fusing family and firm

7/23/2019 Fusing Family and Firm

1/14

Fusing family and firm: Employee perceptions of perceived homophily,

organizational justice, organizational identification, and organizational

commitment in family businesses

Anna F. Carmon a,*, Amy N. Miller b,1, Amber N.W. Raile c,2, Michelle M. Roers d

a Indiana University-Purdue University Columbus, 4601 Central Avenue, Columbus, IN 47203, United Statesb North Dakota State University, P.O. Box 5075, North Dakota State University, Fargo, ND 58105-5075, United Statesc North Dakota State University, P.O. Box 5075, North Dakota State University, Fargo, ND 58105-5075, United Statesd Sanford Health, 801 Broadway Drive, Fargo, ND 58122, United States

1. Introduction

Extant research has established a strong connection between

employee commitment and firm productivity and performance

(Muse, Rutherford, Oswald, & Raymond, 2005; Whitfield & Poole,

1997). Organizational commitment is a strong belief in the

organizations goals and values, a willingness to work on behalf

of the organization, and a desire to maintain membership in the

organization (Porter, Steers, Mowday, & Boulian, 1974). This sense

of commitment among employees could be used not only to

preserve the longevity of their businesses, but also to create a

positive working environment for employees.

Whether employees perceptions of how similar they see

themselves to other employees and whether they believe they

are treated fairly should affect how identified and committed they

become to their organizations. Perceptions of organizationaljustice and homophily likely will contribute to how connected

individuals feel with the family businesses for which they work.

Previous research has found strong, positive correlation between

perceptions of organizational justice and perceptions of co-worker

identification, satisfaction, and support as well as affective

commitment (Byrne, 2003; Cohen-Charash & Spector, 2001;

Colquitt, Conlon, Wesson, Porter, & Ng, 2001). While prior research

has focused on these constructs within the context of non-family

firms, little to no research has addressed them within a family

business context.

Organizational identification has been researched in family

businesses. However, much of previous research only considers the

identity development and identity conflict of family member

employees (Dyer & Whetten, 2006; Reay, 2009; Shepherd &

Haynie, 2009a, 2009b; Sundaramurthy & Kreiner, 2008), as they

are the employees who constantly struggle to manage their family

and organizational identities simultaneously within their family

businesses. However, more than 80% of employees of family

businesses are non-family members (Mass Mutual, 2007). It is

likely that non-family member employees struggle to find theirplace within the family business for which they work as they do

not necessarily have in-group status due to their lack of

membership in the founding family. Therefore, research needs

to consider how this identity development process may affect all

family business employees, not just those related to the founding

family.

Recently, scholars have begun to examine the role of non-

family member employees (Astrachan & Keyt, 2003; Barnett &

Kellermans, 2006; Chua, Chrisman, & Sharma, 2003). Family and

non-family employees compare themselves to each other as

they develop their roles and identities within their organizations.

Journal of Family Business Strategy 1 (2010) 210223

A R T I C L E I N F O

Keywords:

Family business

Interpersonal justice

Informational justice

Homophily

Organizational identification

Organizational commitment

A B S T R A C T

Social identity theory has been applied to many organizational contexts, including family businesses.

However, the current study is one of the first to explore the organizational identification of non-family

member employees. Based on previous research, it seems likely that, for family business employees,

organizational identification mediates the relationship between organizational justice, homophily, and

commitment. This study proposes a model of identification for family business employees based on

these considerations. Although the current study did not confirm the proposed model, an alternative

model is discussed.

2010 Elsevier Ltd. All rights reserved.

* Corresponding author. Tel.: +1 812 348 7213.

E-mail addresses: [email protected](A.F. Carmon),

[email protected](A.N. Miller), [email protected](Amber N.W. Raile),

[email protected](M.M. Roers).1 Tel.: +1 701 231 8586.2 Tel.: +1 701 231 7705.

Contents lists available at ScienceDirect

Journal of Family Business Strategy

j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / j f b s

1877-8585/$ see front matter 2010 Elsevier Ltd. All rights reserved.

doi:10.1016/j.jfbs.2010.10.003
http://dx.doi.org/10.1016/j.jfbs.2010.10.003mailto:[email protected]:[email protected]:[email protected]:[email protected]://dx.doi.org/10.1016/j.jfbs.2010.10.003http://dx.doi.org/10.1016/j.jfbs.2010.10.003mailto:[email protected]:[email protected]:[email protected]:[email protected]://dx.doi.org/10.1016/j.jfbs.2010.10.003


2/14

Non-family employees within family businesses likely provide

additional challenges non-family firms may not face in terms of

creating a strong sense of identification and commitment among

family and non-family members alike.

The development of family and business identities as either

separate or integrated entities of family member employees affects

non-family employees perceptions and work experiences. If the

founding familys family and business identities are seen as

integrated, non-family employees may be given limited access to

managerial opportunities, limited contributions to organizational

decision-making, and limited presence on governing boards

(Lubatkin, Schulze, Ling, & Dino, 2005; Schulze, Lubatkin, & Dino,

2003b). On the other hand, if the founding familys family and

business identities are viewed as separate, non-family employees

are likely to have a larger, more important presence in family

businesses (Sundaramurthy & Kreiner, 2008). Other research

suggests family businesses should develop a family business

meta-identity addressing who they are as a family business,

rather than developing who they are as a family and who they are

as a business separately (Reay, 2009; Shepherd & Haynie, 2009a).If

they are able to do this, family businesses will be more likely to be

able to create a strong sense of organizational identification and

commitment where all family business employees perceive

themselves as not only similar in values, beliefs and attitudes,but as also treated fairly within the organization.

To understand more thoroughly how family business employ-

eescome to identify with andcome to be committedto their family

businesses, we decided to draw on social identity theory (SIT),

based in intergroup theory (Tajfel, 1982). Social identity theory

explains how individuals compare themselves to members of their

social groups as well as to other social groups (Tajfel & Turner,

1986) in order to develop a series of social identities, including

their organizational identities. Family business employees must

compare themselves to both non-family member and family

member employees in order to develop their unique identities and

come to identify with their organizations.

Family business employees are likely to form their perceptions

of organizational identification and commitment based on theirperceptions of how they are treated within their family businesses

and whether or not that treatment varies among employees.

Employees will likely focus on their perceptions of organizational

justice and homophily to determine how connected to their

organizations they feel. Organizational justice is the perception of

whether individuals are treated with respect based on the

decision-makers and the policies in their organizations (Colquitt,

2001; Scott, Colquitt, & Zata-Phelan, 2007). Similarly, individuals

base their perceptions of homophily on how similar they see their

attitudes, values, and backgrounds to others in their organizations

(Lazarsfeld & Merton, 1954; McCroskey, Richmond, & Daly, 1975).

SeeTable 1for more clarification of the current studys constructs.

When employees identify with their family business, they are

likelyto be committed to thefamily business as identification has

been linked to both work satisfaction and lower turnover rates,

and, ultimately, commitment (van Dick et al., 2004). Employees

perceptions that they are members of a cohesive organization

tend to lead to a more productive workplace (Fleming, 2000;

Richter, Van Dick, & West, 2004). Therefore, through addressing

family business employees perceptions of justice and homophily,

both researchers and practitioners may come to better under-

stand their experiences in order to create better working

environments within family businesses. To this end, the current

study tests test whether ones status as a family member or

non-family member employee affects perceptions of organiza-

tional justice, homophily, organizational identification, and

commitment.

Further, it will also test whether organizational identification

mediates the relationship between organization justice and

homophily and organizational commitment.

2. Literature and hypotheses

2.1. Family businesses

Forthe purposes of this study,a family business is definedas: a

business governed and/or managed with the intention to shape

and pursue the vision of the business held by a dominant coalition

controlled by members of the same family or a small number of

families in a manner that is potentially sustainable across

generations of the family or families (Chua, Chrisman, & Sharma,

1999, p. 25). The focus of this definition is the role of the family

within a family business. As such, families as owners are likely to

affect the development of a strong sense of organizational

identification among their employees differently than owners of

non-family firms.

Family business owners have the opportunity to aid their

families and their businesses by developing a family business

meta-identity in order to address who they are as a family

business, rather than developing who they are as a family and who

they are as a business separately (Shepherd & Haynie, 2009a).

Through the development of this meta-identity, family businessesare able to use the intersection of the business and families to

benefitnot only the business, butalso the psychological well-being

of the family (Shepherd & Haynie, 2009b). For instance, family

values such as commitment, loyalty, and supportiveness will likely

benefit the business, while an in-depth knowledge of family

personalities, values and beliefs, and weaknesses will allow family

businesses to make decisions for the benefit of the family. Further,

this meta-identity allows family businesses to develop a sense of

pride in their distinctiveness as a family business ( Reay, 2009, p.

1267). This sense of pride will likely contribute to feelings of

organizational identification, as allemployees are able to develop a

sense of belonging to the family and the business simultaneously,

rather than just to the businesssomething that non-family firms

cannot do.

Table 1

Definitions of study constructs.

Construct Definition Key references

Interpersonal justice Perceptions of how individuals are treated with

respect and dignity by decision-makers

Colquitt (2001)

Informational justice Perceptions of how honestly and thoroughly decision-makers

explain the rationale for their decisions

Colquitt (2001)

Homophily Perceptions of si milarity in attitude, background,

values, and/or appearance

Lazarfeld & Merton (1954),McCroskey et al. (1975),

Rogers & Bhowmik (1970)

Organizational identification Perceptions of shared values, belonging, and

loyalty to an organization

Burke (1937),Cheney (1982, 1983),Mael and

Ashforth (1992)

Organizational commitment Perceptions of how connected an individual is

with an organization

Meyer and Allen (1991, 1997),Mowday et al. (1979),

Porter et al. (1974)

A.F. Carmon et al. / Journal of Family Business Strategy 1 (2010) 210223 211


3/14

Non-family memberemployees have a great deal to offer family

businesses; however, their contributions are often not accepted

without a struggle. Non-family member CEOs and employees are

forcedto notonly understand howto effectively perform their jobs,

but also how to best interact with the founding family ( Mitchell,

Morse, & Sharma, 2003). In order to gain a more complete

understanding of family businesses, it is necessary to consider the

experiences of all family business employees. Family businesses

provide a relevant context for addressing social identity as

employees of family businesses likely make comparisons between

family member and non-family member employees as they

develop their organizational identities. More specifically, SIT

allows us to consider how individuals develop their social

identities based upon comparisons between in-groups and their

related out-groups.

2.2. Social identity theory

Social identity theory is based in intergroup theory, initially

developed by social psychologists to describe the effect of group

membership on intergroup behavior. Intergroup theory further

suggested by categorizing individuals into groups, they would

favor thegroups of which they were part andlook differently at the

groups of which they were not (Billig & Tajfel, 1973; Tajfel, 1982).However, categorization was ultimately found not to be the only

cause of group favoritism (Turner, 1975, 1978, 1981). Rather a

combination of social identity and group categorization lead to a

more complete understanding of intergroup behavior, suggesting

that an individual must not only be categorized into a social group,

he or she must identify with the group before feelings of group

favoritism will develop. It was through these ideas that SIT was

developed.

Social identity theory posits that an individual has a single

personal identity, but multiple social identities (Hogg & Vaughn,

2002; Tajfel & Turner, 1979; Terry, Hogg, & White, 1999). Ones

social identity is formed as an individual places himself or herself

into distinct social categories (Scott, 2007) based on similarities

with these groups (Ashforth & Mael, 1989), such as students,Christians, or females. It is also through this process that

individuals develop their organizational identities, a specific type

of social identity (Pratt, 1998), through comparing themselves

with others in their organizations and reflecting on these

comparisons over time (Albert & Whetten, 1985). For instance,

non-family member employees in family businesses will develop

their organizational identities based upon comparisons with other

non-family member employees as well as family member

employees. Similarly, family member employees will engage in

the same process to develop their organizational identities.

When a group identity is important to an individual, he or she

attempts to distinguish between that socially desirable in-group

and those belonging to the undesirable out-group(Abrams, Hogg,

Hinkle, & Otten, 2005). Both in-groups and out-groups aredetermined byprototypes, or fuzzy sets of interrelated attributes

that simultaneously capturesimilaritiesand structural relationships

within groups and differences between groups, and prescribe group

membership related behavior (Hogg, Abrams, Otten, & Hinkle,

2004, p. 253). For instance, as a non-family member employee feels

more attached to their non-family member colleagues, the more

important being a non-family member employee will become to his

or her overall organizational identity. As an individual feels more

connected to a group, that in-group becomes a more salient portion

of his or her social identity (Brewer, 2001).

An individual begins to make inter-group comparisons after he

or she has developed his or her own social identity. Social identity

theory suggests that individuals evaluate their social identities by

comparing their in-groups to their relevant out-groups (Tajfel &

Turner, 1986). For instance, non-family member employees likely

compare themselves to family member employees and how they

are treated when they are developing their organizational

identities. Generally, favorable comparisons of social groups,

when an individual perceives his or her social group to be distinct

from other social groups in a positive way, are the foundation of a

positive social identity.

Social identities are often developed through social interac-

tions (Li, Xin, & Pittutla, 2002). Therefore, an employees

organizational identity is created through workplace interaction.

Organizational identification occurs when an individual melds the

beliefs of the organization with his or her social identity (Pratt,

1998). If one non-family member employee interacts with

members of the founding family on a daily basis while another

non-family member employee works only with other non-family

member employees, the two are likely to develop different

organizational identities. Social identity theory has been used to

study the organizational context (e.g.,Postmes, Tanis, & de Wit,

2001; Scott, 2007) but has not been used to address family

businesses a great deal (Reay, 2009). Family businesses can be

comprised of both family member and non-family member

employees, potentially forming specific in-groups and out-groups

within the organization.

Employees who identify with their organizations and considerthemselves members of the in-group tend to have positive social

identities, viewing their organizational identity as a beneficial and

constructive part of their personal identity (Richter et al., 2004). If

employees perceive themselves as members of a cohesive

organization, they are less likely to be negatively affected by

inter-group comparisons (Richter et al., 2004). However, if an

employee does not identify with an organization, it can lead him or

her to distrust organizational information or to interpret the

information in a way that reinforces feelings of disidentification

(Dukerich, Kramer, & Parks, 1998).

2.3. Organizational identification

Organizational identification is a specific type of socialidentification that elicits perceptions of shared values, belonging,

andloyalty to an organization(Cheney, 1982). Burke(1937) argued

that individuals naturally respond to division by identifying with

other individuals or groups. Identification essentially becomes the

function of sociality. . .ones participation in a collective, social role

cannot be obtained in any other way (Burke, 1937, p. 144).

Identification, therefore, is important for employees who must

navigate inevitable divides in order to find their fit within a family

business.

Family businesses have the unique opportunity to use their

familiness, the resources resulting from the interaction of the

family and business systems (Habbersohn, Williams, & MacMil-

lan,2003), in order to create a strong senseof identification among

all of their employees. By developing an overall organizationalidentity based on [1_T DIF]familiness, family businesses are able to

develop a strong sense of cohesion among family member

employees and perhaps encourage other employees to buy [_TDDIF]into

their vision and values (Zellweger, Eddleston, & Kellermanns,

2010).

Once family business employees have bought into the ideals

of the family business, the goals and ideas of all employees are

likely to be aligned. This not only encourages participation in

organizational decision making, but also encourages the develop-

ment of a strong sense of identification with the family business

(Zellweger et al., 2010). If family businesses succeed in creating a

sense of belonging and identification for non-family member

employees, the perception that employees have been let into the

family may be enough to create a sense of commitment.

A.F. Carmon et al. / Journal of Family Business Strategy 1 (2010) 210223212


4/14

This sense of commitment is partially based on an obligation to

continue employment with an organization and the perception

that staying with an organization is the right thing to do ( Fields,

2002; Meyer & Allen, 1991). For example, family member

employees may feel identified with and committed to their family

businesses, because they provide their families livelihoods and a

great deal of their family wealth is tied up in the business ( Degadt,

2003). However, non-family member employees do not have the

same obligations to theorganization as a family memberemployee

does. Therefore, it is likely that a non-family member employees

sense of commitment to and identification with their family

businesses will differ from a family member employees sense of

commitment and identification.

As organizational identification increases, it positively influ-

ences a variety of work attitudes, behaviors, and outcomes.

Organizational identification is linked theoretically and empirical-

ly to motivation, decision making, employee interaction, and

length of service (Cheney, 1983); turnover and turnover intentions

(Mael & Ashforth, 1992; van Dick et al., 2004); and job satisfaction

and performance (Carmeli, Gilat, & Waldman, 2007; Cheney,

1983). Organizational identification plays an important role in

employees well-being, satisfaction, and productivity (Gautam, van

Dick, & Wagner, 2004) and therefore has promising potential for

increasing organizational commitment (Scott et al., 1999; van Dicket al., 2004).

One of the primary motives for commitment to an organization

is identifying with the organization (Becker, 1992). Individuals

who embrace their organizations goals and missions through

identification processes are more likely to remain committed to

their organizations than individuals who do not (van Dick et al.,

2004).Ellemers, Spears, and Doosje (1997)found that individuals

who had stronger in-group identification were more likely to feel

commitment to their work groups. Therefore, if an employee

begins to identify with an organization, particularly if they

perceive they are a member of the in-group in this case the

family then it is likely he or she will also begin to feel more

committed to the organization. In order to appreciate family

business employees experiences better, SIT suggests understand-ing their perceptions of similarity and treatment (i.e., organiza-

tional justice) within the organization (Turner, 1975, 1978, 1981).

These perceptions could determine the extent to which employees

meld their organizations views with their social identities to affect

organizational identification. Therefore, perceptions of justice may

have important implications for organizational identification.

2.4. Organizational justice

Organizational justice is the role of fairness as a consideration

in the workplace (Greenberg, 1990, p. 400). While justice is

historically and deeply rooted in society and law, the concept is

vital to organizations. Research links perceptions of justice to an

array of organizational outcomes, including job satisfaction,withdrawal, and organizational citizenship behavior (Colquitt

et al., 2001).The results ofColquitts (2001) meta-analysis establish

that organizational justice has four distinct dimensions: distribu-

tive justice, procedural justice, interpersonal justice, and informa-

tional justice. The current study focuses on perceptions of

interpersonal and informational justice because they are most

related to human interaction withinthe organization as opposed to

organizational structure (Cheung & Law, 2008). Interpersonal and

informational justice, sometimes grouped within a single dimen-

sion as interactional justice, refer to the interpersonal treatment

people receive as procedures are enacted (Colquitt, 2001; Scott

et al., 2007). Interpersonal justice is related to whether or not

decision-makers treat individuals with dignity, respect, and

sensitivity (Colquitt, 2001). Informational justice is related to

how decision-makers openly, honestly, and thoroughly explain the

rationale for their decisions (Colquitt, 2001).

Interpersonal and informational justice are important factors to

family business employees because everyone expects and desires

fair human interaction. Perceptions of organizational justice have

been linked to a variety of work outcomes, including: job

performance, organizational citizenship, organizational commit-

ment, trust, and responses to organizational change (Cohen-

Charash & Spector, 2001; Colquitt et al., 2001; Lipponen,Olkkonen,

& Moilanen, 2004; Tyler & Blader, 2000, 2002; Tyler, Degoey, &

Smith, 1996). By carefully implementing and cultivating interper-

sonal and informational justice, family businesses could likely

enhance the identification of employeesas employees are more

likely to feel a sense of belonging and loyalty to their organizations

if they areinvolvedin thedecisionmaking processes of their family

businesses.

Interactional justice, a combination of interpersonal and

informational justice, has specifically been linked to employees

willingness to engage in more extra-role behaviors as well as

higher levels of job satisfaction than employees who do not

perceiver interactional justice (Masterson, Lewis, Goldman, &

Taylor, 2000). Strong, positive correlations are also found between

interactional justice and co-workeridentification,satisfaction with

co-workers, and co-worker support (Byrne, 2003). While thisresearch was conducted in non-family firms, the results may carry

over to family businesses. The more fairly non-family member

employees feel they are treated, the more likely they are to go

above and beyond their given role in the organization, be more

satisfied with their jobs, and, feel more identified with both their

co-workers and their family businesses in general. However, it is

also possible that the familial relationships within a family

business might disrupt perceptions of just treatment. Non-family

member employees may attribute their treatment in the business

to their non-familial status rather than their own performance.

De Cremer(2002) found that the more similarly individuals feel

they are treated, the more likelythey are to have increasedfeelings

of self-esteems and acceptance concepts closely related to

feelings of satisfaction and identification. Organizational justicehas also been linkedto increased feelings of affective commitment,

very closely related to feelings of organizational identification

(Cohen-Charash & Spector, 2001; Colquitt et al., 2001). Further,

organizational justice is positively related to job satisfaction, rated

job performance, and job attendancethree factors likely to

contribute to individuals feelings of belongingness within an

organization (Lam, Schaubroeck, & Aryee, 2002).

Those family business employees who perceive organizational

justice and homophily at work may be more likely to have

increased perceptions of organizational identification. It is likely

themore similar they feel to andthe more similar they feel they are

treated to other employees, the more connected they will feel to

their overall organizations. Because the identity of many family

businesses is partially based on the identity of the family itself,non-family member employees may especially rely on their

feelings about their treatment to know whether they belong in

the organization because they do not have that shared family

identity.

Identification, in turn, can help these employees understand

their experiences, organize their thoughts, and even make

decisions based on what is best for the company (Cheney,

1983), because of their sense of shared values and missions with

their organizations. Because feelings of organizational identifica-

tion are dependent upon feelings of shared values, belonging, and

loyalty to an organization (Cheney, 1982), as an employee

perceives shared values and goals, his or her perceptions of

organizational identification are likely to be strengthened. This

process may be particularly salient for non-family member



5/14

employees. When a family member employee views the family

business as working toward sustainability and profitability, he or

she is able to view that as working toward his or her own familys

sustainability and profitability; the connection between the

employees values and goals and the organizations values and

goals is very evident. However, for a non-family member

employee, the way in which the goals of the organization are

similar to personal goals must be made more explicit for

identification to occur.

Olkkonen and Lipponen (2006) found that relationship between

procedural and distributive justice and turnover intentions was

mediated by perceptions of organizational identification. Addi-

tionally, work-unit identification mediated the relationship

between interactional justice and work-unit identification. Though

the effects of interpersonal and informational organizational

justice have not been linked directly to perceptions of organiza-

tional identification and commitment, these forms of justice are

likely critical to the success of the family because, by carefully

implementing and cultivating interpersonal and informational

justice, family businesses may enhance the identification of

employees. Further, an employees perceptions of how similar

they are to their fellow employees may also contribute to their

perceptions of identification.

2.5. Homophily

Homophily is ones perception of similarity to another in

attitude, background, values, and appearance (McCroskey et al.,

1975; Rogers & Bhowmik, 1970). Homophily theory posits that

contact between similar people occurs more frequently than

contact between dissimilar people (McPherson, Smith-Lovin, &

Cook, 2001). Lazarsfeld and Merton (1954) observed that

individuals with similar characteristics were more likely to be

friends, thus improving their communication and creating a more

trusting environment and stronger personal relationships. Some

research indicates that when individuals perceive homophily in a

relationship, they are likely to develop positive feelings because of

apparent confirmation of their beliefs, values, or interests (Prisbell& Andersen, 1980)an idea that relates closely to identification.

Further, homophily also helps determine which co-workers

individuals will compare themselves with, listen to, and model

their behavior on within an organization (Lawrence, 2000).

According to SIT, individuals evaluate their social identities by

comparing their in-groups to the relevant out-groups (Tajfel &

Turner, 1986). Homophily may be one of the primary factors in

distinguishing between the in-group of an organization and the

related out-groups as individuals may perceive those who are

similar to them in values and beliefs to be their organizational in-

groups.

Perceived similarity leads to more positive evaluations of other

group members (Rokeach & Mezei, 1966). The more similar to an

individual a group is perceived as being, the more positively thegroup will perceive the group and the groups members

(Henderson-King, Henderson-King, Zhermer, Posokhova, & Chiker,

1997). According to SIT, individuals tend to identify with those

groups that they evaluate positively. If individuals perceive similar

groups andsimilar individuals as more positive, it is likely that that

perceived homophily may lead to identification with that group.

Individuals who are members of groups that have minority-

status within the organization, are likely to seek out other

members of the group to form close bonds. Leonard, Mehra, and

Katerberg (2008)found that individuals who were members of an

ethnic minority group within an organization tended to form

friendships with other members of that ethnic minority group.

There were notable differences between who members of minority

groups befriended even after controlling for relative access to

members of the minority group as compared to access to members

of other minority groupsand the majoritygroup. This suggests that

individuals do tend to form ties and connections with those whom

they perceive as more similar to them. In fact, individuals

perceivedsimilarity to a group is themost importantfactor leading

to identification with the group (Fisher, 1998). Even when

individuals find group membership appealing, they will only

identify with that group if and when they feel that they are similar

to other group members.

Perceptions of homophily are likely linked to how identified

and committed individuals perceive themselves to be to their

organizations. When individuals feel they are part of the in-group

because of perceived homophily, they may be more likely to

identify with the organization. As individuals perceive themselves

to be more like their colleagues, they are more likely to feel as if

they belong to their organization, align themselves with their

organizations goals and values, and to remain committed to their

organizations.

In a study of the United States Army,Mael and Ashforth (1995)

found that certain shared activities, values, and beliefs predicted

how likely a new recruit was to identify with the Army and stay

enlisted in the Army. Those soldiers who engaged in outdoor

activities like hunting or intellectual pursuits like reading for

pleasure beforeenlisting were more likely to have a strong sense oforganizational identification with the Army. Those recruits who

had high levels of organizational identification were also more

likely to stay enlisted in the Army.

This suggests those similar values and shared pursuits between

soldiers, those elements that comprise perceptions of homophily,

likely lead to organizational identification, which in turn may lead

to organizational commitment. Family businesses share similar

characteristics with military organizations, such as a strong focus

on identity, shared values, and commitment. Therefore, these

results would likely be similar for family businesses. Developing a

sense of identification and commitment is likely to be important to

family businesses because there are likely to be sharper and more

clearly delineated boundaries between in-groups and out-groups

of employeesgroup status may be decided based on membershipin the founding family.

Homophily is a prevalent characteristic of organizational

networks (McPherson et al., 2001). Because individuals learn

which in-groups they belong to through feelings of similarity with

these groups (Ashforth & Mael, 1989), organizational members

perceptions of homophily likely affect other organizational out-

comes. Further, as individuals feel more connected to particular

groups, their social identity associated with that group becomes

more salient (Brewer, 2001). Therefore, homophily may be

particularly relevant for family business employees because the

salience of their identities as family or non-family members may

affect their perceptions of organizational identification, and,

ultimately, commitment, depending on their perceptions of the

organizations in-group.

2.6. Hypotheses

An individuals identity is comprised of more than his or her

own in-group associations. Individuals have multiple social

identities which they must constantly negotiate (Hogg & Vaughn,

2002; Tajfel & Turner, 1979; Terry et al., 1999).Within a family

business, family member employees must manage two separate

identities: family identity and worker identity (Shepherd & Haynie,

2009a). These identities must be co-managed in order for family

member employees to function in the work environment. This co-

management results in a family business identity, comprised of

who weare as a familyand who weare as a business. However,

because the family business identity is partially based on the



6/14

family identity, it is likely that non-family member employees will

have a different identification with the organization than an

employee who can claim the family identity.

When the family identity has a strong influence on an

organizational identity, the family identity may become the basis

for organizational decision making (Reay, 2009). While this may

result in positives outcomes for the family business ( Reay, 2009),

non-family member employees will likely react differently than

family member employees to the idea of family identity-based

decision making. It may be more difficult to justify business

decisions on the basis of what is in line with the familys values to

someone who is not a member of that family and may not share

those values.

However, the familys role in the business is not the only factor

that may cause family member employees to relate differently to

the business than non-family member employees. A family-owned

business is usually the main (or at least one of the main) sources of

employment and income for the managing family (Degadt, 2003).

Additionally, for many family business owners, the majority of

their wealth is tied up in the business itself. Therefore, the financial

status of the business is often directly tied to the financial status of

the family (Degadt, 2003). If the business experiences a downturn,

multiple family members may have their wages garnished, default

on their business loans, or need to invest their personal finances tokeep the business going. This likely results in a very different

identification with and commitment to the business.

Because family businesses involve both family member and

non-family member employees, distinct in-groups and out-groups

may be created among employees. SIT argues that individuals

compare themselves to members of their socialgroups as well as to

other social groups when forming their social identities (Tajfel &

Turner, 1986). This is likely thecase for family business employees.

The identity of a family business is directly tied to the familys

identity as well as the business identity (Reay, 2009). Therefore,

family member employees may have a very different identification

process than non-family member employees because they are

already part of the family identity. Additionally, family member

employees frequently have stronger emotional and financial ties toan organization out of necessity than do non-family member

employees, likely resulting in a different level of identification and

commitment (Degadt, 2003). Therefore, the following hypotheses

are posited:

H1: Employee status (i.e., family member vs. non-family

member) will affect perceptions of (a) informational justice,

(b) interpersonal justice, (c) homophily, (d) organizational

identification, and (e) commitment.

H2: Perceptions of organizational identification mediate the

relationship between perceptions of interpersonal justice,

informational justice, and homophily and perceptions of

commitment among non-family member employees in family

businesses.

3. Method

3.1. Participants

Family businesses were targeted for participation through

the local Yellow Pages and a local website advertising small

businesses. Businesses were first targeted through the mail and

then via a phone call from one of the researchers. In order to

determine if the businesses were qualified to participate, they

were first asked if they considered themselves a family business,

meaning that they were family-owned. If they were not, they were

thanked for their time and disqualified from the study. If they were

family-owned, they were then asked if at least twopeople from the

founding family worked for the company, if at least one family

member was in a management role, and if they employed at least

one non-family member employee. Of 64 contacted businesses,

employees from 18 family businesses ranging in size from two to

77 employees were invited to participate in the study. The

organizations had been in business for between five and 104 years,

with an average age of 36.6 years, and were from a variety of

industries (e.g., construction, sales, and professional organiza-

tions). In total, 118 employees from those family business

participated in this study, which was a 39% overall response rate.

There was a great degree of variation between businesses in terms

of participation, ranging from 11% to 75% (11 participants did not

report their organizational affiliation). Of participants, 34 were

family member employees, 76 were non-family member employ-

ees, and eight declined to identify whether they were family or

non-family. Of all respondents, 56 were male, 52 were female, and

10 declined to identify. Their average age was 47 ( SD= 13.40),

ranging from 19 to 75. They had worked for their organizations for

an average of 137.55 months (SD= 122.75 months; or M= 11.46

years,SD =10.23 years), ranging from 3 months to 40 years.

3.2. Measures

Participants completed the survey online (n= 65), by returninga paper copy of the survey distributed via the U.S. or office mail

(n= 34), or by completing hard copies of the survey during

researcher visits (n= 19). One-way ANOVA analysis revealed no

significant differences among the three data collection methods.

The survey took approximately 20 minutes to complete and

included measures of interpersonal justice, informational justice,

attitude homophily, organizational identification, and organiza-

tional commitment. All scales were subjected to validity and

reliability analyses. All items included in the final scales used for

model testing are listed inTable 2along with their standardized

and unstandardized estimates and standard errors. The stan-

dardized factor loadings were used to calculate the reliabilities

for each item, which were used to calculate the average variance

extracted (AVE) for each scale to establish convergent validityand the construct reliability for each scale. More detailed

procedures and results for validity and reliability testing are

detailed below.

3.2.1. Organizational justice

Perceptions of justice were measured using the interpersonal

justice and informational justice scales from Colquitts (2001)

Justice Measure. Responses to all items were measured using a

Likert-type scale, ranging from 1 (to a small extent) to 5 (to a large

extent). The informational justice measure (construct reliabili-

ty = .93, AVE = 71%) consisted of five items. Acceptable percentages

of AVE are above 50% (Hair, Black, Babin, & Anderson, 2010). The

interpersonal justice measure (construct reliability = .90,

AVE = 70%) consisted of four items. Face and content validity ofthe items were assessed by evaluating their consistency, clarity,

and completeness to the intended constructs of interpersonal and

informational justice (seeDeVellis, 2003). Because each subscale

appeared to have both face and content validity, no items were

removed before conducting further validity testing.

Confirmatory factor analysis (CFA) was conducted to examine

construct validity for the two organizational justice factors

with a covariance matrix using LISREL 8.80 (Joreskog & Sorbom,

2006). This model showed acceptable fit to the data (x2 [df= 26,

N= 118] = 51.49, p< .001; GFI = .91; AGFI = .85; IFI = .98; CFI = .98;

RMSEA = .09) (see Colquitt, 2001, for more detailed validity

analysis). In order to determine discriminant validity, we

calculated the squares of the interconstruct correlations and

compared them to the AVEs. As a rule of thumb, the AVE should be



7/14

larger than the squared interconstruct correlation (Hair et al.,

2010). In the current study, all the AVEs were greater than the

interconstruct correlationsdemonstrating discriminant validity.

As evidence of the discriminant validity of the two forms of

justice, the two dimensional solution was a significantly better

fit (Dx2 [1] = 210.9, p < .001) than a unidimensional solution

(x2

[df= 27, N= 118] = 262.39, p< .001; GFI = .67; AGFI = .45;IFI = .89; CFI = .89; RMSEA = .27).

3.2.2. Attitude homophily

Perceived attitude homophily was measured using a modified

version ofMcCroskey et al.s (1975)Attitude Homophily Measure

(AHM). The AHM consists of four items on a seven-point semantic

differential scale. In response to the prompt: Other people in your

organization, participants responded to four adjective pairs

measured on a seven-point semantic differential scale. The scale

showed acceptable reliability (construct reliability = .61, M= 4.21,

SD= 1.41) and only moderate convergent validity (AVE = 36%),

which is consistent with previous research using this scale due to

the limited number of items (McCroskey, McCroskey, & Richmond,

2005). Face and content validity of the items were assessed by

evaluating their consistency, clarity, and completeness to the

intended construct of attitude homophily (see DeVellis, 2003).

Because the scale appeared to have both face and content validity,

no items were removed before conducting further validity testing.

CFA was conducted to determine construct validity for all

measures with covariance matrices using LISREL 8.80 (Joreskog

& Sorbom, 2006). This scale showed only moderate fit to the data(x2 [df= 2, N= 118] = 44.76, p < .001; GFI = .84; IFI = .48; CFI = .46;

RMSEA = .43). While the RMSEA value exceeds the acceptable

range between .05 and .08 (Hair et al., 2010), the other fit indices

are within satisfactory ranges to warrant acceptable goodness of

fit. Further, comprehensive validation across studies and samples

has previously been reported (McCroskey et al., 2005). However,

readers should interpret the findings with care and future research

shall validate this model using alternative data samples and

organizational contexts.

3.2.3. Organizational identification and commitment

Because the distinction between organizational identification

and organizational commitment measures is debated in the

literature, two organizational identification scales and two

Table 2

Survey items and estimates from LISREL 8.80.

Scale Item SFL a UFLb SEc

IPJ Thinking of your immediate supervisor, to what extent has (he/she) treated you in a pol ite manner? .93 0.68 .02

IPJ Thinking of your immediate supervisor, to what extent has (he/she) treated you with dignity? .84 0.61 .02

IPJ Thinking of your immediate supervisor, to what extent has (he/she) treated you with respect? .91 0.72 .02

IPJ Thinking of your immediate supervisor, to what extent has (he/she) refrained from improper remarks or comments? .63 0.67 .09

INFOJ Thinking of your immediate supervisor, to what extent has (he/she) been candid in (his/her) communications with you? .80 0.86 .06

INFOJ Thinking of your immediate supervisor, to what extent has (he/she) explained the procedures thoroughly? .82 0.84 .05

INFOJ Thinking of your immediate supervisor, to what extent were (his/her) explanations regarding the procedures reasonable? .88 0.84 .04INFOJ Thinking of your immediate supervisor, to what extent has (he/she) communicated details in a timely manner? .87 0.97 .05

INFOJ Thinking of your immediate supervisor, to what extent has (he/she) seemed to tailor (his/her) communications to individuals specific needs? .85 0.97 .06

AH Other people in your organization: Are like me. . .Are not like me .76 1.20 .42

AH Other people in your organization: Are different from. . .Are similar to me .09 0.15 .30

AH Other people in your organization: Think like me. . .Do not think like me .86 1.30 .58

AH Other people in your organization: Do not behave like me. . .Behave like me .32 0.47 .26

OIC I am proud to be an employee of this organization. .80 0.75 .05

OIC This organizations image in the community represents me well. .77 0.86 .07

OIC I am glad I chose to work for this organization rather than another company. .81 0.84 .05

OIC I talk up this organization to my friends as a great company to work for. .72 0.94 .11

OIC I have warm feelings toward this organization as a place to work. .80 0.93 .07

OIC I feel that this organization cares about me. .78 0.95 .08

OIC The record of this organization is an example of what dedicated people can achieve. .77 0.94 .08

OIC I find that my values and the values of this organization are very similar. .83 1.12 .08

OI C I would descr ibe this org aniza tion as a lar ge family in wh ich mo st memb ers feel a se nse o f be longing. .69 0.84 .11

OIC I find it easy to identify myself with this organization .85 1.01 .06

OIC I really care about the fate of this organization. .84 0.83 .04

OIEP What this organization stands for is important to me. .85 0.83 .04

OIEP I share the goals and values of this organization. .90 0.91 .03

OIEP My membership in this organization is important to me. .85 0.93 .05

OIEP I feel strong ties with this organization. .86 0.98 .05

OC I am willing to put in a great deal of effort beyond that normally expected in order to help this organization be successful. .68 0.57 .05

OC I feel very little loyalty to this organization. (R) .29 0.45 .29

OC I would accept almost any type of job assignment in order to keep working for this organization. .40 0.74 .39

OC I could just as well be working for a different organization as long as the type of work was similar. (R) .62 0.92 .19

OC This organization really inspires the very best in me in the way of job performance. .66 0.83 .12

OC It would take very l ittl e change in my present circumstances to cause me to l eave thi s organization. (R) .61 0.81 .13

OC I am extremely glad that I chose this organization to work for over others I was considering at the time I joined. .47 0.54 .13

OC Theres not too much to be gained by sticking with this organization indefinitely. (R) .53 0.95 .31

OC Often, I find it difficult to agree with this organizations policies on important matters relating to its employees. (R) .58 0.91 .22

OC For me this is the best of all possible organizations for which to work. .79 1.13 .12

OC Deciding to work for this organization was a definite mistake on my part. (R) .64 0.40 .03

IPJ= from theColquitt (2001)interpersonal justice measure.

INFOJ = from theColquitt (2001)informational justice measure.

AH= from theMcCroskey et al. (1975)attitude homophily scale.

OIC = from the Miller et al. 12-item version of Cheneys (1982) Organizational Identification Questionnaire.

OIEP= from theEdwards and Peccei (2007)organizational identification measure.

OC= from theMowday et al. (1979) organizational commitment measure.

R = item was reverse-coded.a Standardized Factor Loadings.b Unstandardized Factor Loadings.c Standard Error.



8/14

commitment scales were simultaneously submitted to various

validity tests, including content validity and CFA to establish

validity. Perceptions of organizational identification were mea-

sured using the 12-item version of the Organizational Identifica-

tion Questionnaire (OIQ;Cheney, 1982) recommended byMiller,

Allen, Casey, and Johnson (2000). Because the validity of the OIQ

has been questioned (e.g., Miller et al., 2000), the OIQ was

supplemented with another measure of organizational identifica-

tion:Edwards and Pecceis (2007) measure, which has six items.

Both identification measures offerresponse options ranging from 1

(strongly disagree) to 7 (strongly agree).

Organizational commitment was measured using two scales:

Mowday, Steers, and Porters (1979) instrument and the normative

commitment portion ofMeyer and Allens (1997) Organizational

Commitment Scale (OCS). As specified in the literature review, this

study focused on organizational commitment; normative commit-

ment was measured to establish validity due to previous issues with

the organizational commitment and organizational identification

scales. In addition, we consulted the affective commitment portion

of Meyer and Allens OCS and eliminated items that overlapped

across scales prior to distribution of the survey consistent with

the definitions used for each concept (see Table 1) in order to

maintain content validity.

Confirmatory factor analysis withall 35 originalitems specifyinga four-factor solution (x2 [df= 554, N= 118] = 1257.22, p < .001;

GFI = .62; AGFI = .57; IFI = .95; CFI = .95; RMSEA = .10) resulted in

moderate fit.The modification indices provided by LISRELsuggested

that several items were crossloading to more than one latent

construct. One item from each organizational identification scale

and one item from the normative commitment scale were thus

removed to improve model fit for further analysis. Thus, the next

model to be tested was a four-factor, 32-item measurement model.

The fit indexes showed minimal improvement (x2 [df= 458,

N= 118] = 998.81, p< .001; GFI = .65; AGFI= .60; IFI = .95;

CFI = .95; RMSEA = .10).

Because several items between the two organizational identifi-

cation scales were crossloading and they were not conceptually

distinct, the two organizational identification scales were com-bined to test a three-factor, 32 item measurement model (x2

[df= 461, N= 118] = 1080.34, p < .001; GFI = .63; AGFI = .58;

IFI = .95; CFI = .95; RMSEA = .11). Modification indices suggested

that one item from theEdwards and Pecceis (2007)measure was

crossloading with both commitment constructs. Thus, the item

was removed for a fourth analysis (x2 [df= 431,N= 118] = 988.24,

p< .001; GFI = .65; AGFI = .59; IFI = .95; CFI = .95; RMSEA = .11).

Although some of the fit indices failed to meet the rule of thumb

thresholds, no further modifications were suggested by LISREL to

improve model fit.

In order to determine discriminant validity, we calculated the

squares of the interconstruct correlations and compared them to

the AVEs. As a rule of thumb, the AVE should be larger than the

squared interconstruct correlation (Hair et al., 2010). As evidenceof the discriminant validity of the constructs, this three dimen-

sional solution was a significantly better fit (Dx2 [3] = 239.32,

p< .001) than a unidimensional solution (x2 [df= 434,N= 118] =

1227.56, p < .001; GFI = .60; AGFI = .54; IFI = .94; CFI = .94;

RMSEA = .13). While the RMSEA value exceeds the general rule

of thumb of between .05 and .08 (Hair et al., 2010), the other fit

indices are within satisfactory ranges to warrant acceptable

goodness of fit. However, readers should interpret the findings

with care and future research shall validate this model using

alternative data samples and organizational contexts.

Face and content validity of the items were assessed by

evaluating their consistency, clarity, and completeness to the

intended constructs of organizational commitment and organiza-

tional identification (seeDeVellis, 2003) as previously discussed.

The analysis provided two valid measures to test the hypotheses:

one of organizational commitment and one of organizational

identification. Because that factor analysis suggested that organi-

zational andnormative commitment were distinct andthe focus of

this study was organizational commitment, Mowday et al.s (1979)

instrument was used. Unlike other measures of commitment (e.g.,

the affective commitment items on Meyer & Allen, 1997, OCS), this

instrument did not contain items that overlapped with the

organizational identification items used in this study. This

instrument was comprised of 11 items with responses ranging

from 1 (strongly disagree) to 7 (strongly agree). The commitment

scale showed acceptable reliability (construct reliability = .84) and

only moderate convergent validity (AVE = 34%). As noted previ-

ously, all organizational identification items were combined to

form one 15-item organizational identification measure (construct

reliability = .97, AVE = 66%).

4. Results

4.1. Hypotheses

One-way ANOVA was conducted to test H1. Results are shown

inTable 3. Employee status did not result in significantly different

ratings of informational justice, interpersonal justice, attitudehomophily, organizational identification, or commitment. Thus,

results for all organizations and both employee types were

combined in tests of the hypothesized model (Fig. 1) because

individual-level analysis best captured the variation in responses.

4.2. Model testing

The model presented inFig. 1 was operationalized to include

the variables measured in the study (H2). Fig. 2 depicts the

statistical model that was tested.

First,a correlation analysis wasperformed. Results areshownin

Table 4.

Next, LISREL 8.80 was used to evaluate the fit of the predicted

statistical model (see path diagram presented in Fig. 2). Thecovariance matrix used in the analysis is presented inTable 5.

Standardized maximum likelihood parameter estimates are

shown in Fig. 2. None of the paths from the three exogenous

variables (interpersonal justice, informational justice, and

attitude homophily) to organizational identification were signif-

icant. However, the path from organizational identification to

commitment was significant. Overall, the model does not appear

to fit t he data well (x2 [df= 3, N= 118] = 60.78, p < .001;

GFI = .83; AGFI = .14; IFI = .76; CFI = .75; RMSEA = .41). These fit

indices failed to meet the standard rule of thumb of .90 (Hoyle

and Panter, 1995). Thus, the predicted model is not consistent

with the data.

Further, in analyzing the predictedmodel, LISREL 8.80 proposed

a modification that would betterexplain the data. Compared to the

Table 3

Analyses for H1.

Descriptive

statistics

One-way ANOVA

M SD By employee type

F df p

Informational justice 3.87 1.33 2.95 1, 107 .09

Interpersonal justice 4.21 1.28 2.29 1, 107 .13

Attitude homophily 4.21 1.41 0.29 1, 108 .59

Organizati onal identification 6.22 1.02 1.84 1, 108 .18

Organizational commitment 5.61 1.43 2.42 1, 108 .12



9/14

original model, commitment is predicted to mediate the relation-

ship between interpersonal justice, informational justice, and

attitude homophily and organizational identification. LISREL 8.80

was again used to evaluate the fit of this alternative model.

Standardized maximum likelihood parameter estimates are shown

inFig. 3. Overall, this model appears to be a better fit for the data

than the hypothesized model (x2 [df= 3, N= 118] = 4.38, p= .22;

GFI = .99; AGFI = .93; IFI = 1.00; CFI = 1.00; RMSEA = 0.06).

5. Discussion

This study examined how family business employees percep-

tions of organizational justice and homophily are related to

perceptions of organizational identification and commitment

through the lens of SIT. Family member employees and non-

family member employees did not differ significantly in theirperceptions of informational justice, interpersonal justice, attitude

homophily, organizational identification, or commitment. Within

the proposed model, the paths from interpersonal justice,

informational justice, and attitude homophily to organizational

identification were not significant. However, the path from

organizational identification to commitment was significant.

Overall, the predicted model was not consistent with the data.

In relation to the first hypothesis, there were no significant

differences found between non-family member and family

member employees of family businesses for most of the variables

in the study (i.e., interpersonal justice, homophily, organizational

identification, or commitment), which suggests that employee

status within family businesses may not make a difference in

employee experiences in this sample. Because the family

[

Fig. 2. Predicted path model with standardized parameter estimates.

Table 4

Correlations among constructs.

1 2 3 4 5

Informational justice .86** .47** .29* .68**

Interpersonal justice .56** .25* .70**

Attitude homophily .17 .54**

Organizational identification .50**

Organizational commitment

* p< .01.** p< .001.

Table 5

Covariance matrix.

1 2 3 4 5

Informational justice 1.63 1.47 1.01 0.32 1.27

Interpersonal justice 1.78 0.88 0.39 1.28

Attitude homophily 2.01 0.24 1.10

Organizational identification 1.03 0.72

Organizational commitment 2.03

[

Fig. 1. Proposed model.



10/14

businesses studied were small to mid-sized businesses from

relatively homogenous communities, it is possible that all

employees were treated similarly and, therefore, came to identifywith their family businesses in similar ways.

These results may suggest that two distinct social groups,

family member employees and non-family member employees,

maynot always be formedwithin the context of family businesses.

It is possible that the family businesses included in the current

study took advantage of their familiness (Habbersohn et al., 2003)

in order to make their non-family member employees feel as

though they were one of the family and not a an entirely separate

group. Through the development of positive relationships with

members of the founding family, non-family member employees

are likely to develop positive, salient organizational identities.

According to SIT, perceptions of membership in in-groups and out-

groups are not always based on formal divisions such as family

member versus non-family member. However, when thesedivisions become salient for employees, thus forming in-groups

and out-groups. Future research should be conducted to determine

how family businesses develop working environments in which

non-family members are considered parts of the family (i.e., thein-

group).

It is possible the individuals most identified with their family

businesses may have been the individuals who participated in this

study. This falls in line with the findings of SIT suggesting those

individuals who identify most with the organization are most

likely to make larger contributions to the organization. While both

family member and non-family member employees were sur-

veyed, those non-family member employees who chose to

participate may have been those who considered themselves to

not only be most identified with their family businesses, but alsoparts of their organizational in-groups. It is possible the results

only capture the experiences of those considered to be a part of the

organizational in-groups,regardless of employee status, and do not

capture the experiences of those who consider themselves to be on

the fringes of their family businesses. Future research should be

conducted to provide a more complete picture of the work

experiences of all family business employees.

These results may indicate that family businesses are a special

type of organization and may all have similar characteristics. They

may differ in this way from non-family firms, where family does

notnecessarily play as central of a role. Regardless of size, industry,

or gross revenue, all family businesses must negotiate the divide

between family and firm in ways that non-family firms do not

(Dunn, 1999). Therefore, all family business employees may

engage in similar identity negotiation processes, which elicit

similar perceptions of interactional justice, homophily, organiza-

tional identification, and commitment. Future research shouldaddress whether the differences among this studys variables lie

primarilybetween the types of organization (i.e., family businesses

or non-family firms), rather than between specific organizations,

regardless of type.

In relation to the second hypothesis, this study proposed a

model of organizational commitment, suggesting that perceptions

of informational and interpersonal justice and homophily predict

organizational identification and, ultimately, commitment. How-

ever, this model did not serve as a good fit for the data. The only

significant path in the predicted model was between organiza-

tional commitment and organizational identification. These

findings support previous research suggesting that organizational

identification predicts commitment (Scott et al., 1999; van Dick

et al., 2004).As a potential explanation for the results, a one-way ANOVA

using organizational affiliation was conducted for all variables.

Though one-way ANOVA revealed some significant differences

between groups based on organization for interpersonal justice ( F

[17,88] = 2.53, p = .003), informational justice F [17,88] = 2.02,

p= .02, and organizational identification (F [17,89] = 3.07,

p< .001; intra-class correlation coefficients (ICC) did not support

the idea that between organization variance exceeded individual

variance. On the contrary, ICCs for all variables ranged from .07 to

.010, suggesting independence (i.e., random variability within and

between organizations) or heterogeneity (i.e., variability predomi-

nantly within organizations). Thus, these differences were signifi-

cant, but they were not meaningful as further analysis suggested

that there was more variance among individuals than there wasamong organizations. Unequal cell sizes across organizations likely

account for the significant result (Howell, 2006).

Although SIT suggests that family business employees percep-

tions of similarity and treatment (i.e., organizational justice) affect

individuals experiences within their family businesses, these

perceptions likely have an effect on different outcomes than this

study originally predicted. We initially predicted that perceptions

of interpersonal and informational justice may have been

positively related to perceptions of organizational identification.

Social identity theory suggests that perceptions of consistency and

neutrality drive perceptions of justice within organizations;

therefore, decisions within organizations should be based upon

facts and rules, and not on personal opinions or preferences

(Tyler & Blader, 2000, p. 92). Employees may simply believe that

[

Fig. 3.Post hoc path model with standardized parameter estimates.



11/14

their employers treat them fairly, because it is the right thing to

do. More specifically, just because family business employees

perceive themselves as being treated fairly does not necessarily

mean that they will identify with the goals and missions of their

organizations and, ultimately, consider themselves an important

part of the organization.

Similarly, perceptions of homophily were originally thought to

be related to perceptions of organizational identification. However,

homophily is generally considered to affect interpersonal out-

comes (e.g., mutual trust, intimacy, and confirmation of personal

beliefs) (Crary, 1987; Prisbell & Andersen, 1980). Therefore, the

effects of homophily may not extend to perceptions of organiza-

tional identification, which are based upon feelings of belonging

and loyalty to an organization (Cheney, 1982). Just because

employees perceive themselves to be similar to their co-workers

does not necessarily relate to the organization itself and whether

or notthey believe strongly andfeel related to the missions of their

organizations.

Further, there was a near-significant effect between employee

status and informational justice. More specifically, non-family

member employees perceived higher levels of informational

justice than family member employees. This may suggest that

whether an employee is related to the founding family or not may

contribute perceptions to how openly decision-makers communi-cate about their decision-making processes (Colquitt, 2001).

According to SIT, members of the out-group, in the case of family

businesses, non-family member employees, will compare their

treatment within the organization to that of the in-group, in this

case family members. It is likely that non-family member

employees compare the amount of information they are privy to

the amount of information family members have. However, in

order to understand this relationship between informational

justice and employee status, future research should be conducted

with a larger sample of family business employees.

There is also a possible alternative explanation forthese results.

The reference points of family and non-family employees are likely

to be different from each other. Family member employees may

expect to have a great deal of information shared with thembecause they may believe that they are at least psychological

owners of the firm, if not actual owners (Pierce, Rubenfeld, &

Morgan, 1991) On the other hand, non-family employees may not

expect to get a great deal of information from the founding family

and, therefore, may be grateful for any information that is shared

with them. Therefore, for any given level of information shared,

other than complete transparency, family members are more likely

than non-family member employees to perceive themselves as

being unjustly treated.

In analyzing the predicted path model, LISREL proposed a

modification that would better explain the data, one in which

organizational commitment mediates the relationship between

interpersonal justice, informational justice, and organizational

homophily and organizational identification. The first part of thismodel proposes that interpersonal justice, informational justice,

and organizational homophily lead to organizational commitment.

Previous research has shown that perceptions of justice are related

to commitment (Cohen-Charash & Spector, 2001; Colquitt, Conlon,

& Wesson et al., 2001; Tang & Sarsfield-Baldwin, 1996). One of the

three factors of commitment is a desireto maintain membership in

the organization (Porteret al., 1974). Perhaps an individuals desire

to maintain membership in an organization is based, not on his or

her perceptions of the organization and what it is perceived to

offer, but rather on interpersonal perceptions of co-workers.

Therefore, homophily would lead to a desire to maintain

membership in the organization.

The post hoc model suggests that interpersonal justice,

informational justice, and attitude homophily lead to organiza-

tional commitment, which leads to organizational identification.

This supports Herrbachs (2006) research suggesting a positive

correlation between affective organizational commitment, defined

as an employees attachment to his or her organization (Meyer &

Allen, 1991), and the experience of positive affective states at work

(i.e., identification). However, other previous research has shown a

relationship between organizational commitment and identifica-

tion, although the relationship has previously been argued in the

opposite direction with identification leading to commitment (van

Dick et al., 2004). This new model would extend the existing

research by highlighting the potentially multi-directional nature of

relationship between the two constructs.

Because LISRELproposes models based only on their fit with the

current data, assuming that the post hoc model is an accurate

representation of the identification and commitment processes

would be premature because this post hoc model might simply be

an artifact of the current data set. In particular, we do not know the

extent to which a relationship between interpersonal justice,

informational justice, attitude homophily, and organizational

identification is partially or fully mediated or whether it exists

at all. Therefore, research should test both models using data sets

across business types.

6. Conclusion

6.1. Practical implications

This study provides several implications for family businesses.

This study suggests that family member and non-family member

employees do not differ in their perceptions of organizational

justice, homophily, organizational identification, and commit-

ment. Previous research has suggested that employees who

perceive themselves to be a part of a cohesive organization are

likely to develop positive organizational identities as well as to

develop feelings of organization identification (Bartel & Dutton,

2001; Richter et al., 2004). Therefore, in order to develop a positive

sense of organizational identification, family businesses should

treat their employees similarly and strive to develop a positiveworking environment. This can be done by paying particular

attention to organizational justice and fostering a sense homophily

of among employees.

In order to develop a sense of cohesion among employees,

family businesses should ensure they are treating their employees

equally and fairly. Perceptions of organizational justice can be

fostered through not only treating all employees with respect, but

also by thoroughly explaining the organizational decision-making

processes (Colquitt, 2001). Family businesses should be cognizant of

ensuring their non-family members are informed of key decisions

and that they are involved in the organizational processes, even if

they are not privy to family interactions about the business.

Family businesses should strive to encourage a sense of

homophily through encouraging all employees to spend timetogether in both work and social settings in order to get to know

each other better. By encouraging employees to develop positive

relationships, they are likely to see ways in which they are similar

in attitudes, values, and beliefs that will affect their workplace

experiences. By being aware of issues related to organizational

justice and encouraging a sense of homophily among all employ-

ees, family businesses will likely encourage a greater sense of

identification and commitment among their employees.

6.2. Limitations

This study is not without limitations. First, some measurement

issues may have affected these results. There is conceptual overlap

between many organizational commitment measures and organi-



12/14

zational identification measures. Specifically, the items measuring

affective commitment are nearly identical to the items for

organizational identification (see Cheney, 1982; Mowday et al.,

1979). Therefore, the research that formed the basis for our study

may not have clearly distinguished between the commitment and

identification in the operationalization and conceptualization of

the two constructs. Overlapping items were removed from the

scales for the current study, which may explain why the current

studys results differ from some previous research. Though other

indicators for the organizational commitment scale were strong,

the AVE did not meet standard rules of thumb in this study. To

better understand the relationships between organizational

commitment and the other variables in the study, this measure-

ment issue should be addressed in future research. In addition, the

attitude homophily scale showed scale properties similar to those

in other studies that questions its validity and reliabilityan issue

that must be addressed in future research on this construct.

Next, this study used a relatively small sample with several

organizations submitting only one or two responses, which

affected inter-organizational comparisons. A potential explanation

for the limited sample is the fact that data was collected shortly

after a regional flooding event and many businesses and their

employees may have been relocated or involved in the rebuilding

the area following the flood. Future research should address thenew model of organizational identification using a larger sample of

family businesses.

Third, a representative number of family member employees as

compared to non-family member employees were not obtained

from each family business. This sampling bias is likely to have

affected the results of the current study, as discussed above.

Therefore, while difficult, future family business researchers

should strive to achieve more representative samples of family

member and non-family employees from the organizations they

survey.

Fourth, only employees of those family businesses who were

willing to participate in the study were surveyed. The businesses

were told the general nature of our study before agreeing to

participate. Business leaders who perceived high levels of conflictwithin the workplace or that have high turnover rates of

employees might not have chosen to participate in the study. As

such, this study mayonly reflect family businesses where there are

high levels of identification and commitment among most

employees. However, this self-selection bias, while problematic,

is common within family business research (see Kirkwood &

Tootell, 2008; Okoroafono, 1999; Scholes, Westhead, & Burrows,

2008).

Finally, because no compensation for participation in the study

was offered, only those employees with strong opinions about the

organization may have filled out the survey. Therefore, the results

may only reflect the perceptions of those individuals who feel very

negatively or very positively about the organization. This is

especially problematic for this study as it is likely that these strongfeelings are tied to feelings of identification and commitment.

6.3. Future research suggestions

In addition to the suggestions mentioned above, there are

several additional opportunities for future research. First,

researchers should continue to test the post hoc model of

organizational identification for family business employees, where

perceptions of organizational commitment serve as the mediator

between interpersonal justice, informational justice, and homo-

phily and organizational identification. The post hoc model could

also be used to address organizational identification differences

between family businesses and non-family ownedbusinesses. This

research may reveal that these two types of organizations are not

that different. If this is the case, the post hoc model of

organizational identification should be tested using a large sample

of employees from a variety of organizations. Future testing of

this model will allow researchers to determine whether this

model is simply an artifact of the current data set or if it provides a

new means for understanding organizational identification for

family business employees. Additionally, future research should

utilize directional hypotheses based on the findings of the current

study.

Future research should be also be conducted controlling for the

length of tenure of employees, actual shareholding of family

employees, and the perceptions of non-family member employees

regarding monetary compensation and fringe benefits. It seems

that individuals who have been employed by their family

businesses for a long time will have had time to cognitively

process their workplace experiences and may respond differently

than employees who have only been employed for a short time.

Whether family members are active owners of their family

businesses, rather than simply helping the business in their off

time from other employment, will likely affect their levels of

organizational identification and commitment. Further, it is

possible non-family member employee may have higher levels

of commitment to their family businesses because they are

satisfied with their monetary compensation and their fringebenefits, even though they may be treated differently than the

family member employees.

Future research should also continue examining the differences

between non-family member and family employees in family

businesses. While the current study did not reveal significant

differences, future research with a larger sample might. It seems

possible that non-family member employees will identify differ-

ently with their family businesses than family member employees

as their work and family livelihoods may not necessarily be tied to

the family business for which they work. Further, non-family

member employees may not be as likely to buy into the mission

and goals of the family business as family member employees,

whose families likely developed and have worked to maintain the

mission.This study focused on the relationship between family business

employees perceptions of organizational justice and homophily

and their perceptions of organizational identification and commit-

ment. Thepredicted model was notconsistent with thedata. Based

on previous research on organizational commitment, this studys

results are surprising. Compared with employees of other business

types, family business employees may have different reasons for

identifying with business, such as loyalty to the founding family or

membership in the founding family. As such, the relationship

between commitment and identification, as well as the relation-

ship between both commitment and identification and homophily

and justice may differ. Researchers should continue to examine

how family businesses are similar to and different from other

organizational types. Further insight into how family businesseswork, survive, and thrive is essential for a thorough understanding

of organizational functioning.

Acknowledgements

The authors are grateful to three anonymous reviewers and the

editor for their thoughtful feedba

fusing family and firm

Documents