
































Microsoft Word - 135-FINAL.docx


http://journals.sfu.ca/abr  ADVANCES IN BUSINESS RESEARCH 
2016, Volume 7, pages 91-103 

	
	

91 
	

Testing the Effect of LMX and HR System Strength  
on Employee and Work Unit Outcomes 

 
Brian Martinson 
Tarleton State University 

 
John Deleon 

Tarleton State University 
 

A single, large organization with a uniform set of HR practices, was used to test a 
model predicting that HR system strength (HRSS) and leader-member exchange 
(LMX) variables will predict employee perceptions of HR practices (HRPP), which 
in turn predict employee job satisfaction and work unit performance. The study 
findings suggest that HRSS and LMX are positively related to variance in job 
satisfaction and work unit performance, and that part of the variance is related to 
work unit supervisors charged with implementing HR practices. The proposed 
relationships were analyzed using structural equation modeling. 
 
Keywords: LMX, HR system strength, job satisfaction, performance 

 
Introduction 

 
In 1998 the Gallup organization collected data from over 2,500 business units of 24 

companies and gathered responses from 105,000 employees in an attempt to identify what a “strong 
workplace” looked like (Buckingham & Coffman, 1999). Through their meta-analysis, they identified 
six questions having the strongest relationship with four separate measures of firm performance. All 
six questions were evaluations of the employees’ interactions with their supervisors, leading Gallup 
to conclude that the single greatest differentiator of firm performance was the relationship between 
frontline employees and their first-line supervisors. A considerable amount of research has been 
conducted by strategic human resource management (SHRM) scholars on the relationship between 
people management practices and firm performance, both before and after Buckingham and 
Coffman published their results in the New York Times’ Best Seller , “First, Break All The Rules” 
(McGregor, 2007). However, research related to the effect of variance in supervisor’s 
implementation of HR practices is fairly limited.   

Another critical issue within the SHRM research area has been the challenge put forth by 
research scholars to improve the methodological approach to measurements used in the practice-to-
performance research stream. Three of the most frequent and critical calls for improvement have 
been for the utilization of multiple sources of respondents to improve reliability, use of more 
proximal measures to improve validity, and the explication of what goes on in the “black box”  
between practice and performance to better understand how HR practices actually bring about 
increased performance (Boselie, Dietz, & Boon, 2005; Gerhart, Wright, & McMahan, 2000; Gerhart, 
Wright, McMahan, & Snell, 2000; Huselid & Becker, 2000, 2011; P. M. Wright & Boswell, 2002).  



LMX AND HR SYSTEM STRENGTH	

92 
	

This article seeks to contribute to the SHRM research stream by presenting a model and 
testing it with a methodology addressing several of these critical issues. Specifically, it is proposed 
that in order to measure the HR practices that actually influence employee behaviors, researchers 
need to measure the practices at the individual employee level.  It is further proposed that HR 
practices are related to performance through job satisfaction, and a more proximal measure of the 
effect of HR practices is made by measuring employees’ perception of the practices as their 
awareness and evaluation is required in order for the practice to have its intended effect on 
employee behaviors.  Employee perceptions of, and experience with, HR practices comes primarily 
through their interactions with their supervisor.  To test these proposals a model predicting a 
positive relationship between leader-member exchange (LMX) and employee perceptions of HR 
practices is presented that proposes a positive relationship between employee perceptions of HR 
practices and job satisfaction, and ultimately, a positive relationship between job satisfaction and 
work unit performance. To improve the reliability of the results, the model is tested with a single, 
large employer guided by a uniform set of HR policies and practices across 90 units and 40,992 non-
supervisory employees. Following is a review of the literature, presentation of the model, description 
of the methodology, and the results of the analysis. This article ends by summarizing the results and 
discussing the limitations of the study and offering suggestions for future research.  

 
Strategic Human Resource Management and Firm Performance 

Strategic human resource management studies are primarily concerned with understanding 
the relationship between firm strategies, human resource systems, and financial performance (Becker 
& Huselid, 2006; P. M. Wright & McMahan, 1992). The ultimate objective of SHRM studies is to 
identify sources of competitive advantages brought about by the design and implementation of 
human resource systems (Delery & Doty, 1996; Huselid, 1995; MacDuffie, 1995; McMahan, Virick, 
& Wright, 1999; P. M. Wright & McMahan, 1992; P. M. Wright, McMahan, & McWilliams, 1994). 
SHRM studies generally take one of two paths: one considers the relationship between individual, 
unit, and firm level employee attributes that comprise human capital resources and various measures 
of individual, unit, and firm performance.  The other investigates the relationship between the 
existence of human resource management practices and firm performance (Boselie et al., 2005; P. M. 
Wright & McMahan, 2011). This study focuses on the practice to performance research stream of 
SHRM. 

 
HR Practices 

Since Huselid’s (1995) seminal study in which he hypothesized that aligning a system of high 
performance work practices (HPWPs) with competitive strategies leads to reduced employee 
turnover and increased productivity and corporate financial performance, the practice-to-
performance model has inspired well over 100 published empirical studies (Boselie et al., 2005). 
Huselid’s 1995 study used 13 practices identified by the US Department of Labor (1993) as 
contributing to successful employee management and identified two key factors associated with 
employee turnover and productivity. They were employee skills and organizational structures, and employee 
motivation. In addition, he found that a one-standard deviation increase in HPWPs was associated 
with a 7.05% decrease in turnover and $27,044 more in sales and $18,641 and $3,814 more in market 
value and profits on a per employee basis, respectively.  

The impact of Huselid’s (1995) findings suggesting that substantial financial benefits were 
associated with the use of HPWPs had the twin effect of focusing the SHRM research field on the 
practice-to-performance model almost exclusively, and stimulated ruminations on the methodology 
employed to measure HPWPs. It also stimulated contemplations regarding the implied causality of 



Martinson and Deleon	

93 
 

HPWPs  leading to performance, which has yet to be solidly addressed with the exception of 
Schneider, Hanges, Smith, and Salvaggio (2003) who make a compelling case for firm performance 
leading to the use of some HR practices (cf. Boselie et al., 2005; Combs, Liu, Hall, & Ketchen, 2006; 
Gerhart, Wright, McMahan, et al., 2000; P. M. Wright & Boswell, 2002).  

As discussions over methodology and causality continued, scholars were busy searching for 
further support of the proposition that systems of HPWPs could yield improved performance.  
Combs, Liu, Hall and Ketchen (2006) published a meta-analysis of 92 practice-to-performance 
studies in which they found an effect size of .20 suggesting that 20% of the ability to predict 
performance differences among separate organizations could be attributed to their use of HPWPs. 
Their study also suggested that a one standard deviation increase in the use of HPWPs could 
translate into a 4.6% increase in ROA and a 4.4% decrease in turnover. Their meta-analysis focused 
on 13 most commonly studied practices including: incentive compensation, training, compensation 
level, participation, selectivity, internal promotion, HR planning, flexible work, performance 
appraisal, grievance procedures, teams, information sharing, and employment security.  
 
Sustained Competitive Advantage and Human Resource Systems 
 The dominant theory supporting the majority of HR practice studies is the resource based 
view (RBV) of the firm (Boselie et al., 2005). Organizations seek to survive and thrive by attaining a 
sustained competitive advantage over their competitors though resources contained within the 
organization. SHRM scholars have argued that both human capital and HR systems are resources 
that meet the RBV criteria for achieving a human resource based competitive advantage (cf. Lado & 
Wilson, 1994; McMahan et al., 1999; P. M. Wright, Dunford, & Snell, 2001; P. M. Wright & 
McMahan, 1992; P. M. Wright et al., 1994; P. M. Wright, Smart, & McMahan, 1995).  

RBV posits that four attributes of a resource (or capability) can provide a sustained 
competitive advantage given that all four are present in a particular resource. A resource must be 
valuable, rare, imperfectly imitable, and non-substitutable. Furthermore, the introduction of the 
VRIO framework (cf. Barney, 1995; Barney & Wright, 1998) adds that to achieve a competitive 
advantage, an organization must be capable of exploiting its resources through systems and practices 
designed to enable the organization to maximize the utilization of their resources.  

The RBV provides a strong explanation for how one firm may create a strategic advantage 
over another firm through the acquisition and development of human capital and the design and 
implementation of HR practices to manage human capital. However, more is needed to fully 
understand the linkages between people, practices and outcomes. Scholars have begun to develop 
theories addressing how HR practices are implemented, how they influence individual behaviors, 
and how practices interact with dimensions of individual attributes of human capital to yield 
performance improvements. The following is a review of two theories that guide this study: HRSS 
and leader-member exchange (LMX).  
 
Human Resources System Strength 

Bowen and Ostroff (2004) introduced a construct called HR system strength (HRSS), a 
dimension of an HR system that signals to employees in a way that helps them to clearly understand 
desired and appropriate responses to given workplace situations. HRSS helps organizational 
members form a collective understanding of the norms of behavior.  Based on the psychological 
principle of situational strength, it encompasses the extent to which workplace situations induce 
conformity (individuals interpret and react similarly), or are interpreted as ambiguous (individuals 
display a variety of interpretations and responses). HR system strength creates an organizational 
climate that is defined as a shared perception of organizational practices, policies, procedures, 



LMX AND HR SYSTEM STRENGTH	

94 
	

routines and rewards. Ostroff and Bowen define HRSS as, “a strong HRM system contains process 
mechanisms that make the set of practices coherent, salient and distinctive, and visible and 
understandable, the result of which builds consensus among employees about the practices and 
allows for shared perceptions of climate to emerge.” (Ostroff & Bowen, 2015 p. 3). 

According to Bowen and Ostroff (2004), HR practices operate as communications from 
employers to the employees. HR practices are designed to help firms acquire and develop 
employees’ knowledge, skills, abilities and motivations in ways that support the implementation of 
organizational strategies. To create a strong situation the HRM system must be perceived as high in 
distinctiveness, consistency, and consensus. The characteristics of HRM associated with 
distinctiveness include: visibility, understandability, legitimacy of authority, and relevance. 
Consistency is defined as establishing an effect over time through the use of different modalities. 
This is achieved through instrumentality (unambiguous perceived cause-effect relationships), validity 
(consistency between what they purport to do and what they actually do), and consistent HRM 
messages (compatibility and stability in the signals sent by the HR practices). The third characteristic 
of a strong situation, consensus, is the result of agreement among HRM decision makers (message 
senders), and perceptions of the overall fairness of the HRM system. Fairness, according to their 
framework, is determined by the system being congruent with principles of distributive, procedural 
and interactional justice (Bowen & Ostroff, 2004).  Increased levels of HRM system strength leads 
to increased uniform interpretations of messages and events that transpire in the workplace, and in 
turn, leads to the development of shared understandings and mental models resulting in greater 
alignment between intended and perceived policies and practices. They also serve to create an 
“influence situation whereby individuals yield to the message and understand the appropriate ways 
of behaving.” (Bowen & Ostroff, 2004: 213).  
 
Leader-Member Exchange (LMX) and Employee Perceptions 

LMX theory attempts to predict the dynamics associated with the relationship between 
supervisors and subordinates (Graen & Scandura, 1987). It ascribes to the same principles of 
interpersonal behavior found in social exchange theory: feelings of increased trust and gratitude, as 
wells as a sense of obligation as the frequency of interactions increases (Blau, 1964; Greguras & 
Ford, 2006).  Purcell and Hutchinson (2007) point out that, “HR practices perceived or experienced 
by employees will, to a growing extent, be those delivered or enacted by line managers, especially 
front-line managers (FLMs) with direct supervisory responsibility. It is often observed that there is a 
gap between what is formally required in HR policy and what is actually delivered by FLMs.”  Their 
study proposes and tests a model predicting that both front line managers’ (FLMs) behavior and 
employees’ perceptions of HR practices will affect organizational commitment and job satisfaction. 
They find support for their hypotheses suggesting employee perceptions of HR practices and LMX 
relationships are two important sources of variance in measures of organizational commitment and 
job satisfaction. This study seeks to add to their findings by testing the effect LMX behaviors have 
on employees’ perceptions of the use of HR practices and not their assessment of the value of, or 
satisfaction with, the practice as has been tested previously (cf.Khilji & Wang, 2006; Purcell & 
Hutchinson, 2007). This provides two significant contributions. The first contribution is to measure 
the variance between stated HR practices and the implementation of HR practices at the lowest level 
possible, the level at which they are designed to affect employee behavior. The second contribution 
is to measure the effect the primary implementer (the line level supervisor) of HR practices has on 
the perceptions of HR practices. Additionally, it is proposed that these effects will also relate to job 
satisfaction and unit performance, both of which are intended outcomes of the design and 
application of HR practices.  



Martinson and Deleon	

95 
 

Supervisor Role in HR System Strength 
This study proposes that the actions and cognitions of an employee’s supervisor will have a 

direct effect on the employee’s perceptions of the HR practices an organization uses to manage 
employee behavior toward the achievement of organization goals and objectives. This occurs 
through three dimensions. The first, and possibly the most direct, is through HR practices that are 
implemented at the supervisor’s discretion (Purcell & Hutchinson, 2007). Practices such as flextime, 
performance evaluations, and compensation decisions are traditionally administered either by, or 
through consultation with, an employee's direct supervisor. HR practices used to manage employees 
implemented through a supervisor’s discretion are, by their very nature, subject to considerable 
amount of variance between stated and implemented practices, and the supervisor’s interpretation 
and application of a given policy will more directly influence employees’ behavior than the stated 
policy as intended by senior leadership. The variation in supervisors’ interpretation and application 
of policy is a reflection of HRSS. This variation is addressed by Wright and Nishii (2011) as they 
suggest organization unit and organizational variance can be linked to individual outcomes. Ostroff 
and Bowen (2015) suggest that, “Signals sent by the (supervisor implemented) HR practices will be 
interpreted idiosyncratically unless other HR process mechanisms reinforce and thus help to create a 
strong HRM system.”. The quality of the relationship between employees and supervisors also 
affects their perceptions and experience of HR practices. As theorized in the LMX relationship 
research, subordinates have access to better and more information and opportunities as a result of 
their relationships with their super-ordinates (Harris & Kacmar, 2005).   
   To test these predicted relationships, the following hypotheses are presented: 
 

Hypothesis 1 – HRSS is positively related to employee perceptions of HR practices.  
Hypothesis 2 – HRSS is positively related to employee job satisfaction.    
 
In addition, it is proposed that the quality of the relationship between employees and 

supervisors will be related to levels of job satisfaction. Based on social exchange theory (Alfes, 
Shantz, Truss, & Soane, 2013; Blau, 1964; Gilbert, De Winne, & Sels, 2011; Gooty & Yammarino, 
2013), which suggests that higher quality exchange relationships yield increased access to resources, 
support, trust and consideration from and for the supervisor, the quality of the LMX relationship 
will be directly reflected in measures of employee job satisfaction.  
To test this relationships, the following hypotheses are presented: 
 

Hypothesis 3 – LMX is positively related to employee perceptions of HR practices.    
Hypothesis 4 – LMX is positively related to employee job satisfaction.    
 
Organizational leaders are responsible for developing policies and procedures for managing 

a firm’s resources.  As previously discussed, an organization’s human resources have the potential to 
provide an organization with a sustained competitive advantage. Successful organizations attempt to 
select the top available talent and provide employees with opportunities for career development and 
advancement, flexible and satisfying working conditions, and allow them to participate in the 
decisions that affect their workplace.  They also seek to recognize and reward employees with 
financial and other incentives for continuing the employer/employee relationship. It is proposed 
that intended HR practices are designed to create working environments and relationships that 
satisfy the needs and desires of employees and the degree to which employees are aware of these 
practices will be reflected in their level of satisfaction with the employee/employer relationship. To 
test this proposition, the following hypotheses are presented: 



LMX AND HR SYSTEM STRENGTH	

96 
	

Hypothesis 5 – Employee perceptions of HR practices are positively related to employee job satisfaction. 
 
The ultimate goal when designing a system of HR practices is to achieve the maximum level 

of operational performance. HR practices are designed to attract, develop and retain highly 
productive employees. HR practices affect performance by increasing the productivity per unit of 
cost per worker (Huselid, 1995; Ichniowski, Shaw, & Prennushi, 1997; MacDuffie, 1995). HR 
practices also are designed to facilitate the coordination, communication and cooperation among 
coworkers. Creating an HR system that engenders job satisfaction is not mutually exclusive from a 
system that achieves performance. HR systems can, and should, do both. Past research has shown a 
link between job satisfaction and performance (Purcell & Hutchinson, 2007). As an additional test of 
these relationships, the following hypotheses are proposed: 

 
Hypothesis 6 – Employee perceptions of HR practices is positively related to unit performance. 
Hypothesis 7 – Employee job satisfaction is positively related to unit performance. 

 
 
 
Figure 1. Model of Predicted Relationships 
 
 
 
 
 
 
 
 

 
 

 
 
 
 
 
 

 
Method 

 
To test the hypotheses (shown in Figure 1 above) a single large public sector organization 

with employees guided by a single set of formally documented HR practices was used. This helped 
to minimize extraneous variance and is consistent with current HR practice implementation research 
(Khilji & Wang, 2006; Nishii, Lepak, & Schneider, 2008). The sample was randomly selected from 
available responses of the organization’s annual 85 item employee survey. It contained complete 
responses from 40,992 employees across 90 separate agencies. Work unit size ranged from 12 
employees to 2,294 with a mean size of 455 and median of 197. The demographic composition of 
the sample was 55.7% female, 80% white, 11.7% black or African American, 6.6% American or 
Alaska Native, 2.1% Hispanic or Latino, and 1% Asian. The age of the respondents was distributed 
as follows: 42% were in the 50 to 59 year old range, 27.6% were 40 to 49, 15.4% were 30 to 39, 10% 

Job	
Satisfaction 

Unit	
Performance 

HR	Practice	
Perceptions 

HRSS 

LMX 

H2	
+ 

H4	
+ H6	

+ 
H3	
+ 

H5	
+ 

H1	
+ 

H7	
+ 



Martinson and Deleon	

97 
 

were 60 and older, and less than 5% were under the age of 30.  The distribution of respondent’s unit 
tenure suggests that 36% have been employed in their unit 15 to 20 years, 26.5% from 11 to 14 
years, 20.2% from 6 to 10 years, 6.1% from 4 to 5 years and 10.7% 3 years or less. 
 
Measures 

Survey items were measured on a 5-point scale with anchors such as “strongly agree” and 
“very good” assigned a value of 5 and “strongly disagree” and “very poor” assigned a value of 1. 
Using factor analysis with principal axis factoring and Varimax rotation, the factors designed to test 
the hypotheses were confirmed. They include: LMX measured with four items consistent with 
Graen and Uhl-Bien’s (1995) LMX-7 scale (Cronbach's α =.896); sample items include, “I have trust 
and confidence in my supervisor,” and “Overall, how good a job do you feel is being done by your 
immediate supervisor/team leader?”  

HR Practice Perceptions (HRPP) was a composite factor comprised of 6 items (Cronbach's 
α=.922) assessing the existence and use of the following HR practices: internal promotion, 
performance-based pay and rewards, training and needs assessment, employment security, and 
grievance procedure. Sample items included, “Promotions in my work unit are based on merit,” 
“Pay raises depend on how well employees perform their jobs” and “Employees have electronic 
access to learning and training programs readily available at their desk.” Job Satisfaction (4 items, 
Cronbach's α=.805) included items such as, “I like the kind of work I do,” “I know how my work 
relates to the agency’s goals and priorities,” and “The work I do is important.”   

In an organization with high levels of HRSS, it is expected that employees and supervisors 
will have similar evaluations of organizational goals, policies, and priorities as set by senior 
leadership. In order to calculate HRSS first an Evaluation of Leadership (5 items, Cronbach's α=.928) 
variable for every employee was calculated. The sample items included, “Managers communicate the 
goals and priorities of the organization.” and “How satisfied are you with the policies and practices 
of your senior leaders?” Increased distance between employees and their supervisors’ evaluations 
will reflect weak system strength, signaling the potential lack of consistency and consensus (Bowen 
& Ostroff, 2004). It is proposed that weak system strength will increase the variability of the 
implementation of stated HR practices such that an employees’ perceptions and experience of stated 
practice will be a reflection of the degree to which supervisors uniformly communicate and utilize 
the stated practices.  

While it was not possible to match individual supervisors and their direct reports, the 
variation between an individual employee’s perception of organizational policies and those of the 
unit’s supervisors was measured. To calculate the HRSS variable, first the mean of the aggregated 
responses for all unit supervisors (n=8,965) by each unit for the Evaluation of Leadership factor was 
computed (Cronbach's α=.928).  Next, the value was subtracted from each individual unit’s 
employees’ responses taking the absolute value as the difference between employee and supervisor 
perspective. This step captures the distance between employee and supervisor attitudes towards the 
goals, policies, and practices put in place by senior management. Finally, the absolute value was 
multiplied by negative one in order to align the variable with the construct as defined and simplify 
interpretation, i.e., multiplying by negative one makes stronger HRSS to be represented by larger 
values and weaker HRSS to be represented by smaller values. For robustness, a simple subtraction 
operationalization was checked; the model differences were insignificant (p>.10). The Unit 
Performance variable was created using the mean of the aggregated unit managers’ (n=5,583) 
responses to the (4 item) factor measuring unit work quality (Cronbach's α=.703). Sample items 
include:  “The people I work with cooperate to get the job done,” and “How would you rate the 
overall quality of work done by your work group?” 



LMX AND HR SYSTEM STRENGTH	

98 
	

Analysis 
 

 The hypotheses were tested using structural equation modeling within LISREL 8.80. The 
analysis began with a confirmatory factor analysis (CFA) to determine item loadings on the 
appropriate factor to determine how well the data fits the proposed five-factor structure model. 
Model fit ranged from good to fair with ! !"

! = 24,325.76 ! < .01 , CFI=0.947, RMSEA=0.08, 
and SRMR=0.050. To further address concerns of dimensionality, several additional models were 
run to confirm the proposed factor structure. All tests for alternate forms of dimensionality using 
!! difference testing led to the conclusion that the five-factor structure model was most appropriate. 
 In addition, the possibility of common method bias was investigated using a Harmon’s one-
factor test. Model fit (!(!"#)! = 302,971.20,! < .01,!"# = 0.23,!"#$% = 0.26, !"#" =
0.16) dropped drastically with the fit indices indicating a poorly fitting model. Further, a !!   
difference test between the five-factor structure model and a one-factor structure model confirmed 
the model was significantly worse (p<.01); leading to the conclusion that common method bias does 
not significantly affect the analysis. 
 All hypotheses were tested using a full structural model with all factors included. Model fit 
was acceptable with ! !"

! = 24,387.9 ! < .01 , CFI=0.94, RMSEA=0.08, and SRMR=0.050. 
Note that, although the RMSEA is above the more stringent .06 cut-off recommended by Hu & 
Bentler (1998), the RMSEA is below .08 cut-off for reasonable model fit. The full structural model 
(Figure 2) is shown with LISREL ML estimates. 
 

 
Figure 2. Path Diagram and Results 

 
 
 

Results 
 

 Hypothesis 1 is strongly supported in the full structural model. The path is positive and 
significant (β=.15, p<.01), suggesting that HRSS, measured as the separation between employee and 
supervisor evaluations of leadership’s communication of goals and priorities, has a positive effect on 
HR practice perceptions. Stated another way, less dissimilarity between ratings of supervisors and 
employees predicts levels of higher HR practice perceptions. Similarly, hypothesis 2 is strongly 



Martinson and Deleon	

99 
 

supported (β= .04, p<.01), suggesting that HRSS is positively related to employee job satisfaction. 
Standardized path coefficients suggest that the relationship between HRSS and HR practice 
perceptions is stronger than the relationship between HRSS and job satisfaction. 
 Hypothesis 3 is strongly supported (β=.65, p<.01), suggesting that LMX, measured as 
employee evaluations of supervisors, is positively related to HR practice perceptions.  
Additionally, hypothesis 4 proposing that LMX is positively related to job satisfaction finds strong 
support (β=.32, p<.01). Similar to hypotheses 1 and 2, a comparison of the standardized coefficients 
suggests that LMX is more strongly related to HR practice perceptions than to job satisfaction.  
 Hypothesis 5, suggesting that HR practice perceptions is positively related to job satisfaction, 
is strongly supported (β=.28, p<.01). While the effect size is modest, a relationship between 
perceptions of HR practices and job satisfaction is clearly identified. Additionally, hypothesis 6, 
predicting a positive relationship between HR practice perceptions and unit performance also finds 
support (β=.03, p<.01).  However, the effect size is small suggesting that the relationship while 
statistically significant it is of limited practical value. And lastly, hypothesis 7 finds support (β=.00, 
p<.05) with an extremely small effect size. Again, the relationship while statistically significant is 
practically insignificant. A summary of the findings is included in table 1. 
 
 

 
Table 1. Summary of Findings 
 

Hypotheses Results Standardized 
Coefficient 

H1 HRSS is positively related to employee perceptions of HR 
practices.  Supported 0.15** 

H2 HRSS is positively related to employee job satisfaction.  Supported 0.04** 

H3 Leader/Member Exchange is positively related to employee 
perceptions of HR practices. Supported 0.65** 

H4 Leader/Member Exchange is positively related to employee 
job satisfaction.    Supported 0.32** 

H5 Employee perceptions of HR practices is positively related 
to employee job satisfaction. Supported 0.28** 

H6 Employee perceptions of HR practices is positively related 
to unit performance. Supported 0.03** 

H7 Employee job satisfaction is positively related to unit 
performance. Supported 0.00* 

   **Significant at p<.01, *Significant at p<.05 
 
 

Limitations 
 

The results of this study using HR practice perceptions at the individual level with a large 
sample of employees managed under a single set of HR practices contributes to the SHRM research 
area with a unique empirical test of the effects of HR system strength. However, there are some 
limitations affecting the study that need to be discussed.  First, although the study’s sample size is 
large which helps to strengthen arguments of generalizability, because of the nature of the data set, a 



LMX AND HR SYSTEM STRENGTH	

100 
	

pooled cross-sectional analysis method was used. This limits the ability to make causal inferences. 
Second, all data was gathered from a single large public sector firm, which may limit generalizability 
to other firms such as large for-profit corporations and small and medium private organizations. 
Furthermore, because of data constraints, it was not possible to directly match each employee to 
each particular supervisor, which prevents a perfectly dyadic relationship when examining 
congruence between supervisor and employee evaluations of leadership. Additionally, the research 
design was cross-sectional so no assentation about causality can be made. Lastly, the study used 
secondary data not designed specifically for this study which was collected in environments outside 
of the control of the authors. This also may limit some of the inferences that can be made and 
supported by the study results.  

 
Conclusion 

 
The objective with this study is to identify the influence supervisors have on employee 

perceptions of HR practices in order to contribute to the discussion on the methodological 
problems associated with the traditional manner in which HR practices-to-performance process 
research is conducted. In the typical hierarchical organizational structure designed to disseminate 
communications, address uncertainty, and manage individual employees (Galbraith, 1973), the main 
source of information and responsibility for decisions affecting the employer/employee relationship 
is through an employee’s immediate supervisor.  The analysis shows that positive relationships exist 
between LMX and HRSS, and employees’ perceptions of HR policies and practices. This, in turn, is 
related to increased job satisfaction and work unit performance. These results suggest a confirming 
test of the effects of HRSS in relation to the effectiveness of HR practices.  

HR policies and practices are designed and implemented by organizations to maximize 
desired employee behaviors such as job satisfaction and job performance. Uniformly communicating 
and administering practices is essential to achieving the desired outcomes. Supervisors are a critical 
link in the communication and implementation of policies and practices, and contribute to the 
varying degree to which the policies and practices achieve their intended effect. This study suggests 
that researchers and practitioners alike could benefit from further exploring the linkages between 
HR policies and practices designed to affect employee behaviors at the organization level and the 
actual implementation of policies and practices that the targets of the practices, the individual 
employees within an organization, actually experience. The study suggests that a greater difference 
between these two levels equates to a loss of potential performance.  

 These results also support that the concerns raised by scholars regarding the reliability of 
using single, distal sources for identifying the use of HR practices across an organization is well 
placed. When a study uses the perceptions of a single top level manager as the source of an 
organization’s intended HR practices used to influence employee behaviors, it is a bit of a stretch to 
infer that the response is actually connected to measured behavior. A succinct example of the chasm 
that may exist between intended and realized practices may be best illustrated by this simple 
anecdote. One of the authors of this paper was attending a Society of Human Resource 
Management presentation given by Scott Cawood, who at the time was a Vice President with the 
Great Places to Work Institute, producer of Fortune´s "100 Best Companies to Work For” list. At 
this meeting Dr. Cawood asked attendees to provide an example that differentiated a good place to 
work from a great place to work. Lacking a response from the audience, he provided this succinct 
example, “a good place to work has a flextime policy; a great place to work lets you use it.” This 
unambiguous example illustrates that having a policy is not the same as utilizing a policy to achieve a 
desired behavioral outcome.  A future research stream with the potential to improve SHRM 



Martinson and Deleon	

101 
 

researchers’ understanding of practice to performance linkages might be measuring the effects on 
performance brought about by differences between espoused practices and the actual 
implementation of practices. Is this relationship constant, curvilinear, or moderated by some other 
condition? As the field of SHRM matures, it is critical to incorporate improved measures of 
predictor and response variables to allow for better inferences regarding the relationship between 
HR practices and firm performance. This, in turn, will provide scholars and practitioners alike with a 
greater understanding of which practices stimulate desired behaviors and how they work.  

 
References 

 
Alfes, K., Shantz, A. D., Truss, C., & Soane, E. C. (2013). The link between perceived human 

resource management practices, engagement and employee behaviour: a moderated 
mediation model. International Journal of Human Resource Management, 24(2), 330-351. 
doi:10.1080/09585192.2012.679950 

Barney, J. B. (1995). Looking inside for competitive advantage, 49. 
Barney, J. B., & Wright, P. M. (1998). On becoming a strategic partner: The role of human resources 

in gaining competitive advantage. Human Resource Management, 37(1), 31-46.  
Becker, B. E., & Huselid, M. A. (2006). Strategic human resources management: Where do we go 

from here? Journal of Management, 32(6), 898-925. doi:10.1177/0149206306293668 
Blau, P. M. (1964). Exchange and power in social life. New York: Wiley. 
Boselie, P., Dietz, G., & Boon, C. (2005). Commonalities and contradictions in HRM and 

performance research. Human Resource Management Journal, 15(3), 67-94.  
Bowen, D. E., & Ostroff, C. (2004). Understanding HRM-firm performance linkages: The role of 

the "strength" of the hrm system. Academy of Management Review, 29(2), 203-221.  
Buckingham, M., & Coffman, C. (1999). First, break all the rules. New York: Simon and Schuster. 
Combs, J., Liu, Y., Hall, A. T., & Ketchen, D. J. (2006). How much do high performance work 

practices matter?" A meta-analysis of their effects on organizational performance. Personnel 
Psychology, 59, 501-528.  

Delery, J. E., & Doty, D. H. (1996). Modes of theorizing in strategic human resource management: 
Tests of universalistic, contingency, and configurational performance predictions. . Academy 
of Management Journal, 39(4), 802-835.  

Galbraith, J. R. (1973). Designing complex organizations. Reading, Mass.: Addison-Wesley Pub. Co. 
Gerhart, B., Wright, P. M., & McMahan, G. C. (2000). Measurement error in research on the human 

resources and firm performance relationship: Further evidence and analysis. . Personnel 
Psychology, 53(4), 855-872.  

Gerhart, B., Wright, P. M., McMahan, G. C., & Snell, S. A. (2000). Measurement error in research on 
human resources and firm performance: How much error is there and how does in influence 
effect size estimates? Personnel Psychology, 53(4), 803-834.  

Gilbert, C., De Winne, S., & Sels, L. (2011). The influence of line managers and HR department on 
employees' affective commitment. International Journal of Human Resource Management, 22(8), 
1618-1637. doi:10.1080/09585192.2011.565646 

Gooty, J., & Yammarino, F. J. (2013). The Leader–Member Exchange Relationship A Multisource, 
Cross-Level Investigation. Journal of Management, 0149206313503009.  

Graen, G. B., & Scandura, T. A. (1987). Toward a psychology of dyadic organizing Research in 
Organizational Behavior, 9, 175-208.  

Graen, G. B., & Uhl-Bien, M. (1995). Relationship-based approach to leadership: Development of 
leader-member exchange (LMX) theory of leadership over 25 years: Applying a multi-level 



LMX AND HR SYSTEM STRENGTH	

102 
	

multi-domain perspective. The Leadership Quarterly, 6(2), 219-247. doi:10.1016/1048-
9843(95)90036-5 

Greguras, G. J., & Ford, J. M. (2006). An examination of the multidimensionality of supervisor and 
subordinate perceptions of leader-member exchange. Joumal of Occupational and Organizational 
Psychology, 79, 433-465.  

Harris, K. J., & Kacmar, K. M. (2005). Easing the strain: The buffer role of supervisors in the 
perceptions of politics--strain relationship. Journal of Occupational & Organizational Psychology, 
78(3), 337-354. doi:10.1348/096317905x26110 

Hu, L.-t., & Bentler, P. M. (1998). Fit indices in covariance structure modeling: Sensitivity to 
underparameterized model misspecification. Psychological Methods, 3(4), 424.  

Huselid, M. A. (1995). The impact of human resource management practices on turnover, 
productivity, and corporate financial performance. Academy of Management Journal, 38(3), 635-
672.  

Huselid, M. A., & Becker, B. E. (2000). Comment on “Measurement error in research on human 
resources and firm performance”: How much error is there and how does it influence effect 
size estimates?' by Gerhart, Wright, McMahan, and Snell. Personnel Psychology, 53(4), 835-854.  

Huselid, M. A., & Becker, B. E. (2011). Bridging micro and macro domains: Workforce 
differentiation and strategic human resource management. Journal of Management, 37(2), 421-
428. doi:10.1177/0149206310373400 

Ichniowski, C., Shaw, K., & Prennushi, G. (1997). The effects of human resource practices on 
manufacturing performance: A study of steel finishing lines. American Economic Review, 87(3), 
291-313.  

Khilji, S. E., & Wang, X. (2006). ‘Intended’ and ‘implemented’ HRM: The missing linchpin in 
strategic human resource management research. The International Journal of Human Resource 
Management, 17(7), 1171-1189. doi:10.1080/09585190600756384 

Lado, A. A., & Wilson, M. C. (1994). Human resource systems and sustained competitive advantage: 
A competency-based perspective. Academy of Management Review, 19, 699-727.  

MacDuffie, J. P. (1995). Human resource bundles and manufacturing performance: Organizational 
logic and flexible production systems in the world auto industry. Industrial & labor relations 
review, 48(2), 197-221.  

McGregor, J. (2007). Ladies and gents…Marcus Buckingham! News & Insights.  Retrieved from 
http://www.businessweek.com/magazine/content/07_13/b4027050.htm 

McMahan, G., Virick, M., & Wright, P. (1999). Alternative theoretical perspectives for SHRM: 
Progress, problems, and prospects. In L. D. J. B. P. Wright & G. Milkovich (Eds.), (Vol. 
Supplement 4, pp. 99-122). Greenwich, CT: JAI Press. 

Nishii, L. H., Lepak, D. P., & Schneider, B. (2008). EMPLOYEE ATTRIBUTIONS OF THE 
“WHY” OF HR PRACTICES: THEIR EFFECTS ON EMPLOYEE ATTITUDES AND 
BEHAVIORS, AND CUSTOMER SATISFACTION. Personnel Psychology, 61(3), 503-545. 
doi:10.1111/j.1744-6570.2008.00121.x 

Ostroff, C., & Bowen, D. (2015). Reflections on the 2014 Decade Award: Is there strength in the 
construct of HR system strength? Academy of Management Review, amr. 2015.0323.  

Purcell, J., & Hutchinson, S. (2007). Front-line managers as agents in the HRM-performance causal 
chain: Theory, analysis and evidence. Human Resource Management Journal, 17(1), 3-20. 
doi:10.1111/j.1748-8583.2007.00022.x 

Schneider, B., Hanges, P. J., Smith, D. B., & Salvaggio, A. N. (2003). Which comes first: Employee 
attitudes or organizational financial and market performance? The Journal of applied psychology, 
88(5), 836-851. doi:10.1037/0021-9010.88.5.836 



Martinson and Deleon	

103 
 

U.S. Department of, L. (1993). High performance work practices and firm performance. Retrieved from 
Washington, DC:  

Wright, P., & Nishii, L. H. (2011). Strategic HRM and organizational behavior: Integrating multiple levels of 
analysis. Oxford: Blackwell Publishing. 

Wright, P. M., & Boswell, W. R. (2002). Desegregating HRM: A review and synthesis of micro and 
macro human resource management research. Journal of Management, 28(3), 247-276.  

Wright, P. M., Dunford, B. B., & Snell, S. A. (2001). Human resources and the resource based view 
of the firm. Journal of Management, 27(6), 701-721. doi:10.1177/014920630102700607 

Wright, P. M., & McMahan, G. C. (1992). Theoretical perspectives for strategic human resource 
management. Journal of Management, 18(2), 295.  

Wright, P. M., & McMahan, G. C. (2011). Exploring human capital: Putting human back into 
strategic human resource management. Human Resource Management Journal, 21(2), 93-104.  

Wright, P. M., McMahan, G. C., & McWilliams, A. (1994). Human resources and sustained 
competitive advantage: A resource-based perspective. International Journal of Human Resource 
Management, 5(2), 301-326.  

Wright, P. M., Smart, D. L., & McMahan, G. C. (1995). Matches between Human Resources and 
Strategy among Ncaa Basketball Teams. Academy of Management Journal, 38(4), 1052-1074.  

  
 
Brian Martinson is an assistant professor of management at Tarleton State University. 
 
John Deleon is an assistant professor of management at Tarleton State University. 
	


