Purchasing and social responsibility: a replication and extension.

INTRODUCTION

The popular press offers evidence of a diverse range of business improprieties, including questionable financial behavior and violations of safety and human rights responsibilities. These actions underscore the significance of socially irresponsible behavior by organizations, including

negative ramifications for the bottomline. Conversely, socially responsible behavior can lead to greater customer loyalty and increased revenues. A survey by the Reputation Institute of over 21,000 consumers suggests that "almost unanimously, the public says it wants information about a company's record on social and environmental responsibility to help decide which companies to buy from, invest in, and work for" (Alsop 2002, p. B1). Further, socially responsible behavior can also contribute to the bottomline through lower costs. For example, 3M's environmental initiatives from 2000 to 2002 resulted in a 10 percent "improvement in energy efficiency" (3M Environmental, Social, and Economic Sustainability 2004), and Johnson and Johnson's safety program and goals resulted in a 67 percent, 14 percent and 71 percent reduction in fires, serious injuries and lost workdays, respectively, between 1991 and 2002 (Johnson & Johnson Social Responsibility 2004).

The issue of social responsibility at the corporate level extends to the purchasing and supply management function. This is evidenced by the grounded theory development of Carter and Jennings (2002) and the current efforts by the Institute for Supply Management[TM] (ISM) to study and operationalize social responsibility issues into principles for its organization (ISM Principles of Social Responsibility 2004). Additionally, recent work by Carter and Jennings (2004) has placed such formerly stand-alone issues as environmental purchasing, sourcing from minority business enterprise (MBE) suppliers, and other supply management issues such as human rights and safety within a broader conceptual and empirical framework of purchasing social responsibility (PSR). However, the results of the Carter and Jennings study were limited to organizations within consumer products industries. Additional research is needed to extend these findings to a broader range of industries.

As the field of supply management begins to mature into distinct streams of research (Carter and Ellram 2003), replication becomes an important component of maturation of the discipline. Researchers can only have strong confidence in theory when multiple attempts are made to falsify that theory:

     Only when certain events recur ... can our observations be tested.
     We do not take even our own observations quite seriously, or accept
     them as scientific observations, until we have repeated and tested
     them. Only by such repetitions can we convince ourselves that we
     are not dealing with a mere isolated "coincidence," but with events
     which, on account of their regularity and reproducibility, are in
     principle intersubjectively testable (Popper 1959, p. 45).

There seems to be a growing interest in replication research in the behavioral sciences including management in general (Eden 2002), and more specifically the fields of operations and supply management (Frohlich and Dixon 2003). The stance of this article is that replication is needed in supply management research, in order to advance as an academic discipline. In light of the rapidly growing interest of the supply management community in the area of social responsibility, and the need to begin to replicate research in our field, the specific purpose of this article is to replicate the findings of Carter and Jenning's (2004) empirical test of a conceptualization of PSR, by extending their study to a significantly broader group of industries. Such a replication and extension will provide greater assurance that PSR is indeed a higher-order construct that ties together past, stand-alone supply management research.

The next section of the article reviews the broad corporate social responsibility (CSR) literature, followed by more specific literature relating to PSR. This overview leads to an introduction of the five hypotheses tested in the Carter and Jennings (2004) study. Afterward, the methodology is described, followed by an examination of the results of a multigroup second-order confirmatory factor analysis using data from a large-scale survey of supply management professionals in a diverse group of manufacturing and service industries. This is followed by a discussion of the results. The article ends by providing guidance for future investigation in this emerging stream of supply management research.

LITERATURE REVIEW

Corporate Social Responsibility (CSR)

Within the CSR literature as it applies to the general management of an organization, researchers have examined specific dimensions of socially responsible activities, including the advancement of gender and racial workplace diversity (Clair et al. 1997), the effect of business on the natural environment (Hart and Ahuja 1996), contributions to philanthropy and the community (Mallot 1998; Wokutch 1998), human rights (Jennings and Entine 1999), and safety in the workplace (Wokutch 1992). In addition to the focus on these separate activities, several theoretical viewpoints and frameworks have been used to explore CSR from a broader and more holistic perspective. Friedman (1970) alleged that engaging in CSR is indicative of an agency theory dilemma or conflict between the interests of managers and shareholders. In essence, Friedman asserted that businesses are in the business of making money and that only the interests of shareholders should be considered in making managerial decisions. Freeman's (1984) stakeholder perspective suggests that business should be responsible not only to shareholders but to multiple, additional stakeholders, such as regulatory agencies, competitors, consumer advocacy groups and the media, as well as earlier identified stakeholders including customers, suppliers and employees (Ansoff 1965).

Davis (1973, p. 312) presented a classic definition of CSR as "the firm's consideration of, and response to, issues beyond the narrow economic, technical, and legal requirements of the firm." Like Davis, Freeman and Friedman, the work of Carroll (1979, 1991) also suggests that businesses have an economic responsibility to shareholders, but that this is just the base level in a hierarchy of social responsibility, as displayed in Figure 1. At the next level of social responsibility, businesses are expected to comply with legal obligations--the "ground rules"--imposed by governments and regulatory agencies (Carroll 1979, p. 500). Ethical responsibilities are those activities that are expected as part of societal norms but are not codified into law. They are often avoidance activities, such as "not lying" or "not accepting bribes." The final group of activities consists of those actions that are guided by an organization's discretion rather than any legal requirement or ethical norms.

[FIGURE 1 OMITTED]

[FIGURE 2 OMITTED]

Conceptual work by other researchers also suggests a hierarchical ordering of social responsibilities. For example, the model of Wartick and Cochran (1985) acknowledges and incorporates economic performance as the primary, base element of social responsibility, without excluding the other responsibilities defined by Carroll.

This article adopts the conceptualizations of Carroll (1979, 1991) and Wartick and Cochran (1985) in defining CSR as "meeting the discretionary responsibilities expected by society." This definition encompasses the activities relating to (1) diversity, (2) the environment, (3) human rights, (4) philanthropy and community, and (5) safety, as described at the beginning of this section. In addition, this definition is consistent with the PSR conceptualization and empirical findings of Carter and Jennings (2004).

Purchasing Social Responsibility (PSR)

The focus on social responsibility is narrowed in this section, by reviewing literature on purchasing and supply management activities that relate to the five groups of discretionary undertakings described above. The literature exploring environmental purchasing and supply management has relied primarily on case studies (Carter and Dresner 2001; Handfield et al. 1997; Narasimhan and Carter 1998) and survey research (Carter and Carter 1998; Min and Galle 1997). Most of these studies have investigated the antecedents to environmental purchasing, including external drivers such as customers and regulation (Carter and Carter 1998; Melnyk et al. 1999; Min and Galle 1997; Walton et al. 1998) and internal influences such as management support, training, and organizational culture (Carter et al. 1998; Drumwright 1994; Narasimhan and Carter 1998). In addition, Carter et al. (2000) combined survey and archival data and found a significant relationship between environmental purchasing and financial performance, after controlling for organization size, leverage, and primary earnings per share.

The extant supply management literature has examined diversity issues from the standpoint of sourcing from minority- and women-owned business enterprises (M/WBEs). Research in this stand-alone area of PSR has also focused upon the precursors to successfully developing and managing these programs. Internal factors, such as top management support and the inclusion of diversity sourcing criteria in the formal evaluations of supply management personnel, have been shown to be significant antecedents to M/WBE programs (Carter et al. 1999; Dollinger et al. 1991). Additionally, Krause et al. (1999) proposed that some firms' M/WBE programs are set in motion not only by social considerations but also by customer concerns, since minorities can represent sizable and growing market segments.

Scholarly supply management research has been much less common in the areas of human rights, philanthropy and community, and safety. Emmelhainz and Adams (1999) investigated human rights issues including selecting suppliers that pay a living wage and that avoid the use of sweatshop working conditions in their factories. Emmelhainz and Adams suggested that these issues have become relevant to supply managers as a result of greater awareness by consumers and increased regulatory scrutiny.

Most recently, these five stand-alone areas of supply management (diversity, the environment, human rights, philanthropy and community, and safety) have been conceptualized as a more holistic, higher-order construct of social responsibility termed "purchasing social responsibility" (PSR) (Carter and Jennings 2004). This concept of an umbrella construct of PSR is shown in Figure 2. Carter and Jennings adopt Carroll's (1979, 1991) framework and show that these five groups of socially responsible activities are first-order constructs that load on a second-order, PSR construct. Their five hypotheses, corresponding to the five discretionary groups of supply management activities reviewed in this section of the article, are introduced here:

H1a: Purchasing activities relating to the environment constitute a dimension of PSR.

H1b: Purchasing activities relating to diversity constitute a dimension of PSR.

H1c: Purchasing activities relating to human rights constitute a dimension of PSR.

H1d: Purchasing activities relating to philanthropy constitute a dimension of PSR.

H1e: Purchasing activities relating to safety constitute a dimension of PSR.

The next section describes the methodology used to replicate and extend the work of Carter and Jennings in testing these five hypotheses.

METHODOLOGY

A Web-based questionnaire was employed to test the study's hypotheses. The scale items from the Carter and Jennings (2004) study were used as a starting point to develop the measurement items for the current study's constructs. The original scale items are displayed in Appendix A. These items were modified slightly in order to improve applicability to a wider range of industries. This modification of the scale items was an iterative process, derived from the input of a focal member from each of seven firms, representing a diverse group of service and manufacturing operations, along with several additional managers and employees from these companies. The final scale items utilized in the current, Web-based survey are displayed in Appendix B, which also shows the results of a confirmatory factor analysis (CFA) for survey respondents in manufacturing industries. The same scale items are also exhibited in Appendix C, which displays the results of a CFA for respondents in service industries. The results of these CFAs will be discussed in more detail in the next section of the article.

The sampling frame consisted of all ISM members and affiliates with a title of manager or higher (Title 1 and Title 2 members and affiliates). This initial list was sorted and manually searched for duplicate organizations. In instances where the same organization was included more than once in the database, the organizational member with the highest title was maintained and all other potential respondents for that organization were eliminated. In January 2004, ISM e-mailed the survey to this culled sample of 11,119 members and affiliates. A reminder e-mail was sent one week after the initial e-mail. Of the e-mailed requests to complete the survey, 1,809 e-mail messages were returned as undelivered. A total of 1,163 organizations responded to the survey, for a response rate of 12.5 percent.

The responding organizations are divided among manufacturing (45 percent), service (30 percent), and government and education (9 percent), with the remaining 16 percent selecting "other" as a response. Additional demographic information is provided in Table I. The summary statistics displayed in the table suggest reasonable variability in organization size, with approximately 10 percent of firms with revenues of less than $10 million and 13 percent of firms with revenues of more than $5 billion. Similarly, about 19 percent of organizations had an annual spend of less than $10 million, while approximately 12 percent had an annual purchase volume of $1 billion or more.

Nonresponse bias was tested by comparing the responses of early versus late respondents to the survey, where late respondents are deemed to be more like nonrespondents than are early respondents (Armstrong and Overton 1977). A naturally occurring breakpoint between the first and second waves of the survey was used to represent early versus late respondents, and a multivariate t-test was computed using the responses to the survey questions displayed in Appendix B, to determine whether significant differences exist between the early and late respondents. The results indicate that the early respondents do not display statistically significant differences from late respondents (p = 0.4177).

To assess the key informant issue, two global measures of a respondent's ability to competently answer the survey questions were employed: (1) the number of years the respondent has been involved with the purchasing function's socially responsible initiatives, and (2) the degree of involvement on a 1-7 Likert scale, where 1 = only somewhat involved and 7 = very involved (Kumar et al. 1993; Phillips 1981, 1982). A total of 61 respondents were only "somewhat involved" in socially responsible initiatives for one year or less and were eliminated from further analyses.

ANALYSIS

The study's hypotheses were tested via a second-order CFA. Prior to conducting the second-order CFA, a first-order CFA of the five hypothesized dimensions of PSR was performed to ensure that the survey items used to measure these constructs were reliable and valid indicators of the constructs. Since it is reasonable to assume that manufacturing and service organizations might differ in their responses to the survey's questions, the factor structures of these two groups were examined and developed separately (Byrne et al. 1989). The factor structures of the government and education and the "other" groups were not large enough to undergo analyses via CFA and were thus not examined (MacCallum et al. 1992). The measurement models of the manufacturing and service groups are displayed in Appendices B and C, respectively. Those items that did not reliably or validly measure a construct were eliminated from further analyses, based on low factor loadings or high normalized residual values (Anderson and Gerbing 1988; Bagozzi and Yi 1988); these deleted scale items are also displayed in Appendices B and C.

All factor loadings had highly significant t-values in excess of 10.5 (p < 0.0001), providing evidence of convergent validity (Gerbing and Anderson 1988). Fornell and Larcker's (1981) composite reliability measure was used to assess scale reliability. All scales demonstrated acceptable levels of reliability, with composite reliabilities well in excess of the minimum 0.70 level recommended for existing scales (Churchill 1979; Flynn et al. 1990; Van de Ven and Ferry 1978). Finally, discriminant validity was examined by performing one-at-a-time [chi square] difference tests, where cross-construct correlations were large (Bollen 1989). Similarly, the 95 percent confidence interval was examined to ensure that the correlation plus two standard errors did not exceed 1.00 (Anderson and Gerbing 1988). For each test of discriminant validity, the [chi square] value was significantly improved for the unconstrained versus constrained model (p < 0.0001), and no confidence interval exceeded 1.00. These results provide evidence of discriminant validity for the measurement model of the hypothesized dimensions of PSR.

The [chi square]/degrees of freedom ratio, Bentler's (1989) comparative fit index (CFI), Bentler and Bonett's (1980) non-normed fit index (NNFI), and the root mean square error of approximation (RMSEA) (Steigler 1990) were used to assess the fit between the measurement model and the data. An appropriate fit between the measurement model and the data is indicated by a [chi square]/degrees of freedom ratio below 3.00 (Bollen and Long 1993); values above 0.90 on the CFI and NNFI (Bentler 1989; Bentler and Bonett 1980); and a value of 0.08 or less for the RMSEA (Browne and Cudeck 1993). The results of the CFAs for the manufacturing and service groups are displayed in Table II, along with the results from the Carter and Jennings (2004) study. As shown in Table II, all values of the fit indices for the manufacturing and service groups fell within acceptable limits, providing indication of good model fit.

The measurement models developed for the manufacturing and service sectors appear to be identical, with the exception of the scale item "participates in the design of products for recycling or reuse," which loaded on the Environmental construct in the manufacturing group but not in the service group. A more formal comparison of the measurement models was made by performing a two-group CFA of the remaining four constructs (diversity, human rights, philanthropy and community, and safety) across the manufacturing and service sectors, with factor loadings constrained to be equal across the two sectors. None of the constraints were significantly different from zero (p > 0.05), indicating measurement equivalency across sectors.

In concert, the above results provide strong support for the overall reliability and validity of the scale items used to measure the five hypothesized dimensions of PSR.

In order to test the study's hypotheses, the five PSR constructs were next subjected to a second-order CFA for the manufacturing and service groups. The results of these analyses are displayed in Table III. For the manufacturing sector, the [chi square]/degrees of freedom ratio is 2.66, and the values of the CFI, NNFI and RMSEA are 0.96, 0.95 and 0.07, respectively. Quite comparable fit statistics for the service sector are displayed in the next column of the table, with [chi square]/degrees of freedom ratio, CFI, NNFI and RMSEA values of 2.67, 0.95, 0.94 and 0.07, respectively. The t-values associated with the path loadings are significant at the 0.0001 level for each of the five first-order PSR construct, suggesting that these areas are dimensions of the second-order PSR construct, and providing strong support for Hypotheses 1a-1e.

Comparison of Means Between Industry Sectors

Finally, a comparison of scale item means was conducted across industry sectors. Alpha inflation was controlled for by employing the Bonferroni approach for multiple comparisons (Flynn et al. 1990, p. 264). The results of these means comparisons, which are displayed in Table IV, suggest that supply managers in manufacturing sectors are significantly more involved in the environmental, human rights and safety activities than are respondents in service industries. Conversely, supply managers in service industries appear to be significantly more involved in supplier diversity initiatives.

DISCUSSION AND MANAGERIAL IMPLICATIONS

The Carter and Jennings (2004) study was limited to organizations in consumer products industries. The results of this study confirm the earlier findings of Carter and Jennings, and offer additional and more generalizable evidence, across a diverse set of industries, that PSR provides a theoretical umbrella under which a wide range of socially responsible supply management activities can be understood. Like the earlier study, the current findings present strong evidence that PSR is a second-order construct that includes activities surrounding the areas of diversity, the environment, human rights, philanthropy and community, and safety. By using the PSR framework to bring these activities together, a more inclusive picture of the significance of supply management actions with a social dimension can be developed.

From the standpoint of theory and related research, the findings suggest that researchers should recognize the interrelatedness of the PSR dimensions in developing and testing models and frameworks. Pragmatically, this means that researchers not only should review the extant literature in their particular area of interest (e.g., environmental purchasing) but also should consider relevant findings in other areas of PSR. Further, supply management researchers should incorporate the rich theories and empirical findings from the CSR literature in the course of developing, testing and interpreting models that investigate stand-alone dimensions of PSR.

The findings indicate that managers view each of the five individual constructs--diversity, the environment, human rights, philanthropy and community, and safety--as belonging to a higher-order construct of PSR. Supply managers and executives should thus explicitly recognize the interrelatedness of PSR activities and programs and manage these activities accordingly. Lessons learned in one area of PSR, in terms of the barriers, drivers and consequences, might very well apply to other PSR activities (Carter and Jennings 2002, 2004). Supply management executives must ensure that a coordination mechanism exists that allows supply managers of stand-alone activities, such as diversity or the environment, to coordinate and share knowledge.

The equivalency of the measurement models and second-order factor structures across samples is encouraging, as this provides evidence that the PSR scale consists of activities that are generalizable to a wide range of industries. The results also suggest that supply managers in both manufacturing and service industries view PSR activities as being encompassed by a higher-order PSR construct. However, the results displayed in Table IV indicate that the mean levels of PSR activities vary significantly across sectors.

Supply management professionals in manufacturing industries are significantly more involved in environmental, human rights and safety activities than are their counterparts in service industries. Some of these differences might be explained by the larger proportion of purchasing volumes associated with firms in manufacturing versus service industries (Krause and Scannell 2002). For example, relatively high spend is probably required to warrant activities such as "conduct(ing) supplier visits to ensure that suppliers are not using sweatshop labor." Disparities in the operating characteristics of manufacturing versus service industries may explain other significant differences in the mean responses across these two groups. For instance, manufacturers purchase higher proportions of goods versus services (Krause and Scannell 2002) and are thus more likely to engage in packaging reduction activities; safety is often more of an overall, corporate priority for firms with, as opposed to without, manufacturing processes, due to the management of plants and warehouses. Firms in service industries, which conversely purchase a higher proportion of services versus goods, may have a higher proportion of MBE spend because MBE suppliers are themselves more often in service as opposed to manufacturing industries (Auskalnis et al. 1995).

FUTURE RESEARCH

The majority of research in the area of PSR has examined the antecedents to both the individual and collective areas of interest. Obviously, such research is valuable, as it aids managers in identifying the drivers and facilitators that allow the most effective means of implementing PSR. However, there is a huge gap in the investigation of the consequences of PSR. How does PSR affect a firm's bottomline--is it just window dressing, or can it positively impact profitability? How can businesses objectively measure and assess the costs and benefits of PSR activities?

Organizations may be hesitant to initiate or expand PSR programs unless a solid business case can be made. Further, some organizations perceive that socially responsible programs can lead to increased costs. Companies must thus be able to accurately gauge changes in both performance and costs that result from their PSR activities. Future research, using a case study methodology, could investigate how purchasing organizations might use a life-cycle analysis (LCA) and/or a total cost of ownership approach to assess PSR activities (Ellram and Siferd 1998; Ellram and Feitzinger 1997; Nielsen and Wenzel 2002).

Purchasing managers can use the PSR questionnaire, which is available on the ISM Web site (www.ism.ws), to measure their organizations' current involvement in PSR activities, and to benchmark their involvement against those of hundreds of other organizations in the manufacturing and service sectors. Managers might ask multiple employees to complete the ISM survey. Agreement among employees across survey items will provide some evidence of the reliability of responses (Phillips 1981). Conversely, a lack of agreement among employees would warrant investigation and could yield rich insights regarding why disagreements exist across various areas or levels of the supply management function.

While some explanation has been given regarding the reasons why mean differences exist in PSR activities between the manufacturing and service groups, additional research is needed to explore, in an in-depth fashion, why these differences occur. These types of "why" questions would be fitting for a case study methodology (Yin 1994).

Finally, this stream of research in purchasing management in general, and even PSR in particular, has developed to the point that it is in need of additional methodologies to augment the often employed mail surveys and somewhat less often used case studies and interviews. Two additional methodologies that might be applied are laboratory experiments and social network analysis. Laboratory experiments, which in contrast to surveys and field research have very strong internal validity, might be used to study the decision-making processes concerning PSR. This research method could aid in answering such questions as "How do managers decide when and which specific PSR activities to initiate?" and "What are the trade-offs and specific criteria that managers consider when making these decisions?" Social network analysis, which originated in the field of sociology (Tichy et al. 1979), could be used to gain enhanced insights into the cross-functional and even interorganizational interplay that takes place as organizations initiate and continue to manage PSR activities.

Appendix A ORIGINAL QUESTIONNAIRE SCALE ITEMS

Construct                                             Standardized
Reliability (b)                                       Factor Loading (a)

Socially Responsible Purchasing
0.84
The Environment (c)                                        (0.74)
Diversity (c)                                              (0.52)
Human Rights (c)                                           (0.85)
Philanthropy (c)                                           (0.66)
Safety (c)                                                 (0.80)

The Environment (b)
0.86
Currently, our purchasing function: (d)
... uses a life-cycle analysis to evaluate the
    environmental friendliness of products and
    packaging                                              (0.70)
... participates in the design of products for
    disassembly                                            (0.71)
... asks suppliers to commit to waste reduction
    goals                                                  (0.81)
... participates in the design of products for
    recycling or reuse                                     (0.85)

Diversity
0.82
Currently, our purchasing function: (d)
... purchases from minority-/women-owned business
    enterprise (M/WBE) suppliers                           (0.85)
... has a formal M/WBE supplier purchase program           (0.82)

Human Rights
0.86
Currently, our purchasing function: (d)
... visits suppliers' plants to ensure that they
    are not using sweatshop labor                          (0.85)
... ensures that suppliers comply with child labor
    laws                                                   (0.91)
... asks suppliers to pay a "living wage" greater
    than a country's or region's minimum wage              (0.69)

Philanthropy
0.75
Currently, our purchasing function: (d)
... volunteers at local charities                          (0.82)
... donates to philanthropic organizations                 (0.72)

Safety
0.73
Currently, our purchasing function: (d)
... ensures that suppliers' locations are operated
    in a safe manner                                       (0.89)
... ensures the safe, incoming movement of product
    to our facilities                                      (0.62)

(a) Standardized factor loadings of scale items measuring latent
constructs are given in parentheses. All loadings are significant
(t > 18, p < 0.0001).
(b) Composite reliability.
(c) The loadings here represent the loadings of each first-order
construct on the second-order, PSR construct. All loadings are
significant (t > 18, p < 0.0001).
(d) These items were measured on a 7-point Likert scale where 1 = no
extent whatsoever and 7 = very great extent.

Appendix B QUESTIONNAIRE SCALE ITEMS FROM SECOND STUDY-MANUFACTURING
SECTOR

Construct                                             Standardized
Reliability (b)                                       Factor Loading (a)

Socially Responsible Purchasing
0.87
Diversity (c)                                              (0.68)
The Environment (c)                                        (0.76)
Human Rights (c)                                           (0.85)
Philanthropy/Community (c)                                 (0.67)
Safety (c)                                                 (0.81)

The Environment
0.87
Currently, our purchasing and supply management
    function: (d)
... reduces packaging materials (E2)                       (0.68)
... challenges suppliers to commit to waste
    reduction goals (E4)                                   (0.84)
... participates in the design of products for
    recycling or reuse (E5)                                (0.80)
... assesses the environmental responsibility of
    suppliers (E6)                                         (0.83)
... evaluates the environmental friendliness of
    purchased products and packaging (Deleted: large
    standardized residuals with E2, E4 and E5)
... has waste reduction goals for the organization
    (Deleted: large standardized residual with E4)

Diversity
0.82
Currently, our purchasing and supply management
    function: (d)
... purchases from diversity-owned suppliers               (0.83)
... has a formal supplier diversity purchase program
    (D2)                                                   (0.83)
... proactively develops diversity-owned suppliers
    (Deleted: large standardized residuals with D1,
    D2 and S1)
... requires suppliers to implement and support an
    active supplier diversity program (Deleted:
    large standardized residual with D2)

Human Rights
0.90
Currently, our purchasing and supply management
    function: (d)
... conducts supplier visits to ensure that
    suppliers are not using sweatshop labor                (0.82)
... ensures that suppliers comply with child labor
    laws                                                   (0.94)
... requires suppliers to provide fair compensation
    (a living wage) to workers                             (0.85)
... treats people with dignity and respect (Deleted:
    low factor loading)
... assesses key secondary suppliers to ensure
    compliance with human rights policies and goals
    (Deleted: large standardized residual with V5)
... requires suppliers to demonstrate a proactive
    human rights program (Deleted: large
    standardized residuals with V3 and V2)

Philanthropy/Community
0.84
Currently, our purchasing and supply management
    function: (d)
... has volunteers supporting local charities              (0.84)
... donates to community organizations                     (0.86)

Safety
0.76
Currently, our purchasing and supply management
    function: (d)
... ensures that suppliers' locations are operated
    in a safe manner                                       (0.94)
... ensures the safe, incoming movement of product
    to our facilities (S2)                                 (0.60)
... ensures that each employee in our organization
    works in a safe environment (Deleted: large
    standardized residual with S2)

(a) Standardized factor loadings of scale items measuring latent
constructs are given in parentheses. All loadings are significant
(t > 11.17, p < 0.0001).
(b) Composite reliability.
(c) The loadings here represent the loadings of each first-order
construct on the second-order PSR construct. All loadings are
significant (t > 11.68, p < 0.0001).
(d) These items were measured on a 7-point Likert scale where 1 = no
extent whatsoever and 7 = very great extent.

Appendix C QUESTIONNAIRE SCALE ITEMS FROM SECOND STUDY--SERVICE SECTOR

Construct                                             Standardized
Reliability (b)                                       Factor Loading (a)

Purchasing
0.88
Diversity (c)                                              (0.59)
The Environment (c)                                        (0.77)
Human Rights (c)                                           (0.85)
Philanthropy/Community (c)                                 (0.75)
Safety (c)                                                 (0.86)

The Environment
0.87
Currently, our purchasing and supply management
    function: (d)
... reduces packaging materials (E2)                       (0.74)
... challenges suppliers to commit to waste
    reduction goals (E4)                                   (0.91)
... participates in the design of products for
    recycling or reuse (ES: Deleted: low factor
    loading)
... assesses the environmental responsibility of
    suppliers (E6)                                         (0.83)
... evaluates the environmental friendliness of
    purchased products and packaging (Deleted: large
    standardized residuals with E2, E4 and E5)
... has waste reduction goals for the organization
    (Deleted: large standardized residual with E4)

Diversity
0.80
Currently, our purchasing and supply management
    function: (d)
... purchases from diversity-owned suppliers               (0.81)
... has a formal supplier diversity purchase program
    (D2)                                                   (0.82)
... proactively develops diversity-owned suppliers
    (Deleted: large standardized residuals with D1,
    D2, and S1)
... requires suppliers to implement and support an
    active supplier diversity program (Deleted:
    large standardized residual with D2)

Human Rights
0.89
Currently, our purchasing and supply management
    function: (d)
... conducts supplier visits to ensure that
    suppliers are not using sweatshop labor                (0.85)
... ensures that suppliers comply with child labor
    laws                                                   (0.90)
... requires suppliers to provide fair compensation
    (a living wage) to workers                             (0.80)
... treats people with dignity and respect (Deleted:
    low factor loading)
... assesses key secondary suppliers to ensure
    compliance with human rights policies and goals
    (Deleted: large standardized residual with V5)
... requires suppliers to demonstrate a proactive
    human rights program (Deleted: large
    standardized residuals with V3 and V2)

Philanthropy/Community
0.83
Currently, our purchasing and supply management
    function: (d)
... has volunteers supporting local charities              (0.87)
... donates to community organizations                     (0.81)

Safety
0.75
Currently, our purchasing and supply management
    function: (d)
... ensures that suppliers' locations are operated
    in a safe manner                                       (0.93)
... ensures the safe, incoming movement of product
    to our facilities (S2)                                 (0.59)
... ensures that each employee in our organization
    works in a safe environment (Deleted: large
    standardized residual with S2)

(a) Standardized factor loadings of scale items measuring latent
constructs are given in parentheses. All loadings are significant
(t > 10.49, p < 0.0001).
(b) Composite reliability.
(c) The loadings here represent the loadings of each first-order
construct on the second-order PSR construct. All loadings are
significant (t > 7.66, p < 0.0001).
(d) These items were measured on a 7-point Likert scale where 1 = no
extent whatsoever and 7 = very great extent.

Table I RESPONDENT DEMOGRAPHICS

Annual Revenues                     Percent of Respondents

Under $10 million                             9.72%
$10 million to under $50 million             22.63%
$50 million to under $100 million             9.96%
$100 million to under $500 million           20.79%
$500 million to under $1 billion              9.96%
$1 billion to under $5 billion               14.27%
$5 billion to $10 billion                     5.54%
Over $10 billion                              7.13%

Annual Purchase Volume (Spend)      Percent of Respondents

Under $10 million                            19.04%
$10 million to under $50 million             28.95%
$50 million to under $100 million            12.81%
$100 million to under $500 million           19.60%
$500 million to under $1 billion              7.57%
$1 billion to $5 billion                      8.02%
Over $5 billion                               4.01%

Table II MEASUREMENT MODEL FIT STATISTICS

                      Carter &
                      Jennings  Manufacturing  Service  Recommended
                      (2004)    Sector         Sector   Value

[chi square]/degrees
  of freedom          1.388     2.21           2.17     <=3.00
Bentler's CFI         0.97      0.97           0.97     >=0.90

Bentler and Bonett's
  NNFI                0.96      0.96           0.96     >=0.90
RMSEA                 0.05      0.06           0.06     <=0.08

Table III MEASUREMENT MODEL FIT STATISTICS

                      Carter &
                      Jennings  Manufacturing  Service  Recommended
                      (2004)    Sector         Sector   Value

[chi square]/degrees
  of freedom          1.822     2.66           2.67     <=3.00
Bentler's CFI         0.91      0.96           0.95     >=0.90
Bentler and Bonett's
  NNFI                0.90      0.95           0.94     >=0.90
RMSEA                 0.07      0.07           0.07     <=0.08

Table IV MEANS COMPARISON ACROSS SECTORS

                                      Mean           Mean
Scale Item                            Manufacturing  Service

Diversity
... purchases from diversity-owned
  suppliers                           3.58           3.99       S>M**
... has a formal supplier diversity
  purchase program                    2.88           3.41       S>M**
... proactively develops diversity-
  owned suppliers                     2.73           3.21       S>M**

Environment
... reduces packaging materials       4.24           3.39       M>S***
... challenges suppliers to commit
  to waste reduction                  3.17           2.68       M>S**
... assesses the environmental
  responsibility of suppliers         3.17           2.79       M>S*

Human Rights
... conducts supplier visits          3.71           2.94       M>S***
... ensures that suppliers comply
  with child labor laws               3.43           3.00       M>S*
... requires suppliers to provide
  fair compensation ...               2.79           2.84

Philanthropy/Community
... has volunteers supporting local
  charities                           3.61           3.90
... donates to community
  organizations                       3.90           4.10

Safety
... ensures that suppliers'
  locations ...                       4.15           3.40       M>S***
... ensures the safe, incoming
  movement ...                        5.35           4.83       M>S***

Notes: The complete scale items appear in Appendices B and C. All
p-values for means comparisons were adjusted to control for alpha
inflation. Items with significantly different means across industry
sectors are bolded.
* p < 0.05
** p < 0.01
*** p < 0.0001

The author acknowledges the helpful and critical comments contributed by three anonymous Journal of Supply Chain Management reviewers and the editor on previous versions of the article.

REFERENCES

3M Environmental, Social, and Economic Sustainability, www.3m.com/about3m/ sustainability, accessed May 1, 2004.

Alsop, R. "Perils of Corporate Philanthropy--Touting Good Works Offends the Public, but Reticence Is Misperceived as Inaction," The Wall Street Journal, January 16, 2002, p. B1.

Anderson, J. and D.W. Gerbing. "Structural Equation Modeling in Practice: A Review and Recommended Two-Step Approach," Psychological Bulletin, (103:3), 1988, pp. 411-423.

Ansoff, I. Corporate Strategy, McGraw-Hill, New York, NY, 1965.

Armstrong, J.S. and T.S. Overton. "Estimating Nonresponse Bias in Mail Surveys," Journal of Marketing Research, (14:3), 1977, pp. 396-402.

Auskalnis, R.J., C.L. Kethum and C.R. Carter. Purchasing From Minority Business Enterprises: Best Practices, Center for Advanced Purchasing Studies, Tempe, AZ, 1995.

Bagozzi, R.P. and Y. Yi. "On the Evaluation of Structural Equation Models," Journal of the Academy of Marketing Science, (16:1), 1988, pp. 74-94.

Bentler, P.M. EQS Structural Equations Program Manual, BMDP Statistical Software, Los Angeles, CA, 1989.

Bentler, P.M. and D.G. Bonett. "Significance Tests and Goodness-of-Fit in the Analysis of Covariance Structures," Psychological Bulletin, (88:3), 1980, pp. 588-606.

Bollen, K.A. Structural Equations With Latent Variables, Wiley, New York, NY, 1989.

Bollen, K.A. and J.S. Long. "Introduction." In Testing Structural Equation Models, K.A. Bollen and J.S. Long (Eds.), Sage Publications, Newbury Park, CA, 1993, pp. 1-9.

Browne, M.W. and R. Cudeck. "Alternative Ways of Assessing Model Fit." In Testing Structural Equation Models, K.A. Bollen and J.S. Long (Eds.), Sage Publications, Newbury Park, CA, 1993, pp. 136-162.

Byrne, B.M., R.J. Shavelson and B. Muthen. "Testing for the Equivalence of Factor Covarfance and Mean Structures: The Issue of Partial Measurement Invariance," Psychological Bulletin, (105), 1989, pp. 456-466.

Carroll, A.B. "A Three-Dimensional Conceptual Model of Corporate Social Performance," Academy of Management Review, (4:4), 1979, pp. 497-505.

Carroll, A.B. "The Pyramid of Corporate Social Responsibility: Toward the Moral Management of Organizational Stakeholders," Business Horizons, (34:4), 1991, pp. 39-48.

Carter, C.R. and J.R. Carter. "Interorganizational Determinants of Environmental Purchasing: Initial Evidence From the Consumer Products Industries," Decision Sciences, (29:3), 1998, pp. 659-685.

Carter, C.R., L.M. Ellram and K.J. Ready. "Environmental Purchasing: Benchmarking Our German Counterparts," International Journal of Purchasing and Materials Management, (34:4), 1998, pp. 28-39.

Carter, C.R., R. Auskalnis and C. Ketchum. "Purchasing From Minority Business Enterprises: A Cross-Industry Comparison of Best Practices," The Journal of Supply Chain Management, (35:1), 1999, pp. 28-32.

Carter, C.R., R. Kale and C. Grimm. "Environmental Purchasing and Firm Performance: An Empirical Investigation," Transportation Research E, (36:3), 2000, pp. 219-228.

Carter, C.R. and M. Dresner. "Environmental Purchasing and Supply Management: Cross-Functional Development of Grounded Theory," The Journal of Supply Chain Management, (37:3), 2001, pp. 12-27.

Carter, C.R. and M.M. Jennings. "Logistics Social Responsibility: An Integrative Framework," Journal of Business Logistics, (23:1), 2002, pp. 145-180.

Carter, C.R. and L.M. Ellram. "Thirty-Five Years of The Journal of Supply Chain Management: Where Have We Been and Where Are We Going?" The Journal of Supply Chain Management, (39:1), 2003, pp. 38-50.

Carter, C.R. and M.M. Jennings. "The Role of Purchasing in the Socially Responsible Management of the Supply Chain: A Structural Equation Analysis," Journal of Business Logistics, (25:1), 2004, pp. 145-186.

Churchill Jr., G.A. "A Paradigm for Developing Better Measures of Marketing Constructs," Journal of Marketing Research, (16:1), 1979, pp. 64-73.

Clair, J.A., M. Crary, M. McDaniels, D. Spelman, J.D. Buote and T. MacLean. "A Cooperative Inquiry Into Teaching and Taking a Course on 'Managing Diversity,'" Research in Corporate Social Performance and Policy, Supplement 2, 1997, pp. 25-62.

Davis, K. "The Case For and Against Business Assumption of Social Responsibilities," Academy of Management Journal, (16:2), 1973, pp. 312-322.

Dollinger, M.J., C.A. Enz and C.M. Daily. "Purchasing From Minority Small Businesses," International Journal of Purchasing and Materials Management, (27:2), 1991, pp. 9-14.

Drumwright, M.E. "Socially Responsible Organizational Buying: Environmental Concern as a Noneconomic Buying Criterion," Journal of Marketing, (58:3), 1994, pp. 1-19.

Eden, D. "Replication, Meta-Analysis, Scientific Progress, and AMJ's Publication Policy," Academy of Management Journal, (45:5), 2002, p. 841.

Ellram, L.M. and S.P. Siferd. "Total Cost of Ownership: A Key Concept in Strategic Cost Management Decisions," Journal of Business Logistics, (19:1); 1998, pp. 55-84.

Ellram, L.M. and E. Feitzinger, "Using Total Profit Analysis to Model Supply Chain Decisions," Journal of Cost Management, (11:4), 1997, pp. 12-22.

Emmelhainz, M.A. and R.J. Adams. "The Apparel Industry Response to 'Sweatshop' Concerns: A Review and Analysis of Codes of Conduct," The Journal of Supply Chain Management, (35:3), 1999, pp. 51-57.

Flynn, B.B., S. Sakakibara, R.G. Schroeder, K.A. Bates and E.J. Flynn. "Empirical Research Methods in Operations Management," Journal of Operations Management, (9:2), 1990, pp. 250-284.

Fornell, C. and D. Larcker. "Evaluating Structural Equation Models With Unobservable Variables and Measurement Error," Journal of Marketing Research, (18:1), 1981, pp. 39-50.

Freeman, R.E. Strategic Management: A Stakeholder Approach, Pitman/Ballinger (Harper Collins), Boston, MA, 1984.

Friedman, M. "The Social Responsibility of Business Is to Increase Its Profits," The New York Times, September 13, 1970, pp. 122-126.

Frohlich, M. and J.R. Dixon. "Journal of Operations Management Special Issue: OM Replication Research." Journal of Operations Management, (21:3), 2003, pp. 373-374.

Gerbing, D.W. and J.C. Anderson. "An Updated Paradigm for Scale Development Incorporating Unidimensionality and Its Assessment," Journal of Marketing Research, (25:2), 1988, pp. 186-192.

Handfield, R.B., S.V. Walton, L.K. Seegers and S.A. Melnyk. "Green' Value Chain Practices in the Furniture Industry," Journal of Operations Management, (15:3), 1997, pp. 293-315.

Hart, S.L. and G. Ahuja. "Does It Pay to Be Green? An Empirical Examination of the Relationship Between Emission Reduction and Firm Performance," Business Strategy and the Environment. (5:1), 1996, pp. 30-37.

ISM Principles of Social Responsibility, www.ism.ws/SR/index.cfm, Web site accessed May 3, 2004.

Jennings, M.M. and J. Entine. "Business With a Soul: A Reexamination of What Counts in Business Ethics," Journal of Public Law and Policy, (20:1), 1999, pp. 1-88.

Johnson&Johnson Social Responsibility, www.jnj.com/community/environment/performance/index.htm, accessed May 1, 2004.

Krause, D.R., G.L. Ragatz and S. Hughley. "Supplier Development From the Minority Supplier's Perspective," The Journal of Supply Chain Management, (35:4), 1999, pp. 33-41.

Krause, D.R. and T.V. Scannell. "Supplier Development Practices: Product- and Service-Based Industry Comparisons," The Journal of Supply Chain Management, (38:2), 2002, pp. 13-21.

Kumar, N., L.W. Stern and J.C. Anderson. "Conducting Interorganizational Research Using Key Informants," Academy of Management Journal, (36:6), 1993, pp. 1633-1651.

MacCallum, R.C., M. Roznowski and L.B. Necowitz. "Model Modifications in Covariance Structure Analysis: The Problem of Capitalization on Chance," Psychological Bulletin, (111:3), 1992, pp. 490-504.

Mallot, M.J. "An Interview With Keith Davis," Research in Corporate Social Performance and Policy, (15), 1998, pp. 241-250.

Melnyk, S.A., R. Calantone, R. Handfield, R.L. Tummala, G. Vastag, T. Hinds, R. Sroufe, F. Montabon and S. Curkovic. ISO 14000: Assessing Its Impact on Corporate Effectiveness and Efficiency, Center for Advanced Purchasing Studies, Tempe, AZ, 1999.

Min, H. and W.P. Galle. "Green Purchasing Strategies: Trends and Implications," International Journal of Purchasing and Materials Management, (33:3), 1997, pp. 10-17.

Narasimhan, R. and J.R. Carter. Environmental Supply Chain Management, Center for Advanced Purchasing Studies, Tempe, AZ, 1998.

Nielsen, P. and H. Wenzel. "Integration of Environmental Aspects in Product Development: A Stepwise Procedure Based on Quantitative Life Cycle Assessment," Journal of Cleaner Production, (10), 2002, pp. 247-257.

Phillips, L.W. "Assessing Measurement Error in Key Informant Reports: A Methodological Note on Organizational Analysis in Marketing," Journal of Marketing Research, (18:4), 1981, pp. 395-415.

Phillips, L.W. "Explaining Control Losses in Corporate Marketing Channels: An Organizational Analysis," Journal of Marketing Research, (19:4), 1982, pp. 525-549.

Popper, K.R. The Logic of Scientific Discovery, Hutchison, London, 1959.

Steigler, J.H. "Structural Model Evaluation and Modification: An Interval Estimation Approach," Multivariate Behavioral Research, (25), 1990, pp. 173-180.

Tichy, N.M., M.L. Tushman and C. Fombrun. "Social Network Analysis for Organizations," Academy of Management Review, (4:4), 1979, pp. 507-519.

Van de Ven, A.H. and D. Ferry. Measurement and Assessment of Organizations, Wiley, New York, NY, 1978.

Walton, S.V., R.B. Handfield and S.A. Melnyk. "The Green Supply Chain: Integrating Suppliers Into Environmental Management Processes," The Journal of Supply Chain Management, (34:2), 1998, pp. 2-11.

Wartick, S.L. and P.L. Cochran. "The Evolution of the Corporate Social Performance Model," Academy of Management Review, (10:4), 1985, pp. 758-769.

Wokutch, R.E. "An Interview With Lee E. Preston," Research in Corporate Social Performance and Policy, (15), 1998, pp. 227-239.

Wokutch, R.E. Worker Protection, Japanese Style: Occupational Safety and Health in the Auto Industry, ILR Press, Ithica, NY, 1992.

Yin, R.K. Case Study Research, 2nd ed., Sage Publications, Thousand Oaks, CA, 1994.

AUTHOR

Craig R. Carter is assistant professor of supply chain management in the College of Business Administration at the University of Nevada in Reno, Nevada.