Hrm on organizational performance

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.


Several studies have shown a positive association between effective human resource management (HRM) practices and organizational performance ( Becker & Huselid, 1998; Huselid, 1995). Despite several studies that exposed the positive effects of HRM on an organization's performance, the main reason why this study is conducted is that there are several problems that were not fully addressed in previous studies. First, there is no harmony on how and what to measure regarding effective HRM practices. Although it would be impossible to measure each HRM practice carried out at workplaces, HRM researchers and practitioners should agree on the representative HRM items that are the most influential in terms of boosting an organization's lower staff (bottom line). This unsettled issue of measurement items of HRM creates difficulties when applying the findings of such academic studies to the workplace and when comparing results across the studies.

Second problem stems from the level of measurement. As Bamberger and Meshoulam (2000) pointed out, measuring HRM practices at different levels within an organization and between organizations is likely to yield different results. In 1998 Delery studied that assuming that the use of practices across an organization is the same for all levels of employees is probably a false assumption. Given this suggestion, this study attempted to reconcile the limitations that existed in the previous studies by measuring HRM practices at two different levels: managerial and non-managerial employees. The data was analyzed separately for theses two groups.

Third, the most imperative and critical problem is measuring the outcomes of HRM on organizational performance. Organizational performance is well discussed in the HRM literature. However, there is a little theoretical consensus on basic definitions and technical issues. Kanter (1981) stated that the problem is "not how to measure effectiveness or productivity, but what to measure; how definitions and techniques are chosen." Studies about the impact of HRM on organization performance have adopted measures or indicators of organization performance without prior conceptual questioning of what to measure, how such measures are defined, and why they are chosen. In addition, there are limited studies investigating an organization's performance indicators that are influenced by HRM, particularly for those organizations which are at a business unit level. Obviously, private companies and the ones at a unit level do not issue equities; thus, market-value added organizational performance indicators that are examined extensively in HRM literature are not useful for such organizations.

Over the past two decades, a growing number of management and HR professionals have argued that investment in sound HRM practices and programmes pays off for an organization through improved productivity and superior financial performance. Peter Drucker (2002) and several other writers (e.g. Towers, 1992; Pfeffer, 1998; Greer, 2001) suggested that the success of organizations in the present competitive environment largely depended upon the calibre of their human resources and innovative employee management programmes and practices. In a widely cited study, Mark Huselid (1994), for instance, has demonstrated that well-formulated HRM programmes not only reduce employee turnover but also make a substantial contribution to financial performance and productivity (see also Huselid, 1995; Delaney and Huselid, 1996). Several case studies have reached similar conclusions. Anthony et al. (2002), for instance, attribute Southwest Airline's continued success, remarkable level of efficiency and customer satisfaction to its HR strategy and the special bond it has established with its employees. Recent research on best HR practices has shown that the HR function is indeed an important success factor (Thompson and Strickland, 2001; Greer, 2001).

Finally, the other urgent inherent in this study is attributed to the uniqueness of the public sector regulatory authorities and organizations generating their revenues at their own and being regulated by these authorities. Historically, payroll expenses have accounted for the major part of operating expenses in these regulatory authorities. Despite this compelling need, the effectiveness of human resources has not been thoroughly studied in the regulatory authorities. Given this background, the purpose of this study is to develop a conceptual model to measure the effects of HRM on organizational performance.


Taking a lead from the growing body of literature on HRM-performance relationship (e.g. Guest, 1997; Wright et al., 1998; Richardson and Thompson, 1999; Truss, 2001; Greer, 2001; Rehman et al., 2010) a tentative conceptual model has been outlined in Figure 1 that has been developed and tested in this study. In the proposed model, HRM practices are treated as independent variable and Recruitment as intervening variables. With respect to organizational performance the present study, in recognition of the multidimensional nature of the concept, defined performance in terms of achieving superior results in the following three key areas: voluntary annual turnover rates for human resource outcomes, employee productivity for organizational outcomes, and revenue generated for financial accounting outcomes. These performance dimensions are the primary dependent variables in the proposed model.


Figure 1 About Here


For questions of HRM practices, the scale was from 1 (strongly disagree) to 5 (strongly agree). Given the HRM researchers' suggestion (Bamberger & Meshoulam, 2000; Delery, 1998), HRM practices were measured at two different levels: managerial employees and non-managerial employees. Table 1 presents the questions used to measure the independent and dependent variables.


Table 1 About Here


The dependent variable in this study was organizational performance. Organizational performance was measured in three categories: voluntary annual turnover rates for human resource outcomes, employee productivity for organizational outcomes, and revenue generated for financial accounting outcomes. The target population of this study was regulatory authorities in the various sectors of Pakistan. The findings of this study will have implications for other public sector organizations at the government level. Therefore, this study does not intend to generalize the findings to other organizations funded by the government of Pakistan. The sample for this research was drawn from the public sector authorities regulating telecommunication, power, oil & gas, media and corporate, capital and baking sectors of Pakistan and some of the organizations being controlled by these regulators. Fifteen (15) organizations were selected for the research. The platform which helped to choose the sectors and the organizations was the data taken from Pakistan Public Administration Research Centre, Establishment Division, Annual Statistical Bulletin of Autonomous Bodies Employees and website of these organizations. Name of regulatory authorities and organizations selected are given in table 2


Table 2 About Here


Research participants included both managerial and non-managerial staff of these organizations. Fifty (50) questionnaires were distributed in each organization for minimizing the possibility of uncertainty and biasness. 750 questionnaire were distributed in 15 organizations out of which 568 questionnaire were received back response rate remained (76 %) which is given in Table 3


Table 3 About Here


Organizations operating locally were personally visited and the questionnaires were disseminated to the officer / officials and human resource professional of these organizations. Demographic percentage of employees is given in table 4. Total sample size became 568 and results were entered for further analysis by using several statistical techniques. Before hypotheses testing, data purification was carried out. Basic descriptive analyses was done, which included examining for coding errors, normality, skewness, and kurtosis. The purified data was subjected to confirmatory factor analysis (CFA) by means of the Analysis of Moment Structures (AMOS). Hypotheses were tested by using a structural equation modeling (SEM) by means of AMOS.


Table 4 About Here



This research was conducted from very initial level because there was no data available prior to this research. Descriptive statistics were firstly done with the aim to describe the data (Durrheim, 2002b). Further statistical analyses were then utilized to test the research hypotheses to determine if a statistical association existed between the research variables (Bailey, 1987). The computerized statistical package, SPSS Version 12, was used for this purpose. Descriptive statistics enable to obtain an overall picture of the research data and assist by presenting the data in a user-friendly and orderly way (Durrheim, 2002). Descriptive statistics assist in providing a concise description of the quantitative data (Kaplan & Saccuzzo, 2001). For the purpose of research, descriptive statistics was considered for reporting on the profile of the sample.

Descriptive statistics by way of means, standard deviations and reliabilities were calculated for the sample, predictors and criteria. Means represent the average response values whilst standard deviations highlight the degree of variance or distance away from the mean (Durrheim, 2002). A number of correlations were calculated for the purpose of testing the hypotheses and to determine the relationship between the variables. Correlation coefficients (r) were calculated to measure the existence and strength of the statistical relationship between the research variables (Bailey, 1987; Cronbach, 1970; Gekoski, 1964; Kaplan & Saccuzzo, 2001). r=1 indicates a perfect positive relationship whilst r=-1 indicates a perfect negative relationship (Cascio, 1982; Durrheim, 2002b).The sign of the correlation coefficient serves to indicate the direction of the relationship (Howell, 1997) and usually correlations higher than zero but less than 1.00 are found (Anastasi, 1988). Factor analysis technique was used to identify a smaller number of factors underlying a large number of observed variables. Variables that have a high correlation between them and are largely independent of other subsets of variables were combined into a factor.

Regression analysis was carried out to measure the relationship between one dependent variable and several independent variables. Linear regression analysis estimates the coefficients of a linear equation, involving one or more independent variables that best predict the value of the dependent variable. In the present model, there is one dependent variable only. Than, Multiple Regression Analysis, the term was first used by Pearson, is to study more about the association between a number of independent or predictor variables and a dependent or criterion variable was used.

The first factor, labelled "Importance of job analysis" consisted of one item that reflects the importance of job analysis in the selected organization. It accounted for 21.89% of the variance. The second factor consisted of one item that measured frequency of the practice of job analysis, not performed, every year, every two years, after every three years, after four to five years and when it is felt necessary. This factor accounted for 7.501 % of the variance and was labelled "Practice of job analysis." The third factor, labelled "job design" accounted for 5.97% of the variance and consisted of five items that measured an individual's ability to be creative, working style, interference, superior support and suggestions. The fourth factor, labelled "job evaluation" accounted for 5.00% of the variance and consisted of two items that measured an individual's worth of the job and financial needs fulfilment. The fifth factor, labelled "Job Security" consisted of five items that measured an individual's ability to secure his job on performance basis, consistency in job performance and security of job. This factor accounted for 4.14% of the variance. The sixth factor, labelled "job succession planning" accounted for 3.46% of the variance. It consisted of six items that measured an individual's ability to career growth on the basis of performance, knowledge and skills.

The seventh factor, labelled "Recruitment KSA" accounted for 3.40% of the variance. It consisted of knowledge, skills and abilities and matching of job demand with the level of eligibility. The eighth factor, labelled "organizational performance" accounted for 3.20% of the variance. It consisted of three items, voluntary annual turnover rates for human resource outcomes, employee productivity for organizational outcomes, and revenue generated for financial accounting outcomes. The ninth factor, labelled "job satisfaction" accounted for 2.92% of the variance. It consisted of six items that measured an individual's satisfaction upon sense of fulfilment, internal satisfaction, peers support, referral to the others and organization support. The tenth factor, labelled "job retention," accounted for 2.77% of the variance. It consisted of six items that measured an individual's ability to retain his or her job on the basis of pay, expectations, monotony of the job, other job options and changing of the job.

Model fits was evaluated by using the normed chi-square, Tucker-Lewis Index (TLI), Comparative Fit Index (CFI) and Root-Mean-Square-Error of Approximation (RMSEA) (Kenny & McCoach, 2003; Lievens & Keer, 2001). The first CFA indicated a poor fit of the model to the data. Therefore, model improvement techniques were applied (Anderson & Gerbing, 1988), and factor loadings and significance levels were examined. For the managerial group, four items with low factor loadings were eliminated, although all items had positive and significant loadings.

The four items were JA2 (.34), D5 (.33), N2 ( .41 ) and R1 (.37). For the non-managerial group, five items were excluded. The four items were D3 (.27), E2 (.38), S4 (.32), SP1(.31) and R4 (.30). The second CFA comprised 23 items for the managerial group and 24 items for the nonmanagerial group. For the managerial employee group, the TLI was .87, the CFI was .91, the RMSEA was .05, and the normed chi-square was 1.28. All goodness of fit indexes supported the overall quality of the measurement model. For the group of nonmanagerial employees, the second CFA was conducted with 19 variables. The statistics of fit indexes showed a good fit of the model to the data. The normed chi-square was 1.17, which was in the acceptable range. The values of TLI (.92), CFI (.91), and RMSEA (.03) were all indicative of a good fit of the model to the data.

Next, the reliability of the constructs were assessed. Reliability was calculated by using the procedure suggested by Fornell and Larcker (1981), ρη = (Σλyi)2 / {(Σλyi)2 + Σ(εi)}, where ρη represents construct reliability for scale η, λyi reflects standardized factor loadings for scale item yi, and εi indicates measurement error for the scale item γi. The reliabilities for the seven constructs ranged between .78 (JD) and .60 (Retention) for the managerial group and ranged between .72 (JD) and .56 (Retention) for the non-managerial group. With the exception of Retention (.56) for the non-managerial group, all constructs had an acceptable reliability value (Fornell & Larcker, 1981).

The results of hypothesis testing are presented in Table 5, including estimated parameters, critical ratios, p values, and the fit statistics. Each hypothesized model was tested using AMOS for each employee group. The model fit statistics indicated that the hypothesized model achieved acceptable fit in both groups. The findings indicated that the model generally fits the data for each employee group; however, since some paths are not significant in a certain group, it implies that HRM constructs implemented for the two employee groups have different impacts on organizational performance.


Table 5 about here



This paper reveals a diverse aspect of a practical operation of human resource management practices and performance linkage in Pakistan. It shows that human resource planning, the most critical element of personnel function is emerging. The task of acquiring sound and credible employees belongs to the Human Resources Management (HRM). Since, HRM is a known field under the concept of management, it is expected that several theories, models and tools have been developed to promote its manifestations.

In an increasingly competitive and turbulent market, organizations are largely dependent on their employees for success. The challenge of identifying the right man for the right job, individuals to fill public sector positions is to be met. A large number of potential workforce is available in the employment market but the challenge for organizations is to identify and select those candidates who could perform effectively and efficiently.

Selection and the assessments chosen for the selection process should be done on the basis of the requirements of the job. Knowledge is essential on part of the organization in terms of what constitutes good job performance; what kind of knowledge, skills and abilities are required and what measures would be effective in assessing these.

A reliable and valid measurement model was developed for HRM practices covering six HRM domains (job analysis, job design, job evaluation, job security, job succession planning, job satisfaction and job retention) and tested it using CFA. Although this measurement model could not cover each HR practice adopted and implemented by organizations, it included the most representative domains of HRM suggested by HRM researchers ( "Best practices," 1999; Ulrich & Lake, 1990). Furthermore, the relationship between each HRM domain and organizational performance was revealed. Therefore, HR professional and practitioners can adopt a specific HRM practice to boost their organiztion's bottom line. While testing interrelationships among independent variables. It has been theorized that organizational performance indicators influenced one another. Despite numerous studies that suggest interrelationships among organizational performance measures, there were no studies incorporating such complete information. This study examined the interrelationships among the independent variables and found that employee's fs reduced by HRM practices indeed increased their job satisfaction and productivity, ultimately leads to improve the organizational performance. Employee productivity generated more revenue. Therefore, the findings from this study successfully took the theoretical model to a practical level.

Third, the effects of HRM practices on organizational performance were examined separately for managerial and nonmanagerial employees. This analytical approach has very important implications for both academicians and practitioners. For instance, the degree of implementation of HRM practices has been measured at unilateral level. However, as stated earlier in this study, this approach assumes that all employees are equally important. The findings of this study indicate that this assumption is indeed false. The effects of HRM implemented for the managerial employee group were very different from the effects of HRM implemented for nonmanagerial employees in terms of influencing a organization's bottom line. The results have very important implications for the public sector organizations and academicians. As a start, it was established that practitioners should evaluate the bottom line impact of HRM when they execute a certain type of HRM practice. As previously presented, not all HRM practices are likely to have the same impact on various employee groups.

Despite of some limitations, the results of this paprt move about the field of human resource management forward by empirically viewing a link between HR practices like job analysis, job design, job evaluation, job security , job succession planning and job performance. The implications of the distinctive consequence of the human resource practices on organizational performance measures were found remarkable. Altogether this paper makes an effort signifying a substantial positive contribution by HRM to organizational performance.

Generally, the findings of this study were found reliable with the studies conducted in western countries on the contribution of human resource practices related to the performance. Consequently, the worth of present study lies in the reality that it offers a requisite rationale of theoretical models built on the basis of studies conducted in the western organizations. The present study illustrates the efforts looking for establishing a Human Resource management-Performance linkage by developing a specific conceptual framework to demonstrate the progression by which the HR practice and performance might be correlated and by testing it with empirical data from a non-western country. It should be emphasized that the likelihood of HRM- Performance relationship has not been expansively explored in the South-Asia.