsibility on planned and open career and empowering practices. Similarly, there is a 13 per cent variation of the performance management practices explained by MVWO. The significant individual MVWO include work norms, WE and work beliefs, extrinsic work values, and WDABC. Further, the MVWO accounts for 8 and 4 per cent variation of HR practices concerned with generic functional perspective of job-person fit and the job-related competence and rewards, respectively. It reveals that there is a significant impact of work locus of control, WC, and WDABC on the generic functional perspective of job-person fit. Extrinsic work values also have a significant impact on the job-related competence and rewards practices. In brief, the study shows that all the four HRM preferences are influenced by at least one MVWO variable. On the other hand, seven out of eight MVWO are related to at least one of the four HRM preference factors. Interestingly, Table VI also shows that there is a difference between DF and FIF with regard to the impact of MVWO on HRM preferences. This will be discussed separately in a proceeding section.
MVWO of Sri Lankan manufacturing sector employees
Table III shows mean, α, and MICC of MVWO sub-scales. The analysis revealed that the Sri Lankan sample consists of eight distinct MVWO, which can be interpreted as follows.
Individual growth/humanistic beliefs oriented work norms (IGOWN)
Work norms generally refer to the standards people employ when making normative evaluations about work. Sri Lankan sample tends to indicate a fairly balanced orientation towards entitlement and obligatory norms. These norms, however, seem to be dominated by HB which indicate individual growth and development in the job is more important than output. Thus, the factor as a whole represents both positive and negative aspects of work orientations. The high overall mean (5.41) and itemised mean values also indicate that this work orientation is strongly held in the Sri Lankan sample. This observation is supported by the existing literature on cultural values in Sri Lanka, which indicates that Sri Lankans are more inclined to be individualistic than pluralistic or group oriented, which signifies the importance attached to individual concerns rather than organisational concerns and the concern that responsibility is more important than results (Nanayakkara, 1992). The concern for responsibility seems to reflect the obligatory norms or duties towards society, while individual growth oriented work norms (IGOWN) reflect the idea that individualistic orientation is mixed with the concern that responsibility is more important than results. In brief, this factor represent the view that although the respondents believe that their right to work in and their duties towards society, they tend to be oriented towards the belief that individual growth and development in a job is more important than output.
Organisations and positions oriented work ethics and beliefs
The orientation towards work ethics and beliefs (OPOWEB) tends to indicate the belief that work is good in itself and success is a result of personal efforts and therefore, the state of individual independence is much preferred. It also reflects that work takes on meaning only as it affects the organisation and contribution to one's position at work. This orientation is further blended by collective and participative types of beliefs relating to work organisations. Theoretically, this is a very interesting position because it indicates the existence of both positive and constrained work ethics and beliefs simultaneously in a system of work orientations. This provides evidence for Rodger's (1978) theoretical assertion that work ethics are not a single conviction, but a complex of ideals with roots and branches. The relatively high mean (4.97) values of the scales seem to indicate that this work orientation is quite strongly held by the respondents. There seems to be quite similar observations made by previous researchers as well. For example, Nanayakkara (1993) contended that Sri Lankan employees tend to “occupy” positions and emphasise rights to positions based on qualifications rather than fulfilling their roles. Because of power, prestige, and privileges of positions, they may tend to believe that work itself is good as far as it is related to a position and an organisation. In brief, the respondents believe that work is good in itself, but meaningful only if it relates to an organisation and a job. This reflects the fact that people have positive work ethics but in putting them into practice, they believe and depend much on organisations and job positions (rather than orienting towards entrepreneurship).
Status and security oriented upwards striving
This component clearly indicates the respondents' orientation towards advancement and social status with security concern. The security concern means an orientation towards predictive results and averting risks. The overall mean value (4.13) seems to indicate that status and security oriented upwards striving (SSOUS) is quite high in the Sri Lankan sample. Some aspects of this value orientation are congruent with Nanayakkara's (1992) assertion that the class structure in the traditional society (in which a particular occupation, e.g. fishing, was attached to a particular cast) has added a new meaning to work as a means to gain status and power. Drawing inferences from a study of Indian managers by England (1975), the same author contended that Sri Lankan managers were status and personal goal oriented. In brief, employees in the sample tend to exhibit a security and status oriented upward striving.
Extrinsic value orientation (EVO)
This factor represents respondents' orientations toward monetary rewards. In particular, they are oriented towards choosing a job over another because of high salaries and retirement benefits, believing that a good job is a well paying job. However, the concern for earning money by way of doing overtime is not given high value. With regard to the theoretical concerns, all these measures are related to the variables concerned with “attitudes towards earning” in the scale of Wollack et al., (1971). The mean value (2.85) indicates a quite moderate orientation towards earning or extrinsic values. Quite similarly, Nanayakkara (1992) also reports, in response to a question on “why do you work” and “why do others work”, that more than 70 per cent of responses from Sri Lankan managers (public sector) centred around “earning a living”. The significance of attitudes towards earning can also be justified in the context of other developing countries because Furnham (1997) notes that it is quite possible that a fairly large proportion of workers in developing countries are context oriented (high salary, job security, etc.), as they strive to better the livelihood of themselves and their families.
External work locus of control
This value orientation represents two elements of EWLOC: “promotions are usually a matter of good fortune” and “it takes a lot of luck to be an outstanding employee in most jobs”. The rating score, however, indicates that the orientation towards external work locus of control (EWLOC) is moderate (mean 2.85). Earlier research indicates that lack of independence and accepting status quo tendencies, which have also been attributed to the lack of planning tradition in Sri Lanka (Nanayakkara, 1992) may be, at least partially, related to the above representation of EWLOC. In addition, the relationship to nature-subjugation value orientation adopted by Kluckhohn and Strodbeck (1961) is represented in some of the characteristics of external locus of control because it contains items suggesting that “it is better to be lucky than smart”, “success is a matter of good fortune”, and “people should not try to change their fate”. In this sense, it can be argued that the concept of work locus of control is directly represented in and related to national culture values and value measures. Accordingly, the tendency towards work locus of control in Sri Lankan manufacturing sector employees is external.
WC refers to the degree of importance that working has in the life of individuals and to what extent people put work above other aspects of their life. Both these aspects were combined in the analysis to get a general view about WC. As in the case of comparable studies (England and Harpaz, 1990; MOW International Research Team, 1987), WC remains in relatively high position in Sri Lanka (mean 4.38). The analysis also reveals that WC tends to be higher with increased level of education. This may be due to the fact that positive work values are internalised through higher education, which may be the major avenue through which people can feel the importance of work in a culture of collective family environment where children, at least until they married or even after, are largely dependent on their parents earning. Ranasinghe (1996) also reveals that there exists a high degree of WC with a relatively lesser degree of agreement with Protestant ethics in Sri Lankan context.
Working is defined as a burden and constraint
People in this cluster defined working as an unpleasant thing and as something imposed by others. It tends to be unpleasant because it is physically and mentally strenuous and that they believe it has time and place constraint. In other words, they do not consider working as a basic human value or as something that has been internalised into their personality. Simply, these people mean working in negative terms. However, the number of people involved in defining working in this negative term is 164 out of 487 respondents. This evidence seems to be congruent with the idea that many people in Sri Lanka work to live rather than live to work (Nanayakkara, 1993). However, there seems to be some positive sides of this orientation because this negative view of working tends to change with the higher level of education, indicating a more common trend on positive attitudes toward working in the Sri Lankan national culture as indicated below in the second definitional cluster of working.
Working is defined as a social responsibility and contribution (WDASRC)
People in this cluster defined working as something that benefits to them as well as to others in the society. They also believe that working adds values to something for which they are accountable. This seems to reflect that the positive thinking about working seems to have internalised into the value system of these people. The relevant definitional statements include: if it adds values to something, if others profit by it, if you get money for doing it, and if you have to account for it. About 98 per cent respondents have marked their preferences in favour of at least one of the above definitional statements. In addition, there is a tendency to indicate that this work orientation is a strong characteristic of Sri Lankan national culture because of the fact that none of the demographic factors have influenced this factor in this study.
HRM policy-practice preferences of Sri Lankan employees
The cluster structure shown in the Table IV indicates that there are four HRM preference clusters in the Sri Lankan sample. The first cluster consisted of items in which low mean values equate to a “planned and open career and empowering system”. Basically respondents prefer to have planned HRM system that follows formal procedures. They also prefer an open career system that promotes publishing every job promotion, reveal selection criteria to every job applicant, provides a feedback to individual employee, provides an empowered HRM culture that features a significant say in their career path and training, provides opportunities for employees' involvement in decision-making, and for becoming shareholders of the company. In brief, the items in this cluster are generally concerned with a preference for planned and systematic treatment of HRM, especially with regard to selection and promotion. A preference of transparent and open data and involvement of employees in these systems also become evident.
The second cluster consisted of items in which relatively low mean values equate to a “Performance management system concerned with both performance and qualifications based rewards with generic training preferences”. Specifically, respondents prefer to have performance-based selection system for company sponsored training programmes and to have performance-based career development policies. However, this cluster tends to indicate rather inconsistent preferences because of the fact that the respondents' preference to have qualifications and skills-based pay increases. Their preference to have common training programmes for all employees is also somewhat akin to qualifications based pay increase because their focus is mixed with both specific and generic performance management preferences. They also prefer their companies to adopt a formalised procedure to evaluate performance and to discuss results openly with individual employee. In brief, this cluster tends to reveal both qualifications and performance focus built into selection, promotion, and reward system, with generic training focus and a formalised system for performance management.
The third cluster consisted of items in which relatively high mean values equate to a system of “Generic functional perspective on job-person fit”. It reflects preferences for diversity and variety in performing job duties and deciding training and career paths. They also prefer to have training for all employees rather than having it for only new employees or poor performers, to adopt team-based performance appraisal system rather than individual-based system, and to select the best person from inside or outside. In brief, this cluster reveals a system of varied jobs that are designed in relation to different functional context, career system built around different functional paths, prioritised team-based task performance, and the best person sought for the job either internally or externally.
The fourth factor consisted of items in which low means values equate to a system of “Job-related competence and rewards”. This cluster in general reflects reward management practices. The respondents specifically prefer to have opportunities to build their job related-skills rather than behaviour-related skills, complete instructions to do the job, considered what was achieved, and been rewarded financially for special achievements. In brief, this cluster consists of a system of reward management practices that reveals preference for building job-related skills, one best way of doing the job with complete instructions, and a moderate preference for measuring performance on output.
Comparison between DF and FIF
A comparison between DF and FIF was made with regard to MVWO, preferences for HRM practices, and the impact of MVWO on preferences for HRM practices. Since there is no difference between the DF and FIF in MVWO, results are not presented in this paper. Table VI shows that there is a difference between the DF and FIF with regard to the impact of MVWO on HRM preferences, which will be discussed in the following section. Table VII shows that there is a significant difference between the FIF and DF with regard to two HRM preference factors.
The differences seem to indicate that respondents in FIF prefer a relatively less tightly planned and open career and empowering system than those respondents in DF. On the other hand, respondents in DF seem to prefer a more generic functional perspective on job-person fit rather than functional perspective on job-person fit relevant to those in FIF.
The major objective of assessing MVWO in the Sri Lankan context is to examine whether they have an impact on individuals' preferences for HRM policies and practices. We have argued that MVWO is an element of national culture and that people with certain work orientations may prefer their organisation to adopt certain types of corresponding HRM practices.
The findings of this study indicate that there are eight distinct MVWO and four distinct HRM preference factors in relation to the Sri Lankan sample. The impact analysis reveals that all the four HRM preference factors are influenced by MVWO. Therefore, the hypothesis that MVWO influence individual HRM preferences was supported. Of the four HRM factors emerged, MVWO variables account for 17 per cent variation of “planned and open career and empowering practices”. The analysis shows that the greater the orientation towards IGOWN, OPOWEB and WDASRC, the stronger the preferences for loosely planned and less empowered HRM practices. Although the above first two work orientations have both positive and negative aspects, they appeared largely negative as far as entrepreneurship, creativity, and risk-taking orientations are concerned. These characteristics are the real concerns for making positive impact on an organisation where an empowered HR culture exists. This finding seems to indicate that people with relatively negative orientations toward work norms (e.g. IGOWN) WE and work beliefs (e.g. OPOWEB), tend to prefer relatively less formal planning and empowered HRM culture. The low preferences for empowered HRM culture in relation to developing countries reflects the Jaeger's (1990) argument that developing countries' socialisation makes people passive and reactive to tasks. It also addresses an issue raised by Sparrow et al. (1994), that although the empowering aspect of HRM represents a significant trend in a number of US and UK organisations (Lawler, 1991), how important is it to other part of the world. According to the findings of this study and as far as the impact of cultural values on HRM is concerned, the question of adaptability and transferability of formal planning procedures and empowering practices to developing countries become evident.
The study also shows that MVWO variables represent 13 per cent variation of performance management practices. There are four MVWO that influence this HRM preference factor, namely IGOWN, OPOWEB, EWV, and WDABC. It reveals that the higher the orientations towards IGOWN and OPOWEB, the lower the preferences for generic training and qualifications and performance based rewards. In other words, people with these work orientations tend to prefer individual and performance-based appraisal and rewards and job focused training and career development. This finding is congruent with what one could expect from theoretical position of the associated variables. For example, individual growth oriented people want to reach a higher stage of human development (Buchholz, 1978) and therefore, they may prefer individual and performance-based rewards and appraisal practices. On the other hand, organisations and positions orientation includes such element as every one should work hard, success is directly linked to one's own efforts (Buchholz, 1978) and therefore, people with these values orientations may prefer individual-based performance appraisal and rewards and job-based training (Chandrakumara, 2003). Interestingly, OPOWEB seems to be represented in “relations-hierarchical” value orientations adopted by Maznevski and DiStefano (1995) because they include items that suggest hierarchies of authority are best in organisations, rewards and facilities should be designed, accordingly etc. As discussed earlier, people may prefer to acquire or hold positions in organisations because they give them a sense of status, prestige, and privileges as they go up in the hierarchy (Nanayakkara, 1993). Sparrow and Wu (1998) also found that relation-hierarchy value orientation is related to HRM preferences of formal planning procedures, rewarding for specific skills, and training employees for specific needs. The above results of this study largely support these conceptual relationships.
The impact of EVO and WDABC was also found significant in influencing performance management practices in this study. It reveals that the stronger the orientation towards EVO and WDABC, the greater the preferences for generic training and skills and qualifications-based rewards. People with EVO exhibit higher motivation when working in team work settings (Cummings, 1981) and therefore may tend to prefer common or generic training (Chandrakumara, 2003). On the other hand one can conceptually argue that people with WDABC, as it represent negative work orientation, may tend to prefer qualifications and performance-based criteria to be included in performance-based reward system.
Three MVWO have a significant impact on HRM preferences concerned with “Generic functional” perspective on job-person fit, namely EWLOC, WDABC, and WC. This HRM factor basically represents preferences for diversity and variety in performing job duties and deciding training and career paths, and for selecting the best person from inside or outside. The result shows that individuals with EWLOC do not prefer these HRM practices. Spector (1982) also found that externals are more concerned with the social rather than task aspect of job and prefer directive supervision with more formalised task rather than a variety of task. Similarly, one can argue that those who define working in negative terms (e.g. WDABC) tend to work on others' instruction, and prefer standardised and routine tasks. The finding of this study supports this argument by indicating a negative relationship between WDABC and HRM practices concerned with generic functional perspective on job-person fit. In contrast, there is a positive relationship between this HRM factor and WC. It indicates that individuals with high work centrality tend to prefer high task and skill variety by selecting job in different functional areas.
Finally, the HRM preference practices concerned with “job-related competence and reward” was also influenced by MVWO. The result shows that the higher the orientation towards extrinsic values, the lower the preferences for job-related competence and rewards. This reflects preference for building job-related skills, job with complete instructions, and for measuring performance on output. The negative relationship between EVO and this HRM factor may be due to the fact that individuals with extrinsic values tend to prefer HR practices that provides more material benefits rather than concentrating on skills and competence development that may contribute towards increasing performance. Quite similarly, Furnham (1997) notes that it is quite possible that a fairly large proportion of workers in developing countries are context oriented, as they strive to better the livelihood of themselves and their families.
Differences between DF and FIF
As it is generally expected, there is no significant difference between DF and FIF with regard to MVWO. This may be due to the fact that values and work orientations are relatively stable characteristics of individual behaviour (Rokeach, 1973), which can be considered as elements of national culture (Schwartz, 1999; MOW International Research Team, 1987). However, it is found that there is a significant (p<0.05) difference between DF and FIF with regard to preferences for two HRM factors. Respondents in FIF seem to prefer a relatively “less tightly planned and open career and empowering system” than those respondents in DF. On the other hand, respondents in DF seem to prefer a more “generic functional perspective on job-person fit” rather than more functional perspective on job-person fit which is relevant to those respondents in FIF (Table VII).
The comparative result (Table VI) also indicates that the higher the orientation towards three MVWO factors (OPOWEB, WDASRC, SSOUS), the greater the preference for loosely planned and less empowered HRM culture. It is obvious that these three work orientations consist of both positive and negative work ethics and values. What the evidence of this study reveals is that, irrespective of the positive or negative trend of work values, employees in FIF tend to prefer relatively less formal and empowered HRM cultures. Then, the reasons for these preferences seem to be something beyond the differences in value orientations of employees. One of such reasons may be that employees are not encouraged or motivated to involve in empowered HRM cultures in FIF. For example, initial discussion that the researchers had with HR managers regarding actual practices of these companies revealed that three out of four FIF did not offer employees the opportunity to become shareholders of their respective companies. In addition, employees general opinion in the country do not hold very positive attitudes toward survival and long-run existence of FIF because of the fact that some FIF companies have closed their operations after enjoying tax holiday period and other concessions. This may have led employees to create not only unfavourable attitudes toward FIF but also to expect different rewards systems from FIF. For example, a relatively short-run existence of an employer may cause employees to create expectation for more extrinsic rewards such as high pay and obtaining training and technical skills rather than high intrinsic rewards such as an active involvement in an empowered culture. Indeed, Venkata Ratnam (1998) found that MNCs have highly competitive pay and better working conditions than those in indigenous firms in India. Further, obtaining opportunities for workforce training is a national level objective set at attracting FDI in Sri Lankan (The Central Bank of Sri Lanka, 1999). On the other hand, the opinion of top management of local companies in the apparel industry reveals that the closures of companies or changing geographical location from country to country has been an inevitable business strategy in the face of global competition. All these conditions may have contributed to create low preferences for tightly planned career system and employee involvement, and to create high preferences for extrinsic rewards and training and skill development.
One of the theoretical implications of this is that “the type of ownership” has an impact not only on actual HRM practices (Tayeb, 1988; Budhwar and Sparrow, 2002) but also on preferred HRM practices such as empowering and performance management practices. Further, employees' preferences for HRM practices in FIF seem to be influenced by some macro level factors such as long-run existence and host government expectations. The practical implications of this evidence is that HR managers in FIF should pay a particular attention not only to recruit employees with positive work ethics and values and adjust HRM practices according to cultural value orientations but also to deal with unfavourable attitudes toward FIF and different and unique expectations of employees in FIF and host governments.
Limitations of the study
Although MVWO could be related to national culture values, they may not represent all aspects of national cultural value orientations. The data represent only the situation and trends in Sri Lankan manufacturing sector sample of 487 employees. As the manufacturing sector represents employees with relatively low level of education and as the level of education has an impact on some MVWO variables, other sectors in the economy may produce different picture with regard to work and value orientations and HRM preferences. Studies employing self-report measures face the problem of “shared method variance”. This was addressed by using different response format in different section of the questionnaire. To ascertain the level of independence of the measures, the total data set across MVWO and HRM preferences was subjected to principle component analysis and found no MVWO loaded on HRM items and vice versa. Since individual responses were anonymous, method bias due to social desirability is minimised.
The impact of national culture on HRM cannot be taken for granted (Aycan et al., 2000; Hofstede, 1993; Schuler, 2000; Sparrow and Wu, 1998). This study shows that MVWO can be conceptualised as an element of national culture and assessed its impact on HRM policy-practice design choices in Sri Lankan context. This is a step towards helping international HR managers who adjust HRM policies to match employees' expectations. The Sri Lankan sample held four distinct HRM preference factors: planned and open career and empowering system, qualifications and performance-based reward system, generic functional perspective of job person fit, and job-related competence and rewards. All these practices were impacted by MVWO reflecting 17, 13, 8, and 4 per cent variation for the said HRM factors, respectively. This evidence supports culture-related work values relativity of HRM practices and the question of transferability of HR practices such as empowering and performance management practices to developing countries. It was also evident that there could be differences between DF and FIF with regard to preferences for HRM practices and that the type of ownership has an impact not only on actual HRM practices but also on preferred HRM practices. It is therefore important for international managers to know which work and value orientations are held by the people of a particular country or a region and to what extent those values orientations influence HRM design choices. In addition to the impact of MVWO on HRM preferences, macro level issues and attitudes such as existence in the long-run, ability to transfer share-ownership, strategic decision on location, host government expectations, etc., can also have an impact on preferences for HRM practices in FIF. Managers in FIF therefore should pay a particular attention not only to adjust HR practices to fit cultural realities and recruit employees with positive WE and values but also to deal with more macro level issues and unfavourable attitudes toward FIF.
The theoretical exploration of the concept of MVWO and the empirical investigation of Sri Lankan work orientations and HRM preferences raise a number of issues for future research. First, a better understanding of HRM policy-practice design choices can be gained by examining work and value orientations of employees. Second, work and values orientation is a very complex concept in the context of developing countries and cannot be easily categorised or structured by using models developed in developed countries. This calls for more studies in developing country context. The findings not only question the applicability of some HR models which have developed in other developed countries but also in the context of DF and FIF. This suggests an agenda for future research in developing countries and comparative studies between DF and FIF.
Table IParticipating companies and their background
Table IIDemographic characteristics of the respondents
Table IIIMean, α, and MIIC of MVWO factors
Table IVCluster analysis solution, mean, and MIIC
Table VMean, standard deviation, α coefficients, and correlations between work orientations and HRM preferences
Table VIMultiple regression results of work orientations on HRM preferences
Table VIIDifferences in HRM preferences between the DF and FIF
Table AIRotated component matrix – five MVWO variables
Aldenderfer, M.S., Blashfield, R.K. (1984), Cluster Analysis, Sage, Beverly Hills, CA, .
Aycan, Z., Kanungo, R.N., Mendonca, M., Kaicheng, Y., Deller, J., Stahl, G., Kurshid, A. (2000), "Impact of culture on HRM: a 10 country comparison", Applied Psychology: An International Review, Vol. 49 pp.192-221.
Briggs, S.R., Cheek, J.M. (1986), "The role of factor analysis in the development and evaluation of personality scales", Journal of Personality, Vol. 54 pp.106-48.
Buchholz, R.A. (1977), "The belief structure of managers relative to work concepts measured by a factor analytic model", Personnel Psychology, Vol. 30 pp.567-87.
Buchholz, R.A. (1978), "An empirical study of contemporary beliefs about work in American society", Journal of Applied Psychology, Vol. 63 No.2, pp.219-27.
Budhwar, P., Debrah, Y. (2001), Human Resource Management in Developing Countries, Routledge, London, .
Budhwar, P., Sparrow, P.R. (2002), "An integrative framework for understanding cross-national HRM practices", Human Resource Management Review, Vol. 12 No.3, pp.377-403.
Chandrakumara, A. (2003), "The impact of meaning and values of work orientations on predicting HRM policy-practice design choices and contextual and task performance", University of Sheffield Management School, Sheffield, .
Claes, R., Ruiz-Quintanilla, S.A. (1995), "Work meaning patterns in early career", in Peiro, J.M., Prieto, B.F., Melia, J.L., Luque, O. (Eds),Work and Organizational Psychology: European Contribution of the Nineties, Erlbaum Hove/Taylor & Francis, Basingstoke, .
Cray, D., Mallory, G.R. (1998), Making Sense of Managing Culture, International Thomson Business Press, London, .
Cummings, T.G. (1981), "Designing effective work groups", in Nystrom, P.C., Starbrach, W.H. (Eds),Handbook of Organizational Design, Oxford University Press, New York, NY, pp.250-71.
Dickson, J.W., Buchholz, R.A. (1977), "Managerial beliefs about work in Scotland and the USA", Journal of Management Studies, Vol. 14 pp.80-101.
Dose, J.J. (1997), "Work values: an integrative framework and illustrative application to organisational socialisation", Journal of Occupational and Organisational Psychology, Vol. 70 pp.219-40.
England, G.W. (1975), The Manager and His Values, Ballinger, Cambridge, MA, .
England, J.W., Harpaz, I. (1990), "How work is defined: national contexts and demographic and organisational role influences", Journal of Organizational Behaviour, Vol. 11 pp.253-66.
Frost, P.J., Cyr, D.J. (1990), "Selective frames, findings, and futures: a review of the second conference on international personnel and HRM", in Ferris, G.R., Rowland, K.M. (Eds),Research in Personnel and HRM, JAI Press, England, No. 2, .
Furnhan, A. (1996), "Attitudinal correlates and demographic predictors of monetary beliefs and behaviours", Journal of Organisational Behaviour, Vol. 17 pp.375-88.
Furnham, A. (1997), The Psychology of Behaviour at Work: The Individual in the Organisation, Psychology Press, Hove, .
Hamer, R., Cunningham (1981), "Cluster analysing profile data confounded with interrater differences: a comparison of profile association measures", Applied Psychological Measurement, Vol. 5 pp.63-72.
Hofstede, G. (1993), "Cultural dimensions in people management", in Puick, V., Tichy, N.M., Barnett, C.K. (Eds),Globalizing Management, Wiley, New York, NY, .
Jaeger, A.M. (1990), "The applicability of Western management techniques in developing countries", in Jaeger, A.M., Kanungo, R.N. (Eds),Management in Developing Countries, Routledge, London, pp.131-45.
Joiner, T.A. (2001), "The influence of national culture and organisational culture alignment on job stress and performance", Journal of Managerial Psychology, Vol. 16 No.3, pp.1-12.
Kanungo, R.N., Jaeger, A.M. (1990), "Introduction: the need for indigenous management in developing countries", in Jaeger, A.M., Kanungo, R.N. (Eds),Management in Developing Countries, Routledge, London, pp.1-23.
(1994), in Kanungo, R.N., Mendonca, M. (Eds),Work Motivation: Models for Developing Countries, Sage, London, .
Kluckhohn, F.R., Strodbeck, F.L. (1961), Variations in value orientations, Harper & Row, New York, NY, .
Lawler, E.E. (1991), "The new plant approach: a second generation approach", Organisational Dynamics, pp.5-15.
Locke, A.E. (1976), "The nature and causes of job satisfaction", in Dunnette, M.D. (Eds),Handbook of Industrial and Organisational Psychology, Skokie, IL, Rand McNally, pp.1297-350.
McCourt, W. (2001), "Towards a strategic model of employment reforms in developing countries", International Journal of Human Resource Management, Vol. 12 No.1, pp.56-75.
McGaughey, S.L., Iverson, R.D., de Cieri, H. (1997), "A multi-method analysis of work related preferences in three nations: implication for inter- and intra-national HRM", International Journal of Human Resource Management, Vol. 8 No.1, pp.1-17.
Maznevski, M.L., DiStefano, J.J. (1995), "Measuring culture in international management: the cultural perspective questionnaire", paper presented at the Academy of International Business Annual Meeting, Seoul, November, .
Mendonca, M., Kanungo, N.K. (1990), "Performance management in developing countries", in Jaeger, A.M., R.N. (Eds),Management in Developing Countries, Logman, London, .
MOW International Research Team (1987), The Meaning of Working, Academic Press, London, .
Nanayakkara, G. (1992), Culture and Management in Sri Lanka, PIM, University of Sri Jayewardenepura, Sri Lanka, .
Nanayakkara, G. (1993), The Negative Leader: An Analysis of Sri Lankan Cases, Conference on Management Studies, PIM, Colombo, pp.141-53.
Nyambegera, S.M., Sparrow, P.R., Daniels, K. (2000), "The impact of cultural value orientations on individual HRM preferences in developing countries", International Journal of Human Resource Management, Vol. 11 No.4, pp.639-63.
Pallant, J. (2001), SPSS Survival Manual, Open University Press, Philadelphia, PA, .
Pelled, L.P., Hill, K.D. (1997), "Employee work values and organizational attachment in North Mexican maquiladoras", The International Journal of Human Resource Management, Vol. 8 No.4, pp.495-505.
Punnett, B.J. (1998), "Culture, cross-national", in Poole, M., Warner, M. (Eds),The IEBM Handbook of Human Resource Management, International Thomson Business Press, London, pp.9-26.
Ranasinghe, S. (1996), "Work centrality and job satisfaction: application of protestant work ethics in the Sri Lankan context", Sri Lankan Journal of Management, Vol. 1 No.1, pp.19-36.
Rodgers, D. (1978), The Work Ethic in Industrial America, University of Chicago Press, Chicago, IL, pp.1850-920.
Rokeach, M. (1973), The Nature of Human Values, Free Press, New York, NY, .
Ros, M., Schwartz, S.H., Surkiss, S. (1999), "Basic individual values, work values, and the meaning of work", Applied Psychology: An International Review, Vol. 48 No.1, pp.49-71.
Ruiz-Quintanilla, S.A., England, G.W. (1996), "How working is defined: structure and stability", Journal of Organisational Behaviour, Vol. 17 pp.515-40.
Schuler, R.S. (2000), "The internationalisation of HRM", Journal of International Management, Vol. 6 No.2000, pp.239-60.
Schuler, R.S., Jackson, S.E. (1987), "Linking competitive strategy with HRM practices", Academy of Management Executive, Vol. 1 No.3, pp.207-19.
Schwartz, S.H. (1999), "The theory of cultural values and some implications for work", Applied Psychology: An International Review, Vol. 48 No.10, pp.23-47.
Sekaran, U. (2000), Research Methods for Business: A Skill Building Approach, 3rd ed., Wiley, New York, NY, .
Singelis, T.M., Triandis, H.C., Bhawuk, D., Gelfand, M.J. (1995), "Horizontal and vertical dimensions of individualism and collectivism: a theoretical and measurement refinement", Cross-Cultural Research: The Journal of Comparative Social Science, Vol. 29 pp.240-75.
Sparrow, P., Budhwar, P.S. (1997), "Competition and change: mapping the Indian HRM recipe against worldwide patterns", Journal of World Business, Vol. 32 No.3, pp.17-29.
Sparrow, P.R., Wu, P. (1998), "How much do national value orientation really matter? Predicting HRM preferences of Taiwanese employees", Employee Relations, Vol. 20 No.1, pp.26-56.
Sparrow, P.R., Schuler, R.S., Jackson, S.E. (1994), "Convergence or divergence: HR Policies and practices for competitive advantage worldwide", International Journal of Human Resource Management, Vol. 5 No.2, pp.267-99.
Spector, P. (1982), "Behaviour in organisations as a function of employees' locus of control", Psychological Bulletin, Vol. 91 pp.482-97.
Steven, J. (1992), Applied Multivariate Statistics for Social Sciences, 2nd ed., Erlbaum Associates, London, .
Taris, R., Feij, J.A. (2001), "Longitudinal examination of the relationship between supplies-values fit and work outcomes", Applied Psychology: International Review, Vol. 50 No.1, pp.52-80.
Tayeb, M. (1988), Organisations and National Culture, Sage, London, .
The Central Bank of Sri Lanka (1999), Annual Report, Central Bank Printing Press, Colombo, .
The World Bank (2000), Sri Lanka: Recapturing Missed Opportunities (Report No: 20430), The World Bank, Washington, DC, .
Thomas, A.S., Philip, A. (1994), "India: management in an ancient and modern civilization", International Studies of Management and Organization, Vol. 24 No.1, pp.91-115.
Venkata Ratnam, C.S. (1998), "Multinational companies in India", International Journal of Human Resource Management, Vol. 9 No.4, pp.567-89.
Vishnath, W. (1997), "Business process and information systems re-engineering: an analysis in the Sri Lankan context", Sri Lankan Journal of Management, Vol. 2 No.1/2, pp.69-99.
Wang, G.T. (1996), A Comparative Study of Extrinsic and Intrinsic Work Values of Employees in the United States and Japan, The Edwin Mellen Press, New York, NY, .
Weiss, D.J. (1976), "Multivariate procedures", in Dunnette, M.D. (Eds),Handbook of Industrial and Organisational Psychology, Rand-McNally, Chicago, IL, .
Wollack, S., Goodale, J.G., Wijting, J.P., Smith, P.C. (1971), "Development of survey of work values", Journal of Applied Psychology, Vol. 55 No.40, pp.331-8.
Wright, M., Pendleton, A., Robbie, K. (2000), "Employee ownership in enterprises in Africa and Asia", International Journal of Human Resource Management, Vol. 11 No.1, pp.90-111.
Zaffane, R. (1994), "Utility theory and organisation design", American Business Review, Vol. 12 No.1, pp.66-86.
Budhwar, P., Sparrow, P.R. (1997), "Evaluating levels of strategic integration and development of HRM in India", International Journal of Human resource Management, Vol. 8 No.4, pp.476-94.
Ward, J. (1963), "Hierarchical grouping to optimise an objective function", Journal of American Statistical Association, Vol. 58 pp.244-63.