princeton university (cid 195003) · according to sober and wilson (1998), Òthe altruism...

Delivered by Ingenta toPrinceton University Library (cid 75000146)Princeton University (cid 35005133)Princeton University Libraries (cid 1762)PRINCETON UNIVERSITY (cid 195003)IP: 128.112.69.222004..07..26..11..58..

Altruism is a long-standing problem in soci-ology. However, surprisingly little attention

has been paid to two important facts: First, thatthe incidence of altruism varies greatly; andsecond, that altruism is structured, promoted, andmade logistically possible by organizations andinstitutions with a strong interest in producingit. These facts are not unrelated. Helping, giv-ing, or caring is systematically elicited frompeople by organizations that are usually theimmediate recipients of individual goodwill or,at least, the necessary brokers for it. Withoutthese organizations, much of the altruism weobserve would not happen. Altruism, in short,is highly institutionalized.

A social-organizational approach to altruismwas set out by Titmuss (1971), but his pioneer-ing effort has seen little subsequent develop-ment. Instead, theoretical interest in altruismhas mainly centered either on its robustness asa heritable trait or on its reality as a psycho-logical motive. In the former case, researchfocuses on whether altruistic behavior canevolve and survive across generations as altru-ists compete with selfish agents. In the lattercase, research focuses on whether individualshave altruistic motives or identities and, if so,how they acquire them. Evidence from bothfields suggests that altruism is a real, and com-mon, phenomenon. The question of why ratesof altruism should vary across different orga-nizational and institutional settings has not beenwell addressed. Either the problem is not imme-diately relevant to the research program (instudies of evolution) or it must be posed interms of identifying the real intentions of indi-vidual agents (in studies of motivation).

This article reframes the question by con-sidering the literature on motivation togetherwith that on voluntary and nonprofit organiza-tions. I consider altruistic practices in terms of

AAllttrruuiissmm aass aann OOrrggaanniizzaattiioonnaall PPrroobblleemm:: TThhee CCaassee ooff OOrrggaann PPrrooccuurreemmeenntt

Kieran HealyUniversity of Arizona and Australian National University, RSSS

This article presents a social-organizational approach to explaining empirical variation

in rates of altruism. The efforts of organizations are mostly responsible for much of the

altruism seen today, and the substance of these efforts varies. Although research from

social psychology and organizational studies suggests that altruistic action is sensitive to

social context, the link between individual and organizational aspects of altruism has not

been clearly articulated. In particular, our knowledge of “one-shot,” organizationally

managed altruism is limited. I suggest that the factors of organizational resources,

scope, and persistence are likely to generate higher rates of individual altruism in the

absence of long-term relationships that encourage giving behavior. The approach is

applied to the case of cadaveric organ procurement in the United States. The analysis

highlights the central role of organ procurement organizations (OPOs). Quantitative

analysis of OPO procurement rates shows that, while demographic characteristics are

important, OPO resources and scope are important predictors of procurement. The

findings strongly suggest that the ability of organizations to produce contexts for giving

explains a substantial amount of variation in rates of one-shot altruism.

AAMMEERRIICCAANN SSOOCCIIOOLLOOGGIICCAALL RREEVVIIEEWW,, 22000044,, VVOOLL.. 6699 ((JJuunnee::338877––440044))

#1600-ASR 69:3 filename:69304-healy

Direct correspondence to Kieran Healy,Department of Sociology, Social Sciences 404,University of Arizona, Tucson, AZ 85721([email protected]). The author thanks PaulDiMaggio, Bruce Western, Viviana Zelizer, MarkChaves, Ron Breiger, Marion Fourcade-Gourinchas,Joe Galaskiewicz, Ted Gerber, audiences atNorthwestern and Arizona, and the ASR reviewers forcomments on earlier versions of this article.


their frequency of occurrence for individuals onthe one hand, and their degree of social organ-ization on the other. Cases where altruism is arare or “one-shot” event from the individualpoint of view yet also possess a substantialdegree of social organization are an importantbut undertheorized type in both literatures. Iidentify organizational features that ought tobe associated with higher rates of altruism insuch cases and apply them to the case of cadav-eric organ procurement in the United States.Organ procurement has not previously beenexamined in detail from this perspective. Aswith research on altruism more generally, theemphasis has been on the motives and experi-ences of individual donors. My general goal isto develop a middle-range approach to deter-mining why rates of altruism vary, highlightingthe logistical efforts of organizations in foster-ing altruism. Results from the motivational lit-erature are put into a comparative context, andtheory on voluntary organizations is extendedto cases where organizations cannot have hadestablished relationships with donors.

IINNDDIIVVIIDDUUAALL AALLTTRRUUIISSMM AANNDD SSOOCCIIAALLOORRGGAANNIIZZAATTIIOONN

DDEEFFIINNIITTIIOONNAALL QQUUEESSTTIIOONNSS

In everyday usage, an altruistic act is one moti-vated by concern or regard for others ratherthan oneself. Simmons (1991) gives a usefuldefinition: “Although scholars’ definitions dif-fer, most would agree that altruism (1) seeks toincrease another’s welfare, not one’s own; (2) isvoluntary; (3) is intentional, meant to help some-one else; and (4) expects no external reward” (p.3). According to Sober and Wilson (1998), “Thealtruism hypothesis maintains that people some-times care about the welfare of others as an endin itself. Altruists have irreducible other-direct-ed ends” (p. 228).

Operational definitions of altruism are the-ory-laden, in that different research programsplace varying degrees of constraint on whatacts should count as truly altruistic. This meansthere can be reasonable disagreement over howto classify some actions or practices. For exam-ple, from an evolutionary perspective, altruismis a behavioral trait. The problem is to say howa behavior that helps others at a cost to oneselfcould have evolved. This is a question of thereproductive fitness of altruistic agents. Sober

and Wilson (1998) note that “evolutionary biol-ogists define altruism entirely in terms of sur-vival and reproduction. A behavior is altruisticwhen it increases the fitness of others anddecreases the fitness of the actor” (p. 17). Froma social-psychological perspective, altruism isa disposition or identity. The problem is whetherpeople truly are altruistic or whether theiractions are covertly selfish in some way—aquestion of purity of motive.

These contrasting approaches are not neces-sarily in conflict with each another—rather,they are concerned with quite different prob-lems. In the evolutionary approach, altruism istreated as (or as if it were) a heritable trait.Research formally analyzes or numerically sim-ulates the reproductive fitness of agents whobehave altruistically towards other agents.Classic results from Hamilton (1964) and Smith(1964) showed that altruism is a successfulstrategy if organisms maximize their inclusivefitness. More ambitiously, Sober and Wilson(1998) argue that multilevel or group-selectionmechanisms may favor the evolution of altru-ism in populations to a greater degree than hasgenerally been accepted by biologists. Trivers(1971) defined the concept of reciprocal altru-ism, later developed in Axelrod and Hamilton(1981), Smith (1982), and Axelrod (1984).1

Other studies show how altruistic behavior canbe transmitted culturally, in ways analogous tothe biological mechanisms of kin- and group-selection (Allison 1992; Boyd and Richerson1985; Macy and Skvoretz 1998; Mark 2002).These sociological applications of the evolu-tionary approach examine problems of socialorder in a general way, with the focus on iden-tifying the survival chances of prosocial behav-ior. The present article is not directly concernedwith these questions; instead, I examine varia-tion in a kind of altruistic practice and take theexistence of altruism per se as given.

A direct conflict between the evolutionaryand motivational approaches arises only where

338888——––AAMMEERRIICCAANN SSOOCCIIOOLLOOGGIICCAALL RREEVVIIEEWW


1 The “tit-for-tat” (cooperate f irst) strategydescribed in Axelrod (1984) is often thought not tobe altruistic, because the cooperator benefits in thelong run. There is some disagreement about this:Sober and Wilson (1998: 84) argue that the distinc-tion is not built into a formal model of these process-es, but lies instead in their interpretation.


theory demands that ostensibly well-intentionedactions must be selfishness in disguise. Forexample, homo economicus is both rational andwholly self-regarding, so costly altruism isimpossible ex hypothesi. Apparently altruisticacts must confer some kind of benefit, eitherdirectly or simply in the form of a “warm glow.”This view of altruism, it should be noted, aris-es out of the constraints of a particular modelof action and not simple facts about the world.More sophisticated rational-choice models neednot force this point, and have room for a morerecognizable concept of altruism. As Schmidtz(1993) argues, reflectively rational agents (thatis, agents who think about the effects of theirchoices on their preferences) may have reasonsto cultivate an altruistic regard for others in theordinary sense. Having a reason for doing some-thing for someone—even a reason from one’sown point of view—does not disqualify the actfrom being altruistic.

Rather than examine the altruistic authentic-ity of particular actions, I present a middle-range framework for understanding empiricalvariation in kinds of altruistic practices in away that avoids strong assumptions about themotives of individual actors. I begin from thesocial-psychological conception of altruism asintentional, voluntary, and unrewarded actionoriented toward the welfare of others. But myargument here is not in conflict with the generalconcerns of evolutionary approaches to proso-cial behavior, and—apart from suggesting thata narrowly rational-choice view may confuse theissue—it is agnostic on the question of the prop-er model of individual agency.

KKIINNDDSS OOFF AALLTTRRUUIISSTTIICC PPRRAACCTTIICCEE

To provide a framework for the subject, it maybe helpful to juxtapose findings from the socialpsychology of altruism on the one hand, andresearch on voluntary organizations on the other.The two areas of study are related, but theyhave generally progressed without much regardfor each other. Bringing them together allows usto think about variation in altruism in a new way.Consider the two-dimensional space shown inFigure 1, defined on one axis by the degree towhich practices are routine for individuals, andon the other by the degree to which they areinstitutionalized by formal organizations. Thefirst dimension represents the frequency of

altruistic actions from the point of view of indi-viduals. Practices range from rare, one-shotexchanges to common or routine occurrences.The second dimension captures variability insocial organization. Practices here range fromthose that are diffusely or informally regulatedto those that are strongly institutionalized andformally managed. We can place acts of altru-ism within this framework.

Some acts of altruism are neither routine forindividuals nor well institutionalized. Muchexperimental work on motivated altruism isconcerned with situations like this, where peo-ple have a sudden opportunity to act in one-shotevents detached from any strong organization-al context. “Bystander intervention” experi-ments test how subjects react to an unexpectedchance to help a stranger who has had an acci-dent, been the victim of a crime, or is otherwisein need of immediate assistance (Latane andDarley 1970; Austin 1979; Krebs and Miller1985). Studies of help in the context of a rela-tionship are largely absent from this type ofstudy (Simmons 1991). Also absent is a stableorganizational context for action; the aim is totest people’s reactions when they must make aquick decision whether to help someone.

Other practices, such as simple acts of kind-ness or consideration, may be habitual for manyindividuals or a common part of their lives, butthey are not managed by formal organizations.Social theorists have pointed to a general normof reciprocity that grounds actions of this sort(Gouldner 1960; Granovetter 1985: 489–90).Research has examined the possibility thatempathy is a general human trait underlyingthese sorts of actions (Davis 1996; Batson 1987;Hoffman 1981). A related line of research doc-uments actions that cannot reasonably be calledself-interested (Oliner and Oliner 1988; Monroe1998). This approach emphasizes extraordinarycases of altruism rather than the everyday vari-ety. (These studies have come about partly inresponse to the rational-choice critique men-tioned earlier: the researchers are interested infinding cases of altruism that cannot reasonablybe construed as selfishly motivated.)

Social organization is minimal in both ofthese forms of altruism, extending at most to theconventional or routine activities of individuals.But most of the altruism we observe in modernsocieties is more likely than not to have a strong-ly institutionalized aspect, with staffed organi-

OORRGGAANN PPRROOCCUURREEMMEENNTT——––338899



zations working to produce contexts in which itcan happen.

Research on voluntary organizations exam-ines part of this social-organizational dimension.In the United States, many nonprofits operatein environments subject to considerable uncer-tainty over sources of funding, whether gov-ernmental or private (DiMaggio and Anheier1990; Grønbjerg 1993). Much recent researchhas analyzed the structuration of organization-al populations in light of these constraints.Organizations have reacted strategically in var-ious ways to these circumstances (Clarke andEstes 1992; Oliver 1991); they may, for exam-ple, co-opt competitors (Galaskiewicz andBielefeld 1998), diversify into less competitiveareas (Alexander 1998) and attempt to differ-entiate themselves from competitors (Barman

2002). Although studies at the organizationallevel have tended to focus on questions of sur-vival and competition (usually for grant money),some have examined how routine altruistic prac-tices are managed by organizations. This is thethird case shown in Figure 1, where both thedegree of social organization and the frequen-cy of individual practice are relatively high.Studies of rates of charitable giving find thathaving effective “communities of participation”is the best way to channel social norms, priordispositions, or available resources into actualdonations (Schervish and Havens 1997).Similarly, ethnographic evidence shows thatcharitable acts and volunteering are commonwhen they are embedded in the social structureof a community (Eckstein 2001), but at thesame time effective organizations efficiently



Figure 1. Kinds of Altruistic Practice, by Individual Frequency and Degree of Social Organization.


recruit and carefully manage donors (Ostrander1995). Frumkin (2002) notes that “[w]hile vol-unteers remain an important engine drivingnonprofits, most nonprofits use professionalsto manage volunteers, rather than using vol-unteers to manage their organizations” (p.102).

Theory and research from the social-psy-chological and organizational literature, then,has focused on these three kinds of altruisticpractice: one-shot opportunities with no orga-nizational context; conventional or routineactions, also with little in the way of formalsocial organization; and repeated giving man-aged by organizations. The fourth variety sug-gested by our typology—one-shotopportunities for altruism in socially organizedcontexts—has received comparatively littleattention. This is not to say that the logisticalaspects of altruism have gone wholly unno-ticed: research does find that social-organiza-tional context directly affects the likelihood ofaltruism. In particular, the importance of directrequests to potential donors or helpers is wellestablished. Oliner and Oliner (1988) foundthat rescuers of Jews were more likely to havebeen directly asked for help than nonrescuers.Drake, Finkelstein, and Sapolsky (1982) foundthat people were more likely to give blood ifthey were asked directly (see also Piliavin andCallero 1991). The same is true for donationsof money to charity (Clotfelter 1993;Hodgkinson and Weitzman 1992; Jencks 1992;Schervish and Havens 1997).

Despite the recognition that direct requestsincrease the likelihood of altruism, however,the relationship between organizationally pro-duced contexts for giving and variation in ratesof individual altruism remains underexplored,particularly in the case of one-shot events. Thelink between the logistical effort of organiza-tions and the action of individuals is general-ly missing in studies of bystander intervention(where there is no organization) as well as inthe literature on voluntary organizations (wherethe focus is on establishing patterns of giving,or more often on field-level processes of orga-nizational competition). Thus we f ind anempirical and a theoretical gap in the literature.Empirically, we know little about altruisticpractices that are one-shot experiences forindividuals and are also socially managed byorganizations. Theoretically, we should iden-

tify the particular features of the social contextor organizational environment that mightexplain variation in such altruistic acts.

Learning more about one-shot, socially man-aged exchanges is important. Recent studies ofaltruistic practices such as blood donation(Healy 2000), together with those on cognateactivities such as voluntary association(Salamon and Anheier 1994) and civic partic-ipation (Skocpol, Ganz, and Munson 2000),develop the general view that “involvement involunteer activities does not simply springfrom already constituted social groups or fromaggregated individual characteristics” but isstructured by institutions (Schofer andFourcade-Gourinchas 2001, 807). This argu-ment is counter to the view that such prac-tices are to be explained by individual-levelcharacteristics or value orientations (Almondand Verba 1963; Inglehart 1997; Knoke 1986).One-shot exchanges would seem to be the mostlikely case where this argument would notapply—where either the distinctive character-istics of individual altruists or the motivatingforce of generalized norms would be mostimportant. But this need be true only if one-shot exchanges must by definition lack aninstitutional context. This is a plausibleassumption, as the degree of repeated or rou-tine action is usually thought of as the mech-anism of institutionalization rather than aspotentially orthogonal to it. By separating thefrequency of altruistic practice at the individ-ual level from the degree to which contexts forgiving are socially organized, however, we gainleverage on the question of why rates of dif-ferent kinds of altruism vary. Empiricallyestablishing the organizational basis of one-shot altruistic exchanges thus serves two pur-poses. First, i t extends theory on theorganizational and institutional sources of vol-unteering and participation to a challengingcase. Second, it connects the problem ofexplaining individual altruism, which has beendominated by social-psychological explana-tions, to wider debates in economic and polit-ical sociology about the institutional sourcesof individual actions and identities.

LLOOGGIISSTTIICCAALL AASSPPEECCTTSS OOFF AALLTTRRUUIISSMM

From a social-organizational perspective, one-shot altruistic acts are distinguished by the logis-




tical problems that organizations must solve inorder to create a viable context in which giv-ing can happen. Variation in these efforts oughtto generate variation in rates of altruism, inde-pendent of the characteristics of the individualdonors. Just as a sale made by someone in amarket is necessarily also a purchase made bysomeone else, donation is also necessarily pro-curement. Organizations attempt to routinizeand maximize the production of procurementopportunities. Even though these may be one-shot experiences from the individual point ofview, they can be well institutionalized fromthe procuring organization’s perspective.

Although the literature on charitable givingsuggests that asking directly makes it morelikely that individuals will donate, it offers lit-tle guidance as to what organizational featureswill matter in this regard. I suggest three kindsof logistical effectiveness: resources, scope,and persistence. Resources should be posi-tively associated with higher rates of procure-ment because larger, better-funded, orbetter-staffed organizations will find it easierto generate opportunities to give. Scope can bethought of as the reach, or spread, of a procur-ing organization across the range of potentialsites where donors present themselves.Organizations with broader scope will be pres-ent in more places where potential donors arelikely to be found. Resources and scope arerelated but distinct metrics, as an organiza-tion may be large or well funded but narrow infocus, or vice versa. Finally, persistence meas-ures the degree to which an organization willpursue a potential donor once the opportunityis discovered. More persistent organizationsshould be more successful at procuring donors.

I test these hypotheses using the case ofcadaveric organ procurement in the UnitedStates. This case is useful and important for anumber of reasons. The decision to donate isa one-shot exchange from the individual’s pointof view. The organizations that manage theprocess are not in direct competition with oneanother for funds or donors, so complicatingfield-level forces are not directly relevant. Thelogistical forces at work are thus clearer thanin cases where giving is bound up with well-established social relations. In addition, organdonation is a canonical example of individualaltruism: the idealized individual organ donoris a strong trope in public discourse and a

stock example in social theory. A social-orga-nizational treatment of organ donation extendsour general understanding of the sources ofvariation in altruism and contributes to ourknowledge of an increasingly common med-ical therapy.

TTHHEE CCAASSEE OOFF OORRGGAANN PPRROOCCUURREEMMEENNTT

Most organs transplanted in the United Statescome from cadaveric donors.2 A report from theGeneral Accounting Office (1993) describesthe process:

Organ donation is dependent on voluntarism andgenerosity as well as solicitation and decision-making at a time when family members are underthe stress of bereavement. Typically this processbegins at a hospital when a patient is identified asa potential organ donor. .|.|. Once a potential organdonor has been identified, the patient’s family iscontacted by a staff member of either the hospitalor the OPO [organ procurement organization] andthe family is given the opportunity to donate thedeceased’s organs. If the family consents to dona-tion, OPO staff coordinate the remainder of theprocurement activities, including recovering andpreserving the organs and arranging for their trans-port. (pp. 17–18)

Note that it is the next of kin who makes thedecision whether to donate the person’s organs.Cadaveric organ donors have usually died sud-denly, either from a serious nonaccidentalinjury (such as a brain hemorrhage or stroke)or from injuries sustained in a violent accident(such as a motor vehicle accident or gunshotwound). Whether the deceased carried an organdonor card may affect the family’s decision todonate. But in general OPO staff defer to thechoice of the next of kin, who decides whetherto give.3

Donating the organs of a recently deceasedrelative is a difficult choice. The decision is not



2 Kidneys are the main exception. An increasingnumber is obtained from living, related donors. Notethat I do not address the question of commercialtrafficking in organs here. On this topic see, forexample, Scheper-Hughes and Waquant (2002),Scheper-Hughes (2000), and Cohen (1999).

3 The extent to which OPOs actively draw atten-tion to the wishes of the deceased is an empiricalquestion, discussed below with respect to variationin OPO procurement policies.


straightforward. Donor families do not knowwho will benefit from their choice; the vastmajority never learn more than the gender,age, and approximate geographical locationof the recipient. In addition, those deciding todonate receive no reward. In short, the decisionto donate satisfies our definition of altruism inthat it is an intentional, voluntary, unreward-ed action that seeks to increase the welfare ofothers (strangers, in this case).

Most research on organ donation, like the lit-erature on altruism generally, has focused onquestions of individual motivation and cost.The focus has been on discovering the sourcesof opposition to donation and ways of over-coming such opposition. Refusals have beenfound to be due to bad information or mythsabout the donation process (GallupOrganization 1993). Irrational beliefs and fearsthat many people have about organ donationmay be responsible for keeping the procure-ment rate down (Prottas 1994). Thus, for thoseworking with a concept of altruism as moti-vated action, the issue is treated as an indi-vidual-level problem whose solution lies inunderstanding (and perhaps adjusting) themotives of potential donors. A further parallelwith research on motivated altruism in gener-al is the discovery (and rediscovery) of theimportance of situational factors to donation.Simmons, Marine, and Simmons (1977) foundthat kidney donors were more likely to havebeen asked in person to give than nondonors.Zimmerman et al. (2000) found that female rel-atives were more likely to donate a kidneythan male relatives, possibly because womenwere more likely than men to be asked todonate. In the case of cadaveric donors, theimportance of the request process and the needto take care in asking relatives to donate theirnext of kin’s organs have been recognized forsome time (Verble and Worth 2000; Ehrle,Shafer, and Nelson 1999; Gortmaker 1998).There are few studies, however, of the logisti-cal role of OPOs in the process of organ pro-curement, outside of research on the momentwhen consent is requested.4

TTHHEE SSOOCCIIAALL OORRGGAANNIIZZAATTIIOONN OOFF OORRGGAANN

PPRROOCCUURREEMMEENNTT

Some background on the procurement systemis necessary. Through the 1960s and 1970s, itwas taken for granted that human organsbelonged to the surgeons who had removedthem from donors, and these surgeons decidedwho received organs for transplant on the basisof whatever ethical and clinical criteria theysaw fit to apply (Prottas 1994). The federal gov-ernment overhauled the system in 1984 when itpassed the National Organ Transplant Act.Under this law, human organs must be given asgifts in the United States. Organs are consideredto be a public good belonging to the state; theycannot be sold. The United Network for OrganSharing (UNOS) oversees the activities ofOPOs, who help make up its membership.Besides the OPOs, the other two main compo-nents of the transplant system (and also mem-bers of UNOS) are about 60 independenthistocompatibility laboratories and about 270hospital transplant centers. The labs carry outthe tests that allow compatible organs andpatients to be matched up; the transplant cen-ters, of course, are where the surgery actuallytakes place.

Today, a national network of organizationslocates as many potential donors as possible,secures voluntary consent from the next of kin,and distributes the procured organs to patientson waiting lists in eleven regions across thecountry. In 1997, there were sixty-one OPOsoperating in the United States. Each one is ableto procure, store, and deliver organs to transplantcenters, where patients register on waiting lists.Each one is responsible for procurement with-in a particular area. OPO borders are drawn atthe county level and may cross state lines. Inaddition, a few OPOs administer noncontiguousareas. Figure 2 shows the OPO boundaries in theUnited States.

OOPPOO RREESSOOUURRCCEESS,, SSCCOOPPEE,, AANNDD PPEERRSSIISSTTEENNCCEE

How might variation in the donor procurementrate of OPOs be explained? Successful pro-curement depends in part on logistical effort—that is, the process of locating potential donors,creating opportunities to give, and securingconsent. I have argued that three organization-al features help determine the success of theseefforts: resources, scope, and persistence. The



4 But see Klassen et al. 1999, Siminoff and Nelson1999, and Wendler and Dickert 2001 for importantexceptions.


first of these features for OPOs is suggested bytheir administrative spending per annum. Whenmeasured relative to the number of potentialdonors, OPO spending indexes the resourcesthat the organization brings to the procurementprocess. Donor procurement is a resource-inten-sive procedure requiring substantial coordina-tion. The OPO (or its agents) must identifypotential donors, determine that brain death hasoccurred, contact and obtain consent from thenext of kin, and then recover and preserveobtainable organs. Time is of the essence in thisprocess. Resource-rich OPOs, on average,would be expected to be more successful indonor procurement than those with fewerresources.

OPOs are not the only players in the pro-curement process. Almost all deaths that lead toprocurement occur in hospitals. The scope of theorganization across hospitals is therefore ofsome importance; Klassen et al. (1999) found

that hospitals play a key role in the procurementprocess. In some cases, the OPO has a staffmember working at the hospital. More often, amember of the hospital staff is responsible forcontacting the OPO to let the organization knowthat a potential donor has become available.OPOs may make referral agreements with hos-pitals. The more hospitals an OPO has referralagreements with, the greater its scope and themore donors it is likely to procure.

OPOs also vary in their procurement policiesand consent practices. Donor families eithergive or refuse consent for donation (Klassenand Klassen 1996), and it may be assumed thatsuccessful organ procurement is related to strate-gies adopted by OPOs. Families rarely suggestdonation on their own; Powner and Darby(1999) found that “educational interventionsfor health care professionals and a coordinatedrequesting process that includes the organ pro-curement organization and hospital personnel



Figure 2. OPO Boundaries in 1997.

Note: Shaded areas mark the catchment areas of different organ procurement organizations (OPOs), which areidentified by number.

10

37

33

2511

39

52

43

20

15

12

13

16

63


result in a higher number of donations” (p.1225). Requesting procedures may include thepresence of racial or ethnic minority staff todiscuss procurement with minority families,the presence of training programs, and theinvolvement of medical social workers and cler-gy in the procurement process (Siminoff et al.1995). The consent process is a delicate one, andOPOs differ in their policies about asking fam-ilies or next of kin to donate. If some OPOs aremore willing than others to, for instance, men-tion the wishes of the deceased, this may affectthe outcome. Wendler and Dickert (2001) foundsubstantial variation in consent policies. Consentpolicy may be taken as a measure of organiza-tional persistence. The expectation is that thestronger the stated willingness to procure underadverse circumstances, the higher the procure-ment rate.

DDEEMMOOGGRRAAPPHHYY OOFF PPRROOCCUURREEMMEENNTT

Successful procurement is not simply a functionof organizational factors, of course. Becausethey administer geographical areas with wide-ly varying populations, OPOs’ procurementrates are affected by features of their catchmentarea. Some populations are more likely thanothers to yield potential donors regardless of thesocial organization of the procurement system.It is important to control for these backgroundforces. The research literature on organ dona-tion and donor motivation points toward a num-ber of important demographic variables.Specifically, population density, racial compo-sition, the poverty rate, and the degree of edu-cational attainment within an OPO’s catchmentarea should all affect the number of potentialdonors.

POPULATION DENSITY. Some OPOs adminis-ter relatively small, densely populated regions.Others service much larger, more sparsely pop-ulated areas. Because concentrated populationsare easier to manage and are more likely to pro-vide opportunities for donation the procure-ment rate should rise with population density.

RACE. Support for organ donation is knownto vary by race; African-Americans are lesslikely to donate than whites (Ehrle et al. 1999).It is unclear whether this is due mainly to beliefs

that the transplant system is unfair to minorities(Kasiske et al. 1991), a more general distrust ofthe medical system (Randall 1996), or ineffec-tive methods of request on the part of OPOs(Gortmaker 1996). Recent studies have dis-covered evidence that African-Americans faceinstitutionalized barriers to organ transplanta-tion. African-American patients are much lesslikely than whites to express an interest in, belisted for, or receive a transplant (Alexanderand Seghal 1998; Kasiske et al. 1991). Thissuggests that lower rates of donation and trans-plantation among African-Americans are notsimply a matter of irrational individual belief,but may reflect a more structural exclusion fromthe medical system. The expectation is that thehigher the black population within an OPO’sservice area, the lower the procurement rate.

POVERTY. The logistical demands of donorprocurement begin at the point when a personbecomes a potential donor. Once the initialevent happens—a car accident or a shooting, forexample—the victim must be found, quicklybrought to a hospital, and placed on a ventila-tor. The longer the gap between initial injury andsubsequent hospitalization, the less likely it isthat procurement will be successful. OPOs thatserve wealthier counties are more likely to havethe necessary resources and facilities availableto them (and the hospitals they work with) tosuccessfully manage this task. OPOs servingpoorer counties will tend to procure fewer of thepotential donors that become available to them.

EDUCATION. Survey data show a higher levelof support for organ donation among more edu-cated people (Gallup Organization 1993;Southeastern Institute of Research 1994),though there is no direct evidence that this trans-lates into consent to procure when the occa-sion arises. There could be a sizable gap betweenabstract support for organ donation and theactual decision as next of kin to allow procure-ment to go ahead. But we should still expect thatOPOs serving more educated populations to dobetter on average than others.

UNMEASURED FACTORS. A number of otherfeatures of OPO catchment areas might have aneffect on the procurement rate; their effect is notestimated in my analysis.




First, the age distribution of the populationmight be a factor. In the early days of organtransplantation, surgeons were reluctant to usethe organs of older potential donors. An OPOwith a disproportionately older population mighttherefore be expected to procure fewer donors.As surgical techniques improved and the organshortage worsened, however, transplant teamshave drawn on as much of the available pool aspossible. In 1997, 19.7 percent of cadaverdonors were aged 17 years or less; 52.3 percentwere aged 18–49; and 28 percent were aged 50or older. In exploratory analyses, the proportionof the population aged between 5 and 60 was notassociated with the procurement rate (r = –0.14).

Second, religious participation and affilia-tion might also affect one’s likelihood of sup-porting donation. Existing research on religionhas focused on the theological or traditionalobjections some religions have or have had withorgan donation (Twersky, Gold, and Jacob 1991;Kelly and Wiest 1991). There has been noresearch on whether the geography of religiousaffiliation drives variation in donation rates.The literature on religious attitudes toward organdonation might suggest that areas with Jewish,Conservative Catholic, and some Protestantdenominations will have lower donation rates.I used county-level data on religious affiliationand activity to examine whether there was alink between religion and procurement. Thereligion data were obtained from the AmericanReligion Data Archive (Bradley et al. 1992).Exploratory analyses did not show any strongeffects. This is consistent with available opin-ion data. A 1993 survey (Gallup Organization1993, section 6) found that “[r]eligious barriersto organ donation do not appear to be wide-spread.” Such religious objections as there werecame mainly from Black and Hispanic respon-dents, who were “much more likely to report thatorgan donation is against their religion (14 per-cent and 13 percent agree, respectively) than arewhite respondents (4 percent agree).”

Third, states have implemented several kindsof laws that may affect organ procurement rates.Soon after the passing of the National OrganTransplantation Act, states began to provide theorgan donor cards on the back of drivers’ licens-es. In the late 1980s, “required request” lawswere introduced in most states. This legislationrequires hospitals to consult with the potentialdonor’s next of kin should the patient be near

death. These developments are generally agreednot to have increased donation rates (Norris1990), and a number of studies have shown thatabout a quarter of the time, eligible familiesare not offered the option to donate (Gortmaker1996). Several states have passed “routine noti-fication” laws, which require that all deaths,actual or imminent, be referred to the localOPO. There is evidence that this has increasedprocurement rates in some areas (Nathan 1999;Shafer et al. 1998). Because their diffusion wasso rapid, the effects of these laws are notassessed here.

Few studies have analyzed variation in OPOprocurement rates. None is entirely satisfacto-ry. Prottas (1989) noted the importance of theorganization of procurement and suggested anumber of different measures of OPO effec-tiveness. Evans, Orians, and Ascher (1992) triedto estimate the number of potential donors butdid not analyze procurement data. Siminoff andNelson (1999) studied the efficiency of OPOsbut confined themselves to a particular UNOSregion. A study by Ozcan, Begun, andMcKinney (1999) focused on organizationalmeasures of efficiency but did not control forany structural variables. The analysis presentedhere brings together measures of structural andorganizational forces affecting organ procure-ment and tests their importance using a goodmeasure of the procurement rate across thewhole population of procurement organizations.

DDAATTAA AANNDD MMEETTHHOODDSS

To make a sensible comparison between OPOswe must first estimate a standardized procure-ment rate. To do so for any OPO in a givenyear, we need to know the number of donors itactually procured and the number of cadaversthat it could have procured—that is, the truenumber of potential donors. The former figureis known with certainty, as UNOS tracks allthe organ donors in the country. The latter fig-ure must be estimated.

Following a study by the General AccountingOffice (1997), I have assumed that the best esti-mate of the potential donor pool is the in-hos-pital death rate adjusted for circumstance orcause of death. Because organs suitable fortransplantation must in general be undamagedand undiseased and quickly obtained after death,many causes or circumstances of death rule out




the possibility of donation. This is not quite aperfect measure, because the classification sys-tem5 does not always give enough informationto say with certainty whether a particular patientwas a donor candidate or not. Nevertheless, thisadjusted death rate is the best available denom-inator for calculating the procurement rate.

The CDC-WONDER database (maintainedby the Centers for Disease Control andPrevention) provided counts of all deaths bycounty for 1997. These data were filtered usingthe criteria reported in General AccountingOffice (1997) to yield an estimate of “DonorEvaluable Deaths.” (The rate calculated withthis number is sometimes called the “DonorExtraction Rate,” or DER.) The dependent vari-able in the analysis is therefore the absolutenumber of donors procured by an OPO in 1997divided by the number of evaluable deaths andmultiplied by a thousand. This county-levelmeasure of evaluable deaths was summed for allcounties administered by each OPO, thus aggre-gating observations for 3,142 counties to 61OPOs.

This measure of the dependent variable con-trols for selection into the pool of potentialdonors. For example, motor vehicle accidentsaccounted for 26 percent of organ donors in1997. The death rate from road accidents var-ied from 4 to 18 cases per 100,000 people in thesame year. Thus, OPOs that operate in areaswith high rates of road deaths can expect tohave a higher procurement rate. A measure ofroad accident fatalities is not included in thepresent analysis, however, because its effect isalready controlled for in the denominator ofthe dependent variable.

Measures of population density (per squarekilometer), percentage of black population (in1996), percentage of poor (in 1993), and per-centage of population with a college degree (in1996) are available by county from the CensusBureau. They were aggregated to the OPO levelin the same way as the dependent variable.County-level population statistics were summedto the OPO level in order to calculate the rates.The original data are published in the CensusBureau’s U.S. County Data for 1998.

Information about the resources of OPOswas compiled from Medicare reimbursementreports for 1997, which contain data on thespending, staff, and cost structures of the organ-ization together with information on the costsof organ procurement. Procurement costs—including fees paid to hospitals, surgical teams,tissue-typing laboratories, and so on—are cal-culated by OPOs and, in the case of kidneys, arereimbursed by the federal government. Thesedata were not used as a measure of the resourcebase, because these costs are an accountingmeasure that depends directly on the number ofdonors procured. To avoid a spurious correlationof this sort, spending was calculated from theTotal Administrative Expenses worksheet ofthese reports and is expressed here in hundredsof dollars per capita. Administrative spendingincludes the salaries of the OPO staff and thecost of data processing and accounting, travel,employee professional education, public rela-tions, and official vehicles, as well as othermiscellaneous expenses. It is neither a balance-sheet measure of the direct cost of acquiringdonors nor an index of the reimbursementsassociated with successful kidney procurementand therefore is not simply a function of the pro-curement rate.

Data on Referring Hospitals come fromAppendix 5 of GAO report hrd-93-56 (U.S.General Accounting Office 1993). These figuresare for 1991–92. Some OPOs changed theirname or merged with others between then and1997. Information from UNOS was used to rec-oncile the two lists. Because different hospitalstreat different sorts of patients and not allpatients are equally likely to become potentialdonors, I count the number of referrers perthousand in-hospital evaluable deaths.

Data on OPO procurement policies were col-lected by Wendler and Dickert (2001).6 Theyconducted a telephone survey of OPOs.Although they did not observe organizationalpractices directly, the respondent was chosen bythe executive director of each OPO as the per-son most familiar with the organization’s con-sent practices.7 The questionnaire presented



6 I am grateful to the authors for kindly making thisdata available to me.

7 Wendler and Dickert (2001) note, “A total of 26(43 percent) respondents were the OPOs’ executivedirectors, 19 (31 percent) were procurement or organ

5 The International Classification of Diseases, 9thRevision— Clinical Modification, or ICD-9CM code.


cases where procurement might or might not bepursued by the OPO, varying the strength ofboth the potential donor’s and the donor family’ssupport for donation. Respondents used a five-point scale to say how likely their OPO wouldbe to procure the organs. I combined theseresponses into a measure of the overall strengthof the policy. The higher the score on this vari-able, the more likely the OPO is to express awillingness to procure across a wide range ofadverse circumstances (that is, in the face ofopposition from either the donor or the next ofkin).

Table 1 provides descriptive statistics foreach variable.

The data set includes all of the OPOs in thecontinental United States as of 1997. The rela-tively small number of observations means thatregression estimates might be sensitive to out-liers. A number of observations are missing forthe spending and referral measures, whichreduces the valid N in the analysis. The modelsthat follow take these issues into account in twoways. First, I report the results of models usinga robust MM-estimator. This method yields amore conservative estimate of the effects thanregular OLS. These models were estimatedusing the rlm function in R (Venables and Ripley2002, 161–63; R Core Development Team2003). The results are substantively the same asan OLS regression.

Estimating the robust model gives valid N of44 rather than 61, due to missing data. Giventhat the deleted cases are missing observationson only one or two variables and the number ofcases is small, it is preferable to make use of allof the available data to estimate a model ratherthan delete 17 cases. I used multiple imputationto predict values for the missing data. The algo-rithm applied here uses additive regression andpredictive mean matching to impute missingvalues. This process is repeated multiple timesusing bootstrap resampling. Bootstrapping theimputation process generates a large sample of“new” data sets with imputed values. Regression

coefficients are then calculated by fitting themodel to the multiply imputed data sets andaveraging the results. Variance and covarianceestimates are weighted to account for the factthat the model is estimated from partly imput-ed data. The multiple imputation and modelestimation procedures are described in moredetail in Harrell (2001, 47–50, 69–70) andimplemented in his Hmisc library for R.

RREESSUULLTTSS

Regression results are shown in Table 2. Model1 shows a robust regression of the procurementrate on the structural and demographic vari-ables. Population density has a significant pos-itive effect on the procurement rate. The percentof an OPO’s catchment area that is black, poor,or college-educated has a negative effect onprocurement; all three of these factors are sig-nificant at conventional levels. Together, thesevariables explain just over 30 percent of theobserved variation in the procurement rate.

Model 2 shows the same variables as model1, with the organizational measures added. Thestructural and demographic controls generallyretain their sign and magnitude. Population den-sity and racial composition significantly affectthe procurement rate. A 10 percent increase inpopulation density per square kilometer increas-es the procurement rate by just over 0.6 of apoint. A percentage point increase in the num-ber of African-Americans in an OPO area isassociated with a statistically significant dropin the procurement rate of just over a quarter-point. The poverty rate is still negatively asso-



Table 1. Means and Standard Deviations of Variables

Variable Mean SD

Procurement rate 040.11 012.35Population density (km2) 186.19 571.49Percent black 009.65 010.78Percent poor 015.69 004.69Percent college 015.23 004.10OPO spending 029.29 019.50Referring hospitals 038.03 019.85OPO policy 034.43 006.11

Note: Procurement rate, population density, organ pro-curement organization (OPO) spending, and referringhospitals are standardized per million population or perthousand evaluable deaths, as appropriate.

recovery coordinators, 11 (18 percent) were directorsof procurement, and 5 (8 percent) were chief execu-tive officers. A total of 51 (84 percent) had beenemployed by their current OPO for 3 or more years,and 41 (67 percent) had been employed by their cur-rent OPO for more than 6 years” (p. 331).

Delivered by Ingenta toPrinceton University Library (cid 75000146)Princeton University (cid 35005133)Princeton University Libraries (cid 1762)PRINCETON UNIVERSITY (cid 195003)IP: 128.112.69.222004..07..26..11..58..ciated with procurement, but is no longer sig-

nificant at p < 0.05.The percentage of the population that is col-

lege educated retains its significantly negativeeffect. This finding is contrary to expectations.Survey studies report substantially strongersupport for organ donation among more edu-cated people, and education tends to predictcharitable giving and other kinds of altruism.There is, however, no strong evidence that thebehavior of more educated people followsdirectly from their attitude toward donation.Further, outcomes are measured here at theOPO level rather than the individual level, sowe cannot be confident that individual prefer-ences explain this effect. A mechanism con-sistent with the argument that the organizationalbasis of procurement is very important is thatbetter-educated people are less likely to die incircumstances conducive to procurement (suchas motorcycle accidents and gunshot or stabwounds). Consequently, areas with more-edu-cated people might have relatively fewerprospective donors among all deaths. Recall,however, that regional variation in cause ofdeath is already accounted for in the construc-tion of the dependent variable, so the negativeeffect of education seems to remain to be fully

explained.8 One possibility is that the effect isthe result of outlying observations. A sensitiv-ity analysis suggested that the “percent col-lege” measure was the variable most sensitiveto the presence of a small number of cases inthe data, but its effect was always negative.

The three organizational measures show inter-esting patterns. Taken together, they raise the R2

of the model from 33 to 53 percent. OPOresources are positively and significantly relat-ed to the procurement rate. A 10 percent increasein spending raises the procurement rate by near-ly 0.9 points. Organizational scope also has astrong effect on procurement. A 5-point increasein referrer density raises the procurement rateby about 0.75 points. These results stronglysuggest the importance of the OPOs’ logisticalcapability in raising the procurement rate.

In contrast, the measure of persistence—OPO policy—is not significant, and even theweak effect is not in the expected direction.Two interpretations may be suggested. First, it



Table 2. Models Predicting Donor Procurement Rates

Model

Variable .01 .02 .03

(Intercept) 52.907** 3.109 1.331(5.10) (.19) (.08)

Population density (log) 5.174** 6.574** 6.319**(3.14) (4.27) (4.39)

Percent black –.325* –.298* –.333*(–2.24) (–2.20) (–2.57)

Percent poor –.937* –.616 –.582(–2.53) (–1.65) (–1.77)

Percent college –1.134* –.957* –.840*(–2.51) (–2.30) (–2.12)

OPO spending (log) 9.359** 9.243**–3.04 –2.96

Referrers .149* .183**–2.31 –2.88

OPO policy –.040 –.011(–.20) (–.05)

Adjusted R2 .336 .534 .443Valid N .61 .44 .61

Notes: Models 1 and 2 employ a resistant MM-estimator. Model 3 employs multiple imputation. See text fordiscussion. T-values are in parentheses below coefficients.* p < .05; **p < .01 (two-tailed test)

8 In an otherwise identical model with donors permillion population as the dependent variable, theeducation effect is stronger and more significantlynegative than those in Table 2.


may be that the measure used here is not pick-ing up the underlying variations in organiza-tional practice. The answers the OPOs gavewhen asked about policy may not accuratelyreflect the way they operate. But, second, sincethose designing the survey instrument behindthis variable took considerable care to ensureaccurate responses, the consent policies ofOPOs may not in fact have a strong effect on theprocurement rate.

Model 3 shows the results when all the avail-able data are used and missing values are imput-ed. The pattern of results is broadly the same asthat in model 2, with all of the variables retain-ing their magnitude and significance.

DDIISSCCUUSSSSIIOONN

The organizational and institutional basis ofvariation in altruistic practices is a theoretical-ly important but underexplored area. The pro-duction of altruistic action can be thought of asa resource extraction problem for organizations,a problem that they will solve more or lesseffectively. Thus, individuals’capacity for altru-ism and the social organization of procurementare not separate questions, but rather two aspectsof the same process. As organizations createcontexts for giving they generate altruistic actiondifferentially across populations. Rather thansimply drawing from sui generis donor popu-lations, they help create them. Previous theoryand research suggested that social context wasimportant, but did not address just which orga-nizational features mattered in effectively gen-erating donation.

The decision to donate the organs of one’snext of kin can rightly be seen as an individualchoice made in terrible circumstances. But theopportunity to make that choice is created by anetwork of organizations whose job it is to findcandidates for organ procurement, elicit analtruistic action from the next of kin, and get theorgans to wherever the allocation system saysthey need to go. These organizations do their jobmore or less effectively, in more or less favor-able circumstances. Investment in the logisticsof procurement has a strong influence on thenumber of donors procured. My study showsthat even essentially one-shot exchanges are, toa significant degree, organizational accom-plishments. There is also some evidence that, forthis kind of altruism, the role of individual char-

acteristics (such as education) differs fromlonger-term patterns of giving, though thisremains an open question given the data.

The social organization of altruism is ofcourse a larger question than the particular caseof organ procurement. It is a long-standing,though muted, theme in social theory. The soci-ological analysis of altruism is dominated by theidea that it is a scarce resource unevenly dis-tributed across individuals. Yet this misses adistinctive aspect of the problem. Titmuss (1971)argued that “[t]he ways in which society organ-izes and structures its social institutions .|.|. canencourage or discourage the altruistic in man”(p. 225). Following his lead, Singer (1973) sug-gests that we can think of a quality like altru-ism as being a capacity or skill that becomesmore available with regular use. Hirschman(1992) qualifies this idea, arguing that suchqualities “exhibit a complex, composite behav-ior: they atrophy when not adequately practicedand appealed to .|.|. yet will once again makethemselves scarce when preached on and reliedon to excess” (p. 157). Each of these theoristshighlights the way in which a basic capacity foraltruistic action may be structured and devel-oped by the organizational and institutionalenvironment. Yet there has been little investi-gation of the particular circumstances and con-ditions under which this happens. One-shotaltruistic practices are of particular interest inpart because they seem to lack a significantinstitutional dimension and to depend largely onthe individuals involved. But I have argued thatthe one-shot case, in particular, reveals the logis-tical efforts required to constitute populationsof donors. This adds a novel dimension to ourknowledge of the institutional underpinningsof individual identities and opportunities foraction.

Future research could further investigateaspects of the organ procurement system orapply the approach outlined here to other formsof altruism. In the case of organ donation, weknow little about the dynamics of procurementover time or the way different parts of the orga-nizational system interact with one another andwith individual donors. The typology I presenthere may help clarify the different contexts inwhich altruistic actions take place and may raisecomparative questions about why differentactions become institutionalized in differentways. We can ask why different forms of altru-




ism—such as blood donation, charitable dona-tions of money in various forms, voluntarywork, and so on—come to be institutionalizedin different ways, or why the same practice isorganized differently in different places.Although the emphasis here has been on theorganizational dimension, it would be a mistaketo see a stark opposition between individualactions and organizational practices. Without thereasons that make it meaningful for an individ-ual to donate, providing the necessary oppor-tunities to do so would be futile. Yet it is also anerror to focus entirely on the individual, par-ticularly if this means attending only to whethera particular action is truly altruistic. Framing theproblem in this way draws attention away fromthe organizational mechanisms through whichdifferent kinds of donor populations are created.

The dominance of the concept of self-inter-est, both as a starting point for social theory andan assumption about interaction in everydaylife (Miller and Ratner 1996), tends to make stu-dents of altruism look like partisan defenders ofit. It is clear that we should not be content sim-ply to show that altruism happens. A middle-range approach to explaining the organizationalsources of variation in altruism complementsour growing understanding of the socially sit-uated character of self-interested action(Fligstein 2001; Granovetter 1985). To draw ananalogy, few would agree that the discovery ofa general tendency in people to “truck, barterand exchange one thing for another” (Smith2000 [1776]: 14) is the end point of research onself-interest. Rather, it is the beginning of studyinto the institutions and organizational systemsthat surround and sustain this basic disposition.

Kieran Healy is Assistant Professor of Sociology atthe University of Arizona, and Research Fellow at theAustralian National University’s Research School ofSocial Sciences. His main interests are in economicsociology and the sociology of culture. He is com-pleting a book on blood and organ donation in theUnited States and Europe. Currently, he is studyingthe structural sources of cross-national variation inorgan procurement rates within OECD countries;the institutionalization of markets in human goods;and the relationship between hierarchy, informationtechnology, and volunteering in the world of OpenSource Software development.

RREEFFEERREENNCCEESS

Alexander, G. C. and A. R. Seghal. 1998. “Barriersto Cadaveric Renal Transplantation among Blacks,Women, and the Poor.” Journal of the AmericanMedical Association 280:1148–52.

Alexander, Victoria. 1998. “EnvironmentalConstraints and Organizational Strategies.” Pp.272–90 in Private Action and the Public Good,edited by Walter W. Powell and Elisabeth S.Clemens. New Haven, CT: Yale University Press.

Allison, Paul. 1992. “The Cultural Evolution ofBeneficial Norms.” Social Forces 71:279– 301.

Almond, Gabriel and Sidney Verba. 1963. The CivicCulture. Princeton, NJ: Princeton University Press.

Austin, W. 1979. “Sex Differences in BystanderIntervention in a Theft.” Journal of Personalityand Social Psychology 37:2110–20.

Axelrod, Robert. 1984. The Evolution of Cooperation.New York: Basic Books.

Axelrod, Robert and W. D. Hamilton. 1981. “TheEvolution of Cooperation.” Science 211:1390–96.

Barman, Emily. 2002. “Asserting Difference: TheStrategic Response of Nonprofit Organizations toCompetition.” Social Forces 80:1191–1222.

Batson, C. D. 1987. “Prosocial Motivation: Is It EverTruly Altruistic?” Pp. 65–122 in Advances inExperimental Social Psychology, vol. 20, edited byL. Berkowitz. New York: Academic Press.

Boyd, Robert and Peter J. Richerson. 1985. Cultureand the Evolutionary Process. Chicago, IL:University of Chicago Press.

Bradley, Martin B., Norman M. Green, Jr., Dale E.Jones, Mac Lynn, and Lou McNeil. 1992.Churches and Church Membership in the UnitedStates 1990. Atlanta, GA: Glenmary ResearchCenter.

Clarke, Lee and Carroll Estes. 1992. “Sociologicaland Economic Theories of Markets andNonprof its: Evidence from Home HealthOrganizations.” American Journal of Sociology97:945–69.

Clotfelter, C. 1993. “On Trends in Private Sources ofSupport for the US Non-Profit sector.” Voluntas4:190–95.

Cohen, Lawrence. 1999. “Where It Hurts: IndianMaterial for an Ethics of Organ Transplantation.”Daedalus 128:135–66.

Davis, Mark. 1996. Empathy: A Social PsychologicalApproach. Boulder, CO: Westview Press.

DiMaggio, Paul and Helmut Anheier. 1990. “TheSociology of Nonprof it Organizations andSectors.” Annual Review of Sociology 16:137–59.

Drake, A. W., S. N. Finkelstein, and H. M. Sapolsky.1982. The American Blood Supply. Cambridge,MA: MIT Press.

Eckstein, Susan. 2001. “Community as Gift-Giving:Collectivistic Roots of Volunteerism.” AmericanSociological Review 66:829–51.




Ehrle, R. N., T. J. Shafer, and K. R. Nelson. 1999.“Referral, Request and Consent for OrganDonation: Best Practice—a Blueprint for Success.”Critical Care Nurse 19:21–33.

Evans, R., C. Orians, and N. Ascher. 1992. “ThePotential Supply of Organ Donors.” Journal of theAmerican Medical Association 267:239–46.

Fligstein, Neil. 2001. The Architecture of Markets.Princeton, NJ: Princeton University Press.

Frumkin, Peter. 2002. On Being Nonprofit: AConceptual and Policy Primer. Cambridge, MA:Harvard University Press.

Galaskiewicz, Joseph and Wolfgang Bielefeld. 1998.Nonprofit Organizations in an Age of Uncertainty:A Study of Organizational Change. New York:Aldine DeGruyter.

Gallup Organization. 1993. The American Public’sAttitudes toward Organ Donation andTransplantation: A Survey Conducted for thePartnership for Organ Donation. Boston, MA:Partnership for Organ Donation and HarvardSchool of Public Health.

Gortmaker, S. L. 1996. “Organ Donor Potential andPerformance: Size and Nature of the Organ DonorShortfall.” Critical Care Medicine 24:432–39.

———. 1998. “Improving the Request Process toIncrease Family Consent for Organ Donation.”Journal of Transplant Coordination 8:210–17.

Gouldner, Alvin W. 1960. “The Norm of Reciprocity:A Preliminary Statement.” American SociologicalReview 25:161–78.

Granovetter, Mark. 1985. “Economic Action andSocial Structure: The Problem of Embeddedness.”American Journal of Sociology 91:481–510.

Grønbjerg, Kirsten. 1993. Understanding NonprofitFunding: Managing Revenues in Social Serviceand Community Development Organizations. SanFrancisco, CA: Jossey Bass.

Hamilton, W. D. 1964. “The Genetical Evolution ofSocial Behavior. I, II.” Journal of TheoreticalBiology 7:1–52.

Harrell, Frank E. 2001. Regression ModelingStrategies. New York: Springer.

Healy, Kieran. 2000. “Embedded Altruism: BloodCollection Regimes and the European Union’sDonor Population.” American Journal of Sociology105:1633–57.

Hirschman, Albert. 1992. “Against Parsimony.” Pp.142–60 in Rival Views of Market Society.Cambridge, MA: Harvard University Press.

Hodgkinson, V. and M. Weitzman. 1992. Giving andVolunteering in the United States: Findings froma National Survey. Washington, DC: IndependentSector.

Hoffman, M. L. 1981. “Is Altruism Part of HumanNature?” Journal of Personality and SocialPsychology 40:121–37.

Inglehart, Ronald. 1997. Modernization andPostmodernization: Cultural, Economic and

Political Change in Forty-Three Societies.Princeton, NJ: Princeton University Press.

Jencks, Christopher. 1992. “Who Gives to What?” Pp.321–39 in The Nonprofit Sector: A ResearchHandbook, edited by Walter W. Powell. NewHaven, CT: Yale University Press.

Kasiske, B. L., J. F. Neylan, R. R. Riggio, G. M.Danovtich, L. Kahana, S. R. Alexander, and M. G.White. 1991. “The Effect of Race on Access andOutcome in Transplantation.” New EnglandJournal of Medicine 324:320–27.

Kelly, David and Walter Wiest. 1991. “ChristianPerspectives.” Pp. 199–221 in New Harvest:Transplanting the Body and Reaping the Benefits,edited by C. Don Keyes and Walter E. Wiest.Clifton, NJ: Humana Press.

Klassen, A. C., D. K. Klassen, R. Aronoff, A. G.Hall, and J. Braslow. 1999. “OrganizationalCharacteristics of Solid-Organ Donor Hospitalsand Nondonor Hospitals.” Journal of TransplantCoordination 9:87–94.

Klassen, Ann C. and David K. Klassen. 1996. “Whoare the Donors in Organ Donation? The Family’sPerspective in Mandated Choice.” Annals ofInternal Medicine 125:70–73.

Knoke, David. 1986. “Associations and InterestGroups.” Annual Review of Sociology 12:1–21.

Krebs, D. K. and D. T. Miller. 1985. “Altruism andAggression.” Pp. 1–71 in Handbook of SocialPsychology II, edited by G. Lindzey and E.Aronson. New York: Random House.

Latane, B. and J. M. Darley. 1970. “SocialDeterminants of Bystander Intervention inEmergencies.” Pp. 13–27 in Altruism and HelpingBehavior: Social Psychological Studies of SomeAntecedents and Consequences, edited by J.Macaulay and L. Berkowitz. New York: AcademicPress.

Macy, Michael and John Skvoretz. 1998. “TheEvolution of Trust and Cooperation BetweenStrangers: A Computational Model.” AmericanSociological Review 63:638–60.

Mark, Noah. 2002. “The Cultural Evolution ofCooperation.” American Sociological Review67:323–44.

Miller, Dale T. and Rebecca Ratner. 1996. “ThePower of the Myth of Self-Interest.” Pp. 25–48 inCurrent Societal Issues In Justice, edited by L.Montada and M. Lerner. New York: Plenum Press.

Monroe, Kristen Renwick. 1998. The Heart ofAltruism. Princeton, NJ: Princeton UniversityPress.

Nathan, H. M. 1999. “Pennsylvania Looks forAnswers to the Organ Donor Shortage.” Pressrelease. Gift of Life Donor Program, Pittsburgh,PA.

Norris, M. K. 1990. “Required Request: Why It HasNot Significantly Improved the Donor Shortage.”Heart and Lung 19:685–86.




Oliner, S. P. and P. M. Oliner. 1988. The AltruisticPersonality: Rescuers of Jews in Nazi Europe.New York: Free Press.

Oliver, Christine. 1991. “Strategic Responses toInstitutional Processes.” Academy of ManagementReview 16:145–79.

Ostrander, Susan. 1995. Money for Change: SocialMovement Philanthropy at Haymarket People’sFund. Philadelphia, PA: Temple University Press.

Ozcan, Y. A., J. W. Begun, and M. M. McKinney.1999. “Benchmarking Organ ProcurementOrganizations: A National Study.” Health ServicesResearch 34:855–74.

Piliavin, Jane Allyn and Peter L. Callero. 1991. GivingBlood: The Development of an Altruistic Identity.Baltimore, MD: Johns Hopkins University Press.

Powner, David J. and Joseph M. Darby. 1999.“Current Considerations in the Issue of BrainDeath.” Neurosurgery 45:1222–27.

Prottas, Jeffrey. 1989. “The Organization of OrganProcurement.” Journal of Health Politics, Policy,and Law 14:41–55.

———. 1994. The Most Useful Gift: Altruism andthe Public Policy of Organ Transplants. SanFrancisco, CA: Jossey Bass.

R Core Development Team. 2003. R: A Language forStatistical Analysis and Computing. Vienna,Austria: R Foundation for Statistical Computing.

Randall, V. R. 1996. “Slavery, Segregation, andRacism: Trusting the Medical System Ain’t AlwaysEasy! An African-American Perspective onBioethics.” St. Louis University Public Law Review15:181–235.

Salamon, Lester M. and Helmut K. Anheier. 1994.The Nonprofit Sector in Comparative Perspective:An Overview. Baltimore, MD: Johns HopkinsUniversity Press.

Scheper-Hughes, Nancy. 2000. “The Global Trafficin Human Organs.” Current Anthropology41:191–211.

Scheper-Hughes, Nancy and Loïc Waquant, eds.2002. Commodifying Bodies. Thousand Oaks, CA:Sage.

Schervish, Paul G. and John J. Havens. 1997. “SocialParticipation and Charitable Giving: A MultivariateAnalysis.” Voluntas 8:235–60.

Schmidtz, David. 1993. “Reasons for Altruism.”Social Philosophy and Policy 10:52–68.

Schofer, Evan and Marion Fourcade-Gourinchas.2001. “The Structural Contexts of CivicEngagement: Voluntary Association Membershipin Comparative Perspectice.” AmericanSociological Review 66:806–28.

Shafer, T. J., J. Orlowski, R. N. Ehrle, R.Gruenenfelder, K. Davis, and D. A. Reyes. 1998.“Two Years Experience with Routine Notification.”Paper presented at the Annual Meeting of theNorth American Transplant CoordinatorsOrganization, August 10, New York.

Siminoff, L. A., R. M. Arnold, A. L. Caplan, B. A.Virnig, and D. L. Seltzer. 1995. “Public PolicyGoverning Organ and Tissue Procurement in theUnited States.” Annals of Internal Medicine123:10–17.

Siminoff, Laura A. and Kristine A. Nelson. 1999.“The Accuracy of Hospital Reports of OrganDonation Eligibility, Requests and Consent: ACross-Validation Study.” Joint CommissionJournal on Quality Improvement 25:129–36.

Simmons, Roberta G. 1991. “Altruism andSociology.” Sociological Quarterly 32:1–22.

Simmons, Roberta, Susan Klein Marine, and RichardSimmons. 1977. Gift of Life: The Social andPsychological Impact of Organ Transplantation.New York: Wiley.

Singer, Peter. 1973. “Altruism and Commerce: ADefense of Titmuss against Arrow.” Philosophy andPublic Affairs 2:312–20.

Skocpol, Theda, Marshal Ganz, and Ziah Munson.2000. “A Nation of Organizers: The InstitutionalOrigins of Civic Voluntarism in the United States.”American Political Science Review 94:527–46.

Smith, Adam. 2000 [1776]. The Wealth of Nations.New York: Modern Library.

Smith, J. Maynard. 1964. “Group Selection and KinSelection.” Nature 201:1145–46.

———. 1982. Evolution and the Theory of Games.New York: Cambridge University Press.

Sober, Elliot and David Sloan Wilson. 1998. UntoOthers: The Evolution and Psychology of UnselfishBehavior. Cambridge, MA: Harvard UniversityPress.

Southeastern Institute of Research. 1994. GeneralConsumers: American Attitudes toward theAllocation of Organ Transplantation. Richmond,VA: United Network for Organ Sharing.

Titmuss, Richard. 1971. The Gift Relationship: FromHuman Blood to Social Policy. New York: Vintage.

Trivers, R. L. 1971. “The Evolution of ReciprocalAltruism.” Quarterly Review of Biology 46:35–57.

Twersky, Abraham, Michael Gold, and Walter Jacob.1991. “Jewish Perspectives.” Pp. 187–98 in NewHarvest: transplanting the body and reaping thebenefits, edited by C. Don Keyes and Walter E.Wiest. Clifton, NJ: Humana Press.

U.S. General Accounting Off ice. 1993. OrganTransplants: Increased Effort Needed to BoostSupply and Ensure Equitable Distribution ofOrgans. Report #GAO/HRD-93-56. Washington,DC: General Accounting Office.

———. 1997. Organ Procurement Organizations:Alternatives Being Developed to More AccuratelyAssess Performance. Report #GAO/HEHS-98-26.Washington, DC: General Accounting Office.

Venables, W. N. and B. D. Ripley. 2002. ModernApplied Statistics with S. 4th ed. New York:Springer.




Verble, Margaret and Judy Worth. 2000.“Overcoming Families’Fears and Concerns in theDonation Discussion.” Progress in Transplantation10:155–60.

Wendler, Dave and Neal Dickert. 2001. “The ConsentProcess for Cadaveric Organ Procurement: How

Does It Work? How Can It Be Improved?” Journalof the American Medical Association 285:329–33.

Zimmerman, D., S. Donnelly, J. Miller, D. Stewart,and S. E. Albert. 2000. “Gender Disparity in LivingRenal Transplant Donation.” American Journal ofKidney Diseases 36:534–40.



princeton university (cid 195003) · according to sober and wilson (1998), Òthe altruism...

Documents