《Developing and maintaining clients’ trust through institutional mechanisms in online...》中英文对比文献翻译_这一领域已有研究的共识英文

《Developing and maintaining clients’ trust through institutional mechanisms in online service markets for digital entrepreneurs: A process model》中英文对比翻译

引用：Dua W, Mao J Y. Developing and maintaining clients’ trust through institutional mechanisms in online service markets for digital entrepreneurs: A process model[J]. Journal of Strategic Information. Systems 27 (2018) 296–310.

ABSTRACT
Crowdsourcing intermediaries, such as Upwork and ZBJ, are powering a new force of digital entrepreneurs by creating online service markets where service providers seek contracts from clients all over the world. An important reason for this flourishing digital entrepreneurship is that the intermediaries establish institutional mechanisms (IMs) that develop clients’ trust in digital entrepreneurs. However, insights into how intermediaries achieve it remain limited. Although the e-Commerce literature has generated rich findings on intermediary-based trust development, its findings are based on online product markets and thus may not be readily applicable to online service markets. More specifically, these findings are mainly focused on various IMs that facilitate initial trust formation based on variance models. However, a successful service transaction requires not only the initial trust but further development and maintenance of that trust, because it is a cooperative process between clients and vendors. By conducting an in-depth case study of ZBJ, the largest crowdsourcing intermediary in China, this study develops a three-phase process model, whereby the intermediary deploys different sets of IMs that initiate, augment, and maintain clients’ trust in vendors. Our findings make important theoretical contributions to the literature on digital entrepreneurship, crowdsourcing, and e-Commerce trust development.

摘要：

诸如Upwork和ZBJ之类的众包中介机构正在通过创建在线服务市场来推动数字企业家的新生力量，在这些市场中，服务提供商从世界各地的客户那里寻求合同。数字企业蓬勃发展的一个重要原因是，中介机构建立了制度机制(IM)，以发展客户对数字企业家的信任。然而，对于中介机构如何实现这一目标的了解仍然有限。虽然电子商务文献对基于中介的信任开发产生了丰富的研究结果，但其研究结果是基于在线产品市场的，因此可能不容易适用于在线服务市场。更具体地，这些发现主要集中在各种便于基于方差模型建立初始信任的IM上。然而，成功的服务事务不仅需要最初的信任，还需要进一步的开发和维护这种信任，因为它是客户和供应商之间的合作过程。本研究通过对中国最大的众包中介ZBJ的深入案例研究，建立了一个三阶段的流程模型，中介通过部署不同的IMs来启动、增强和维护客户对供应商的信任。我们的发现对数字企业、众包和电子商务信任发展的文献做出了重要的理论贡献。

Introduction
By creating online service markets where self-employed service providers sell their services to clients all over the world, crowdsourcing intermediaries, such as Upwork and ZBJ, have been touted as a wellspring of entrepreneurial activities; the phenomenon is commonly known as the “gig economy” (Dishman, 2017; Shadpour, 2018; Sundararajan, 2015). Previously, gigs were only for musicians, while most people looked for stable jobs that offered fixed salaries. Today, more and more skilled workers are earning income from self-employment in a side gig, and some have transformed their gig into a new business venture (Kaufman, 2013; Shadpour, 2018). For example, ZBJ, China’s largest crowdsourcing intermediary, hosts over 10 million service providers, such as computer programmers, website developers, animators, artists, and designers and small/medium-sized enterprises.

1.引言

通过创建在线服务市场，个体服务提供商向全球各地的客户销售服务，Upwork和ZBJ等众包中介被吹捧为企业活动的源泉;这种现象通常被称为“零工经济”(Dishman, 2017;Shadpour, 2018;Sundararajan, 2015)。以前，只有音乐家才会做零工，而大多数人都在寻找有固定工资的稳定工作。如今，越来越多的熟练工人通过兼职赚取自营职业的收入，一些人已经将他们的兼职转变成新的商业风险(Kaufman, 2013;Shadpour, 2018)。例如，中国最大的众包中介ZBJ，拥有超过1000万的服务提供商，包括电脑程序员、网站开发者、动画师、艺术家、设计师和中小企业。

These service providers can be seen as digital entrepreneurs, who “pursue the opportunities based on the use of digital media and other information and communication technologies” (Davidson and Vaast, 2010, p.2). First, they pursue opportunities afforded by crowdsourcing intermediaries’ digital platforms (Majchrzak and Malhotra, 2013). Second, they deliver solutions in the form of digital artifacts, such as software packages and websites (Nambisan, 2017). Digital entrepreneurship is an emerging field in information systems (IS) research. Crowdsourcing intermediary-enabled entrepreneurship falls into the subcategory of online market-enabled entrepreneurship. Studies in this subcategory have focused on the actions of individual entrepreneurs (Gregg and Walczak, 2008; Reuber and Fischer, 2010) and entrepreneurial communities (Avgerou and Li, 2013; Leong et al., 2016), but have overlooked the role of intermediaries in enabling entrepreneurship.
这些服务提供商可以被视为数字企业家，他们“基于使用数字媒体和其他信息通信技术来寻求机会”(Davidson和Vaast, 2010, p.2)。首先，他们追求众包中介的数字平台提供的机会(Majchrzak和Malhotra, 2013)。其次，他们以数字产品的形式提供解决方案，如软件包和网站(Nambisan, 2017)。数字企业家精神是信息系统（IS）研究中的一个新兴领域。 众包中介推动的企业家精神属于在线市场推动的企业家精神的子类别。这一细分领域的研究主要关注个体企业家的行为(Gregg和Walczak, 2008;(Reuber and Fischer, 2010)与创业社区(Avgerou and Li, 2013;Leong et al.， 2016)，但忽略了中介机构在促进创业中的作用。

Our study intends to fill this gap by examining how crowdsourcing intermediaries enable entrepreneurs from a trust development perspective. Trust is defined as the willingness to depend upon and make oneself vulnerable to the trustee (Mayer et al., 1995). Trust is critical to online transactions, because it is an effective mechanism for mitigating uncertainty (Datta and Chatterjee, 2008; Jarvenpaa et al., 2000). Uncertainty can increase clients’ transaction costs and discourage them from engaging in crowdsourcing (Ye and Kankanhalli, 2015). It is difficult for individual entrepreneurs to gain clients’ trust, because they have no prior interaction and lack face-to-face communication with clients. Third-party institutional mechanisms (IMs) should be put in place for trust development. The crowdsourcing literature has also recognized that intermediaries play an important role in developing clients’ trust in vendors (Feller et al., 2012; Ye and Kankanhalli, 2015; Zogaj et al., 2014), but insights into how they achieve it remain limited.
我们的研究试图从信任发展的角度来研究众包中介如何帮助企业家，从而填补这一空白。信任被定义为依赖和使自己易受托人的意愿（Mayer等，1995）。信任对在线交易至关重要，因为它是减少不确定性的有效机制(Datta和Chatterjee, 2008;Jarvenpaa等，2000)。不确定性会增加客户的交易成本，阻碍他们参与众包(Ye and Kankanhalli, 2015)。个体企业家很难获得客户的信任，因为他们没有事先的互动，也缺乏与客户面对面的交流。应当建立第三方机构机制来发展信任。众包文献也认识到中介机构在发展客户对供应商的信任方面起着重要作用(Feller et al.， 2012;Ye and Kankanhalli, 2015;(Zogaj et al.， 2014)，但对他们如何实现这一目标的了解仍然有限。

The e-Commerce literature has generated rich findings on how intermediaries develop clients’ trust, in particular through the deployment of IMs that punish opportunistic behaviors and engender cooperation (Gefen et al., 2003). However, these findings are derived from online product markets and may not be readily applicable to online service markets. Moreover, existing findings are mainly focused on IMs that facilitate initial trust formation only (Ba and Pavlou, 2002; McKnight et al., 2002; Pavlou, 2002; Pavlou and Gefen, 2004), based on variance models. Whereas initial trust can lead to a transaction in online product markets, a successful service transaction also requires further development and maintenance of that trust, because services are delivered via a cooperative process between clients and vendors (Cheng et al., 2017; Jarvenpaa and Leidner, 1997; Lewicki and Bunker, 1996). In this case, crowdsourcing intermediaries should develop and maintain clients’ trust throughout the transaction, and a process model is needed. Therefore, we derive our research question as follows, “How do crowdsourcing intermediaries develop and maintain clients’ trust in vendors throughout a service transaction?”
电子商务文献对中介机构如何发展客户的信任产生了丰富的研究成果，特别是通过部署IMs来惩罚机会主义行为和产生合作(Gefen et al.， 2003)。然而，这些发现来自在线产品市场，可能并不适用于在线服务市场。此外，基于方差模型，现有发现主要集中在仅促进初始信任形成的IM（Ba和Pavlou，2002； McKnight等，2002； Pavlou，2002； Pavlou和Gefen，2004）。虽然最初的信任可以导致在线产品市场上的交易，但成功的服务交易也需要进一步发展和维护这种信任，因为服务是通过客户和供应商之间的合作过程交付的(Cheng et al.， 2017;Jarvenpaa和Leidner, 1997;《路易斯基与邦克》，1996)。在这种情况下，众包中介应该在整个交易过程中发展和维护客户的信任，这需要一个流程模型。因此，我们的研究问题如下:“众包中介如何在整个服务交易中发展并维持客户对供应商的信任?”

Based on an in-depth case study of ZBJ, we develop a three-phase process model, whereby the intermediary initiates, augments, and maintains clients’ trust in vendors by deploying three different sets of IMs. This study makes three important theoretical contributions. First, it contributes to the digital entrepreneurship literature, in particular studies on online market-enabled entrepreneurship, by examining the enabling role of intermediaries through a trust development perspective. Second, it contributes to the crowdsourcing literature by developing a process model of how intermediaries develop and maintain clients’ trust in vendors, and by explaining how intermediaries facilitate the crowdsourcing process. Third, it contributes to the e-Commerce trust development literature by extending intermediary-enabled trust development from online product markets to online service markets, and adding a process view.
在对ZBJ进行深入的案例研究的基础上，我们开发了一个三阶段的流程模型，其中，中介机构通过部署三套不同的IM来发起，增强和维护客户对供应商的信任。本研究有三个重要的理论贡献。首先，它通过从信任发展的角度审视中介机构的促进作用，为数字企业文献，特别是对在线市场驱动型企业的研究做出了贡献。其次，它通过开发中介如何开发和维护客户对供应商的信任的过程模型，并解释中介如何促进众包过程，从而为众包文献做出贡献。第三，它通过将支持中介的信任开发从在线产品市场扩展到在线服务市场，并添加了流程视图，为电子商务信任发展文献做出了贡献。

Literature review
Digital entrepreneurship and crowdsourcing
Today, digital technologies have significantly reduced the cost of starting a company and accelerated the rate of expansion, thus creating a boom in digital entrepreneurship. Digital entrepreneurship has provided IS scholars many research opportunities. According to Del Giudice and Straub (2011), “The fascinating research questions that can be asked are endless” (p. 6). These questions lie at the intersection of entrepreneurship and various digital technologies (Nambisan, 2017), such as online markets (e.g., Avgerou and Li, 2013; Gregg and Walczak, 2008), social media (e.g., Fischer and Reuber, 2011, 2014), and technology ecosystems (e.g., Ceccagnoli et al., 2012; Jarvenpaa and Standaert, 2017; Nambisan and Baron, 2013).

2.文献综述

2.1数字创业和众包

如今，数字技术大大降低了创办公司的成本，加快了扩张速度，从而创造了数字创业的热潮。数字创业为学者提供了许多研究机会。Del Giudice和Straub(2011)认为，“可以提出的迷人的研究问题是无止境的”(第6页)。这些问题位于创业和各种数字技术的交叉点(Nambisan, 2017)，如在线市场(如Avgerou和Li, 2013;Gregg和Walczak, 2008)，社交媒体(如Fischer和Reuber, 2011, 2014)，和技术生态系统(如Ceccagnoli等，2012;Jarvenpaa和Standaert, 2017;Nambisan和Baron, 2013)。

Because crowdsourcing intermediaries create online service markets for service providers to seek contracts from clients all over the world, intermediary-enabled entrepreneurship is subset of of online market-enabled entrepreneurship. Crowdsourcing intermediaries are similar to traditional e-Commercial platforms, such as Amazon and eBay, but intended for service exchange rather than product exchange. Crowdsourcing intermediaries may have a broader impact toward entrepreneurship than traditional online product markets, because they have further lowered the barriers to entrepreneurship. Online product transactions still require transportation and thus exclude entrepreneurs who do not have access to logistic infrastructures (Leong et al., 2016). In contrast, service transactions require no such infrastructures. For example, some marginalized communities in India have successfully started new ventures on Upwork and moved out of poverty by taking orders from western clients (Agrawal et al., 2013). As noted above, existing research on online market-enabled entrepreneurship has mainly focused on the activities of entrepreneurs and their communities, but overlooked the role of intermediaries. Our study intends to fill this gap by examining how crowdsourcing intermediaries enable entrepreneurs from a trust development perspective.
由于众包中介为服务提供商创建了在线服务市场，以从世界各地的客户那里寻求合同，因此，中间型创业是在线市场型创业的一个子集。众包中介类似于传统的电子商务平台，如亚马逊和eBay，但其目的是服务交换而不是产品交换。与传统的在线产品市场相比，众包中介可能会对创业产生更广泛的影响，因为它们进一步降低了创业的障碍。在线产品交易仍然需要运输，因此将无法使用物流基础设施的企业家排除在外（Leong等，2016）。相反，服务交易不需要这种基础设施。例如，印度的一些边缘社区已经成功地在Upwork上建立了新的企业，并通过接受西方客户的订单摆脱了贫困(Agrawal et al.， 2013)。如上所述，现有的关于以市场为导向的网上创业的研究主要集中在企业家及其社区的活动上，而忽视了中间商的作用。我们的研究试图从信任发展的角度来研究众包中介如何帮助企业家，从而填补这一空白。

The term “crowdsourcing” was coined by Howe (2006) to refer to the act of assigning tasks that have traditionally been performed by internal employees to a large group of undefined people outside an organization. Crowdsourcing has grown into a broader concept that encompasses four major types of arrangement: Contests, collaborative communities, complementors, and labor markets (Boudreau and Lakhani, 2013). Labor markets can also be referred to as service markets, because they“match buyers and sellers of services and employ conventional contracting for services rendered” (p. 10).
“众包”一词是Howe（2006）提出的，是指将传统上由内部员工执行的任务分配给组织外部的一大堆未定义人员的行为。众包已经发展成为一个更广泛的概念，它包含了四种主要类型的安排:竞赛、协作社区、互补和劳动力市场(Boudreau和Lakhani, 2013)。劳动力市场也可以被称为服务市场，因为它们“为服务的买卖双方牵线搭桥，为提供的服务采用传统的合同制”(第10页)。

Service markets can be further divided into two types, labor-intensive versus knowledge-intensive. The former is represented by Amazon Mechanical Turk (MTurk), enabling transactions of simple, mundane tasks such as data processing and image tagging (Deng et al., 2016; Karger et al., 2014). For example, in MTurk, vendors complete a task at the terms proposed by clients, and clients do not communicate much with vendors except for the initial job description (Chen and Horton, 2016). In contrast, the latter is represented by Upwork, Freelancer.com, and ZBJ, transacting tasks that require domain knowledge, innovation, and collaboration between clients and vendors. Examples include website and software development, and logo and animation design (Feller et al., 2012; Qi and Mao, 2016; Ye and Kankanhalli, 2015).
服务市场可以进一步分为两种类型：劳动密集型和知识密集型。 前者由Amazon Mechanical Turk（MTurk）代表，可实现简单，日常任务的事务，例如数据处理和图像标记（Deng等，2016； Karger等，2014）。 例如，在MTurk中，供应商按照客户提出的条件完成任务，除了最初的工作描述之外，客户与供应商之间的沟通不多（Chen and Horton，2016）。相比之下，后者由Upwork、Freelancer.com和ZBJ表示，它们处理的任务需要领域知识、创新以及客户和供应商之间的协作。例如网站和软件开发，以及logo和动画设计(Feller et al.， 2012;齐茂，2016;Ye and Kankanhalli, 2015)。

The two types of markets require different approaches to developing and maintaining trust because they deal with different tasks and transaction processes. Furthermore, because digital entrepreneurship is more likely to flourish in knowledge-intensive markets than in labor-intensive ones (Davidson and Vaast, 2010), this study examines the former. Therefore, in the reminder of this paper, the term crowdsourcing specifically refers to service markets, and knowledge-intensive ones in particular.
这两类市场需要不同的方法来发展和维持信任，因为它们处理不同的任务和交易过程。此外，由于数字创业更有可能在知识密集型市场蓬勃发展，而不是在劳动密集型市场，因此本研究研究了前者。 因此，在本文的提醒中，“众包”一词专门指服务市场，尤其是知识密集型市场。

Trust is critical to crowdsourcing, because it is effective in mitigating uncertainty, which increases clients’ transaction costs and discourages them from engaging in crowdsourcing (Ye and Kankanhalli, 2015). Clients in online service markets face more uncertainty than their counterparts in online product markets. For example, unlike products, services are customized based on clients’ specific requirements, and uncertainty can arise in the process of specifying these requirements and communicating them to vendors (Ye and Kankanhalli, 2015). This uncertainty is further exacerbated by the fact that many crowdsourcing clients lack domain knowledge to articulate their requirements and misunderstandings are common between geographically separated clients and vendors (Qi and Mao, 2016).
信任对于众包是至关重要的，因为它可以有效地减少不确定性，而不确定性增加了客户的交易成本，并阻碍他们参与众包(Ye和Kankanhalli, 2015)。在线服务市场的客户比在线产品市场的客户面临更多的不确定性。例如，与产品不同，服务是根据客户的特定需求定制的，在指定这些需求并与供应商沟通的过程中会产生不确定性(Ye和Kankanhalli, 2015)。许多众包客户缺乏领域知识来阐明他们的要求，而且地理位置不同的客户和供应商之间经常会产生误解(Qi和Mao, 2016)。，这一不确定性进一步加剧了。

In addition, unlike products, services are produced through client-vendor cooperation, rather than by vendors independently. Uncertainty can arise from the cooperation, especially when the two parties lack face-to-face communication (Jarvenpaa and Leidner, 1997). For example, clients face uncertainty about vendors’ commitment because they are unable to monitor vendors effectively. This uncertainty, if not mitigated properly, can damage trust built in earlier phases and lead to the early termination of contracts (Veltri et al., 2008; Whitten and Leidner, 2006).
此外，与产品不同的是，服务是通过客户-供应商合作而不是由供应商独立完成的。合作会产生不确定性，特别是当双方缺乏面对面的交流时(Jarvenpaa and Leidner, 1997)。例如，客户无法确定供应商的承诺，因为他们无法有效地监控供应商。这种不确定性，如果没有得到适当的缓解，可能会损害在早期阶段建立的信任，并导致合同的提前终止(Veltri et al.， 2008;Whitten和Leidner, 2006)。

Because of service-specific uncertainty, gaining clients’ trust becomes more important. However, self-employed service providers face great difficulty in gaining clients’ trust on their own, because they are geographically separated from clients and have no prior interaction with them. The crowdsourcing literature shows that intermediaries play an important role in helping vendors gain clients’ trust (Feller et al., 2012; Ye and Kankanhalli, 2015; Zogaj et al., 2014). However, little is known about how the role is achieved.
由于特定于服务的不确定性，赢得客户的信任变得更加重要。然而，个体经营的服务提供商在获得客户的信任方面面临着很大的困难，因为他们与客户在地理上是分开的，之前没有与客户的互动。众包文献表明，中介机构在帮助供应商获得客户信任方面发挥了重要作用(Feller et al.， 2012;Ye and Kankanhalli, 2015;(Zogaj等，2014)。然而，很少有人知道这一角色是如何实现的。

In general, the majority of the crowdsourcing literature has taken either a client or vendor perspective. Few studies have taken the intermediary perspective (Zogaj et al., 2014). For example, Feller et al. (2012) state that “While the commercial success of such brokerages (intermediaries) indicates the potency of arguments concerning the importance of crowds, little is known about the operation of such brokerages” (p. 218). Similarly, Ye and Kankanhalli (2015) state that “Future researchers could investigate the performance of crowdsourcing platforms in facilitating the process of crowdsourcing” (p.106–107). Our study responds to the call for research by examining how crowdsourcing intermediaries develop and maintain clients’ trust in vendors throughout a service transaction. To do so, we first resort to the e-Commerce literature, which has generated findings on intermediary-based trust development. These findings can guide our exploration and provide opportunities for theoretical contributions.
总的来说，大多数众包文献都是从客户或供应商的角度出发的。很少有研究采用中介视角(Zogaj et al.， 2014)。例如，Feller等人(2012)指出“虽然这类经纪公司(中介机构)的商业成功表明了关于群体重要性的论点的力量，但是人们对这类经纪公司的运作知之甚少”(第218页)。同样，Ye和Kankanhalli(2015)指出“未来的研究者可以研究众包平台在促进众包过程中的表现”(p.106-107)。我们的研究响应了研究的号召，研究众包中介如何在整个服务交易中发展和保持客户对供应商的信任。为此，我们首先求助于电子商务文献，这些文献产生了关于基于中介的信任开发的研究结果。这些发现可以指导我们的探索，为理论贡献提供机会。

Intermediary-enabled trust development in e-Commerce
A major reason that small vendors can flourish in e-Commerce markets despite the inherent uncertainty is the presence of intermediaries such as eBay and Amazon, which develop clients’ trust in them (Datta and Chatterjee, 2008). There are two paths by which an intermediary can develop clients’ trust in vendors. First, because trust can be transferred, an intermediary can acquire clients’ trust in its brand and then transfer that trust to its vendors; second, an intermediary can develop effective IMs, which reduce opportunistic behaviors and ensure successful transactions (McKnight et al., 2002). These two paths complement each other. On the one hand, trust in an intermediary underlies the perceived effectiveness of its IMs; on the other hand, effective IMs increase the trustworthiness of the intermediary (Pavlou and Gefen, 2004). Research has focused on the second path, because IMs for trust development are an area in which innovative solutions can be found (Pavlou and Gefen, 2004). In contrast, trust in an intermediary is similar to trust in any target, which comes from familiarity, reputation, and benevolent behaviors (Jarvenpaa et al., 2000). In the remainder of this paper, the term clients’ trust refers to clients’ trust in vendors, and we focus on examining how intermediaries enable vendors to develop and maintain that trust.

2.2电子商务中基于中介的信任发展

尽管存在内在的不确定性，小供应商仍能在电子商务市场中蓬勃发展的主要原因是诸如eBay和Amazon之类的中介机构的存在，它们建立了客户对其的信任（Datta和Chatterjee，2008年）。中介可以通过两种途径来发展客户对供应商的信任。首先，因为信任可以被转移，中介可以获得客户对其品牌的信任，然后将这种信任转移给供应商;其次，中介可以开发有效的IMs，减少机会主义行为，确保交易成功(McKnight et al.， 2002)。这两条路相辅相成。一方面，中介的信任是中介IM有效性的基础;另一方面，有效的IMs增加了中介的可信度(Pavlou and Gefen, 2004)。研究集中在第二种途径上，因为信任发展的IMs是一个可以找到创新解决方案的领域(Pavlou和Gefen, 2004)。相比之下，对中介的信任与对任何目标的信任相似，都来自于熟悉、声誉和慈善行为(Jarvenpaa et al.， 2000)。在本文的其余部分中，术语“客户的信任”指的是客户对供应商的信任，我们将重点研究中介体如何使供应商能够发展和维护这种信任。

Research has identified several effective IMs, such as escrow services, credit card guarantees, monitoring services, accreditation systems, and feedback mechanisms. Escrow services allow clients to pay vendors only when they receive and approve goods. Similarly, credit card guarantees protect clients by offering recourse in cases of fraud. Many credit card companies offer clients zero liability in online fraud (Pavlou and Gefen, 2004). Monitoring services monitor vendor activities and punish activities that do not adhere to generally accepted rules (Pavlou, 2002). Some research categorizes monitoring services as part of escrow services, because escrow services involve monitoring vendors’ activities and their agreements with clients and settling disputes in face of conflicts (Pavlou and Gefen, 2004). Accreditation systems attest to vendors’ ability and induce cooperative behaviors by encouraging vendors to upgrade their accreditation. Accreditation can be granted by both an independent authority and an intermediary which uses it as a means of rewarding high-quality vendors and weeding out problematic ones (Luo, 2002; McKnight et al., 2002). Finally, a feedback mechanism allows clients to evaluate the trustworthiness of a vendor based on what other clients say about them (Ba and Pavlou, 2002).
研究确定了几个有效的IMs，如托管服务、信用卡担保、监控服务、认证系统和反馈机制。第三方托管服务只允许客户在收到并批准货物后向供应商付款。同样，信用卡担保通过在欺诈情况下提供追索权来保护客户。许多信用卡公司在网络诈骗中为客户提供零责任(Pavlou和Gefen, 2004)。监控服务监控供应商的活动，并惩罚不遵守普遍接受的规则的活动(Pavlou, 2002)。一些研究将监控服务归类为第三方托管服务的一部分，因为第三方托管服务包括监控供应商的活动及其与客户的协议，以及解决冲突时的争端(Pavlou和Gefen, 2004)。认证系统通过鼓励供应商提高他们的认证来证明供应商的能力和诱导合作行为。认证可以由一个独立的权威机构和一个中介机构授予，中介机构将其作为奖励高质量供应商和淘汰问题供应商的一种手段(Luo, 2002;McKnight等人，2002)。最后，反馈机制允许客户根据其他客户对供应商的评价来评估供应商的可信度(Ba和Pavlou, 2002)。

Despite their different forms, these IMs share similarities in the approaches that they develop trust. Our review synthesizes four approaches, which can be used to guide our exploration in the crowdsourcing context. First, IMs can develop trust by providing clients with guarantees. For example, escrow services guarantee that clients do not have to pay until they receive the products in good shape; credit card guarantees ensure that clients do not lose their money in online frauds. Accreditation is also a form of guarantee, which ensures a vendor’s ability and expected future behavior (Luo, 2002).
尽管这些IM具有不同的形式，但它们在建立信任的方法上具有相似之处。 我们的综述综合了四种方法，可用于指导我们在众包环境中的探索。首先，IMs可以通过为客户提供担保来建立信任。例如，第三方托管服务保证客户在收到完好无损的产品之前不必付款;信用卡担保确保客户不会在网络诈骗中损失钱财。认证也是一种保证形式，保证了供应商的能力和预期的未来行为(Luo, 2002)。

Second, IMs can develop trust by influencing trusting beliefs. Integrity trusting belief refers to the extent to which the trustee adheres to a set of principles which the trustor finds acceptable. For example, escrow services can lead to integrity trusting belief by creating incentives for vendors to refrain from opportunistic behaviors (Fang et al., 2014). Ability trusting belief refers to the extent to which the trustee has the expertise to meet the trustor’s needs. For example, a feedback mechanism can lead to ability belief by providing a reputation system that reflects a vendor’s ability (Dou et al., 2013; Ho et al., 2017). Benevolence trusting belief refers to the extent to which the trustee will act fairly and not take advantage of the trustor when given the chance (Mayer et al., 1995). Benevolence trusting belief is less relevant to the online market, because it requires trustee and trustor to have prior attachment (Ba and Pavlou, 2002).
其次，IM可以通过影响信任信念来发展信任。诚信信任信念是指受托人坚持一套委托人认为可以接受的原则的程度。例如，第三方托管服务可以通过激励供应商避免机会主义行为，从而导致诚信信任信念的产生（Fang等，2014）。能力信任信念是指受托人具有能够满足其需求的专业知识的程度。例如，反馈机制可以通过提供一个反映供应商能力的声誉系统而导致能力信念(Dou et al.， 2013;Ho等，2017)。仁爱信任信念是指当给受托人机会时，受托人会公平行事而不会利用委托人的程度(Mayer et al.， 1995)。仁爱信任信念与网络市场的相关性较小，因为它要求受托人和委托人事先有依恋关系(Ba和Pavlou, 2002)。

Third, IMs can develop trust by giving clients more control, which makes vendor opportunism irrational. Legal bonds are an effective formal control in offline markets, but they are less effective in online markets (Pavlou, 2002). An important reason for this is that the legal stance of an online market is ambiguous, and executing legal enforcement is difficult (Gefen et al., 2003). Instead, intermediaries should institute rules and procedures to monitor participants’ activities and punish activities that do not adhere to them. Intermediaries can also leverage market-driven informal control, such as a feedback mechanism. A feedback mechanism adds a level of control, because clients’ negative comments can serve as harsh punishment that reduces vendors’ future sales and price premium (Ba and Pavlou, 2002).
第三，IM可以通过给予客户更多的控制权来建立信任，这使供应商的机会主义变得不合理。合法的债券在线下市场是一种有效的形式控制，但在网络市场中却不那么有效(Pavlou, 2002)。一个重要的原因是，网络市场的法律立场是模糊的，难以执行法律（Gefen等，2003）。相反，中介机构应该制定规则和程序来监督参与者的活动，并惩罚不遵守规则和程序的活动。中介机构还可以利用市场驱动的非正式控制，如反馈机制。反馈机制增加了一定程度的控制，因为客户的负面评论可以作为严厉的惩罚，降低供应商的未来销售和价格溢价(Ba和Pavlou, 2002)。

Fourth, IMs can develop trust by providing clients with more information. Information asymmetry is an important reason that vendors engage in opportunistic behaviors. This is a typical principal-agent issue, where vendors (agents) have more information about their characteristics, products, and practices than clients (principals), who lack effective means to monitor vendors (Pavlou et al., 2007). The literature shows that meaningful and timely information sharing between clients and vendors promotes trust (Jarvenpaa and Leidner, 1997). Feedback mechanisms and accreditation are important sources of external information that helps clients mitigate information asymmetry. More information also differentiates vendors. Such differentiation increases the sense of trustworthiness of the vendor community; failure to do so causes high-quality vendors to leave the market, leaving the lemon vendors (Akerlof, 1970).
第四，IMs可以通过向客户提供更多信息来建立信任。信息不对称是供应商进行机会主义行为的重要原因。这是一个典型的委托代理问题，其中供应商(代理)比客户(委托人)拥有更多关于其特征、产品和实践的信息，而客户(委托人)缺乏有效的手段来监控供应商(Pavlou等人，2007)。文献表明，客户和供应商之间有意义且及时的信息共享可以促进信任(Jarvenpaa和Leidner, 1997)。反馈机制和认证是帮助客户缓解信息不对称的重要外部信息来源。更多的信息也可以区分供应商。这种区分增加了供应商社区的信任感;不这样做会导致高质量的供应商离开市场，留下柠檬供应商(Akerlof, 1970)。

Despite the rich findings about various effective IMs, existing findings are derived from online product markets and thus may not be readily applicable to online service markets, for the following two reasons. First, the extant findings are focused on IMs that facilitate initial trust formation based on variance models. However, a successful service transaction requires not only initial trust but further development and maintenance of that trust. Therefore, a process model is needed. This process model should reflect the dynamic view of trust, whereby trust starts small and gradually increases as clients and vendors get to know each other (Cheng et al., 2017; Jarvenpaa and Leidner, 1997; Lewicki and Bunker, 1996). Although some extant studies have also taken a dynamic perspective by examining trust development across multiple transactions, for example how trust in an early transaction affects a subsequent transaction (Fang et al., 2014; Kim et al., 2009; Ou et al., 2014; Zahedi and Song, 2008), trust development within a transaction remains obscure.
尽管关于各种有效的IMs有丰富的发现，但现有的发现来自于在线产品市场，因此可能不容易适用于在线服务市场，原因有以下两个。首先，这些发现主要集中在各种便于基于方差模型建立初始信任的IM上。但是，成功的服务交易不仅需要初始信任，而且需要进一步发展和维护该信任。因此，需要一个流程模型。这个过程模型应该反映信任的动态视图，即信任从很小开始，随着客户和供应商的相互了解而逐渐增加(Cheng et al.， 2017;Jarvenpaa和Leidner, 1997;《路易斯基与邦克》，1996)。虽然现有的一些研究也从动态的角度考察了跨多个事务的信任发展，例如早期事务中的信任如何影响后续事务(Fang et al.， 2014;Kim等人，2009;欧等，2014;(Zahedi和Song, 2008)，交易中的信任发展仍然是模糊的。

Second, the literature has mainly focused on IMs that reduce product-specific uncertainty (Angelika et al., 2012; Pavlou et al., 2007). As noted, clients in crowdsourcing face new types of uncertainty that are specific to service transactions, such as progress uncertainty (Veltri et al., 2008; Whitten and Leidner, 2006). Therefore, prior IMs are inadequate and service-specific IMs should be examined. For example, periodical evaluation is an effective IM that can be used to mitigate progress uncertainty in the cooperation phase (Das and Teng, 1998).
其次，文献主要集中在降低产品特定不确定性的IM上（Angelika等，2012； Pavlou等，2007）。如前所述，众包领域的客户面临着针对服务交易的新型不确定性，如进度不确定性(Veltri et al.， 2008;Whitten和Leidner, 2006)。因此，以前的IMs是不充分的，应该研究特定于服务的IMs。例如，定期评估是一种有效的即时通讯工具，可以用来缓解合作阶段的进展不确定性(Das和Teng, 1998)。

To fill these two gaps, our study intends to build a process model of how crowdsourcing intermediaries develop and maintain clients’ trust in vendors throughout a service transaction. In constructing the process model, we pay particular attention to various types of uncertainty that are specific to service transactions and IMs that reduce them.
为了填补这两个空白，我们的研究打算建立一个流程模型，研究众包中介如何在整个服务交易中发展和保持客户对供应商的信任。在构建流程模型时，我们特别关注特定于服务事务和减少它们的IMs的各种类型的不确定性。

Methods
This research adopts a single case study methodology for three reasons. First, its research question is a “how” type and is thus better answered through an inductive method (Walsham, 2006). Second, since this study aims to break new theoretical ground, a case study method is more appropriate due to its strength in exploring new conceptual arguments (Eisenhardt, 1989). Third, a single case study provides rich narratives that allow researchers to distinguish different phases conceptually and build a process model (e.g., Huang et al., 2017; Montealegre, 2002).

3.方法

本研究采用单一个案研究法，原因有三。首先，它的研究问题是“如何”类型的，因此通过归纳方法更好地回答(Walsham, 2006)。其次，由于本研究的目的是要打破新的理论基础，因此案例研究方法更适合探索新的概念论证(Eisenhardt, 1989)。第三，单个案例研究提供了丰富的叙述，使研究人员能够在概念上区分不同的阶段，并建立一个过程模型(如Huang等，2017;蒙特,2002)。

The case setting
We selected ZBJ as the case site. ZBJ was founded in 2006 by six former journalists who were inspired by Alibaba’s success in building online product markets, such as Taobao.com and Tmall.com, and aspired to grow ZBJ into the Alibaba of the service industry. ZBJ started by organizing contests and tenders, but a major breakthrough came in 2013, when it introduced the online store model. In the online store model, every vendor was given a virtual store where they could display their skills, prior projects, and customer feedback. Clients screen these stores to select a vendor with whom to work. In 2015, more than 90% of transactions in ZBJ were made via this model. By that year, ZBJ had accumulated over 10 million service providers and over 5 million clients, and the total volume of service transactions was worth RMB¥ 7.5 billion (about US$ 1.13 billion). Popular services include website development, software development, animation development, logo design, and marketing campaign development.

3.1案例设置

我们选择ZBJ作为案例站点。ZBJ成立于2006年，创始人是六名前记者，他们受到阿里巴巴成功打造淘宝网(Taobao.com)和天猫(Tmall.com)等在线产品市场的启发，立志将ZBJ打造成服务行业的阿里巴巴。ZBJ最初是组织竞赛和投标，但在2013年有了重大突破，推出了在线商店模式。在在线商店模型中，每个供应商都有一个虚拟商店，在那里他们可以展示他们的技能、以前的项目和客户的反馈。客户筛选这些商店以选择与之合作的供应商。2015年，ZBJ 90%以上的交易都是通过这种模式完成的。到那一年，ZBJ已经积累了超过1000万的服务提供商和超过500万的客户，服务交易总额达到75亿人民币(约合11.3亿美元)。流行的服务包括网站开发、软件开发、动画开发、标志设计和营销活动开发。

Although some established companies also use ZBJ, the majority of service providers are small-scale entrepreneurs. Many are fresh college graduates who have been taking on projects via ZBJ since they were at school, skilled workers who earned selfemployment income from a side gig, and small offline service providers who intended to grow their customer base beyond their local communities. These service providers are digital entrepreneurs, because both their acquisition of new business opportunities and delivery of solutions rely on digital technologies (Davidson and Vaast, 2010). First, in terms of acquiring new business opportunities, they are enabled by the intermediary’s digital platform, which not only connects them to clients all over the world but also helps them gain clients’ trust through IT-enabled IMs, such as escrow services and a feedback mechanism. Second, in terms of solution delivery, they rely on digital technologies as well, because their solutions take the form of digital artifacts, such as software packages and websites, and they use digital tools such as Java and HTML to produce such solutions.
虽然一些老牌公司也使用ZBJ，但大多数服务提供商都是小规模的企业家。他们中的许多人都是刚毕业的大学生，从上学起就通过ZBJ接受项目;他们中的许多人都是靠兼职赚取自营收入的技术工人;这些服务提供商都是数字企业家，因为他们获取新的商业机会和提供解决方案都依赖于数字技术(Davidson和Vaast, 2010)。首先，就获得新的商机而言，中介机构的数字平台为他们提供了支持，该平台不仅将它们与世界各地的客户联系起来，而且还帮助他们通过支持it的IMs(如托管服务和反馈机制)来获得客户的信任。其次，在解决方案交付方面，他们也依赖于数字技术，因为他们的解决方案采用数字产品的形式，如软件包和网站，并且他们使用数字工具，如Java和HTML来产生这样的解决方案。

A major challenge faced by these small entrepreneurs is how to gain clients’ trust. Vendors whom we interviewed all acknowledged that they struggled to gain clients’ trust in the open market. Several vendors had their own sales websites, but few orders came from them. According to these vendors, closing a deal on ZBJ was much simpler than doing it on their own websites, because clients were more inclined to trust them and less concerned about payment. Clients whom we interviewed concurred. They added that were it not for ZBJ, they would not have considered working with unknown online vendors, because there was too much uncertainty involved. For ZBJ, trust development is paramount. In fact, the company’s vision is to “Become China’s most trustworthy online service market.”
这些小企业家面临的一个主要挑战是如何赢得客户的信任。我们采访过的供应商都承认，他们很难在开放市场中获得客户的信任。一些供应商拥有自己的销售网站，但很少收到订单。 这些供应商认为，在ZBJ上完成交易比在自己的网站上完成交易简单得多，因为客户更倾向于信任他们，而不太关心付款。我们采访的客户对此表示赞同。他们补充说，如果不是ZBJ，他们不会考虑与未知的在线供应商合作，因为有太多的不确定性。对于ZBJ来说，信任的发展是最重要的。事实上，该公司的愿景是“成为中国最值得信赖的在线服务市场”。

In addition to the traditional IMs such as escrow services and feedback mechanisms, the company has established IMs that are specific to service transactions. One example is the services provided by transaction counselors. Due to their lack of domain expertise, many clients lack confidence in vendors’ proposals during the negotiation phase. To overcome this issue, ZBJ established transaction counseling department, whose purpose is to help clients evaluate proposals and facilitate effective negotiation. This initiative was highly successful, and ZBJ later expanded transaction counselors’ services into requirements analysis and post-contract moderation. In 2016, the transaction counselor department had become one of the largest departments in ZBJ, consisting of over 100 employees. Many of the clients whom we interviewed reported that the availability of transaction counselors was an important reason why they trusted the unknown vendors.
除了传统的IMs(如托管服务和反馈机制)之外，公司还建立了特定于服务事务的IMs。交易顾问提供的服务就是一个例子。由于缺乏领域的专业知识，许多客户在协商阶段对供应商的建议缺乏信心。为了解决这个问题，ZBJ成立了交易咨询部，其目的是帮助客户评估提案，促进有效的谈判。这个项目非常成功，ZBJ后来将交易顾问的服务扩展到需求分析和合同后的审核。2016年，交易咨询部已成为ZBJ最大的部门之一，拥有员工100余人。我们采访的许多客户报告说，交易顾问的可用性是他们信任未知供应商的一个重要原因。

Effective IMs for trust development also attract high-quality vendors. With the support of the transaction counselors, clients are able to better assess vendors’ ability and high-quality vendors stand out. Several of the vendors whom we interviewed also had stores in Taobao, but they were shifting more attention to ZBJ. They stated that an important reason for this was that ZBJ’s IMs were more specific to service transactions. High-quality vendors in turn attract more clients. Today, ZBJ is a household name in the service industry. There is a general perception among the public that: “If you buy a product, go to Taobao or Tmall; if you buy a service, go to ZBJ.”
用于信任发展的有效IM也吸引了高质量的供应商。 在交易顾问的支持下，客户能够更好地评估供应商的能力，从而使高品质的供应商脱颖而出。我们采访过的几个供应商也在淘宝上有商店，但是他们将更多的注意力转移到ZBJ上。 他们表示，这样做的重要原因是ZBJ的IM更特定于服务交易。 高品质的供应商又吸引了更多的客户。 今天，ZBJ在服务行业已是家喻户晓的名字。 公众普遍认为：“如果您购买产品，请去淘宝或天猫购物； 如果您购买服务，请前往ZBJ。”

Data collection
Data collection started in March 2015 and ended in September 2017. We paid four visits to ZBJ’s headquarters. The senior managers were supportive of our research, granting us access to key employees and helping us schedule interviews with clients and vendors. We interviewed 34 informants from ZBJ, seven from clients, and eight from vendors (for the informant list, please see Appendix A). This sampling structure reflects the intermediary perspective of our research. Some informants participated in multiple interviews. Each interview lasted between 45 min and one hour (for sample interview questions, see Appendix B). Interviews were conducted till we reached the point of theoretical saturation, where information started to repeat existing findings (Eisenhardt, 1989). All interviews were recorded and transcribed. The transcripts filled 655 pages.

3.2数据收集

数据收集开始于2015年3月，结束于2017年9月。我们四次参观了ZBJ的总部。高级经理们支持我们的研究，允许我们接触关键员工，并帮助我们安排与客户和供应商的面谈。我们采访了34位来自ZBJ的线人，7位来自客户，8位来自供应商(线人名单见附录A)。这种抽样结构反映了我们研究的中介视角。一些线人参加了多次采访。每次访谈持续45分钟至1小时(样本访谈问题见附录B)。进行访谈直到我们达到理论上的饱和点，那里的信息开始重复现有的发现（Eisenhardt，1989）。所有访谈均被记录和转录。这份记录长达655页。

Informants at ZBJ included senior managers, middle managers, and line employees. This distribution eliminated elite bias (Pan and Tan, 2011). The senior managers included the Chief Operating Officer (COO) and two Vice Presidents (VP), who were also the cofounders. The middle managers and line employees came from three major departments: transaction counseling, vendor management, and recourse. These departments are directly related to vendor-client relationships and involved in ZBJ’s efforts to develop clients’ trust. At the end of each interview, we asked the informant to recommend other individuals who could offer important information for understanding ZBJ’s role in trust development.
ZBJ的线人包括高级经理、中层经理和一线员工。这种分布消除了精英偏见(Pan和Tan, 2011)。高级经理包括首席运营官(COO)和两位副总裁(VP)，他们也是联合创始人。中层管理人员和一线员工来自三个主要部门:交易咨询、供应商管理和资源管理。这些部门与供应商-客户关系直接相关，并参与了ZBJ发展客户信任的努力。在每次访谈结束时，我们都会要求被调查者推荐其他能够提供重要信息的人，以帮助我们理解ZBJ在信任发展中的作用。

Interviews with clients and vendors were used to triangulate information collected from ZBJ (Eisenhardt, 1989; Klein and Myers, 1999). For example, some ZBJ informants highlighted that their ability to recommend capable vendors was an important reason why clients trusted vendors, but clients responded indifferently to this statement. We thus dropped it. We also paid particular attention to informants who were both clients and vendors. In an online product market, a buyer can easily become a seller by selling their personal possessions (Sun, 2010), but in an online service market, certain skills are necessary to become a service provider. Among our informants, two vendors had outsourced tasks to other vendors and thus assumed the role of both vendor and client. We analyzed their interview transcripts and found no apparent differences between their opinions and those of other interviewees regarding ZBJ’s role in trust development.
对客户和供应商的访谈被用来对从ZBJ收集的信息进行三角测量(Eisenhardt, 1989;克莱因和迈尔斯，1999)。例如，一些ZBJ的线人强调，他们推荐有能力的供应商的能力是客户信任供应商的一个重要原因，但是客户对这种说法反应冷淡。因此，我们放弃了它。 我们还特别注意既是客户又是供应商的线人。在网络产品市场中，买家可以通过出售个人物品轻松地成为卖家(Sun, 2010)，但在网络服务市场中，成为服务提供商需要具备一定的技能。在我们的线人中，有两个供应商已经将任务外包给其他供应商，因此承担了供应商和客户的角色。我们分析了他们的访谈记录，发现他们对ZBJ在信任发展中的作用的看法与其他受访者没有明显的差异。

Data also came from archival data and observations. Archival data included ZBJ’s internal documents and public information. Examples of internal documents included progress reports, training presentation slides, and arbitration records; examples of public information included news articles and analyst reports. The public information was rich, given that ZBJ was a high-profile company that had received a great deal of media attention. The archival data amounted to 712 pages. In December 2016, one of the authors was invited to give a keynote speech at a large trade conference organized by ZBJ. At the conference, the research team attended presentations and panel discussions given by ZBJ’s senior managers, clients, vendors, government officials, and industry analysts. This experience allowed us to collect more information and to develop a more in-depth understanding of the research context (Klein and Myers, 1999).
数据也来自档案数据和观察。 档案数据包括ZBJ的内部文件和公共信息。 内部文件的示例包括进度报告，培训演示幻灯片和仲裁记录； 公共信息的示例包括新闻文章和分析师报告。由于ZBJ是一家备受瞩目的公司，受到了媒体的广泛关注，所以它的公开信息非常丰富。档案资料共712页。2016年12月，作者之一受邀在ZBJ举办的大型贸易会议上发表主题演讲。在会议上，研究团队参加了ZBJ的高级经理、客户、供应商、政府官员和行业分析师的演讲和小组讨论。这一经历使我们能够收集更多的信息，并对研究背景有更深入的理解(Klein和Myers, 1999)。

Data analysis
We used open, axial, and selective coding in the data analysis (Corbin and Strauss, 1990). For an example of data structure derived from data analysis, please refer to Fig. 1. First, we started the open coding by conceptually labeling the data to capture ZBJ’s efforts to develop and maintain clients’ trust in vendors. The research team met regularly to review emerging concepts and ensure coding consistency (Klein and Myers, 1999). For example, when we asked ZBJ employees how they maintained clients’ trust in vendors during the cooperation phase, we derived concepts such as Perspective Taking, Communication Bridge, and Alerts for Potential Conflicts.

3.3数据分析

我们在数据分析中使用了开放、轴向和选择性编码(Corbin和Strauss, 1990)。由数据分析得到的数据结构示例如图1所示。首先，我们通过概念上标记数据来开始开放编码，以捕获ZBJ开发和维护客户对供应商的信任的努力。研究小组定期开会审查新出现的概念并确保编码的一致性(Klein和Myers, 1999)。例如，当我们询问ZBJ的员工在合作阶段如何维护客户对供应商的信任时，我们得到了一些概念，例如“观点接受”，“沟通桥梁”和“潜在冲突警报”。

Second, we started the axial coding by grouping similar concepts from the open coding into major themes (Corbin and Strauss, 1990). The grouping process was also informed by the literature. Continuing with the example above, the three concepts were grouped into a theme named Harmonious Conflict Resolution. This construct was proposed by Goo et al. (2009) in a study on IT outsourcing relationship government. It refers to “The extent to which parties achieve mutually satisfying resolutions of their conflicts and, thus, disagreements are replaced by agreement and consensus” (p. 125). Our data suggest that in the crowdsourcing context, harmonious conflict resolution is an important IM deployed by the intermediary to maintain trust. This mechanism has not been discussed in the literature on e-Commerce trust development and is thus an extension of the literature.
其次，我们开始轴向编码，将类似的概念从开放编码组织成主要的主题(Corbin和Strauss, 1990)。分组过程也从文献中获悉。继续上面的示例，这三个概念被分组为一个名为“和谐冲突解决”的主题。Goo等人(2009)在IT外包关系政府研究中提出了这一概念。它指“各方在何种程度上实现冲突的相互满意的解决，从而以协议和协商一致意见取代分歧”(p. 125)。我们的数据表明，在众包环境中，和谐的冲突解决方案是中介机构为维护信任而部署的重要IM。这一机制在电子商务信任发展的文献中尚未被讨论，因此是文献的延伸。

Third, we started the selective coding by integrating themes from the axial coding into a coherent model. The objective was to create a model that ascertains the phenomenon theoretically. This phase was conducted by constantly comparing data, emerging themes, and the literature (Corbin and Strauss, 1990). Ultimately, we derived a three-phase process model. This emergent model was presented to key informants from ZBJ, clients and vendors, who confirmed that it was an accurate reflection of the reality and provided feedback that helped us to refine it (Pan and Tan, 2011).
第三，我们通过将轴向编码的主题整合到一个连贯的模型中来开始选择性编码。目的是创建一个从理论上确定现象的模型。这个阶段通过不断比较数据、新兴主题和文献来进行(Corbin和Strauss, 1990)。最后，我们导出了一个三相流程模型。这个紧急模型被提交给了来自ZBJ、客户和供应商的关键线人，他们确认这是对现实的准确反映，并提供反馈帮助我们完善它(Pan and Tan, 2011)。

Findings
Our data analysis reveals a three-phase process model whereby a crowdsourcing intermediary initiates, augments, and maintains clients’ trust in vendors and, as a result, facilitates a successful transaction (see Fig. 2). As shown by the literature, an intermediary develops clients’ trust by developing IMs that mitigate uncertainty. Therefore, in each phase, our analysis focuses on uncertainty and the IMs that mitigate it.

4.发现

我们的数据分析揭示了一个三阶段的过程模型，在此过程中，众包中介启动、增强和维护客户对供应商的信任，从而促进成功的交易(见图2)。因此，在每个阶段，我们的分析集中在不确定性和减轻不确定性的IMs上。

Initiating trust in the vendor selection phase
A service transaction starts with the vendor selection phase, where a client screens the vendor community and selects one to work with. In this phase, clients face uncertainty about vendors’ integrity and ability. Nearly all of the clients whom we interviewed raised their concern about vendors’ opportunistic behaviors, for example a vendor taking their money without properly delivering the service purchased. Another major concern was whether vendors had the ability to fulfill their requirements.

4.1在供应商选择阶段启动信任

服务交易从供应商选择阶段开始，在此阶段，客户筛选供应商社区并选择要使用的社区。在这个阶段，客户面临供应商的诚实性和能力的不确定性。我们采访的几乎所有客户都对供应商的机会主义行为表示了担忧，例如供应商在没有正确交付所购买的服务的情况下就拿走了他们的钱。另一个主要问题是供应商是否有能力满足他们的需求。

To mitigate vendor uncertainty, ZBJ has deployed several IMs. The key ones include escrow services, a feedback mechanism, and an accreditation system. Clients feel more secure and comfortable about transferring money through the escrow services offered by ZBJ. In the early days, some vendors tried to persuade clients to settle payment offline so that they could avoid the commission charged by ZBJ, which was about 20% of the payment. Vendors also promised clients a share of the commission. Many of the clients whom we interviewed received such offers and most rejected them, because it was too risky to transfer money to vendors directly. Moreover, after transferring the money, clients lose control over vendors. Today, such solicitation from vendors is less common. Vendors have realized that the escrow services can help them quickly establish trust with clients. A website development vendor stated:
“You can show clients your business licenses and various certificates to demonstrate that you are legitimate, but clients still do not trust you. They trust ZBJ with their money and so it is much easier to close a deal on ZBJ.”
为了减轻供应商的不确定性，ZBJ部署了多个IM。关键包括托管服务，反馈机制和认证系统。通过ZBJ提供的托管服务转账，客户会感到更安全、更放心。在早期，一些供应商试图说服客户离线付款，这样他们就可以避免ZBJ收取的佣金，大约是付款的20%。供应商还向客户承诺将分享佣金。我们采访的许多客户都收到了这样的报价，但大多数都拒绝了，因为直接把钱转给供应商风险太大。并且在转移资金之后，客户就对供应商失去了控制。如今，来自供应商的这种要求已经不那么普遍了。供应商已经意识到托管服务可以帮助他们快速建立与客户的信任。一个网站开发供应商声明:

“你可以向客户展示你的营业执照和各种证书，以证明你是合法的，但客户仍然不信任你。他们把钱交给了ZBJ，所以更容易达成交易。”

The escrow services allow clients to pay vendors periodically based on vendors’ performance. For a regular project, payment is broken into three installments: the initial, interim, and final payment. Initial payment is automatically paid at the beginning of the contract, but clients authorize the interim and final payments only when they are satisfied with the deliverables.
第三方托管服务允许客户根据供应商的表现定期向供应商付款。对于一个常规项目，付款分为三个部分:初始付款、中期付款和最终付款。初始付款在合同开始时自动支付，但客户只有在对交付物满意时才授权中期和最终付款。

The escrow services also include a recourse service. Clients can request recourse when they are not satisfied with the final output. Simple requests are handled by the transaction counselors. Some requests are more complex, because they involve conflicts and require arbitration. For example, a client may deem that a vendor has failed to provide a service in line with prior agreements, while the vendor claims that they have. Arbitrations are handled by designated committees. The committees include members from ZBJ, the client community, and the vendor community to ensure fair verdicts. ZBJ has built a reputation for fair recourse and arbitration procedures, and this reputation has attracted both clients and vendors. Although the procedures are fair, once a client lodges a recourse request or is involved in an arbitration, they have already lost trust in the vendor. To prevent clients from reaching this stage, ZBJ attempts to resolve conflicts harmoniously before they escalate. We will describe this mechanism in the cooperation phase.
托管服务还包括追索服务。当客户对最终的输出结果不满意时，可以请求追索权。简单的请求由交易顾问处理。有些请求比较复杂，因为它们涉及冲突，需要仲裁。例如，客户可能认为供应商没有按照之前的协议提供服务，而供应商则声称他们有。仲裁由指定的委员会处理。委员会包括来自ZBJ、客户社区和供应商社区的成员，以确保公平的裁决。ZBJ在公平申诉和仲裁程序方面建立了声誉，这种声誉吸引了客户和供应商。虽然程序是公平的，一旦客户提出追索权要求或参与仲裁，他们已经失去了对供应商的信任。为了防止客户达到这个阶段，ZBJ试图在冲突升级之前和谐地解决冲突。我们将在合作阶段介绍这一机制。

The second effective IM in the vendor selection phase is a feedback mechanism, which allows potential clients to assess a vendor’s integrity. All the clients we interviewed stated that they relied heavily on previous clients’ feedback in making purchase decisions. They believed that other clients had no incentive to provide inaccurate information. Many stated that they would not consider a vendor who had no feedback comments or only a few. Feedback also allows clients to assess a vendor’s ability. To make assessment more effective, ZBJ links feedback comments to specific projects. According to the head of the transaction counseling department:
“A website development vendor may be known for her PHP development skills, but may not be good at H5 development. It’s important to link comments to specific projects.”
供应商选择阶段的第二个有效IM是反馈机制，它允许潜在客户评估供应商的诚实性。我们采访的所有客户都表示，他们在做出购买决定时非常依赖以前客户的反馈。他们认为其他客户没有提供不准确信息的动机。许多人表示，他们不会考虑没有反馈意见或只有少数反馈意见的供应商。反馈还允许客户评估供应商的能力。为了使评估更有效，ZBJ将反馈意见链接到特定的项目。据交易咨询部门负责人表示:

“一个网站开发供应商可能因为她的PHP开发技能而出名，但可能不擅长H5开发。将评论与具体项目联系起来很重要。”

Given the importance of the feedback, there were cases where vendors created a fake client account to transact with themselves and leave positive reviews. In response, ZBJ imposed severe penalties on such fraud. A vendor’s store can be shut down for several months.
考虑到反馈的重要性，有些情况下，供应商会创建一个虚假的客户帐户来与自己进行交易，并留下正面的评论。作为回应，ZBJ对此类欺诈行为实施了严厉的惩罚。一个供应商的商店可以关闭几个月。

Another IM deployed in this phase is the accreditation system, which ranks vendors on 81 levels. The accreditation is based on a sophisticated methodology that considers nearly all factors related to a vendor’s credibility, such as the number of transactions, revenue, customer satisfaction, and number of recourses. However, despite its sophistication, clients still prefer the feedback comments. Many of the clients we interviewed stated that they were less interested in vendors’ accreditation than what other clients had said about them. However, when two vendors have received similar feedback comments, clients will choose the one with the higher accreditation level.
The escrow services, feedback mechanism, and accreditation system build clients’ initial trust in unknown vendors and lead clients to the negotiation phase.
此阶段中部署的另一个IM是认证系统，该系统将供应商排名在81个级别上。 该认证基于一种复杂的方法，该方法几乎考虑了与供应商信誉相关的所有因素，例如交易数量，收入，客户满意度和资源追索数量。然而，尽管它很复杂，客户仍然喜欢反馈意见。我们采访的许多客户表示，他们对供应商认证的兴趣不如其他客户对他们的评价。然而，当两个供应商收到类似的反馈意见时，客户将选择更高认证级别的供应商。

第三方托管服务、反馈机制和认证体系建立了客户对未知供应商的最初信任，并将客户引入谈判阶段。

Augmenting trust in the negotiation phase
In this phase, clients negotiate with the chosen vendor concerning deliverables and contract prices. Unlike products, services are customized based on clients’ specific requirements and those requirements are often vague and evolving. Uncertainty thus arises. We label the uncertainty arising from vague and evolving requirements requirement uncertainty.

4.2在谈判阶段增加信任

在此阶段，客户与选定的供应商就交付物和合同价格进行协商。与产品不同的是，服务是根据客户的特定需求定制的，而这些需求通常是模糊和不断发展的。因此产生的不确定性。我们将由模糊的和不断发展的需求引起的不确定性标记为需求不确定性。

Several clients stated that the results of more standard services, such as translation and copy editing, were in general predictable and satisfactory. However, the results of services that required more customized requirements were less predictable. One client illustrated this by comparing two projects that he had recently completed. The first was a Quick Response (QR) code development project in which he knew exactly what he wanted and the project specifications were clear. The project was successful, and the final output even exceeded his expectations. The second project was in website development, and he was less certain about the requirements. The vendor he worked with did not help him much in clarifying the requirements, and the final output did not meet his expectations.
一些客户说，比较标准的服务，例如翻译和编辑，其结果一般是可以预测和令人满意的。然而，需要更多定制需求的服务的结果是不那么可预测的。一位客户通过比较他最近完成的两个项目来说明这一点。第一个是一个快速响应(QR)代码开发项目，在这个项目中，他确切地知道他想要什么，并且项目规范是清晰的。项目很成功，最终的产出甚至超出了他的预期。第二个项目是网站开发，他对需求不是很确定。与他一起工作的供应商并没有帮助他明确需求，最终的输出也没有达到他的期望。

Most clients are not domain experts in the services they purchase. They lack the domain knowledge to clearly articulate their requirements. Many clients start negotiations with a vague requirement, such as “building an excellent website that reflects our company’s vision.” Because of the vague requirements, clients also find it difficult to assess vendors’ proposals. Many clients suspect that the proposals they receive do not address their requirements, but are most profitable for vendors. Moreover, the inability to translate requirements into clear specifications leaves clients concerned about whether vendors have understood them and can think about the project from their perspective. In line with this, some clients prefer to work with vendors in the same city so that they can discuss the requirements in face-to-face meetings.
大多数客户并不是他们所购买服务的领域专家。他们缺乏领域知识来清楚地表达他们的需求。许多客户开始谈判时都有一个模糊的要求，比如“建立一个反映公司愿景的优秀网站”。“由于模糊的要求，客户也发现很难评估供应商的建议。许多客户怀疑他们收到的提案没有满足他们的需求，但对供应商来说是最有利可图的。此外，由于无法将需求转换为清晰的规范，客户会担心供应商是否理解了它们，是否能够从他们的角度考虑项目。与此相一致，一些客户更喜欢与同一城市的供应商一起工作，这样他们可以在面对面的会议中讨论需求。

Some uncertainty comes from the fact that requirements evolve. For example, many software development clients only have a basic idea at the beginning and lack a clear blueprint. These clients are concerned about whether vendors will be too restrictive in their scope and charge them for every additional change beyond the initial scope. A client looking for a website development service stated:
“Vendors want us to give them everything clearly so that they can just spend 1 or 2 days to make 1 or 2 thousand RMB. But it’s difficult to give all the specifications up front. If I can do so, I probably don’t need their help.”
需求不断演变这一事实带来了一些不确定性。例如，许多软件开发客户在一开始只有一个基本的想法，缺乏一个清晰的蓝图。这些客户关心的是供应商是否在他们的范围内有太多的限制，并对超出初始范围的每一个额外的变更收取费用。一个寻求网站开发服务的客户说:

“供应商希望我们把所有的东西都给他们，这样他们就可以用一两天的时间赚到1到2千元。但是很难预先给出所有的规格。如果我能做到这一点，我可能就不需要他们的帮助了。”

ZBJ has deployed IMs to reduce requirements uncertainty and to facilitate transaction success. The key IMs include requirements analysis services and contract formation services. ZBJ deploys transaction counselors to help clients produce concrete requirements. Sometimes, adjustments are needed, because the requirements are unrealistic. A transaction counselor for website and software development stated that:
“Last week, I was approached by a client who wanted to build an App, but only had a budget of 5 k. A vendor asked for 100 thousand, which was not unreasonable, but the client felt it was outrageous. After analyzing his requirements, I recommended a Wechat corporate account, which was sufficient for his needs and much cheaper”
ZBJ已经部署了IMs来减少需求的不确定性并促进事务的成功。关键的IMs包括需求分析服务和合同形成服务。ZBJ配备交易顾问，帮助客户提出具体要求。有时，需要进行调整，因为需求是不现实的。一个网站和软件开发的交易顾问说:

“上周，一个客户找到我，他想开发一款应用，但预算只有5000美元。一个供应商要价10万美元，这不是不合理的，但客户觉得这太离谱了。”在分析了他的需求后，我推荐了一个微信公司账户，这足以满足他的需求，而且更便宜。”

Moreover, most clients do not have the skills necessary to draft a complete service contract, in the form of a service level agreement (SLA). They do not trust SLAs drafted by vendors either. In this case, ZBJ provides SLA templates which can translate client requirements into SLAs. SLAs are critical to successful transactions. According to a vendor manager:
“Some vendors don’t spend much time on scoping projects. They just rush into an agreement. But later there are all kinds of issues and they may say, ‘I cannot do this or that, because you did not mention this before’.”
此外，大多数客户都不具备以服务水平协议(SLA)的形式起草完整服务合同所需的技能。他们也不相信供应商起草的sla。在这种情况下，ZBJ提供了SLA模板，可以将客户需求转换为SLA。sla对于成功的事务至关重要。一位供应商经理说:

“有些供应商不会在确定项目范围上花太多时间。他们只是匆匆达成了协议。但后来出现了各种各样的问题，他们可能会说，‘我不能做这个或那个，因为你之前没有提到这个’。”

Each service type has its own SLA template. The design of the templates benefits from ZBJ’s experience in conflict resolution. In areas that are likely to lead to conflicts, the templates request specifications. Take website development as an example. The template requests specific information about the number of pages, the maximum number of concurrent users, the upgrade schedule, and the ownership of source codes. Many first-time clients are not aware of the issue of source code ownership, and most assume that they own the source codes. When clients claim source codes, some vendors request additional payment. This has caused many conflicts. Now, the ownership terms are included in the SLA templates. In this case, clients are better informed and there are fewer conflicts concerning this matter.
每种服务类型都有自己的SLA模板。模板的设计得益于ZBJ在冲突解决方面的经验。在可能导致冲突的区域中，模板要求规范。以网站开发为例。模板请求关于页面数量、并发用户的最大数量、升级计划和源代码所有权的特定信息。许多初次使用的客户不知道源代码所有权的问题，并且大多数客户认为他们拥有源代码。当客户要求获得源代码时，一些供应商要求额外的付款。这引起了许多冲突。现在，所有权条款已包含在SLA模板中。 在这种情况下，可以更好地通知客户，与此相关的冲突也更少了。

Transaction counselors are also part of the contract formation services. They play an important role in moderating prices. Because services are not standardized, two vendors can propose very different prices. For example, the price for a website development project can range from a few hundred to a few thousand RMB. This makes clients anxious about whether some vendors are taking advantage of them. Transaction counselors monitor contract prices and intervene if a vendor proposes an unreasonable price. These counselors also ensure that contracts have the flexibility to handle future changes. These flexible arrangements are often based on implicit terms, and transaction counselors act as witnesses to those terms.
The requirements analysis services and contract formation services facilitate vendors in further developing clients’ initial trust in them and this augmented trust leads clients to the cooperation phase.
交易顾问也是合同形成服务的一部分。他们在调节价格方面起着重要作用。由于服务没有标准化，两个供应商可以提出非常不同的价格。例如，一个网站开发项目的价格从几百元到几千元不等。这让客户担心一些供应商是否在利用他们。交易顾问监控合同价格，并在供应商提出不合理的价格时进行干预。这些顾问还确保合同具有处理未来变化的灵活性。这些灵活的安排通常是基于隐性条款，交易顾问是这些条款的见证人。

需求分析服务和合同形成服务有助于供应商进一步发展客户对他们的最初信任，这种增强的信任将客户带入合作阶段。

Maintaining trust in the cooperation phase
Unlike products, services are not readily available for shipment after the contract is signed. They require cooperation between clients and vendors. In this phase, clients face uncertainty about progress. We label this uncertainty about project status and vendor commitment progress uncertainty. Clients feel insecure when they are not aware of a project’s status. Many client interviewees stated the importance of being on the same page as vendors. While established service providers have standard updating procedures, small vendors often fall short in this aspect, in particular those founded by college students or fresh college graduates. In our interviews, clients often praised these young service providers for their creativity and enthusiasm, but expressed concern about their lack of professionalism. One client stated that:
“I once worked with a college student who halted the development for two weeks, because he had an exam to prepare for … But that boy was a very good programmer. He now works for Microsoft.”

4.3在合作阶段保持信任

与产品不同的是，合同签订后并不容易提供服务。他们需要客户和供应商之间的合作。在此阶段，客户面临进度的不确定性。我们将这种不确定性标记为项目状态和供应商承诺进度的不确定性。当客户不知道项目的状态时，他们会感到不安全。许多客户受访者表示了与供应商保持一致的重要性。虽然现有的服务供应商有标准的更新程序，但小供应商在这方面往往做得不够，特别是那些由大学生或应届大学毕业生创建的供应商。在我们的采访中，客户经常称赞这些年轻的服务提供商的创造力和热情，但对他们缺乏专业精神表示担忧。一个客户说:

“我曾经和一个大学生一起工作，他停止了开发两个星期，因为他要准备一场考试……但那个男孩是一个非常好的程序员。”他现在在微软工作。”

Second, clients also have concerns regarding vendors’ commitment during project implementation. Several clients reported post contract shirking. According to ZBJ employees who manage vendors, an important reason for such shirking is that vendors tend to underestimate the workload and take on projects that exceed their capacity. A client who purchased a photo editing service stated that:
“I once had over 20 pictures that needed to be edited. The first few were edited well, but then the quality started to decline. Later, I learned that the first batch were edited by the director, who then passed the project to a new employee.”
其次，客户还担心供应商在项目实施期间的承诺。一些客户报告称，他们在合同到期后逃避责任。ZBJ负责管理供应商的员工表示，这种逃避的一个重要原因是，供应商往往会低估工作量，承担超出其能力的项目。一位购买了照片编辑服务的客户说:

“我曾经有20多张照片需要编辑。最初的几部编辑得很好，但后来质量开始下降。后来我了解到，第一批是由主任编辑的，然后他把项目交给了一名新员工。”

Several IMs have been deployed to reduce progress uncertainty. The key ones include periodical evaluation and harmonious conflict resolution. ZBJ provides clients with an online management tool to monitor project progress. The tool allows clients to establish check-up points and link check-up results to payment. Vendors who fail a check-up will not receive subsequent payments. If a client deems that a project is not going in the intended direction or that the vendor is unable to complete the task, they can cancel the project and stop further payment. When a vendor has failed several check-ups in its portfolio, ZBJ does not allow it to take on new projects and requests it to focus on projects already underway.
已部署了多个IM，以减少进度不确定性。 关键包括定期评估和解决冲突。 ZBJ为客户提供了一个在线管理工具来监视项目进度。该工具允许客户建立检查点，并将检查结果与付款链接起来。没有通过检查的供应商将不会收到后续付款。如果客户认为项目没有按照预期的方向进行，或者供应商无法完成任务，他们可以取消项目并停止进一步付款。当一个供应商在其产品组合中几次检查失败时，ZBJ不允许它接受新的项目，并要求它专注于已经在进行的项目。

Some clients lack the domain expertise to evaluate progress. Such clients can seek help from transaction counselors. One transaction counselor stated that:
“We usually step in at the middle of a project to check whether the vendor has the ability to deliver and whether the deliverables are satisfactory. Before the deadline, we may step in again to prevent issues that may cause conflicts.”
一些客户缺乏评估进度的领域专业知识。 这些客户可以寻求交易顾问的帮助。 一位交易顾问说：

“我们通常会在项目进行到一半时介入，检查供应商是否有能力交付，以及交付结果是否令人满意。”在最后期限之前，我们可能会再次介入，防止可能导致冲突的问题。”

Moreover, conflicts are inherent in interorganizational relationships like outsourcing, given the factors of goal divergence and vendors’ tendency toward opportunism. If conflicts are not resolved properly, they can damage trust and cause early termination of contracts. There are two types of conflict resolution. One is through arbitration. As mentioned above, once a client reaches this stage, they have already lost trust in the vendor. The other is through harmonious conflict resolution, which aims to form a consensus and achieve mutually a satisfactory resolution (Goo et al., 2009). The key to harmonious conflict resolution is mitigating misunderstandings and creating a shared understanding. For example, in software development projects, it is common for a client to request a seemingly trivial change which in fact requires major technical reconfigurations. It is reasonable for a vendor to refuse such a change, but the client may see it as shirking and start to doubt the vendor’s commitment. In this case, transaction counselors can act as mediators to resolve the misunderstanding.
此外，考虑到目标分歧和供应商的机会主义倾向等因素，冲突在外包等组织间关系中是固有的。如果冲突没有得到妥善解决，它们会损害信任并导致合同提前终止。有两种解决冲突的方法。一是通过仲裁。如上所述，一旦客户到了这个阶段，他们就已经对供应商失去了信任。另一种是通过和谐的冲突解决，其目的是形成共识，实现双方满意的解决(Goo et al.， 2009)。和谐解决冲突的关键是减少误解和建立共识。例如，在软件开发项目中，客户通常要求进行看似微不足道的更改，而实际上这需要进行重大的技术重新配置。供应商拒绝此类更改是合理的，但客户可能会认为此行为是在逃避，并开始怀疑供应商的承诺。在这种情况下，交易顾问可以充当调解人来解决误解。

Several transaction counselors used marriage as a metaphor in referring to the cooperation phase and described their role as a marriage counselor. As a marriage counselor, a transaction counselor also encourages clients and vendors to communicate as much as possible to reach a mutual understanding and to avoid misunderstandings. For example, when a client and its vendor have not communicated for a while, a trigger is sent to the responsible transaction counselor, who will then look into it and prevent conflicts from emerging or escalating. According to one senior transaction counselor:
“Based on our experience, if a client and a vendor have not talked for a while, it is not a good sign. They are usually not happy with each other, and a conflict will follow if we don’t take action. ”
The periodical evaluation and harmonious conflict resolution enable vendors to maintain clients’ trust during the cooperation phase. A trusted interorganizational relationship leads to fruitful results and a successful transaction.
一些交易顾问在提到合作阶段时使用婚姻作为隐喻，并描述了他们作为婚姻顾问的角色。作为婚姻顾问，交易顾问也鼓励客户和卖方尽可能多地沟通，以达成相互理解，避免误解。例如，当客户与其供应商一段时间未进行通信时，将向负责的交易顾问发送触发器，然后由后者进行调查并防止冲突发生或升级。一位资深交易顾问说：

“根据我们的经验，如果客户和供应商有一段时间没有谈过，那不是一个好兆头。他们通常彼此不满意，如果我们不采取行动，就会发生冲突。 ”

定期评估和和谐的冲突解决方案使供应商在合作阶段可以保持客户的信任。值得信赖的组织间关系可以带来丰硕的成果和成功的交易。

Discussion
In this section, we discuss our findings in light of earlier related work. The discussion serves two purposes. First, it explains how the process model is corroborated by the literature, to show that the model can be generalized beyond this single case. Second, it explains how the process model extends the literature. This section consists of two parts. The first describes the process model of intermediary-based trust development, which consists of IMs that are specific to service transactions. Next, it describes these IMs in detail.

5.讨论

在本节中，我们将根据之前的相关工作来讨论我们的发现。讨论有两个目的。首先，它解释了流程模型是如何被文献证实的，以表明该模型可以推广到这个单一案例之外。其次，它解释了流程模型如何扩展文献。本节包括两个部分。 第一部分描述了基于中介的信任开发的过程模型，该模型包含特定于服务交易的IM。 接下来，将详细描述这些IM。

A process model of intermediary-based trust development
The process model of intermediary-based trust development and maintenance involves three phases which correspond to the three phases of a service transaction: Vendor selection, negotiation, and cooperation. This model confirms our postulation that a crowdsourcing intermediary should be involved in trust development throughout the entire transaction, rather than solely in the vendor selection phase. This is a major difference between online product markets and online service markets. In online product markets, trust built in the vendor selection phase can lead to the purchasing decision (Gefen et al., 2003; McKnight et al., 2002), but in service markets, this initial trust needs to be further augmented before a purchasing decision can be made.

5.1基于中介的信任开发过程模型

基于中介的信任开发和维护的流程模型包括三个阶段，分别对应于服务事务的三个阶段:供应商选择、谈判和合作。这个模型证实了我们的假设，即一个众包中介应该参与整个交易的信任开发，而不仅仅是在供应商选择阶段。这是在线产品市场和在线服务市场的主要区别。在在线产品市场中，在供应商选择阶段建立的信任会导致购买决策(Gefen et al.， 2003;但是在服务市场，在做出购买决定之前，这种最初的信任需要进一步增强。

Although product transactions may also involve negotiation, the negotiation is usually about price haggling, which involves limited uncertainty (Ou et al., 2014). In contrast, negotiation in service transactions involves high requirements uncertainty. Research has shown that difficulties in requirements specification are an important source of uncertainty faced by clients (Ye and Kankanhalli, 2015). This uncertainty is further accentuated by miscommunication between clients and vendors (Qi and Mao, 2016). Also, clients face uncertainty about whether vendors will be too restrictive in the service scope and charge them excessively for changes beyond the initial requirements (Koh et al., 2004). Unlike product transactions, many requirements in service transactions are constantly evolving (Conboy, 2009) and hard to specify accurately and completely upfront. Effectively mitigation of the requirements uncertainty further augments initial trust, and this augmented trust leads to the purchasing decision. Otherwise, requirements uncertainty could disrupt trust development and prevent clients from signing contracts.
虽然产品交易也可能涉及谈判，但谈判通常是关于价格的讨价还价，这涉及有限的不确定性(Ou et al.， 2014)。而服务交易中的谈判则包含了较高的需求不确定性。需求规范中的困难是客户面临的不确定性的重要来源（Ye and Kankanhalli，2015）。客户和供应商之间的沟通不畅进一步加剧了这种不确定性（Qi和Mao，2016）。此外，客户还面临不确定性，即供应商是否会在服务范围内限制过多，并对超出初始需求的变更收取过高的费用(Koh et al.， 2004)。与产品交易不同，服务交易中的许多需求是不断发展变化的(Conboy, 2009)，并且很难准确和完全地预先指定。有效地降低需求的不确定性进一步增强了初始信任，而这种增强的信任会导致购买决策。否则，需求的不确定性可能会破坏信任开发，并阻止客户签署合同。

After a contract is signed, clients and vendors start to cooperate. This cooperation phase does not exist in online product purchase, because products are manufactured without clients’ involvement. The cooperation phase is also filled with uncertainty, especially when there is no face-to-face communications (Jarvenpaa and Leidner, 1997). Our study identifies progress uncertainty, which concerns project status and vendor commitment. This uncertainty has been discussed in the IT outsourcing literature. For example, Liu et al. (2011) find that many vendors fail to keep their clients informed of progress and lose clients’ trust as a result. Similarly, Whitten and Leidner (2006) find that many clients feel that vendors engage in post-contract shirking and thus lose confidence in vendors’ commitment. The mitigation of progress uncertainty maintains the trust built in earlier phases and ensures the final success of the transaction.
签订合同后，客户和供应商开始合作。 在线产品购买中不存在此合作阶段，因为产品的制造没有客户的参与。 合作阶段也充满不确定性，尤其是当没有面对面的交流时（Jarvenpaa和Leidner，1997）。我们的研究确定了进度不确定性，该不确定性与项目状态和供应商承诺有关。例如，Liu et al.(2011)发现，许多供应商未能及时告知客户进展情况，从而失去了客户的信任。类似地，Whitten和Leidner(2006)发现，许多客户认为供应商从事合同后推诿，因此对供应商的承诺失去信心。减轻进度不确定性可维持早期阶段建立的信任，并确保交易最终成功。

IMs specific to service transactions
To initiate trust, ZBJ deploys escrow services, a feedback mechanism, and an accreditation system. These IMs have been the subject of prior studies on online product markets (Ba and Pavlou, 2002; Pavlou, 2002; Pavlou and Gefen, 2004). The consistency can be explained by the fact that these IMs are deployed to mitigate vendor uncertainty, which is an issue common to both online product and service markets. One nuanced difference is that escrow services are not based on lump-sum payment, as in online product markets (Pavlou and Gefen, 2004), but based on installment payment, whereby clients authorize periodical payment when they are satisfied with the periodical deliverables.

5.2特定于服务交易的IM

为了启动信任，ZBJ部署了托管服务，反馈机制和认证系统。这些IMs是先前关于在线产品市场的研究的主题(Ba和Pavlou, 2002;Pavlou, 2002;Pavlou和Gefen, 2004)。可以通过以下事实来解释一致性：部署这些IM可以减轻供应商的不确定性，这是在线产品和服务市场共同面临的问题。 一个细微的差别是，代管服务不是像在线产品市场那样基于一次性付款（Pavlou和Gefen，2004），而是基于分期付款，即客户对定期交付物感到满意时可以授权定期付款。

In contrast, IMs deployed in the negotiation and cooperation phases have not previously been addressed in the literature. These IMs are deployed to mitigate requirements and progress uncertainty, which are specific to service transactions. Although these IMs have not been examined in the literature, the ways in which they develop or maintain trust can be corroborated by the literature. As a result, they have the potential to be generalized beyond this single case.
相比之下，在谈判和合作阶段部署的IMs在以前的文献中没有被提及。部署这些IMs是为了减少特定于服务事务的需求和进度不确定性。虽然这些IMs还没有在文献中进行研究，但它们发展或维持信任的方式可以得到文献的证实。因此，它们有可能被超越此单一案例而被推广。

To mitigate requirements uncertainty, ZBJ deploys requirements analysis services and contract formation services. Although vendors offer requirements analysis, clients may not trust the results. Research has shown that analysis by a neutral third party is considered more trustworthy (Lim et al., 2011). These ZBJ services also enable clients to articulate their requirements clearly and reduce misunderstandings between clients and vendors (Qi and Mao, 2016). Misunderstandings are a common factor that disrupts trust development (Goo et al., 2009). These requirements analysis services also reduce the information asymmetry between clients and vendors, and thus alleviate the principal-agent issue (Pavlou et al., 2007). They enable clients to uncover vendors’ hidden intentions and to better assess vendors’ ability. The former leads to integrity trusting belief and the latter to ability trusting belief (Mayer et al., 1995).
为了减少需求的不确定性，ZBJ部署了需求分析服务和合同形成服务。尽管供应商提供需求分析，但是客户可能不相信结果。研究表明，中立第三方的分析被认为更可信(Lim et al.， 2011)。这些ZBJ服务还使客户能够清楚地表达他们的需求，减少客户和供应商之间的误解(Qi和Mao, 2016)。误解是破坏信任发展的一个常见因素(Goo et al.， 2009)。这些需求分析服务还减少了客户和供应商之间的信息不对称，从而缓解了委托代理问题(Pavlou et al.， 2007)。它们使客户能够发现供应商隐藏的意图，并更好地评估供应商的能力。前者导致诚信信任信念，后者导致能力信任信念(Mayer et al.， 1995)。

Contract formation services reduce requirements uncertainty by specifying requirements and vendors’ responsibilities in SLAs (Mayer et al., 1995). Unlike SLAs in offline markets, these SLAs are not enforced by governmental agencies, but by the intermediary itself, which monitors vendors’ activities and penalizes activities that deviate from agreements (Pavlou, 2002). This is consistent with a prior finding that traditional legal protection is not effective in online markets (Pavlou and Gefen, 2004). To ensure that contracts have flexibility and clients are not charged excessively for every change beyond the initial requirements, transaction counselors act as witnesses to the formation of flexible arrangements, which are often based on unwritten, implicit terms, known as “psychological contracts” (Koh et al., 2004). By endorsing these psychological contracts, ZBJ provides guarantees to their fulfillment (McKnight et al., 2002).
合同形成服务通过在sla中指定需求和供应商的职责来减少需求的不确定性(Mayer等人，1995年)。与线下市场的sla不同，这些sla不是由政府机构执行的，而是由中介机构本身执行的，中介机构监控供应商的活动并惩罚偏离协议的活动(Pavlou, 2002)。这与之前的发现一致，即传统的法律保护在网络市场上是无效的(Pavlou和Gefen, 2004)。为了确保合同具有灵活性，并且客户不会因为超出最初要求的每一次变更而被收取过高的费用，交易顾问充当了灵活安排形成的见证人，这种灵活安排通常基于不成文的、隐含的条款，被称为“心理契约”(Koh et al.， 2004)。通过认可这些心理契约，ZBJ为其实现提供了保证(McKnight et al.， 2002)。

To mitigate progress uncertainty, ZBJ develops periodical evaluation and harmonious conflict resolution. ZBJ encourages clients to establish multiple check-up points and to pay vendors periodically based on the check-up results. This periodical evaluation is similar to the monitoring services used in online product markets (Pavlou, 2002), but differ in that it tracks periodical rather than final deliverables; as a result, it affords clients more control over the process (Kirsch, 2004). Transaction counselors also help clients who lack domain expertise to evaluate progress. This is similar to the mediated model of outsourcing, where clients engage thirdparty intermediaries to moderate post-contract execution (Mahnke et al., 2008). Periodical evaluation is also a means of sharing timely and meaningful information between vendors and clients. According to Jarvenpaa and Leidner (1997), such information sharing promotes trust in the interorganizational relationship.
为了减少进展的不确定性，ZBJ开发了定期评估和和谐冲突解决。ZBJ鼓励客户建立多个检查点，并根据检查结果定期向供应商付款。这种定期评估类似于在线产品市场中使用的监控服务(Pavlou, 2002)，但不同之处在于它跟踪定期的而不是最终的交付品;因此，它为客户提供了对过程的更多控制(Kirsch, 2004)。这类似于外包的中介模式，在这种模式下，客户雇佣第三方中介来缓和合同签订后的执行(Mahnke et al.， 2008)。定期评估也是供应商和客户之间及时共享有意义信息的一种方式。根据Jarvenpaa和Leidner(1997)，这种信息共享促进了组织间关系中的信任。

In harmonious conflict resolution, transaction counselors serve as mediators who help clients and vendors take each other’s perspective and reach a mutual understanding. Research has shown that harmonious conflict resolution promotes trust (Goo et al., 2009). This resolution is provided neither by vendors nor clients, as in the literature, but by an intermediary. ZBJ’s transaction counselors are able to provide harmonious conflict resolution for two reasons. First, they are third-party referees who are trusted by both clients and vendors. Given that trust can be transferred, clients’ trust in the referee can be transferred to vendors (Mayer et al., 1995). Second, transaction counselors are ambidextrous employees who have both business domain knowledge and technical knowledge. The literature has shown that these boundary spanners are effective in resolving conflicts and maintaining trust (Du and Pan, 2013; Levina and Vaast, 2008). Harmonious conflict resolution enables clients to gain a deeper understanding of vendors which enables them to better control vendors (Carson et al., 2003).
在和谐的冲突解决中，交易顾问充当调解人的角色，帮助客户和供应商从对方的角度出发，达成相互理解。研究表明，和谐的冲突解决促进信任(Goo et al.， 2009)。与文献中一样，这个解决方案不是由供应商或客户提供的，而是由中介提供的。ZBJ的交易顾问能够提供和谐的冲突解决有两个原因。首先，他们是客户和供应商都信任的第三方审查员。考虑到信任可以转移，客户对裁判的信任可以转移到供应商(Mayer等人，1995)。其次，交易顾问是既具有业务领域知识又具有技术知识的能手。文献表明，这些边界管理人员在解决冲突和维持信任方面是有效的(Du和Pan, 2013;Levina和Vaast, 2008)。和谐的冲突解决能使客户对供应商有更深的了解，从而更好地控制供应商(Carson et al.， 2003)。

Conclusions
We conclude this study by highlighting the theoretical contributions and practical implications of the findings, and by discussing limitations and future research directions.

6.结论

我们通过强调研究结果的理论贡献和实际意义，并通过讨论局限性和未来的研究方向来总结这项研究。

Theoretical and practical contributions
The process model of crowdsourcing intermediary-enabled trust development and maintenance is the key contribution of this study. In terms of theoretical contributions, it contributes to the digital entrepreneurship, crowdsourcing, and e-Commerce trust development literature.

6.1理论与实践贡献

众包中介信任开发与维护的过程模型是本研究的关键贡献。就理论贡献而言，它为数字创业，众包和电子商务信任发展研究做出了贡献。

First, our model contributes to the digital entrepreneurship literature by examining the enabling role of intermediaries from a trust development perspective. The literature has recognized that online markets are an important digital infrastructure that enables entrepreneurs. However, extant studies have mainly focused on the entrepreneur perspective and examined how entrepreneurs exploit opportunities afforded by online markets (Avgerou and Li, 2013; Gregg and Walczak, 2008; Leong et al., 2016; Reuber and Fischer, 2010), whereas the means by which intermediaries enable entrepreneurship remains obscure. Our model reveals that an intermediary can enable digital entrepreneurship by deploying IMs that initiate, augment, and maintain clients’ trust in digital entrepreneurs.
首先，我们的模型通过从信任发展的角度考察中介机构的促进作用，为数字创业文献做出了贡献。文献已经认识到在线市场是使企业家得以发展的重要数字基础设施。然而，现有的研究主要集中在企业家视角，考察企业家如何利用网络市场提供的机会(Avgerou和Li, 2013;Gregg和Walczak, 2008;Leong et al.， 2016;(Reuber and Fischer, 2010)，而中介机构促进创业的方式仍不清楚。我们的模型显示，中介可以通过部署IMs来启动、增强和维护客户对数字企业家的信任，从而实现数字创业。

Second, our model contributes to the crowdsourcing literature by examining how an intermediary develops clients’ trust. The extant literature has shown that trust is a major antecedent to clients’ intention to engage in crowdsourcing and that intermediaries play an important role in trust development (Feller et al., 2012; Ye and Kankanhalli, 2015; Zogaj et al., 2014). However, it is not clear how intermediaries achieve it. Our process model shows that an intermediary can develop and maintain clients’ trust in vendors through a three-phase process. Each phase involves a set of IMs that address a specific type of uncertainty. Moreover, the prior literature has treated uncertainty as a unitary construct (Ye and Kankanhalli, 2015), but our model divides uncertainty into three types and examines IMs for each. Research has shown that a fine-grained understanding of uncertainty can guide better development of a digital platform (Pavlou et al., 2007). In addition, the majority of the crowdsourcing literature has taken either a client or vendor perspective, but overlooked the intermediary perspective (Zogaj et al., 2014). Our process model fills this gap. In particular, through the perspective of trust development and maintenance, our process model answers questions about how intermediaries operate (Feller et al., 2012) and facilitate the crowdsourcing process (Ye and Kankanhalli, 2015).
其次，我们的模型通过研究中介机构如何发展客户的信任来为众包文献做出贡献。现有文献表明，信任是客户参与众包的主要前提，中介机构在信任发展中扮演重要角色(Feller et al.， 2012;Ye and Kankanhalli, 2015;(Zogaj等，2014)。然而，目前还不清楚中介机构是如何实现这一目标的。我们的流程模型表明，中介可以通过一个三阶段的流程来开发和维护客户对供应商的信任。每个阶段都涉及一组针对特定不确定性类型的IM。此外，之前的文献将不确定性视为一个单一的结构(Ye and Kankanhalli, 2015)，但我们的模型将不确定性分为三种类型，并分别考察了IMs。研究表明，对不确定性的细致理解可以更好地指导数字平台的开发(Pavlou等人，2007)。此外，大多数众包文献都是从客户或供应商的角度出发，而忽视了中介的视角(Zogaj等，2014)。我们的流程模型填补了这一空白。特别是，通过信任开发和维护的视角，我们的过程模型回答了关于中介如何运作(Feller et al.， 2012)和促进众包过程(Ye and Kankanhalli, 2015)的问题。

Third, our model contributes to the e-Commerce trust development literature by extending intermediary-enabled trust development from online product markets to online service markets. Findings regarding the online product markets are mainly based on variance models and focus on IMs that facilitate initial trust formation (Ba and Pavlou, 2002; Gefen et al., 2003; McKnight et al., 2002; Pavlou, 2002; Pavlou and Gefen, 2005). They have overlooked trust development and maintenance, which are critical to service transactions. Our process model shows that to ensure successful service transactions, intermediaries initiate, augment, and maintain clients’ trust throughout the transaction process. Moreover, the prior literature has mainly focused on IMs that are specific to product transactions (Angelika et al., 2012; Pavlou et al., 2007). Our model identifies new IMs that are specific to service transactions and that mitigate two types of uncertainty specific to service transactions. We do not claim that our process model covers all IMs for trust development and maintenance in online service markets. We suggest that an important contribution of our process model is that it offers a new process perspective to explore IMs for trust development.
第三，我们的模型通过将中介支持的信任开发从在线产品市场扩展到在线服务市场，对电子商务信任开发文献做出了贡献。关于在线产品市场的发现主要基于方差模型，并关注于IMs，这有助于初始信任的形成(Ba和Pavlou, 2002;Gefen等，2003;McKnight等人，2002;Pavlou, 2002;Pavlou和Gefen, 2005)。他们忽略了对服务事务至关重要的信任开发和维护。我们的流程模型显示，为了确保成功的服务事务，中介体在整个事务流程中发起、增强和维护客户的信任。此外，之前的文献主要关注针对产品交易的IMs (Angelika et al.， 2012;Pavlou等人，2007)。我们的模型确定了特定于服务交易的新IM，这些IM减轻了特定于服务交易的两种不确定性。我们并不声称我们的流程模型涵盖了在线服务市场中用于信任开发和维护的所有IMs。我们认为，我们的流程模型的一个重要贡献是，它提供了一个新的流程视角来探索用于信任开发的IMs。

In terms of practical implications, our model can benefit three categories of stakeholders in crowdsourcing: Intermediaries, clients, and vendors. First, crowdsourcing intermediaries can use our model to deploy IMs that develop and maintain clients’ trust in vendors, and thus create a flourishing online service market. In light of the expanding gig economy, we expect to see more crowdsourcing intermediaries emerge. For example, in Europe, emerging crowdsourcing intermediaries such as Jovoto and Twago are gaining popularity. Also, we expect to see more intermediaries emerging in specialized service areas. One example is Catalent, a crowdsourcing intermediary for high-end management consulting services. These emerging intermediaries still lack established IMs for trust development and maintenance, and will find our model useful. Our model differs from prior trust development models in providing stepwise prescriptions and in covering not only initial trust formation but also the further development and maintenance of that trust in a transaction.
在实际意义上，我们的模型可以使众包中的三类利益相关者受益：中介，客户和供应商。首先，众包中介可以使用我们的模型来部署IM，以开发和维护客户对供应商的信任，从而创建繁荣的在线服务市场。随着零工经济的发展，我们预计会出现更多的众包中介。例如，在欧洲，Jovoto和Twago等新兴众包中介正变得越来越受欢迎。此外，我们希望看到更多的中介机构出现在专门的服务领域。一个例子是Catalent，这是一家提供高端管理咨询服务的众包中介公司。这些新兴的中介仍然缺乏用于信任开发和维护的已建立的IMs，并且会发现我们的模型非常有用。我们的模型与以前的信任开发模型不同，它提供了逐步的处方，不仅涵盖了初始信任的形成，而且还涵盖了交易中该信任的进一步开发和维护。

Second, clients who intend to engage in crowdsourcing can also benefit from our study. They can use our process model to evaluate online service markets and avoid those that do not have effective IMs. Such markets are likely to be filled with lemon sellers and opportunistic behaviors (Akerlof, 1970). In contrast to previous e-Commerce trust development studies, our model sensitizes clients to IMs beyond the initial trust formation phase.
其次，有意参与众包的客户也可以从我们的研究中受益。他们可以使用我们的流程模型来评估在线服务市场，并避免那些没有有效IMs的市场。这样的市场很可能充斥着柠檬卖家和机会主义行为(Akerlof, 1970)。以前的电子商务信任发展研究相比，我们的模型使客户对超出初始信任形成阶段的IM保持敏感。

Third and lastly, digital entrepreneurs pursuing opportunities afforded by crowdsourcing intermediaries can also benefit from our findings. Similar to clients, they can use the process model to evaluate online service markets. Joining an online market and building a professional identity within it require a significant amount of time and effort. Therefore, entrepreneurs need to choose a market carefully. Markets with effective IMs for trust development are likely to be sustainable in the long term. In addition, digital entrepreneurs can use our process model as a toolkit to analyze the dynamics of an online service market, and to devise strategies that better leverage the resources of the market (Thomas and Autio, 2012). For example, knowing that an intermediary provides contract formation services for trust development, vendors can work closely with those contract formation assistants to gain clients’ trust.
第三，也是最后，寻求众包中介提供的机会的数字企业家也可以从我们的研究中受益。与客户类似，他们可以使用流程模型来评估在线服务市场。加入一个在线市场并在其中建立一个专业身份需要大量的时间和精力。因此，创业者需要谨慎选择市场。对于信托发展而言，具有有效IMs的市场很可能是长期可持续的。此外，数字企业家可以使用我们的流程模型作为工具包来分析在线服务市场的动态，并设计更好地利用市场资源的策略(Thomas和Autio, 2012)。例如，了解到中介机构为信任发展提供合同形成服务，供应商可以与那些合同形成助手紧密合作以获取客户的信任。

Limitations and future research
Findings of this research must be considered in light of their limitations, which point to important directions for future research. First, the process model is derived from a single case study, which makes statistical generalizability impossible. A fruitful direction for future research is to verify our model across different crowdsourcing intermediaries. Nevertheless, we posit that the model is generalizable at the analytical level (Lee and Baskerville, 2003), because its constructs and arguments are corroborated by the literature. Moreover, the focus of such single case study is to generate new theoretical insights, rather than statistical generalizable conclusions.

6.2局限性和未来研究

必须根据它们的局限性来考虑这项研究的结果，这些局限性为将来的研究指明了重要的方向。 首先，过程模型是从单个案例研究中得出的，这使得统计通用性变得不可能。 未来研究的一个富有成果的方向是在不同的众包中介机构之间验证我们的模型。 尽管如此，我们认为该模型在分析水平上是可推广的（Lee和Baskerville，2003年），因为其构造和论点得到了文献的证实。 而且，这种单案例研究的重点是产生新的理论见解，而不是统计上可概括的结论。

Second, the model may not be readily applicable to labor-intensive service markets, such as MTurk. An important reason is that uncertainty in these markets is arguably smaller and different from in the knowledge-intensive markets. For example, on the one hand, clients in MTurk face little requirements uncertainty, because tasks such as data processing and image tagging have clear specifications. Therefore, IMs such as requirements analysis services are not critical, if needed at all. On the other hand, clients in MTurk face uncertainty that does not exist in the knowledge-intensive markets. For example, in image tagging, low-paid service providers often submit arbitrary answers just for the sake of collecting their fee (Karger et al., 2014). A future research could investigate how labor-intensive markets, such as MTurk, develop and maintain trust and compare the results with the findings of this study.
其次，该模型可能不适用于劳动密集型服务市场，如MTurk。一个重要的原因是，这些市场的不确定性可以说比知识密集型市场要小，而且也有所不同。例如，一方面，MTurk中的客户面临的需求不确定性很少，因为数据处理和图像标记等任务都有明确的规范。因此，如果需要的话，需求分析服务等IMs并不重要。另一方面，MTurk的客户面临着知识密集型市场中不存在的不确定性。例如，在图像标记中，低报酬的服务提供者常常为了收取费用而提交任意的答案(Karger et al.， 2014)。未来的研究可以调查劳动密集型市场，如MTurk，如何发展和维持信任，并将结果与本研究的结果进行比较。

Third and lastly, our study has taken a client perspective on trust development. Although the client perspective dominates the eCommerce trust development literature, the vendor perspective is also important (Sun, 2010). In fact, the vendor perspective is especially important for service transactions, because service vendors are as vulnerable as clients, if not more so. For example, creative outputs such as design and writing can be easily stolen by opportunistic clients (Deng et al., 2016); service vendors cannot resell rejected solutions like product vendors reselling returned products, because services are custom-made and not fungible. Moreover, unlike product vendors, service vendors have limited capacity and should only dedicate it to serving trustworthy clients (Horton, 2017). Therefore, future research could examine how crowdsourcing intermediaries develop vendors’ trust in clients.
第三，也是最后，我们的研究以客户的视角来看待信任发展。虽然客户视角主导了电子商务信任开发文献，但供应商视角也很重要(Sun, 2010)。实际上，供应商的视角对于服务事务尤其重要，因为服务供应商和客户一样脆弱，甚至更脆弱。例如，设计和写作等创造性输出很容易被机会主义客户窃取(Deng et al.， 2016);服务供应商不能像产品供应商那样转售退回的产品，因为服务是定制的，不可替代。此外，与产品供应商不同，服务供应商的能力有限，只能将其用于服务值得信赖的客户(Horton, 2017)。因此，未来的研究可以检验众包中介如何发展供应商对客户的信任。