Introduction

Each year thousands of students from every walk of life graduate from colleges and universities across the country. This accomplishment can come with many accolades including praise from one’s family and friends, better job opportunities in the graduates chosen field, and of course the title of alumnus or alumna from the institution they just departed. Some of these alumni will form ever-lasting relationships with their alma maters which will bring about a lifetime of post-graduation involvement. Others will take the diploma, leave the institution and never look back. No matter what level of involvement each individual decides to undertake, all of them will share the same categorical designation as an alumnus(a) and will forever be bonded by that recognizable identity.

The role of alumnus(a) can either bring forth great expectations for continued involvement with the institution or no responsibility at all. This depends on both the individual who bears the title and their views on the role in which they have been granted upon graduation. The purpose of this study is to assess how college and university alumni view their role with these institutions after graduation and how that perception relates to behaviors of support. Several questions arise that serve as the framework for this investigation. Do college and university graduates incorporate the role of alumnus or alumna into their own sense of identity? Are the expectations of the alumni role part of a social recognition dimension perpetrated by friends and family or is it simply defined by the institutions and how they would like you to behave? How do alumni role identities relate to institutional support behaviors such as charitable giving, volunteering and other support behaviors? This project will explore each of these questions by means of a sociological theoretical base and from previous literature in the areas of role identities and alumni support.

Alumni Support

The term “alumni” refers to individuals who have either attended and/or have graduated from an institution of higher education (Neufeldt 1996). Beyond the characteristics that grant this title, having membership in an alumni group does not warrant any formal expectations or responsibilities. Historically, colleges and universities were very open to alumni support yet relied upon the graduates themselves to initiate personal involvement (Miller 1993; Rudolph 1990). The alumni were fairly generous providing several types of support including alumni societies, governing boards, athletic assistance and financial contributions. In the modern college and university, alumni are still called upon to support their alma maters in many different ways. The changing scenery of the administrations have caused the institutions to place vast amounts of resources into offices such as university advancement, development, and alumni relations to keep the alumni connected with the institution after graduation (Rowland 1986). The desire for alumni support is still a priority amongst most colleges and universities, yet today, the initiation of involvement must start with the institutions.

The need for support in the modern college or university is especially prevalent in regards to institutional fund raising. Charitable contributions to colleges and universities in the United States peaked in 2008 with over $31 billion in donations being recorded (Council for the Aid to Education 2009). This amount comes after a steady incline of contributions has been documented in these institutions for several years indicating a vast interest in giving across many different constituent groups. The largest group of individuals to give consisted of college and university alumni who contributed over $8.7 billion alone which again, has been on the rise for several years in a row. Although the amounts contributed to colleges and universities in the U.S. have seen steady increases over the years, the actual number of alumni who have contributed continues to fall. This trend has caused institutional administrators to question the motives behind this decline and how they can stop it from making a large impact on their institutional budgets.

Research on alumni giving to colleges and universities has historically focused on individual or institutional characteristics which could increase or decrease someone’s willingness to make charitable donations after graduation. Individual factors such as age (Bruggink and Siddiqui 1995; Lindahl and Winship 1992; Weerts and Ronca 2007), income level (Bruggink and Siddiqui 1995; Clotfelter 2003; Taylor and Martin 1995), satisfaction with one’s collegiate experience (Clotfelter 2003; Gaier 2005; McDearmon and Shirley 2009; Tom and Elmer 1994) and involvement in institutional activities after graduation (Bruggink and Siddiqui 1995; Gaier 2005; Lindahl and Winship 1992; Taylor and Martin 1995; Weerts and Ronca 2007) have been found to increase the likelihood of an alumnus(a) making a gift. Institutional factors like expenditures per full-time student (Baade and Sundberg 1996) and national ranking (Holmes 2009; Liu 2006) have also been found to positively influence alumni giving. Although this line of work has resulted in some valuable information for institutions to use for alumni involvement purposes, most of the research has not been able to produce a meaningful look into how alumni identify with their college or university after graduation.

An article written by Frey (1981) summed up the above argument best when he stated “universities probably know little about their alumni. They presume opinions, beliefs, and preferences, yet they almost never conduct scientific research into the matter” (p. 46). Since this time an increase in research on college and university alumni has been conducted, yet most has focused solely on factors related to an individual’s willingness to make donations. Although this research can be useful to the institutions when developing alumni solicitation strategies, it only scratches the surface into the deep, internal processes that are used by alumni when making the decision to give or even be involved in other ways.

One study by Weerts and Ronca (2007) looked beyond just charitable giving and explored multiple ways alumni can be supportive after graduation. Their study looked to “create a profile of ‘supportive’ university alumni who are likely to be generous in the use of both their time and money” (p. 21). An alumnus(a) who is considered to be a supporter in their study, was one who had both volunteered in some capacity with the institution after graduation and has made a charitable financial contribution. Weerts and Ronca found that alumni who fell into the supporter category actually expected to be involved in supporting the institution. Although, the results of this study do not pinpoint any true meaning behind being an alumnus(a), it does suggest that there are behavioral differences between those who expect to be supportive after graduation and those who do not.

The purpose of this study is to assess dimensions of college and university alumni role identity and how that identity relates to behaviors of post-graduation support. In more practical terms, the current study will investigate how college and university graduates use the role of alumnus or alumna in the formation of their own sense of self. This identification process will be compared to behaviors of institutional support to see if there is a relationship between increases in alumni role identity and the amount of support provided to the institution after graduation. This study uses a comprehensive theoretical base to examine several dimensions of alumni role identity including role identity salience, social and institutional expectations and alumni involvement behaviors.

Theoretical Framework

Stryker’s (1980/2002)Footnote 1 symbolic interactionism is the theoretical position being used to guide this study. According to Stryker, individuals use social cues and perceived expectations to develop a sense of identity and behavior patterns for each role they have been assigned. For example, a college student may wear certain clothing, go on a spring break vacation, and attend social events in order to fit the popular view of how an undergraduate student is supposed to behave. All of these behaviors depend upon the cues and expectations that this particular individual has used to develop their sense of identity as a student. In regards to alumni relations in colleges and universities, making financial donations, volunteering for events, and recruiting new students are just a few of the expectations that have been historically developed by the institutions for their graduates to be considered as a supporter of their alma mater (Weerts and Ronca 2007). Although the expectation may be set, no research has ever tried to discover if college and university alumni use these expectations and behaviors to develop their own sense of identity as an alumnus or alumna.

The theory of symbolic interactionism was first developed by Blumer (1969) and was developed from the writings of America pragmatic philosophers such as John Dewey and George Herbert Mead (Stryker and Statham 1985). The theory consisted of various principles on how an individual’s behaviors are dictated by social interactions. According to Blumer (1969), people do not just react to the actions of others; they interpret and define those actions and respond based on the meanings that have been attached through their own perceptions. Blumer’s three core principles which make up symbolic interactionism included meaning, language, and thought. These principles serve as the basis on how individuals develop their own sense of self and how they interact with others in all social communities.

Stryker’s (1980/2002) symbolic interactionism theorizes that human behavior is dependent upon a named or classified world which is determined by the physical, biological, and social environment for which one resides. These classifications bring forth expectations that fuel the behavior of a person and set a guideline to determine social interactions within the environment. Stryker’s theory of symbolic interactionism uses “position” to refer to any socially recognized category of actors. According to the theory, “positions serve to cue behavior and to act as predictors of the behavior of persons who are placed into a category” (p. 57). The term “role” within this theoretical framework is used for the expectations that correspond with the position in question. Using the current study as an example, the position of alumnus(a) in regards to a college or university refers to an individual who has graduated from an institution. This position has been socially recognized throughout the history of higher education. The role of alumni in relation to an individual’s alma mater refers to the behaviors that are expected to be performed by the person who occupies that position.

Stryker’s (1980/2002) theory does take into consideration the concept of choice that can motivate an individual’s use of role behaviors. According to Stryker, “the degree to which roles are made rather than simply played, as well as the constituent elements entering the construction of roles, will depend on the larger social structures in which interactive situations are embedded” (p. 55). Social behavior in this context is not simply given by role expectations but is the product of a role-making process. The enactment or performance of a role is variable in that there is some choice in whether or not to perform expectations attached to a position. For example, the role of a father within American society is variable depending on the individual in question. Biologically any male who produces an offspring is automatically placed into the social position of a father. However, the amount of support, care, and other social expectations that may be attached to the position is dependent on the individual and their willingness to act out those behaviors and fulfill the role expectations. This part of the theory is extremely relevant to the current project in which the role of alumnus(a) is granted to every college or university graduate; however, the level in which the individual will act out the expectations of that role is what will be tested in this research.

Stryker’s (1980/2002) symbolic interactionism framework also incorporates the development of individual identities. Stryker uses both the terms identity and role identity interchangeably due to the influence role behavior plays on an individual’s sense of self. According to Stryker, “identities are parts of self, internalized positional designations. They exist insofar as the person is a participant in structured role relationships” (p. 60). In this context, an individual could have many identities which are only limited by the number of role relationships someone is willing to be involved in. For example, a woman could have multiple identities such as a wife, mother, attorney, golfer, artist, or any other roles that she uses to compose her sense of self.

To incorporate both the individual choice to act out role expectations and its relationship to identity development, Stryker developed the concept of identity salience (Stryker 1980/2002; Stryker and Serpe 1994). Identity salience is defined as the readiness to act out the behaviors expected for a particular role which formulates identity. For example, a student identity is shown to be highly salient when a college-aged man or woman talks about their studies, grades, and classroom experiences on dates, interactions with friends outside of school, or with coworkers at their jobs. Low salience for the same scenario would be a student who does not share classroom related activities during other non-academic related situations or hides their status from others. In each case, the individual does fit into the role of a student, however major variations can be observed in how likely they are to incorporate that role into their personal identity.

Past studies have examined alumni relationships with institutions of higher education using organizational identity models as a means of conceptualizing an individuals willingness to support their college or university after graduation (Mael and Ashforth 1992). Mael and Ashforth used social identity theory to define organization identification as “the perception of oneness with or belongingness to an organization, where the individual defines him or herself in terms of the organization(s) in which he or she is a member” (p. 104). According to the results of their study, organizational identification was significantly related to all their hypothesized outcomes including making financial contributions to an alma mater and advising one’s own children to attend the same college or university.

Although the current study takes a similar approach as Mael and Ashforth (1992), several conceptual differences can be identified which place this study into a separate category of alumni research. First, their theory is based upon social identification which is “the perception of belongingness to a group classification” (p. 104). An individual under this definition perceives to be an actual member of a group and shares in its successes and failures. The difference between this and the theoretical framework used in the current study, is that this brand of identification is purely perceptual and not based upon an individual’s behaviors relative to the group. The classifications used in Stryker’s (1980/2002) symbolic interactionism brings forth behavioral expectations based upon the perceived role an individual partakes in a group setting.

A second different between the two frameworks is that organizational identification, as conceptualized by Mael and Ashforth (1992), is only relative when compared to individuals in other classifications. The example they use is, “the category of male becomes meaningful only in relation to the category of female” (p. 105). Symbolic interactionism (Stryker, 1980/2002) states that the enactment of a role is determined by choice in which the individual in question determines their own status within a particular classification dependent upon their behaviors. As stated above, the role of a father is variable and only determined by the actions the man takes towards this positional designation. In this case, the role of the mother is not necessary for the role of the father to be conceptualized. The study takes these differences and applies them to the role of alumnus(a) to examine if role identity can produce the same results of institutional support found in Mael and Ashforth’s research (1992).

Role Identity

Role identity as a conceptual framework has been used in research regarding altruistic behaviors. In 1985, Peter Callero studied dimensions of role identity salience on a sample of blood donors to investigate how measures of self-definition and social relationships relate to the voluntary behavior of the donation process. In his study, Callero stated that a role identity is more salient if “it is more representative of the self and consequently one’s self-definition” (p. 204). This hypothesis was based upon the work of Stryker (1980/2002) and other theorists who define identity salience as the readiness of an individual to act out a particular role which formulates their own sense of self. In the study, Callero (1985) found that individuals with high blood donor role identity salience were found to be more likely to define themselves as a regular blood donor and more likely to engage in the corresponding behavior. The results also stated that blood donor role identity salience, self-definition as a regular blood donor, and friendships contingent upon donating all have independent positive effects on the act of donating. This suggests that all dimensions hypothesized by Callero can impact an individual’s willingness to act out the behavior which is assigned to a particular role.

Callero’s (1985) study was an interesting development in the study of role identities because it went beyond just looking at how behavior can be related to one’s self-definition in regards to a particular role. Callero was also interested in the social expectations that accompany a role definition. In his study, he hypothesized that identity salience towards the role of blood donor will increase has the person in question perceives that others in their social circle expect them to act out the behaviors associated with that particular role. The results found a strong correlation between the social expectations of others and a person’s salience towards the blood donor role identity. This is an interesting finding which suggests that our social relationships can actually enhance our willingness to act out the behaviors which are attached to our own self-identified roles.

In 1999, researchers Lee, Piliavin, and Call conducted a study similar to Callero (1985) in order to assess dimensions of role identity on blood donors but added other forms of support to the model including volunteering and charitable donations. Their study looked to assess the effects of perceived expectations, modeling, personal norms, and past behavior on role identity and the intentions to act out support behaviors listed above. The authors found that perceived expectations, modeling, past behaviors, and personal norms were all significant predictors of blood donor role identity in the form of donations and intentions to support with money or time. Giving blood in the past was the most significant predictor in their study. These results suggest that altruistic behaviors could be developed by modeling others who have completed the behavior in the past and having a sense of obligation to the recipients of the individual’s support. It also suggests that once an individual has completed the behavior, it may become a part of their personal identity which could cause them to act out that identity in the future.

The current study models the work of both Callero (1985) and Lee et al. (1999) on a sample of university alumni. This study will examine how support behaviors such as charitable giving, attending alumni events, serving on institutional boards and others can be affected by several dimensions of role identity including institutional expectations, social expectations and identity salience. One major difference between the current study and those investigating blood donors is the nature of the role in question. In order to be labeled as a blood donor, an individual must first act out the donation behavior that corresponds with that role definition. This is different from college and university graduates who are given the title of alumnus or alumna upon receipt of their diploma and does not correspond with any other behaviors such as charitable giving to the institution, volunteering, or other acts of institutional support. In both cases, a role identity can be established by the individual as defined by Stryker (1980/2002). However, in the case of the blood donor, a voluntary act must be performed before that individual integrates the role into their sense of self. In regards to alumni, the role is assigned without any type of altruistic behavior towards the institution. This study will assess the level of role identity of participants and how it relates to institutional support behaviors.

Method

A survey research instrument was previously created to assess the role identities, expectations, and support behaviors of college and university alumni (McDearmon and Bradley 2010). The identity salience and social expectation measurement was adopted from the study of Callero (1985) and Lee et al. (1999) who developed scales to assess the role identities of blood donors. The instrument used in the current study has been modified to fit the role identity of college and university alumni. Each section of the survey including the support behaviors and the role identity scales will be discussed in the following paragraphs.

Support Behaviors

The first section of the survey inquired about alumni support behaviors. As mentioned in the beginning of this study, the focus of most research concerning alumni is primarily to assess charitable giving; however, as concluded by Weerts and Ronca (2007), more behaviors can be completed in order for an alumnus(a) to be considered a supporter. In this study, multiple support behaviors are being assessed with this instrument in order to see if relationships exist between the behaviors themselves and the dimensions of role identity development.

In order to include behavioral items that were appropriate for the specific institution used in this study, the author relied on members of the alumni association staff along with activities listed on their association’s website for support behaviors that were assessed. The author determined that attending alumni events, attending athletic events, membership on a university committee, volunteering at university events, joining the alumni association and making financial contributions were all behaviors that were common amongst supportive alumni at this institution and would be a good fit with the research literature used in the development of this project. Each item was assessed with a dichotomous yes or no answer choice to be used in the overall assessments.

Role Identity Scales

The next three sections of the survey were used to determine an individual’s role identity as a college or university alumnus(a). The term role identity, first coined by McGall and Simmons (1978) was defined as “the character and the role that an individual devises for himself (or herself) as an occupant of a particular social position” (p. 68). Stryker (1980/2002) incorporated the notion of role identity into his theoretical position because it stresses “the tie between components of the self and locations in the social structure” (p. 130). Three dimensions of role identity have been identified in the literature and will be assessed using a 6-point agree/disagree Likert-type scale. Details regarding each dimension and the scale are discussed below.

The first role identity scale used in the survey was a measure of alumni role identity salience. Five items that were originally designed to study the salience of blood donors were modified to fit the role of a college or university alumnus(a). The five items are listed below:

  • Being an alumnus(a)Footnote 2 is something I often think about (Salience1).

  • I really don’t have any clear feelings about being an alumnus(a) (Salience2).

  • For me, being an alumnus(a) means more than just contributing money or time (Salience3).

  • Being an alumnus(a) is an important part of who I am (Salience4).

  • I would feel lost if I were not an alumnus(a) (Salience5).

Based upon the work of Callero (1985) the role identity salience scale assessed how the role is associated with an individual’s self-concept. In the development of his study, Callero stated that “when a role identity is salient it is more representative of the self and consequently one’s self-definition” (p. 204). The variance in role identity salience has been found to be associated with the behaviors that accompany the particular social position in question (Callero 1985; Santee and Jackson 1979; Stryker and Serpe 1982). Therefore, for this study, the role of alumni was analyzed in relation to the expectations that colleges and universities perceive for the alumni to support the institutional mission and the behaviors an alumnus(a) exhibits which meet those expectations (Weerts and Ronca 2007).

The second scale of role identity assessed the social expectations of being a college or university alumnus(a). Again, the items used in the blood donor studies were modified to fit the role of alumnus(a). This scale was used to examine the relationship between the identity salience of the alumnus(a) to the perceived social expectation that corresponds with that role:

  • Many people think of me as being an alumnus(a) (Social1).

  • Other people think that being an alumnus(a) is important to me (Social2).

  • It is important to my friends and family that I am an alumnus(a) (Social3).

  • It does not matter to most people that I am an alumnus(a) (Social4).

  • Many people I know are not aware that I am an alumnus(a) (Social5).

According to Callero (1985), “salient role identities have implications for social relations in that they announce to others who we are” (p. 205). They are not only used for social relationships but also for our desire to act on the overall expectations that a particular role carries. Stryker (1980/2002) wrote on this phenomenon in his theory when he stated that greater identity salience leads to more stable relationships with others based upon the role identity in question. He also acknowledged that stronger salience towards a role identity would lead to greater opportunities to act out behaviors associated with a particular role.

The last dimension assessed in the survey was perceived alumni role expectations. Although, this type of measurement was not used in the studies of Callero (1985) and Lee et al. (1999), role expectations are one of the basic themes which Stryker (1980/2002) states will determine how individuals perceive their roles. The five items are listed below:

  • As an alumnus(a), it is my duty to support the university through financial contributions (donations or gifts) (Role1).

  • As an alumnus(a), it is my duty to support the university through volunteering (Role2).

  • As an alumnus(a), I am expected to attend alumni events (on- and off-campus) (Role3).

  • As an alumnus(a), it is my duty to serve on a university board or committee (Role4).

  • As an alumnus(a), I am expected to attend athletic events (Role5).

These items were used to assess the expectations that individuals have about their own role as an alumnus(a) which is based on both Stryker’s theoretical position and the expectancy theory assessments found in the research of Weerts and Ronca (2007). Their expectancy theory “argues that alumni have expectations about future events and that their giving and volunteer behavior reflects these expectations” (p. 23). Weerts and Ronca’s research found that the perceived expectation to participate after graduation does increase an alumnus(a)’s willingness to engage in those supportive behaviors. These items tie into the theoretical framework since, according to Stryker (1980/2002), behavior is the product of a role-making process, initiated by expectations invoked in the process of defining situations.

Sample and Survey Procedures

The current study took place through the alumni association at a large, public research university in the Midwest. The sampling frame for this study consists of alumni who graduated from the institution with a bachelor’s degree between the years of 1940 and 2009. A random sample was selected from the alumni population who had an email address on file with the alumni association. The overall sample consisted of 8,987 alumni contacts.

The method used in this study was a self-administered, web-based survey using the software available at surveymonkey.com. Alumni in the sample were sent an email invitation which stated the purpose of the study and provided a link to the secure survey website. The email address used to administer the survey was generated from a university domain in order to minimize the invitation being marked as spam. Out of the sample, 688 individuals responded and completed the survey. Using the traditional method of calculating the response rate, the result for this study was 13 %. However, using an electronic survey format differs from the mail option especially in regards to how response rates can be calculated. According to a report by O’Malley (2010), about one in five emails sent with permission from the target address still fails to reach an electronic inbox. Some email can be delivered to a “junk” or “bulk” email folder which collects undesired messages while others simply bounce back to the addressee similar to a letter getting returned in the mail. In the research highlighted by O’Malley, 16.3 % of the emails sent were missing or not delivered at all with no message regarding whether it did bounce back to the sender or some other notification of non-delivery. This issue is in competition with the United States Postal Service (2010) who guarantees that if delivery of a parcel is not possible, the piece will be returned to the sender in a matter of only a few days. At this point, the electronic mail systems have not developed this same ability.

Therefore, when the email invitation for the current study is not opened or bounced back to the email server, there is no guarantee that the message was successfully delivered. In this case, the response rate for this study will be based upon the email open rate, which for the first invitation was 1,365. With 417 surveys completed out of this group, the response rate was 30.5 %. In order to increase the number of survey participants, a second email reminder was sent to the same sample population. This round of invitations resulted in an additional 271 surveys being completed for a total of 688. The final response rate based upon the open emails was 46.1 % after invalid and bounced-back email addresses were removed. It should be noted that the email reminder was sent to the entire sample population including those who had taken the survey during the first round. Email and IP addresses of survey participants were not collected due to privacy concerns. The second email instructed individuals to disregard the message if they had already participated. However, there was no way to determine how many of those who opened the email the second time fit into this category. Therefore, the 46.1 % response rate may be a low estimation of the actual percentage of alumni participants.

Results

Since this was a first exploration into alumni role identity as an area of research, Cronbach alpha procedures were conducted on the scaled items to ensure reliability of the instrument. The Cronbach alpha is a measurement of the internal reliability (Cronbach 1951) and was performed to see if the modifications to Callero’s (1985) original scales to fit the role of college or university alumni would result in the same levels of reliability found in the earlier studies. The results found that the instrument exceeded the Cronbach alpha levels found in Callero’s (1985) and Lee et al. (1999) studies (see Table 1).

Table 1 Role identity scales Cronbach alpha results
Full size table

Descriptive statistics for all items in the survey were generated to provide the author with a visual representation of the responses and displayed in Tables 2 and 4. Responses to the salience scale indicate that the overall sample slightly agrees that the role of alumnus(a) fits into their sense of self. This suggests that alumni from the university being studied use this role after graduation as a piece of their own personal identities which will be a pivotal piece of information for the remaining analyses.

Table 2 Role identity scales descriptive statistics
Full size table

The results from the social expectations and role expectations scales do not fall into the agreement side of the scale; however, the results from the correlation analysis (see Table 3) to the salience scale show positive associations. For the social expectations scale, the results are consistent with the results found in Callero’s (1985) study indicating that those who agree more with the social expectations tied to a particular role, have greater role identity salience towards being an alumnus(a).

Table 3 Role identity scale correlations
Full size table

Descriptive statistics for all behavior items were generated which can be found in Table 4. Several of the behavior items displayed extreme variation between those respondents who completed the examined behaviors and those who did not. For example, item committee member (B3) out of the respondents only 86 indicated they had completed this behavior compared to 569 who did not. Similar results were found in items volunteer student event (B4) and volunteer alumni event (B5). After viewing the descriptive statistics, the author determined that responses on behavior items of attending an on-campus event (B1), attending and off-campus event (B2), attending an on-campus athletic event (B6), attending an off-campus athletic event (B7), membership in the alumni association (B8), and making a financial gift (B9) were considered adequate measures to work into the remaining analyses.

Table 4 Behavior items descriptive statistics
Full size table

Role Identity and Support Behaviors

The next results to be presented will assist in answering the research question: How do levels of role identity as a college or university alumnus(a) affect support behaviors? In order to assess how dimensions of role identity affect these behaviors, independent samples t tests were conducted to compare the survey responses of the scaled role identity items to the behavior items. All behavior items were entered in as independent variables with group labels as 1 = yes for those who indicated they had completed the behaviors and 0 = no for those who did not. Summation results for each role identity scale were entered in as dependent variables. According to the results, significant differences in the responses between groups 1 and 0 were found for all role identity scales. The results found that those in group 1, or the respondents who completed the support behaviors, answered more favorably on all role identity scale items. Table 5 displays the group statistics for both categories.

Table 5 Behavior and role identity t test group statistics
Full size table

The largest difference in the mean scores of both the salience and social scales were found in the items regarding attendance at athletics events. Those respondents who reported that they have attended one or more of these events scored higher on the role identity salience dimension in this study and the social expectations which are attached to the alumni role definition. The fact that both scales resulted in similar responses is not surprising with the strong correlation which was present between the two scales. It is interesting to see that the largest difference in role identity can be found in reference to athletic events and not any of the other behaviors. This could suggest a possible link between college athletics and the desire for alumni to use their connection with the institution as a part of their own personal identity.

Like the other two scales, results for the role expectations dimension also found significant differences between those who completed the support behaviors and those who did not. The largest difference between the two response categories can be found in the attending an on-campus event and making a financial gift items. Once again, this is not necessarily a surprise finding. The connection between event attendance and charitable giving to a college or university has been well documented in the literature (Taylor and Martin 1995; Verner et al. 1998; Weerts and Ronca 2007). However, this is the first connection between the two variables that suggests one may possibly increase the expectations to complete the other. More research will be needed before any definite conclusion on this matter can be made.

To complete the assessment on the relationship between support behaviors and the dimensions of role identity, binary logistic regression models were employed using the behavior items as the dependent variables (see Table 7). The predictors (independent variables) used in the regression models include all three summated scales of role identity. For this particular regression model, the Hosmer and Lemeshow test confirmed that the data adequately fit the model. Collinearity diagnostics were also completed on the independent variables due to the correlations between the items found in Table 3. Table 6 summarizes the results which found no multicollinearity issues since all of the variance inflation factor values were relatively low. The results of the regression analysis found that the predicting value of each role identity dimensional scale varied by the individual behaviors and will be discussed in the following paragraphs.

Table 6 Collinearity statistics
Full size table

The results of the regression analysis in regards to the salience scale reveal that the scale was a predictor of at least 5 of the 6 behaviors. The only missing behavior was attendance at on-campus events which also had the lowest mean difference between the yes and no responses from the t tests. For the other behaviors, these results suggest that an increase in a person’s role identity salience towards being an alumnus(a) can also increase their likelihood of completing one or more of the support behaviors. These results along with the positive correlation found between the salience and expectations scale adds to the argument that alumni role identity could affect institutional support.

In regards to being an adequate variable for predicting support behaviors, the social expectations scale is unique since it is not behavior specific. The items in this scale were designed to assess how an alumnus(a) perceives other people’s views on that particular role definition (this compared to blood donors who were assessed on how others see the person acting out the behavior of donating). Even when no behavior is needed to maintain the role of alumnus(a), it is interesting to see that the social scale did serve as a significant predictor for attending athletic events. This could suggest that the social expectations of being an alumnus(a) could revolve around college athletics. This is not necessarily a surprise since the popularity of college athletics in the U.S. seems to be growing every year. Once again, we have to look back at the correlations which found a strong positive association between social expectations and role identity salience. Although, it cannot be determined from this study which variable more strongly affects the other, these findings put together could indicate that an alumnus(a)’ affiliation with the institution’s athletics program may result in greater identity towards this role and therefore more support.

Table 7 Regression results
Full size table

How the expectation dimension relates to the support behaviors makes up the next set of regression results. The model indicates that the summated role expectation scale serves as a predictor for all support behaviors, with the exception of attending an off-campus athletic event (B7). These results suggest that those who expect to be supportive after graduation are more likely to act out one or more of the behaviors.

Much like the t test results, interpreting the role expectations scale is a different process when compared to the salience and social scales. The results from the overall summated role expectation scale can only tell us so much in regards to actual expectations towards the support behaviors since each individual scaled item questioned one particular behavior. Therefore, additional backwards step-wise likelihood ratio regression procedures were employed to examine the role expectations item-by-item. Table 8 displays the results from these models where only the role expectation item that was present in the last step of the procedure is displayed per behavior. According to the regression models all of the role expectation scaled items are predictors of the corresponding behavior. For example, item Role3 [as an alumnus(a), I am expected to attend alumni events (on- and off-campus)] was found to be a significant predictor for both behavior items regarding event attendance (B1 and B2). These results suggest a positive relationship between the expectation to perform the behavior as an alumnus(a), and the likelihood they will actually perform the behavior.

Table 8 Item-by-item role expectations regression results
Full size table

Two of the items from the regression results display some interesting findings. Item association member-B8 (are you a member of the Alumni Association) was unique since there was no role expectation item related to joining an alumni organization. However, in the results, both the expectation to support the institution financially (Role1) and to attend alumni events (Role3) were predictors of being an association member. These results do correspond with association activities since paying dues and attending alumni functions is normal part of these organizations. Item financial gift-B9 (have you ever made a financial contribution to the university, such as a donation or gift) had the strongest predicting value with the corresponding item, Financial contributions-Role1 (as an alumnus(a), it is my duty to support the university through financial contributions). For every one unit increase in responses to the financial contributions item, participants responded three times more likely to be donors to the institution.

The findings in regards to the role expectations dimension are significant, not only because it provides evidence that alumni graduate with expectations to support the institution but also because of the relationship this dimension has to the role identity salience. When referencing back to the correlations table, we can see that there is a positive correlation between the scales. Although this correlation is not has strong as the one seen between salience and social expectations, it does add to the argument that increases in alumni role identity can also increase someone’s expectations to be supportive of their alma mater. This corresponds to the definition of identity salience by Stryker (1980/2002) which states that when someone has an salient identity towards a role, they are more willing to act out the behaviors that are expected from that role definition. According to the results of this study, even if the behaviors are designated by an external force (i.e. a college or university) and not contingent upon membership into that role designation, an individual with high identity salience may follow the suggested expectations and act upon them in order to maintain membership in that role category.

Discussion

Research on college and university alumni has started to increase over the last decade, mostly driven by the desire for institutions to increase the financial support that can be derived from their former students. Although this research has uncovered some very interesting developments, none of it has specifically looked at how the alumni view their role with the college or university after graduation. The purpose of this study was to assess dimensions of college and university alumni role identity and how that identity relates to behaviors of post-graduation support. According to the results, those who completed one or more of the support behaviors examined in this study responded more favorably to the alumni role identity scales. Favorable responses to these scales were designed to suggest an overall increase in role identity for being an alumnus(a) of the institution used in this study. Therefore, the results suggest that alumni with increased role identity may be more likely to support their institution through attending events, volunteering, joining the alumni association and making financial contributions.

These results are consistent with previous studies on role identity in relation to blood donors where the authors found that those who had increases in blood donor role identity were more likely to engage in the corresponding behaviors (Callero 1985; Lee et al. 1999). However, studying the role identity of college and university alumni is fundamentally different than the study on blood donors. Specifically the difference comes from the nature of how someone is placed into each of the social categories. One is only labeled a blood donor if they have actually donated their own blood one or more times. In the case of college and university alumni, most institutions attach that label to any individual who has attended and/or graduated from one of the degree programs housed within that particular institution. This is vastly different because in the case of alumni, no behaviors beyond attending an institution are needed in order to maintain the role of being an alumnus or alumna including institutional support. Many institutions desire their alumni to provide support in many different arenas; however, actually following through with those behaviors is not contingent upon an individual maintaining this particular role for as long as they wish.

The difference described above is important in this discussion because the support behaviors used in this study are voluntary and, again, not required to maintain someone’s status as being an alumnus(a). However, it is interesting to see that those who responded more favorably on the scales of alumni role identity do choose to engage in one or more of the behaviors. These findings correspond to the theoretical position used in the design of this study where role identities stress the tie between components of a person’s sense of self and locations of that person within the larger social structure (McGall and Simmons 1978; Stryker 1980/2002). It only makes sense that for someone who uses the role of alumnus(a) more steadily in their own identity are more likely to engage in the behaviors that are deemed socially appropriate to increase their own status amongst the college or university. More specifically, Stryker’s (1980/2002) theory of symbolic interactionism states that those who have a salient identity towards a particular role are more likely to act out the behaviors associated with that social designation. In the current study, it was found that those who responded higher on the role identity salience scale towards being an alumnus or alumna, more frequently engaged in the support behaviors.

One question that still remains unchartered throughout this investigation, is what behaviors (if any) constitute the role definition of a college or university alumnus(a)? Weerts and Ronca (2007) suggested in their study that those alumni who were labeled as supporters both made financial gifts to the university and volunteered in some capacity. This suggests that there is a subset of the alumni population who desire to be supportive to their alma mater; yet in the same study, they found there was another population of alumni who did nothing in regards to supporting the institution. Therefore, since there are two separate entities who occupy the role of alumni with differing behaviors, the question now turns to the idea of expectations of the role behaviors and how that dimension affects alumni identification. Stryker’s (1980/2002) theory does include the fact that expectation from both the individual occupying the role in question as well as the social forces which surround the individual (or the role designation) bring about a reasonable definition of what it means to be a member of a certain category. In other terms, if someone in a role expects to behave in a certain way and/or has a social community who expects him or her to behave a certain way, then they will act out those behaviors as a means of maintaining the role recognition. The current study acted upon that theoretical concept to assess how both social and role expectations feed into the identity salience towards to the role of alumnus(a) and the willingness act out the institutional support behaviors. The results found that those who experience greater social expectations towards being an alumnus(a) (from family, friends, and/or peers) have greater alumni role identity salience and are more likely to engage in the support behaviors. However, it should be noted that the social expectations being assessed in this study were not behavior specific and only relevant for a general role definition of a college or university alumnus(a).

Even lacking in a direct relationship to the support behaviors, the social expectations dimension was included because, theoretically, symbolic interactionism is designed to be a framework for the analysis of social interaction and role theory that serves to bridge the social structure and the social person (Stryker and Statham 1985). Stryker states that, “one is committed to a social role to the degree that extensive and intensive social relationships are built upon that role” (p. 345). Callero (1985) recognized this connection as well when he found that “salient role identities have implications for social relations in that they announce to others who we are” (p. 205). His study found that those who had friendships that were centered around the act of blood donation were more likely to continue engaging in the behavior and had more salient identities toward that particular role. Results from the current study found that those who answered more favorably on the social dimension scale completed the support behaviors more frequently. This suggests that the social side of the role identity development could affect an individual’s willingness to engage in one or more of the institutional support behaviors.

Specific role expectations were also assessed in this study to counter-act the shortfall created with the ambiguity of the social expectation items. Theoretically, behavior is the outcome of the role-making process, initiated by expectations invoked by early definitions of the situation (Stryker and Statham 1985). The third dimension of alumni role identity used in this study examined the relationships between the participant’s expectations toward a role behavior and their response to the actual completion of that behavior. The results found that those who had higher levels of alumni role identity salience expected in greater frequency to be supportive of the university. The results also found the specific expectations served as predictors to the corresponding behavior. As mentioned above, the aspect of identity salience toward a role is defined as the readiness to act out the behaviors that correspond with that role in order to place it within the individual’s sense of self (Callero 1985; Stryker and Serpe 1994). The results found in this study suggest that those alumni who portray a salient identity toward the role of an alumnus(a) are more likely to utilize the expectations of that role and support the institution through institutionally defined behaviors including event attendance, joining the alumni association, and charitable giving.

Limitations and Directions for Further Research

Several limitations of the study methodology were identified which need to be addressed before providing final conclusions. First, this study was designed as an exploratory investigation into role identity and its relationship to alumni relations in colleges and universities. Research of this nature has never been conducted towards alumni populations and the results remain highly theoretical in nature. However, it should be noted that the results from the reliability testing suggest that using this instrument created for this study does fit well into the conceptual framework and could be used for further investigation into this field of inquiry.

The second limitation specifically deals with the nature of the survey items regarding the support behaviors. For this study, only dichotomous answer choices were presented for all of the behavioral items presented to the participants. With the range of participants’ graduation date ranging from the mid-1940s to 2010, the responses to the behaviors items do not provide any indication on when the behavior was completed. For example, one participant could have made only one gift to the university 50+ years ago but still answered “yes” to the items which correspond to financial contributions. Although, the author does not see this as a theoretical problem in regards to the studies approach, it would be suggested that further research into this area create more time-sensitive behavioral items along with frequencies, amounts, and longitudinal ranges.

Since this study was highly theoretical and exploratory in its design, several future research projects should be conducted to examine the more practical implications that can be taken from the results. The results of this study found that those who displayed greater alumni role identity were more likely to complete the institutional support behaviors. This was a primary finding of the current study; however, the data that collected was unable to determine what could cause individuals to increase or decrease alumni role identity. Previous studies have found that satisfaction as a student, institutional involvement after graduation, and other factors increase the potential for financial support but that does not necessarily translate into increased role identity. More research is needed to determine what factors and characteristics of both the alumni themselves and their alma maters could lead to increased alumni role identity. It would also be worthwhile to investigate if differences in student experiences result in increased or decreased alumni role identity. These types of studies could be the cornerstone of expanding this research agenda to be more practical for colleges and universities and their alumni relations efforts.

Additional studies can be developed using the same theoretical position as this one but expanding the definitions of alumnus(a) to fit a more modern reality of colleges and universities. For example, many graduates may have more than one degree from multiple universities which by definition, makes them alumni of both institutions. One interesting study could look at the differences in loyalties towards these institutions and degree types by means of role identity salience. Another suggestion would be an investigation into the differences in alumni role identity for public versus private institutions. If a noticeable difference was to be discovered, then an exploration into what would cause one type of institution to produce greater alumni role identity would be an interesting inquiry and add to the literature in alumni relations and higher education.

Conclusion

The current study was the first of its kind to examine role identity in the context of how college and university alumni use their role after graduation in their own sense of self and how it also affects institutional support behaviors. The results of this study suggest that college graduates do vary in how they use the role of alumni in their personal identities could affect their involvement with the institution after graduation. One thing that can be derived from this study is the importance of looking at the personal relationships that alumni can have with their alma maters and how those relationships can greatly impact both the individual alumnus(a) and the future of the college or university in question.

With the decrease in state and federal support of colleges and universities reported almost daily in the news, it is only logical that these institutions will look to their graduates for greater levels of support. In order to make this possible, institutional advancement professionals need to find ways to let all alumni know that they are still important parts of their college or university even if they have not stepped onto campus in years. This could lead to more support and potentially new sources of revenue that can fill the gaps left by the diminishing amounts from government entities. Helping alumni use this role in the development of their own personal identities can move U.S. colleges and universities forward well into the future.