The literature was first consulted to identify items addressing the content of the six main constructs that form AO. 16 items were identified that reflect the type of activities mentioned by the respondents. All the interviews were then reviewed carefully to generate items that could be included in the final questionnaire. 34 items were generated, then a questionnaire containing the 50 items designed to assess the measure's properties was emailed to a total of 10 academics in four different countries: the UK, USA, Australia and Malaysia. Respondents were asked to review the items based on clarity, relevance and specificity (Church and Waclawski, 2001). After receiving four questionnaires and on the basis of the detailed comments, some items were modified and others were eliminated.
The revised scale was then subjected to a second stage of pretesting by asking five head of DAROs in the UK HE sector to further comment on the research instrument. Very few comments were made and very few concerns were noted. The final structured questionnaire consisted of 50 items and respondents were asked to express their level of agreement based on a five-point Lickert scale running from very strongly disagree up to very strongly agree. “Not Applicable” options were included. Alumni engagement was measured using six items, while resource attractions was measured using 5 items (see table1). However, because HE is slow moving, both constructs will be measured over a three-year period instead of the traditional one-year period.
Insert Table1 Here
The population was all UK higher education institutions with university status. No sampling procedures were carried out as surveying all 129 universities was feasible. Of the 129 questionnaires distributed to the directors of development and alumni relations via Surveymonkey, 71 were returned, which resulted in a response rate of 55%. After excluding three questionnaires due to their partial completeness, 68 usable ones remained for the final data analysis. Data collection lasted for nearly three months.
Data analysis: Validating First-Order Indicators
Alumni Orientation Index (AOI) was validated at three stages. In the first stage, the SEVEN first-order reflective indicators were validated: strong cases for support, event management, publication management, social media management, alumni database management, intrafunctional coordination and interfunctional coordination. In the second stage, the ONE second-order formative construct was validated: Coordinating Alumni Activity. In the last stage, AOI was validated as a third order construct consisting of one second-order formative constructs: Coordinating Alumni Activities (interfunctional coordination and intrafunctional coordination), and five first-order reflective constructs: strong cases for support, event management, publication management, social media management, alumni database management.
To validate the seven first-order indicators, exploratory factor analysis (EFA) was first performed to unearth and determine the questions or statements that appear to best measure the various dimensions of AO (Garson, 2008). It is important to note that the appropriate size of sample to perform EFA ranges from minimum 50 to maximum 500 and the optimal number of observations range from minimum 2:1 to minimum 20:1. Nonetheless, it was impossible to perform the EFA on the SEVEN first-order constructs simultaneously (i.e., seven-factor model) for two reasons. First, the sample size was relatively small (68), and second, the number of items was relatively high (50). Therefore, the current researcher followed Malhotra (2004) and Hair et al.’s (2006) suggestions in dividing the items into groups to ensure that the ratio of observations per item for each analysis was at least 4:1 (Malhotra, 2004). This approach is also acceptable because the constructs are conceptually related (Hair et al., 2006) under the umbrella of “prospective student orientation”. Therefore, the items were divided into groups to ensure that the number of observations per item for each analysis was either 4:1 or 5:1. As such, the SEVEN constructs were divided into three groups. The first group consisted of PM and INTRACO. The second consisted of EM and INTERCO. The last consisted of CS, ADM and SMM. After dividing the items into groups, the EFA was performed using the principal component analysis technique and A Varimax rotation to initially-extracted factors. Items that either had communality (<0.5), loading (< 0.55) and cross loadings (>0.3) were removed.
This purification process which resulted in reducing the items from 50 to 48 identified two new factors. More specifically, the items that measure “Event Management” loaded significantly on two factors. Items EM1, EM2, EM3, EM4 and EM5 loaded on one factor, while items EM6, EM7, EM8, EM9 and EM10 loaded on another factor. This means that “Event Management” is a second-order formative construct rather than first-order reflective construct. Hence, the original variable “Event Management” was divided into two new variables and they were labelled accordingly. The items EM1 to EM5 were grouped into a new variable called “Financing Event Management” (FEM) and were labelled as follows: FEM1, FEM2, FEM3, FEM4 and FEM5. Whereas, the items EM6 to EM10 were grouped into a new variable called “Promoting Best Practices of Event Management”, (PBPEM) and were labelled as follows: PBPEM1, PBPEM2, PBPEM3, PBPEM4 and PBPEM5.
Similarly, the items that measure “Publication Management” loaded significantly on two factors. Items PM1, PM2, PM3, PM4 and PM5 loaded on one factor, while these items PM6, PM7, PM8, PM9 and PM10 loaded on another factor. This means that “Publication Management” is a second-order formative construct rather than first-order reflective construct. Thus, items PM1 to PM5 were grouped into a new variable called “Financing Publication Management” (FPM) and were labelled as follows: FPM1, FPM2, FPM3, FPM4 and FPM5. Similarly, the items PM6 to PM10 were grouped into a new variable called “Promoting Best Practices of Publication Management”, (PBPPM) and were labelled as follows: PBPPM1, PBPPM2, PBPPM3, PBPPM4 and PBPPM5. This suggests that the original SEVEN first-order reflective indicators have become NINE first-order reflective indicators.
The 48 items of the NINE factors had communalities ranging from 0.500 to 0.858 and loaded significantly (loadings ranging from 0.555 to 0.873) on their respective factors, suggesting satisfactory factorability for all the items. Moreover, for each group, the p-values for the Bartlett's test for Sphericity was below 0.05, and the Kaiser-Meyer-Okline (KMO) measure of sampling adequacy was above the threshold of 0.60, indicating satisfactory factorability for all the items (see table2 and 3). In conclusion, the initial findings of the EFA show that the NINE first-order indicators have clear factor structures.
Insert Table2 Here Insert Table3 Here
Table4 reports the reliability of the NINE first-order constructs using Cronbach's alpha (Churchill, 1979). The overall coefficient alpha is (0.959), which is much higher than the threshold (0.70). In addition, the NINE scales revealed satisfactory levels, ranging from 0.745 to 0.921, and all exceeded the suggested minimum threshold (>0.70). Regarding composite reliabilities, table 9.26 shows the NINE scales ranged from 0.854 to 0.941, exceeding the recommended threshold value of 0.8. These findings indicate that the NINE scales are internally consistent and have acceptable reliability in their original form
Insert Table 4 Here With regard to convergent validity, which measures the extent to which items on a scale are in theory linked (Harris et al., 2010), it was assessed by observing the average variance extracted (AVE) index using SmartPLS (Wetzels et al., 2009). Table5 shows that the AVE for all the NINR first-order indicators exceeded the minimum threshold value of 0.5, explaining more than 50% of the variance in their observable measures. The AVE of the first-order constructs ranged from 0.556 to 0.762.
Insert Table 5 Here
Regarding discriminant validity, which refers to the extent to which a latent variable A is different and unique from other latent variables (e.g., B, C, D) (Bagozzi et al., 1991), it was assessed by the Fornell-Larcker criterion using SmartPLS (Fornell and Larcker, 1981). Table6 shows that the root AVE values of all the NINE first-order reflective indicators were greater than the corresponding off-diagonal correlations. These findings suggest that the NINE first-order reflective indicators have adequate discriminant validity.
Insert Table 6 Here
Data Analysis: Validating Formative Constructs
To assess formative constructs (second-order and third-order constructs), the researcher constructed Alumni Orientation Index (AOI) as hierarchical latent variable model using the repeated indicator approach and used Mode B measurement and inner path weighting scheme (Becker et al., 2012; Henseler et al., 2009). In addition, three criteria: external validity, formative outer weights and multicollinearity (Wetzels et al., 2009; Bagozzi et al., 1991) were used to validate formative measurement models of AOI.
Table7 shows that the results of the three different tests reveal that the three second-order formative constructs (EM, PM and CAA) are valid constructs and can be used to validate AOI as a third-order formative construct. The same procedures were used to validate AOI as a third-order formative constructs (see table 8). Based on the statistical analysis of the NINE first-order indicators, the THREE formative second-order construct and AOI as a third-order formative construct, it is confirmed that AOI is a third-order construct that consists of three second-order formative constructs and three first-order reflective indicators. Therefore, H1 is rejected.
However, in order to understand how development and alumni offices (DAROs) perceive and practice AO, the mean scores of AO and the NINE first-order indicators were calculated. The findings reveal that the mean score of AO was moderate at (3.48) out of a perfect score of (5.00) and the median was higher, at (3.63). The standard deviation was only (0.67), indicating that respondents’ perceptions on how alumni oriented the service practices of their respective institutions did not have much difference. The minimum and maximum values for overall AO were 1.00 and 4.42 respectively.
However, of the NINE first-order constructs, ‘Cases for Support’ (3.86) and ‘Social Media Management’ (3.78) had the highest mean values. This indicates that the highest priority of DAROs is to ensure that cases for support: communicate to alumni the importance and the impact of their donation and the wider role the university plays in society; are developed based on consultation with key internal and external stakeholders; and match individual preferences concerning which areas, projects and services to support. The results also indicate that: having a well-developed social media strategy, using and developing social media accounts to foster networking among alumni, administering social media accounts by a professional team, regularly updating social media accounts with fresh, timely and tailored content to alumni, and measuring social media efforts using qualitative/quantitative metrics are on the top priorities of DAROs.
Conversely, DAROs put less emphasis on ‘Promoting Best Practices of Event Management’ (3.65), ‘Alumni Database Management’ (3.59) and ‘Interfunctional Coordination’ (3.44). Other indicators also accounted for low priorities of DAROs and these are: ‘Intrafunctional Coordination’ (3.26), ‘Financing Publication Management’ (3.21), “Financing Event Management” (3.20) and ‘Promoting Best Practices of Publication Management’ (3.12). The relatively high standard deviation of ‘Financing Event Management’ and ‘Financing Publication Management’ (0.93162) and (0.92139), respectively, indicates that need for DAROs to allocate adequate resources (e.g., financial resources, human resources, dedicated marketing activities, etc.) to: diversify publications (3.41), enhance the quality of publications (3.26), guide and assist regional alumni groups worldwide on planning and managing their events (3.07), coordinate an optimum number of events in the countries where the masses of alumni are located (3.04), ensure that alumni magazines are delivered to the largest number of alumni (2.98), coordinate a wide range of alumni events at national level (2.95) and international level (2.97), and ensure that the frequency of e-communications with alumni is optimum (2.76).
Discussion and Conclusions
The study contributes to the marketing literature through responding to the several calls made to develop AO measurements that are sufficient in capturing the comprehensive nature of a truly market-oriented operating philosophy. Previous researchers used five items on average to measure the level of AO, whereas, this study develops 48 items. The study also shows that MO is constituent-specific. Such an approach recognises the distinct constituencies when developing, understanding and implementing MO. In addition, this approach, which advances our understanding of the differences between constituency groups, shows that the components that constitute the concept of MO would certainly vary from one constituency group to another. However, such issues, as far as the researcher’s knowledge is concerned, have never been taken into account in prior research. The majority of previous researchers used either standardized items or items that shared great similarities to measure the level of MO against multiple constituency groups. This approach, which maybe said to add little to our understanding on the inherent differences between different constituency groups, did not recognise that the marketing practices and approaches directed towards one group would differ significantly from those towards other groups. Furthermore, it did not recognise that organisations will have to customize their marketing practices in order to shape and influence their relationships with each group.
Additionally, the study contributes to MO literature through broadening the domain of MO. It shows that MO is not about adding as many constituency groups as possible to broaden the domain of the concept, but about identifying the set of dimensions that capture the comprehensive nature of the concept towards each constituency group. In other words, the study demonstrated that the concept can be extended to include at its foundation the set of activities that organisations perform towards each constituency group, rather than on the number of constituency groups that need to be included when conceptualising and measuring the concept.
Accordingly, the current study confirms AO is a third-order formative construct that consists of three second-order formative constructs and three first-order reflective constructs. It shows that the measurement instrument for AOI is the 48-item which can be validly and reliably measured using the NINE multi-item components of: Case for Support; Alumni Database Management; Social Media Management; Financing Event Management; Financing Publication Management; Promoting Best Practices of Event Management; Promoting Best Practices of Publication Management; Intrafunctional Coordination and Interfunctional Coordination. In addition, the NINE first-order indicators of AO, which demonstrated the construct and criterion validated, were closely related and exhibited a high level of interdependence and strong convergence towards the overall index (i.e. AO). These close correlations and convergent relationships are in line with other market orientation studies (e.g., Gray et al., 1998; Jaworski and Kohli, 1993).
The study suggests a strategic index called Alumni Orientation for the HE sector. The instrument that measures this index is intentionally developed to build more market-driven institutions. This managerial tool, which can contribute to yielding a pool of performance indicators in HE, can be effectively used to monitor the level of commitment of HE institutions in serving the alumni market. The scores for the various index components will offer key information on the different service practices that have to be improved in order to enhance constituencies’ values and their level of satisfaction.
In addition, for the sake of simplicity, individual indicators that form the index can be computed (i.e. the average value of that indicator per item) for each school/faculty. They will be able to give an indication of how alumni-oriented the university is, in terms of serving their target markets with respect to a particular indicator. Moreover, the aggregated index value of all the indicators that form the index will give an overall picture of the level of orientation in serving customers. University leaders can use these indices as a yardstick, on which improvement efforts can be focused. In other words, the development of the parsimonious AO instrument should help managers and deans to pinpoint areas of weakness and enable them to make corrective action swiftly. Furthermore, university leaders need to be constantly advised that AOI is multi-dimensional constructs that consist of integral indicators. The high correlations and interdependence among the indicators that form the index imply that it has to be executed as a whole rather than piecemeal. In short, all the indicators that form AOI need to be embraced simultaneously because it bears a holistic philosophy.
Andreoni, J. (1989). Giving with impure altruism: Applications to charity and Ricardian equivalence. Journal of Political Economy 97, 1447-1458.
Arnett, D., German, S., and Hunt, S. (2003). The identity salience model of relationship marketing success: The case of nonprofit marketing. Journal of Marketing, 67: 89–105.
Bagozzi, R., and Phillips, L. (1991). Assessing Construct Validity in Organizational Research. Adminstrative Science Quarterly, 36(3), 421-58.
Becker, J., Klein, K., and Wetzels, M. (2012). Hierarchical latent variable models in PLS-SEM: guidelines for using reflective-formative type models. Long Range Planning, 45, (5–6), 359–394.
Brady, E., Brace-Govan, J., Brennan, L., and Conduit, J. (2011). Market orientation and marketing in nonprofit organizations. Indications for fundraising from Victoria. International
Journal of Nonprofit and Voluntary Sector Marketing, 16(1), 84–98.
Church, A., and Waclawski, J. (2001). Designing and using organizational surveys: Seven step process. San Francisco: Jossey-Bass
Churchill, G. (1979). A paradigm for developing better measures of marketing constructs, Journal of Marketing Research, 19: 491-504.
Corley, G., and Gioia, D. (2004). Identity Ambiguity and Change in the Wake of a Corporate
Cunningham, M., and Cochi-Ficano, K. (2002). The determinants of donative revenue flows from alumni of higher education: An empirical inquiry. The Journal of Human Resources 37(3), 540-569.
Denzin, K., and Lincoln, Y. (2011). Handbook of Qualitative Research. Thousand Oaks: Sage Publications, Inc.
Dwairi, M., Akour, I., and Sayyer, W. (2012). Antecedents And Consequences Of Market Orientation In Public Sector In The Kingdom Of Bahrain. International Conference on Business,
Economics, Management and Behavioral Science, 499-505.
Fearn, H. (2010). Recession hits US private fundraising, but others do better. Retrieved from:
Flavian, C., and Lozano, J. (2006). Organisational antecedents of market orientation in the public university system. International Journal of Public Sector, 19 (5), 447-467.
Fornell, C., and Larcker, D. (1981). Evaluating structural equation models with unobservable variables and measurement error. Journal of Marketing Research, 18(1), 39-50.
Frumkin, P., and Kim, M. (2001). Strategic positioning and the financing of non-profit organizations: Is efficiency rewarded in the contributions marketplace? Public Administration
Review, 61: 266-275.
Gonzalez, L., Vijande, M., and Casielles, V. (2002). The market orientation concept in the private nonprofit organisation domain. International Journal of Nonprofit and Voluntary Sector
Marketing, 7(1), 55 – 67.
Gray, B., Matear, M., Christo, B Phil, M. (1998). Developing a Better Measure of Market Orientation. European Journal of Marketing, 32(9/10), 884-904.
Hair, J., Black, W., Babin, B., Anderson, R., and Tatham, R. (2006). Multivariate data analysis. 6th ed. New Jersey: Prentice Hall.
Hammond, L., Webster, L., and Harmon, H. (2009). The impact of market orientation toward students and student/faculty ratios on performance excellence in the case of AACSBinternational member schools. Academy of Marketing Studies Journal, 13(2), 45-58.
Harrison, W., Mitchell, S., and Peterson, P. (1995). Alumni donations and colleges’ development expenditures: Does spending matter? American Journal of Economics and Sociology, 54(4): 397-
Hemsley-Brown J., and Oplatka, I. (2010). Market Orientation in Universities: A comparative study of two national higher education systems. International Journal of Educational Management, 24(3), 204- 220.
Henseler, J., Ringle, C., and Sinkovics, R. (2009). The Use of Partial Least Squares Path Modeling in International Marketing. Advances in International Marketing 20: 277-320.
Hochman, H., and Rodgers, J. (1973). Utility interdependence and income transfer through charity. In transfer in an Urbanized Economy. (e.d. Kenneth E. Boulding). Belmont.
Huang, Y. (2001). OPRA: A Cross-Cultural, Multiple-Item Scale for Measuring Organization- Public Relationships. Journal of Public Relations Research, 13(1), 61-90.
Jarvis, C., MacKenzie, S., and Podsakoff, P. (2003). A critical review of construct indicators and measurement model misspecification in marketing and consumer research. Journal of Consumer
Research, 30 (2), 199-218.
Jaworski, B., and Kohli, A. (1993). Market orientation: antecedents and consequences. Journal of Marketing, 57, July: 53-70.
Keating, B., Pitts, R., and Appel, D. (1981). United Way contributions, coercion, charity or economic self-interest? Southern Economic Journal 47: 816-823.
Kimery, K., and Rinehart, S. (1998). Markets and Constituencies: An Alternative View of the Marketing Concept. Journal of Business Research, 43: 117-124.
Kotler, P., and Armstrong, G. (2008). Principles of Marketing (12th Ed). Pearson, Prentice-Hall. Upper Saddle River, New Jersey, NJ.
Law, M., Stewart, D., Letts, L., Pollock, N., Bosch, J., and Westmorland, M. (1998). Guidelines for critical review form-qualitative studies. Qualitative Review Form Guidelines. Retrievedfrom: www.srs-mcmaster.ca/Portals/20/pdf/ebp/quanguidelines.pdf
Leslie, L. L., and Ramey, G. (1988). Donor behavior and voluntary support for higher education institutions. Journal of Higher Education 59(2), 115-132.
Lovelock, L., and Weinberg, C. (1989). Public and Nonprofit Marketing (2nd eds). Redwood City: The Scientific Press.
Malhotra, N. (2004) Marketing Research: An Applied Orientation (4th ed.). Pearson Education, Inc: New Jersey.
Menon. A., Bharadwaj S., Adidam P., and Edison S. (1999). Antecedents and consequences of marketing strategy making: A model and a test. Journal of Marketing, 63 (April), 18-40.
Padanyi, P., and Gainer, B. (2004). Market orientation in the nonprofit sector: taking multiple constituencies into consideration. Journal of Marketing Theory and Practice, 12(2) 43-58.
Pavicˇic, J., Alfirevic, N., and Mihanovic, Z. (2009). Market orientation in managing relationships with multiple constituencies of Croatian higher education. High Education, 57:191-
Rivera-Camino, J. & Molero Ayala, V. (2010). Market Orientation at Universities. Construct and Exploratory Validation. Innovar, 20(36), 125-138.
Shapiro, B. (1973). Marketing for non-profit organizations. Harvard Business Review, 51(5),
Shapiro, B. (1988). What the Hell is 'Market Oriented'? Harvard Business Review 66: 119-125.
Siu, N., and Wilson, R. (1998). Modelling market orientation: An application in the education sector. Journal of Marketing Management, 14(4), 293-323.
Voon, B. (2007). SERVMO: A Measure for Service-Driven Market Orientation in Higher Education. Journal of Marketing for Higher Education, 17(2), 216-237.
Wasmer, D., and Bruner, G. (2000). The antecedents of the market orientation in higher education. Journal of Marketing for Higher Education, 9(2), 93-105.
Weerts, D., and Ronca, J. (2007). Profiles of supportive alumni: Donors, volunteers, and those who “Do it all”. International Journal of Educational Advancement, 7(1), 20-34.
Wetzels, M., Odekerken-Schroder, G., and van Oppen, C. 2009. Using PLS Path Modeling for Assessing Hierarchical Construct Models: Guidelines and Empirical Illustration. MIS Quarterly,
Zebal, M., and Goodwin, D. (2012). Market orientation and performance in private universities.