ArticlePDF Available

Inventing Together: The Role of Actor Goals and Platform Affordances in Open Innovation

Authors:

Abstract and Figures

With ubiquity of the Internet and social platforms, open innovation (OI) opportunities now extend to individuals with creative ideas and interests in innovation. Understanding why individuals are willing to engage in open innovation and how their diverse goals affect their participation is important for assessing the viability of various OI models and to inform platform design. In this paper, we develop a theoretical model that examines the impact of three categories of human goals-extrinsic, intrinsic and internalized extrinsic-on actors' continuous intentions to participate in three general categories of open innovation behaviors-ideation, collaboration and socialization. The model also considers how perceived platform participation affordances mediate the influence of goals on these innovation behaviors. We validate this goals-affordances-behavior model via a field survey of participants on a Social Product Development (SPD) platform. By theorizing and empirically examining how goals influence participation in the SPD context, our study advances knowledge about open innovation behaviors, provides a foundation for future research across various OI models, and highlights practical insights for OI platform design.
Content may be subject to copyright.
1
Abhari, K., Davidson, E., & Xiao, B. (Forthcoming). Inventing Together: The Role of Actor Goals and Platform Affordances in
Open Innovation. Journal of the Association for Information Systems.
INVENTING TOGETHER: THE ROLE OF ACTOR GOALS AND
PLATFORM AFFORDANCES IN OPEN INNOVATION
Kaveh Abhari
Fowler College of Business, San Diego State University, kabhari@sdsu.edu
Elizabeth J. Davidson
Shidler College of Business, University of Hawai'i at Manoa, edavidso@hawaii.edu
Bo Xiao
Shidler College of Business, University of Hawai'i at Manoa, boxiao@hawaii.edu
Abstract. With ubiquity of the Internet and social platforms, open innovation (OI)
opportunities now extend to individuals with creative ideas and interests in innovation.
Understanding why individuals are willing to engage in open innovation and how their
diverse goals affect their participation is important for assessing the viability of various OI
models and to inform platform design. In this paper, we develop a theoretical model that
examines the impact of three categories of human goals––extrinsic, intrinsic and
internalized extrinsic––on actors’ continuous intentions to participate in three general
categories of open innovation behaviors––ideation, collaboration and socialization. The
model also considers how perceived platform participation affordances mediate the
influence of goals on these innovation behaviors. We validate this goals-affordances-
behavior model via a field survey of participants on a Social Product Development (SPD)
platform. By theorizing and empirically examining how goals influence participation in the
SPD context, our study advances knowledge about open innovation behaviors, provides
a foundation for future research across various OI models, and highlights practical
insights for OI platform design.
Keywords: Open innovation, social product development, ideation, collaboration,
socialization, goals, perceived participation affordances, behavioral intention, innovation
platform.
2
“Marisa has a million ideas but only a few minutes to spare. She had an idea for a brand-new product her
kids would love, sonaturallyshe shared it on the Quirky invention platform. Talented renderers,
sketchers, and toy enthusiasts in our community helped strengthen her idea submission. In turn, she shared
some of her Influence (i.e., a cut of the product revenue) with the people that helped out the most.”
(Quirky.com)
1 INTRODUCTION AND MOTIVATION
The term open innovation (OI) commonly refers to an array of business models that rely on
creative resources external to a firm to join in and contribute to new product or service
development. With the growth of the Internet and social technologies, OI opportunities now extend
beyond large firms and their organizational customers to include individuals with creative ideas
and interest in innovation (Leenders & Dolfsma, 2016; Muninger, Hammedi, & Mahr, 2019;
Roberts & Piller, 2016; Urbinati, Chiaroni, Chiesa, & Frattini, 2020). OI platforms such as
InnoCentive, NineSigma and IdeaScale thus build on the long-held promises of democratizing
and energizing OI (Acar, 2019; Bogers et al., 2017) by attracting, channeling and maintaining the
creativity of diverse individuals.
Despite the general appeal of OI business models, a number of OI platforms have
struggled to sustain the interest of individuals in their communities and to move beyond their initial
successes and become economically viable enterprises (Bogers, Chesbrough, & Moedas, 2018;
Kohler & Nickel, 2017; Majchrzak, Malhotra, & Zaggl, 2020). OI typically draws on collaborative
as well as competitive interests, on individual incentives for creativity as well as individuals’
community engagement and altruistic support (K. Boudreau & Lakhani, 2009; Füller, Hutter, Hautz,
& Matzler, 2014). Individuals thus bring a variety of goals to OI communities that motivate their
behavior on the OI platforms (von Krogh, Haefliger, Spaeth, & Wallin, 2012); these goals may not
be harmonious among community members or aligned with the platform sponsor’s intentions
(Abhari, Davidson, & Xiao, 2019; Füller et al., 2014; Liao & Xu, 2020). OI platform sponsors are
then challenged to balance their appeal across competitive, collaborative and social goals, so as
to attract individuals with high-quality, creative ideas while also engaging participants in refining
3
others’ ideas and building the innovation community (Füller et al., 2014; Nambisan, Siegel, &
Kenney, 2018; West, 2020).
Understanding how participants’ goals align with and influence their innovation behaviors,
so as to engage them in the spectrum of innovative activities, is important for OI sponsors to
realize the full potential of OI (Bogers et al., 2017; Kohler & Chesbrough, 2020; Randhawa, Wilden,
& Hohberger, 2016). This is particularly important for OI platforms designed to engage external
actors broadly in innovation processes or activities that rely on varied goal-oriented behaviors
(Bogers et al., 2017; Kohler & Chesbrough, 2020). Previous studies have articulated the general
logic of OI, described the workings of some well-known networks, and examined the benefits of
engaging external actors in new product development (Hossain, Islam, Sayeed, & Kauranen,
2016; Randhawa et al., 2016). However, these studies have not examined closely or fully the
goals of individual innovators who participate in OI platforms or how their diverse goals influence
different types of innovation behaviors (Randhawa et al., 2016). Studies of open source
communities or virtual customer communities do provide some insights on individuals’ goals
relevant to OI (e.g. Füller, 2006; Nambisan & Baron, 2009; von Krogh et al., 2012), but these and
other OI business models differ substantively in innovation focus, incentives, activities, community
structure and governance. Thus, their empirical findings cannot be generalized across OI contexts
without further study and theoretical abstraction (cf. Lee & Baskerville, 2003).
In this paper we develop and test a goals-affordances-behavior model that renders the
influence of actors’ goals on their innovation behaviors on an OI platform. Our model advances
understanding of technology affordances for OI (cf. Nambisan, Lyytinen, Majchrzak, & Song, 2017)
by theorizing the mediating influence of perceived participation affordances in goals-behavior
relationships. The context of our study is Social Product Development (SPD) (Abhari, Davidson,
& Xiao, 2020; Annosi, Marzi, Ciampi, & Rialti, 2020). SPD is an OI model in which individual actors
(actor, hereafter) may engage in the spectrum of innovation activities from early ideation to
product development on digital platforms that employ social mechanisms to build participation
4
(Forbes, Schaefer, Panchal, & Han, 2019; Han & Yang, 2019), by drawing on a broad range of
actors with diverse goals and behaviors. Understanding this diversity is critical to portray a holistic
picture of OI participation (Forbes et al., 2019). SPD, with its mix of actor goals and participation
opportunities (Coelho, Nunes, & Vieira, 2016), thus offers such a rich context for our study.
The remainder of this paper is organized as follows. First, we consider the general
dimensions of IO participation behaviors and how they are manifest in the SPD context. We then
develop our research model to explain how actor goals and SPD participation affordances
influence actors’ participation in these innovation activities. Next, we describe our empirical study
methods to assess the research model and present results of a survey of SPD participants. We
conclude with the implications of our study and opportunities for future research.
2 RESEARCH BACKGROUND: OPEN INNOVATION BEHAVIORS
A variety of business models have developed from the basic tenets of OI (Bogers et al., 2018;
Hjalmarsson, Juell-Skielse, & Johannesson, 2017). Open source communities are a commonly
studied model in the Information Systems (IS) field (Germonprez, Levy, Kendall, & Kendall, 2020;
Liao & Xu, 2020; Yeliz Eseryel, Wie, & Crowston, 2020). Other well-known models include
crowdsourcing (cf. Jian, Yang, Ba, Lu, & Jiang, 2019), innovation marketplaces and idea contests
(cf. Camacho, Nam, Kannan, & Stremersch, 2019), and user innovation and virtual customer
environments/communities (cf. Ma, Lu, & Gupta, 2019). Social product development (SPD) is an
emerging OI model that presents a diverse array of OI activities and processes from proposing
new concepts to fully developing solutions (Abhari et al., 2020; Annosi et al., 2020). Activities are
open broadly to individuals interested in innovation, and most interactions among participants are
mediated by social networking-like platforms.
OI models differ from one another in terms of the business model and goals, community
incentives for participation, and community governance. However, a review of actor activities on
a variety of OI platforms suggests these platforms engage individuals in three general categories
5
of innovation behaviors: ideation, collaboration, and socialization (Abhari et al., 2020; Annosi et
al., 2020). Each of these dimensions have been cited in the OI literature to explore actor behavior
(Bhimani, Mention, & Barlatier, 2019; Dingler & Enkel, 2016; Gama, 2019; Gloor, 2006; Hofman,
Faems, & Schleimer, 2017; Kohler & Nickel, 2017). Ideation behaviors entail identifying innovation
opportunities and providing solutions such as new products or services (Schemmann, Herrmann,
Chappin, & Heimeriks, 2016). Collaboration behaviors involve cooperation or partnership
between actors to find or improve solutions, where the extent of collaboration depends on the
complexity of requirements, and expected quality of solutions (Taran, Boer, & Lindgren, 2015).
Socialization behaviors concern actors’ relationships with each other, within and beyond specific
innovation tasks, which build and sustain the innovation community (Dingler & Enkel, 2016).
Together, these categories suggest a conceptual framework of OI behaviors sufficiently
general to apply across individual level OI models (cf. Johns, 2017). Appendix A compares
notable OI models along these and other dimensions. Appendix B presents an overview of SPD
innovation processes, based on our in-depth study of SPD platforms (Abhari et al., 2020) as well
as other researchers’ descriptions of SPD business logic (Allen, Chandrasekaran, & Basuroy,
2018; Annosi et al., 2020; Forbes et al., 2019). In what follows, we highlight how ideation,
collaboration and socialization behaviors are manifested in OI processes and activities, focusing
on our study context (SPD).
2.1 Ideation on OI platforms
Ideation is the most common behavior that can be observed across OI platforms and includes
processes for submitting new ideas (Gama, 2019; Schreier, Fuchs, & Dahl, 2012). OI sponsors
are typically designated the “problem owner” who invites external actors to contribute to pre-
defined problems or specific innovation tasks. Financial rewards motivate ideation in innovation
marketplaces and idea contests (cf. Camacho et al., 2019), whereas in other OI models members’
motivation to ideate depends in part on community members’ dedication to the community’s goals,
6
as in open source communities (Germonprez et al., 2020), or their identification with a brand,
such as in virtual customer environments (Hsieh & Chang, 2016; Schlagwein & Bjørn-Andersen,
2014).
In the SPD context, innovation sponsors do not specify the problems or opportunities
within their designated market domain, such as consumer products (Peterson & Schaefer, 2014).
Instead, SPD actors initiate innovation projects by proposing new product concepts, which the
community and the SPD platform owner then screen (Annosi et al., 2020). Ideation is a
competitive activity, as only a few ideas are selected for development (Kornish & Ulrich, 2014).
After the product launch, the names of lead ideators may be added to the product portfolio and
package, recognized within the community by the sponsor (e.g., via success stories), and
acknowledged by other community members, thus contributing to the ideators’ reputation. As
such, the SPD model places actors at the heart of ideation (Abhari et al., 2020). It is the most
financially rewarding activity in SPD, because successful ideators may share profits from product
sales or licensing with the innovation sponsor (Annosi et al., 2020).
2.2 Collaboration on OI platforms
OI requires some degree of collaborative effort to move a new idea through the innovation process.
For example, in ideation competitions, collaboration may be established between the ideator(s)
and the competition sponsors (Liao & Xu, 2020). OI models may also extend ideation activities
beyond new idea submission to involve the lead ideator with community participants in the further
development of a new idea (Stanko, 2016). On some OI platforms the community collaboratively
selects and then develops and refines new product ideas (Schreier et al., 2012). These
engagement processes can support ideators in the development or refinement of their initial ideas
(Camacho et al., 2019; Piller & Ihl, 2013). When OI platforms allow for both ideation and
collaboration activities, the processes and incentives for these behaviors are typically distinct.
7
SPD platforms engage actors systematically in a broad range of collaboration activities
from product development to commercialization (D. Wu, Rosen, Panchal, & Schaefer, 2016).
Collaboration activities can be as simple as voting to advance a new product idea or as complex
and multilayered as working on product feature development. This diversity helps engage actors
interested in or experienced with different innovation processes, while the openness of
collaborative activities allows actors to participate in various types of projects (Han & Yang, 2019).
SPD actors can also invite other actors to help with tasks such as rendering or specification of a
new product concept. Some collaborative activities (e.g. voting) require little to no specific
experience or skills, allowing a broad array of community members to participate in collaborative
activities for fun and enjoyment (Kornish & Ulrich, 2014). Unlike ideation, only a few collaborative
activities allow the SPD actors involved to earn financial reward (Abhari et al., 2020).
2.3 Socialization on OI platforms
Individuals engage in socialization behaviors on OI platforms to enjoy being part of the community
and to network with other creative individuals. Socialization supports, and may even precede,
ideation and collaboration by helping members learn about and become part of the community
(Carillo, Huff, & Chawner, 2017; Dingler & Enkel, 2016). For example, creating profiles and
sharing project portfolios helps individuals display their qualifications and experiences as well as
to seek out like-minded members, facilitating actor-to-actor networking. On many OI platforms,
actors use social features to build a profile, connect to other members (by “following” or
messaging them), ask questions, share experience, communicate to team up, and explore
opportunities to network and to learn from others (Corral de Zubielqui, Fryges, & Jones, 2019).
SPD platforms rely on social mechanisms to build a community with shared interests,
professional relationships, and trust among its members (West, 2020). Bringing people from
different disciplines and backgrounds into an SPD community, and facilitating their socialization
and experiential communication, are essential to the innovation process (Abhari & Davidson, 2016;
8
Annosi et al., 2020; Carillo et al., 2017; Coelho et al., 2016; Muninger et al., 2019). Socializing
behaviors are distinct from but also support project-focused ideation and collaboration in SPD
since they allow actors to learn and experiment within the community by exchanging information
and networking with other professionals with shared interests (Abhari et al., 2020; Annosi et al.,
2020). SPD actors can also anticipate, request, and evaluate other actors’ competencies and
contributions through social interactions, rather than only via experiences on particular innovation
projects (Annosi et al., 2020; R. D. Evans, Gao, Martin, & Simmonds, 2018). Socializing behaviors
may contribute to the SPD platform’s success by engaging actors in knowledge-sharing (Bhimani
et al., 2019; Kornish & Ulrich, 2014) even though actors do not earn specific rewards from
participating in social activities, and some may simply join in discussions or observe activities.
3 THEORETICAL FOUNDATIONS AND HYPOTHESIS DEVELOPMENT
In this paper, we develop a research model to investigate how actors’ goals influence their
behavioral intention to contribute to OI in the SPD context. Our research model brings together
self-determination theory with technology affordance theory to identify the main goals that drive
actors’ OI behaviors, and to explicate how actors’ perceptions of possible means for ideation,
collaboration and socialization (participation affordances) affect their subsequent behavior and
thus mediate the influence of goals on OI behaviors. We conceptualize innovation behaviors in
the SPD context in terms of actors’ continuous intention to participate in the three categories of
behavioral activities outlined above (ideation, collaboration, and socialization) in line with the
implementation intention perspective (Gollwitzer, 1999). An implementation intention is a self-
regulatory strategy in the form of an "if-then plan" that specifies the when, where, and how portions
of goal-directed behavior (Gollwitzer & Brandstatter, 2008; Moors, Boddez, & De Houwer, 2017).
Accordingly, we view intention to ideate, collaborate, and socialize as the means (i.e. the
implementation intention) to fulfill the relevant goals of the actor, when actors perceive ideation,
collaboration and socialization possibilities (cf. Brandstatter, Lengfelder, & Gollwitzer, 2001). We
9
also account for how actors’ behavioral intentions can be actualized through the technological
features of the SPD platform, given that implementation intentions are driven by actors’ perception
of opportunities to perform the behavior (Pavlou & Fygenson, 2006).
3.1 Goals and OI Behaviors in SPD
Appealing to goals relevant to participants is critical to attracting and sustaining actor engagement
in online communities such as OI platforms (Acar, 2019; Kohler & Nickel, 2017). Goals indicate
what actors are trying to accomplish by participating in community activities and can help explain
their OI behaviors (Khansa, Ma, Liginlal, & Kim, 2015). OI contexts such as SPD encourage
participation in a wide array of innovation activities and thus draw on the spectrum of innovation
behaviors through which participants may satisfy their varied goals (Füller et al., 2014).
Effectively appealing to, incentivizing, and facilitating actors’ goal attainment through an
OI business model and platform can be problematic. Actors’ goals may be in conflict with each
other, or otherwise may not align well with the OI sponsor’s intended business model (Füller et
al., 2014). For instance, Liao and Xu (2020) highlight how appealing to software developers’
financial goals through ideation competitions conflicted with some of the participants’ goals for
collaborative open source software development, alienating some contributors. Füller et al. (2014)
identifed six groups of OI actorsmaster, idea generator, efficient contributor, socializer, passive
commentator, and passive idea generatorwho differ in their goals and thus differ in OI behaviors
and quality of submissions. Prior studies also recognized conflicting OI behaviors when
participation is driven by both pecuniary and non-pecuniary motivations (Abhari et al., 2019; West,
2020). These studies suggest that necessary first steps to inform an OI business model and
platform design are to examine how actor goals influence their various innovation behaviors, to
assess possible (mis)alignment of actor goals, incentives, and desired business OI outcomes,
and to adjust business rules, processes and platform design accordingly.
10
The influences of individual’s goals on their behavioral intention and their actual behavior
have been extensively explored and well established in psychological and socio-cognitive
literatures (e.g. Conner & Armitage, 1998; Gollwitzer & Sheeran, 2008; Ostlund & Balleine, 2010).
Goal theories, such as goal content theory (Austin & Vancouver, 2005), goal achievement theory
(Harackiewicz, 2002), goal-directed behavior theory (Aarts & Elliot, 2012), goal-orientation theory
(Nicholls, 1984), self-determination theory (Deci & Ryan, 1985, 2000), and goal-setting theory
(Locke & Latham, 2002), all provide theoretical foundations to help explain why and how actors
contribute to OI initiatives through their innovation behaviors.
Self-determination theory (Deci & Ryan, 1980) has been widely applied to examine the
motivations of software developers in open source communities (von Krogh et al., 2012). In self-
determination theory, goals are often categorized in two higher-order categories: extrinsic and
intrinsic (Sebire, Standage, & Vansteenkiste, 2009; Vansteenkiste, Lens, & Deci, 2006). Extrinsic
goals refer primarily to external stimuli relevant to the goal context (e.g., gaining financial rewards)
whereas intrinsic goals reflect the actorspersonal values. However, individuals may internalize
some extrinsic goals due to their personal significance (Acar, 2019; Vansteenkiste et al., 2006),
so that “internalized extrinsic” goals have external value but are also internalized. This is because
they are more autonomous (having personal significance) than controlled (having external
significance) (Deci & Ryan, 1985; Mack & Landau, 2020; Vansteenkiste et al., 2006). Internally
regulated goals do not fully depend on an external contingency to drive a particular behavior.
In this study, we adopt these higher-order categories of extrinsic, intrinsic, and internalized
extrinsic from extant self-determination theory literature and earlier research on open source
development (Mack & Landau, 2020; von Krogh et al., 2012). These goal categories can be further
contextualized to investigate OI (and SPD) behaviors. Self-determination theory highlights various
types of aspirations (goals) that actors may be pursuing, such as financial wealth, recognition or
fame, attractive image, personal development, meaningful relationships, community contributions,
and psychophysical fitness (Deci, Olafsen, & Ryan, 2017). Appendix C provides a summary of
11
goal types examined in IS literature relevant to OI behaviors. Monetary gain and recognition
(extrinsic goals), learning and entrepreneurship (internalized extrinsic goals), and enjoyment,
altruism and socialization (intrinsic goals) are key actor goals in the SPD context.
3.2 Perceived Participation Affordances and OI Behaviors in SPD
Research on goal-directed behavior demonstrates that having a goal is just the first step toward
goal attainment, since implementational problems need to be solved successfully to achieve a
goal through actions taken (Gollwitzer & Brandstatter, 2008; Pavlou & Fygenson, 2006). Goal
theory posits that “goals direct attention and effort toward goal-relevant activities and away from
goal-irrelevant activities. This effect occurs both cognitively and behaviorally” (Locke & Latham,
2002, p. 706). In computer-mediated environments such as SPD platforms, goal-oriented action
requires the actor to perceive the features of the platform that enable that action (Khansa et al.,
2015; Nagy & Neff, 2015) and then to act through these features to pursue their goals on the
platform. Goal-directed actors are more likely to look for opportunities for actions that are
consistent with their goals (Locke & Latham, 2002) and to process implementation possibilities
selectively according to their goals (Guinote, 2007, 2008).
In this study, we draw on the concept of technology affordances to theorize the
relationships between actor goals, their perception of participation possibilities that are enabled
by OI platform technology, and innovation behaviors. Markus and Silver (2008) define technology
affordances as “the possibilities for goal-oriented action afforded to specified user groups by
technical objects” (p. 622). In IS research, the affordance concept has been used to address
conceptually the myriad possibilities for action that technology features enable to different users
(Fayard & Weeks, 2014; Grgecic, Holten, & Rosenkranz, 2015; Karahanna, Xu, Xu, & Zhang,
2018; Majchrzak, Faraj, Kane, & Azad, 2013; Strong et al., 2014). Technology affordance is thus
a relational concept arising from the varied possibilities for goal-directed action that a
technological artifact provides specific actors through its features and capabilities (Anderson &
12
Robey, 2017; Majchrzak et al., 2013; Volkoff & Strong, 2013). Moreover, Evans et al. (2017)
argues that affordances are not binary but have gradations in the degree of action that is afforded,
while Anderson & Robey (2017) also suggest the potency or strength of an affordance varies with
users, tasks, and context.
Hence, we argue that an actor’s goal-directed behavior enacted through an OI platform
depends in part on how the actor’s goals influence her perception of the platform’s affordances
(cf. Cañal-Bruland & van der Kamp, 2009; Jankowiak, 2017; Lawler III, 1973). If affordances
intended by designers (i.e., features that enable participation in the platform’s IO activities)
correspond to an actor’s situated goals, the actor is more likely to notice and to act on those
affordances (Locke & Latham, 2002; Stoffregen, 2004). Since affordances present the actor with
possible means to accomplish goal-oriented actions, the stronger an actor’s goals are to
undertake an action, the more likely she is to attend to or notice (perceive) relevant affordances
in the environment and to assess the extent to which the perceived affordance might (or might
not) implement her goals through the actions afforded (Brandstatter et al., 2001; Gollwitzer &
Brandstatter, 2008; Moors et al., 2017). An actor who is not highly motivated to undertake an
afforded action is less likely to attend to (perceive) affordances present in the environment to do
so (Downes et al., 2017).
3.3 Hypotheses
From these theoretical bases in the affordance and goal theory literatures, we argue that an
actor’s goals influence her perceptions of SPD participation affordances, which presents the actor
with a means to realize her goals (implementation possibility), and the resulting implementation
intention may then increase the actor’s propensity to take the afforded SPD actions in order to
fulfill her goals. Accordingly, we posit a series of hypotheses that explain the direct relationships
between goals and OI behavior plus the mediating roles of perceived participation affordances in
SPD. Figure 1 depicts this research model, which we develop in the following sections.
13
Figure 1. Goals-Affordances-Behavior Model
3.3.1 Goals and Ideation in SPD
Ideation entails submission of new ideas (Schemmann et al., 2016). SPD platforms invite any
community member to submit new product ideas for consideration by the innovation community,
and the platform depends on members’ active participation in ideation to initiate innovation
projects (Annosi et al., 2020). SPD actors have freedom to propose new ideas; if successful, they
can receive rewards and recognition (Coelho et al., 2016). Financial rewards can play a significant
role in driving ideation on SPD platforms, since they foster the perception of fairness and
exchanged-based justice (Fernandes & Remelhe, 2016). For example, the SPD platform Quirky
shares revenue from products that move to market with the community members who contributed
the idea; the prospects of earning revenue from their inventive ideas are a major attraction for
some community members (Abhari et al., 2020).
Intrinsic
Goals
Extrinsic
Goals
Internalized
Goals
Perceived
Socialization
Affordances
Perceived
Ideation
Affordances
Perceived
Collaboration
Affordances
Continuous
Intention to
Socialize
Continuous
Intention to
Ideate
Continuous
Intention to
Collaborate
Actual
Contribution*
Money
Recognition
Entrepreneurship
Learning
Socialization
Enjoyment
Altruism
H9a
H9c
H9b
H1
H2
H3
H5
H4
Mediation Effects:
H6a: EXG àPIA àCII
H6b: IEG àPIA àCII
Mediation Effects:
H7a: IEG àPCA àCIC
H7b: ING àPCA àCIC
Mediation Effect:
H8: ING àPSA àCIS
Perceived Participation
Affordances
Goals
*Follow-up Study
Behavioral Intention
14
Opportunities to achieve recognition are also strong motivations for ideation (Acar, 2019).
Successful ideation can lead to public, community, and peer recognition, as well as enhance
professional reputation and eventually increase the actor’s labor market value (von Krogh et al.,
2012). In SPD, the platform owner can heighten this potential outcome by recognizing and
publicizing community members for successful ideation. For instance, platform owners could rank
them or add their names to product profiles and packaging. Through ideation activities, SPD
actors can thus pursue recognition goals to establish a standing as designers or inventors.
Therefore, we posit that monetary compensation and formal recognition goals, individually
and jointly, stimulate ideation behaviors in SPD. Self-determination theory classifies wealth and
fame as extrinsic goals, because they are focused on obtaining external rewards (external
regulation), such as financial rewards, positive evaluations of others, or other external
manifestations of worth (Deci & Ryan, 2000). From this perspective, extrinsic goals are the
reasons (controlled motive) for actors to engage in SPD ideation activities, due to the potential for
gaining financial success and receiving recognition. Thus, we posit:
Hypothesis 1 Extrinsic goals positively influence continuous intention to ideate.
Prior research has validated the relationship between desire for learning new skills and
intention to ideate (Acar, 2019; Fernandes & Remelhe, 2016; Mack & Landau, 2020). Submitting
new product ideas allows SPD community members to try out new ideas, to get feedback on their
market value, and to learn about the process of developing new products. Additionally, an SPD
platform stimulates entrepreneurship, by providing possibilities to acquire information about
potential market solutions and to help commercialize innovation (Battistella & Nonino, 2012).
These professional development goals (i.e., learning and entrepreneurship) can be
classified as internalized extrinsic goals (Deci et al., 2017) that can be fulfilled through ideation
activities (Mack & Landau, 2020; Ståhlbröst & Bergvall-Kåreborn, 2011; von Krogh et al., 2012).
Personal development such as learning through ideation allows actors to partially or fully
internalize the initially external regulation of behavior because of the personal significance of the
15
outcomes (Vansteenkiste et al., 2006). If actors are supported to learn from or self-endorsed by
their ideation successes or failures, they are more likely to persist in their attempts (Camacho et
al., 2019) and develop an inherent interest in ideation, presumably because their need for
competence and self-acceptance (attractive self-image) can be satisfied (Deci & Ryan, 1980). We
expect both learning and entrepreneurship goals act as internalized extrinsic goals that are
positively related to an actor’s continuous intention to ideate on SPD platforms, and therefore:
Hypothesis 2 Internalized extrinsic goals positively influence continuous intention to ideate.
3.3.2 Goals and Collaboration in SPD
Active collaboration among actors to address innovation challenges and to find or improve
solutions is a core premise of OI (Füller et al., 2014). SPD platforms are designed to enable
systematic collaboration among community members (Annosi et al., 2020), such as refining ideas,
developing technical solutions, evaluating prototypes, and critically assessing product features
and design (R. D. Evans et al., 2018) and may even extend to participating in activities such as
market research on some SPD platforms (Annosi et al., 2020; Kohler & Nickel, 2017).
Pursuing extrinsic goals is less compelling through collaborative SPD activities, since
external rewards are primarily associated with ideation. Instead, collaborative behaviors are
typically motivated by goals with some degrees of internalization, such as learning (Mack &
Landau, 2020). Perceived cognitive benefits, such as competence building, develop through
pursuing internalized extrinsic goals in OI activities (Verleye, 2015). Learning goals such as
personal or professional development are important motivators for collaboration in OI
communities generally (Acar, 2019; von Krogh et al., 2012). On SPD platforms, community
members can acquire new knowledge, enhance competencies, and gain first-hand experience by
working collaboratively on innovation projects initiated by others. Entrepreneurial goals can also
be realized by sharing knowledge collaboratively (Nambisan et al., 2018), for instance in SPD by
contributing to community efforts to improve products or to bring new products to the market.
Therefore,
16
Hypothesis 3 Internalized extrinsic goals positively influence continuous intention to collaborate.
Self-determination theory distinguishes the role of intrinsic goals from extrinsic and
internalized goals (Vansteenkiste et al., 2006). Intrinsic goals motivate actors to engage in an
activity for its own sake and are associated with high involvement and interaction behaviors such
as collaboration (Deci et al., 2017). Socialization, altruism, and enjoyment are identified as
intrinsic goals (Acar, 2019; Mack & Landau, 2020; von Krogh et al., 2012), which are primarily
process-related and can be achieved through the enjoyment of the task or obligation-based
satisfaction through meeting morals and values (Li, Tan, & Teo, 2012; von Krogh et al., 2012).
SPD community members may seek enjoyment by participating in collaborative innovation
activities such as finding solutions to problems and helping others in the process of problem-
solving (Acar, 2019; Füller, 2010). They may also find excitement and fun by interacting with
likeminded people in a highly creative environment (Fernandes & Remelhe, 2016). Socialization
and networking are intrinsic goals that help maintain actor participation (Acar, 2019; Dingler &
Enkel, 2016) by establishing collaborations on common fields of interest and developing
professional relationships (Mack & Landau, 2020). Altruism is another facet of intrinsic goals
evident in OI activities (von Krogh et al., 2012). For instance, altruism based on belief in the
community’s goals is important for collaboration in open software communities (von Krogh et al.,
2012). In SPD, collaborative activities such as voting for a product’s name is altruistic, as these
activities benefit the community as a whole but the actor cannot expect explicit individual utility
from her contribution (Füller et al., 2014). Thus, we posit that intrinsic goals can motivate actors
to collaborate on SPD platforms even without (or with limited) external rewards:
Hypothesis 4 Intrinsic goals positively influence continuous intention to collaborate.
3.3.3 Goals and Socialization in SPD
Socially-oriented actors may join SPD platforms to experience inherent satisfaction by interacting
with others in the community to establish a common field of interest and to develop professional
friendships and relationships (Fernandes & Remelhe, 2016; Füller et al., 2014; Mack & Landau,
17
2020). Social exchanges are the first steps toward establishing subsequent working relationships
with other actors (Peterson & Schaefer, 2014), for instance, by assessing other actors’
competency, ideas, and knowledge before engaging in any collaboration (Bstieler & Hemmert,
2015). Furthermore, actors may find the process of socialization with other like-minded individuals
fun, enjoyable, and fulfilling (Salehan, Kim, & Kim, 2017). Actors who find the process of helping
others enjoyable may also show a higher intention to socialize through knowledge-sharing (Zhao,
Detlor, & Connelly, 2016). Thus, intrinsic goals such as socialization, enjoyment, and altruism are
particularly applicable to SPD platforms, and these intrinsic goals can be realized by participating
in socialization behaviors (Acar, 2019; Fernandes & Remelhe, 2016). Therefore,
Hypothesis 5 Intrinsic goals positively influence continuous intention to socialize.
3.3.4 The Mediating Role of Perceived Participation Affordances
SPD platforms provide a variety of features that present participation affordances to OI actors.
Typically, ideation behaviors are afforded through features such as online idea submission,
revision, and resubmission forms, and idea development or visualization tools (Hossain & Islam,
2015; Muninger et al., 2019). Collaboration behaviors generally rely on platform features such
evaluation and ranking forms, product improvement tools, and social survey tools (R. D. Evans,
Gao, Martin, & Simmonds, 2015; Muninger et al., 2019). Socialization behaviors are enabled by
features such as profile creation and management pages, following other actors, and peer-to-
peer and group messaging (Abhari et al., 2020). Some platform features enable actors’ behaviors
beyond the initial intentions of the platform designers, while platform actors may ignore or
underutilize other features (Karahanna et al., 2018).
Prior research has persuasively argued for the direct influence of affordances on actors’
engagement with social platforms (Majchrzak et al., 2013; Sutcliffe, Gonzalez, Binder, & Nevarez,
2011; Treem & Leonardi, 2012). However, this assumes the actor perceives the affordances,
since an artifact cannot exist as a stimulus for intentional actions unless it is received as a
perceptual phenomenon first (Ayer, 1972; Barkin, 2003). Moreover, an actor looks for and attends
18
to action possibilities consistent with their goals (Guinote, 2007, 2008; Locke & Latham, 2002).
The perceived affordances thus present the actor with possible ways to realize goals by taking
the afforded actions (Lu & Cheng, 2012). When perceiving the possibilities for action brings into
focus an implementation intention to achieve the actor’s goals, it may heighten their intention to
take an afforded action (Gollwitzer & Brandstatter, 2008). We theorize this as the mediating
influence of actors’ perceived participation affordances on goal-behavior relationships (Kaptelinin,
Nardi, Hall, & Irvine, 2012) following the logic of hypotheses 1–5.
First, extrinsic goals (money and recognition) are most directly related to ideation activities
in SPD. We expect that the stronger an actor’s extrinsic goals, the more likely the actor attends
to opportunities afforded to participate in ideation activities on the SPD platform; the stronger the
perception of these ideation affordances, the more likely the actor is to develop the behavioral
intention to engage in ideation. For instance, an SPD actor whose primary goal is to earn financial
rewards is likely to focus on possible ideation actions, and thus to attend to (perceive) ideation
affordances in features such as submission, revision, and resubmission forms and visualization
tools as the means to submit new ideas (ideate). Therefore,
Hypothesis 6a Perceived ideation affordances mediate the relationship between extrinsic goals
and continuous intention to ideate.
Via ideation activities on SPD platforms, actors can also learn about new product development
processes, assess their own competencies, evaluate their own new business ideas in a low-risk
environment, and gain insight into entrepreneurship processes. Actors with internalized extrinsic
goals such as learning and entrepreneurship are more likely to perceive ideation affordances and
actualize these affordances to submit new ideas for personal or professional development
purposes. We expect that the stronger an actor’s internalized extrinsic goals, the more likely the
actor searches for and attends to opportunities afforded to participate in ideation activities on the
SPD platform and thus the more likely the actor is to develop the behavioral intention to engage
in ideation. Therefore,
19
Hypothesis 6b Perceived ideation affordances mediate the relationship between internalized
extrinsic goals and continuous intention to ideate.
SPD actors can also fulfill their internalized extrinsic goals for learning and entrepreneurship by
participating in collaboration activities when they take advantage of platform features to contribute
to collaboration activities for products undergoing development. We argue that internalized
extrinsic goals will render the possibilities to collaborate more salient to actors and consequently
trigger a higher intention to engage in collaboration with other actors. Hence, we expect that the
stronger an actor’s internalized extrinsic goals, the more likely the actor attends to opportunities
afforded to participate in collaboration activities on the SPD platform and thus intends to actualize
these afforded collaborative actions. Therefore,
Hypothesis 7a Perceived collaboration affordances mediate the relationship between
internalized extrinsic goals and continuous intention to collaborate.
Collaboration behaviors on SPD platforms can satisfy intrinsic goals such that SPD actors with
intrinsic goals (i.e., socialization, enjoyment, or altruism) will be more likely to perceive possibilities
to collaborate and will be more willing to actualize possibilities to collaborate with others on the
platform. The stronger an actor’s intrinsic goals, the more likely the actor will attend to afforded
opportunities to engage in collaboration behaviors on the SPD platform and then to take these
collaborative actions; therefore,
Hypothesis 7b Perceived collaboration affordances mediate the relationship between intrinsic
goals and continuous intention to collaborate.
Like other online communities, SPD platforms offer opportunities for spending time with like-
minded people that are not only enjoyable but facilitate networking and altruistic actions to further
community development. Actors with intrinsic goals are more likely to attend to an perceive
platform features that enable or support socialization, and consequently are more likely to utilize
these features to socialize with other community members. Therefore,
20
Hypothesis 8 Perceived socialization affordances mediate the relationship between intrinsic
goals and continuous intention to socialize.
3.3.5 Intention to Contribute & Actual Contribution in SPD
In this study, actual contribution refers to measurable individual actions to generate new product
ideas and to work with others to improve these ideas. These actions have direct implications for
the effectiveness of an SPD platform, which depends on members to submit and improve new
product ideas (Camacho et al., 2019; Füller et al., 2014). The relationship between intention and
behavior has been extensively validated in prior technology acceptance literature (Venkatesh,
Morris, Davis, & Davis, 2003). Accordingly, we expect that actors’ continuous intention to ideate,
collaborate, and socialize predicts individuals’ actual contribution. Therefore,
Hypothesis 9 Continuous intention to ideate (H9a), collaborate (H9b), and socialize (H9c)
positively influence actual contribution.
4 STUDY DESIGN AND METHODS
To validate our research model, we conducted a field survey to gather data on actors’ goals for
participating in SPD, perception of participation affordances, continuous intention to contribute,
and actual contribution in an SPD platform. The survey was administrated at two points in time, a
month apart, to test the relationship between continuous intention to contribute and actual
contribution. We used the Partial Least Squares (PLS) modeling technique to assess the
measurement model and structural model (Joseph F Hair, Hult, Ringle, & Sarstedt, 2013).
4.1 Research Setting
Launched in 2009, Quirky (quirky.com) is one of the first companies to implement the SPD open
innovation model on a social media platform (Coelho et al., 2016; Kornish & Ulrich, 2014; Piller,
Vossen, & Ihl, 2012). Quirky provided an ideal setting for our study, as we were interested in an
SPD platform that offers diverse ideation, collaboration, and socialization opportunities and
21
platform tools (see Appendix B). Quirky’s business model is based on soliciting new product ideas
for broad categories of consumer products and sharing a portion of the sales revenue (referred
to as ‘influence’ credits) with individual innovators who contributed to product ideation. Quirky
invites community members to propose new product concepts and participate in the evaluation
and selection of viable product ideas. Online community members can also participate in the co-
development (e.g., product selection, design, development) of the socially validated
ideas.
Once
an idea is selected for manufacturing, the community members can participate in
commercialization activities. Quirky launches new products through either licensing partnerships,
or a direct-to-market path via ecommerce or traditional retail distribution. At the time our research,
the Quirky SPD platform of 600,000 members had collaboratively developed and launched more
than 150 consumer products. Quirky declared bankruptcy in 2014 due to uncontrolled
manufacturing costs (Key, 2017; Kohler & Nickel, 2017), but the platform was reinstated with
minimal changes in the SPD platform activities. As of January 2021, quirky.com has more than
1.3 million members.
4.2 Model Specification and Survey Measurements
Our literature review on OI models, review of previous SPD studies, and our direct observations
of the quirky.com platform (Abhari et al., 2020) informed the operationalization of the core
constructs in our research model––goals, affordances, intention, and contribution––in relation to
the key SPD activities in which actors may engage (Johns, 2017). Table 1 lists the construct
definitions and measurement items we used for the field survey (also, see Appendix C).
Goals. We developed a hierarchical structure of goals that was theoretically informed and
empirically determined and contextualized. Based on our understanding of the interrelationship
(or lack thereof) between various goal measures suggested by self-determination theory, we
modeled actors’ extrinsic, internalized extrinsic, and intrinsic goals as three second-order
constructs measured by seven reflective first-order constructs (Wetzels, Odekerken-Schröder, &
22
Van Oppen, 2009): monetary gain, recognition, learning, entrepreneurship, enjoyment, altruism,
and socialization (see Appendix D). This hierarchical view of goals has been commonly used in
the self-determination literature due to several empirical advantages (Sarstedt, Hair, Cheah,
Becker, & Ringle, 2019). First, higher-order constructs reduce the number of path model
relationships, thereby achieving model parsimony (Polites, Roberts, & Thatcher, 2011). Second,
the bandwidth-fidelity dilemma can be addressed by higher-order constructs (Cooke & Michie,
2001). The bandwidth-fidelity dilemma refers to the tradeoff between variety of information
(bandwidth) and thoroughness of testing to obtain more certain information (fidelity) (Salgado,
2018). Third, a higher-order construct provides a comprehensive measurement by drawing items
from multiple domains and demonstrating construct dimensionality and nomological validity
(Mowen & Voss, 2008). Fourth, higher-order constructs help reduce collinearity among indicators
(Hair Jr, Sarstedt, Ringle, & Gudergan, 2017). Finally, higher-order goal constructs are deemed
more likely to predict actors’ intention while maintaining a parsimonious view of goals suggested
by self-determination theory. The measurement items for the seven lower-order goal constructs
were mainly adapted from existing studies on OI and contextualized for SPD (Acar, 2019;
Antikainen, Mäkipää, Ahonen, & M, 2010; Battistella & Nonino, 2012; Kahnert, Menez, & Blättel-
Mink, 2012; Mack & Landau, 2020) (see Appendix D).
To decide whether the goal constructs should be modeled formatively or reflectively, we
evaluated causality, sensitivity, exchangeability, inclusivity and predictability of the constructs
(Cenfetelli & Bassellier, 2009; Jarvis, Mackenzie, & Podsakoff, 2003; Petter, Straub, & Rai, 2007;
Windeler, Maruping, & Venkatesh, 2017). Accordingly, we modeled extrinsic and internalized
extrinsic goals as formative constructs and intrinsic goal as a reflective construct. First, self-
determination theory suggests that each lower order measure of externally-inspired goals can
partially predict (define) the relevant higher order goal construct (i.e. the direction of causality is
from lower-order goal measures to higher-order goal construct) (Ryan & Deci, 2000). For extrinsic
goals (the higher-order goal construct), money and recognition (the two lower-order goal
23
constructs) are distinct goals, each explaining one form of external rewardsexternal regulation
and introjected regulation—but together providing a comprehensive understanding of extrinsic
goals in the OI context (Ryan & Deci, 2000). Likewise, learning and entrepreneurship manifest
identified regulation and integrated regulation, two distinct forms of internalized extrinsic goals
(Ryan & Deci, 2000; von Krogh et al., 2012). Together they capture identified and integrated
regulations (cf. Petter et al., 2007) that are sufficiently inclusive to capture the essence of
internalized extrinsic goals (cf. Diamantopoulos & Winklhofer, 2001) relevant in the SPD context.
Second, extrinsic and internalized extrinsic goal constructs are also sensitive to the
exclusion of any lower-order construct for reason of being inclusive. For example, learning and
entrepreneurship goals are two different ways in which actors internalize the regulation of
externally significant goals (von Krogh et al., 2012). Learning relates to personal development
whereas entrepreneurship relates to professional development; neither by itself captures the full
meaning of internalized extrinsic goals relevant to SPD. Third, change in one of the lower-order
formative goal construct does not necessarily imply an equal change in the other lower-order
constructs. Entrepreneurship and learning, similar to monetary gain and recognition, cannot be
substituted for each other because each may change independently (J. Zhang, Zhang, Song, &
Gong, 2016). Therefore, omitting one of them would alter the definition and comprehensiveness
of the higher-order construct. There is also empirical evidence that these lower-order goals are
not interchangeable (J. Zhang et al., 2016).
Finally, literature suggests actors’ psychological needs (autonomy, competence or
relatedness) and goal-directed behavior are respectively antecedents and consequences of the
goal constructs (Ciani, Sheldon, Hilpert, & Easter, 2011; Deci et al., 2017). The dimensions of
extrinsic goals (monetary gain and recognition) and internalized extrinsic goals (learning and
entrepreneurship) can satisfy different needs and result in different behaviors (Ryan, Sheldon,
Kasser, & Deci, 1996). In this regard our empirical case studies of SPD (including interviews with
and observation of SPD members’ discussions) (Abhari et al., 2020) as well as our multi-stage
24
survey instrument development process provided additional assurance that these dimensions
capture the most influential extrinsic and internalized extrinsic goals for SPD actors.
In light of the arguments presented above, we modeled extrinsic and internalized extrinsic
goals as formative constructs. However, we are not able to make this claim for intrinsic goals––
enjoyment, altruism and socialization––since they are associated with inherent human emotional
needs (intrinsic regulation) (Ryan & Deci, 2000; Ryan et al., 1996). Therefore, we modeled
intrinsic goals as a reflective second-order construct.
Table 1. Table of Contextualized Construct Definitions
CONSTRUCTS
Extrinsic Goals
Monetary Gain (MNG)
Recognition (RCN)
Internalized
Extrinsic Goals
Entrepreneurship (ENT)
Learning (LRN)
Intrinsic Goals
Altruism (ALT)
Enjoyment (ENJ)
Socialization (SCL)
Perceived
Participation
Affordances
Perceived Ideation
Affordances (PIA)
Perceived Collaboration
Affordances (PCA)
Perceived Socialization
Affordances (PSA)
Continuous
Intention
Continuous Intention to
Ideate (CII)
Continuous Intention to
Collaborate (CIC)
Continuous Intention to
Socialize (CIS)
Behavior
Actual Contribution
Affordances. We modeled perceived participation affordances as three first-order reflective
constructs, namely perceived platform ideation affordances, perceived platform collaboration
affordances, and perceived platform socialization affordances (see definitions in Table 1). Each
25
affordance construct relates to the key action possibilities identified by prior research on OI and
contextualized for this study of SPD (Mathiesen, Bandara, & Watson, 2013; Olapiriyakul &
Widmeyer, 2009; Tan, Tan, & Pan, 2016). Since the constructs and their measurements were
newly developed for this survey, we assessed face validity and content validity using two
independent expert panels
1
. The refined measurement instrument was then pilot tested with
Quirky members before the official field survey.
Intention. The reflective measurement items for continuous implementation intention to
contribute, including continuous intention to ideate, continuous intention to collaborate, and
continuous intention to socialize, were adapted to reflect the SPD context from previous studies
on continuous behavioral intention in virtual collaborative communities (Bhattacherjee &
Premkumar, 2001; Chen, 2007; Y. Zhang, Fang, Wei, & Chen, 2010).
Actual contribution. Actual contribution was measured with (a) the number of new ideas
submitted and (b) the number of products ‘influenced’ (collaborated on) through the collaboration
process. We used two sources to collect data about actual contribution: (a) respondents’ self-
reports of their contributions via a follow-up survey, and (b) respondents’ actual contributions (i.e.,
number of new product idea submitted and influenced as a collaborator) that we observed in the
Quirky profiles of study participants who had voluntarily provided their Quirky public user profile.
4.3 Pretest and Pilot Study
A pre-test was designed to assess respondent concerns and questionnaire-related issues. The
survey questionnaire was circulated among 30 researchers familiar with both the concepts and
the context of this study to solicit feedback on the wording and presentation of the questions. We
1
We used card-sorting technique in which a panel of judges sorted the measurement items suggested by the literature and
contextualized for Quirky into separate affordance categories based on the similarities and differences among items. Then, based on
their placement and comments, the items were re-examined, and ambiguous items were modified or eliminated. The sub-scales were
then combined into an overall instrument for a second round of card sorting exercise (Moore & Benbasat, 1991). See also Abhari,
Davidson, & Xiao (2017) for a full account of development of the SPD affordances measurement instrument.
26
invited five active members of the Quirky community for follow-up interviews to assess the
questions, resulting in further adjustments of the survey instrument. We then conducted a pilot
study via online survey to assess the proper functioning of the survey instrument. The pilot study
collected data from 72 randomly selected Quirky members and helped establish the required
reliability and validity for all the constructs.
4.4 Field Survey
We collected data for the field survey from a random sample of Quirky members. From 600,000
potential respondents at the time of study, 1,000 Quirky members were randomly selected (based
on a unique six-digit user ID defining their profile URLs) and invited via direct message to
participate in an online survey. As an incentive, all those who submitted complete responses were
offered a $10 gift card of their choice. Of the 320 Quirky members who responded, we used 264
responses after screening out incomplete data or data from respondents with less than one-month
experience with Quirky. To identify actors with prior SPD experience on Quirky, respondents were
first asked to indicate their own experience with Quirky in terms of ideation, collaboration, and
socialization. After that, respondents were presented a series of Likert-type questions to assess
their goals to contribute to Quirky, their perception of participation affordances, and continuous
intention to ideate, collaborate, and socialize. Finally, respondents’ demographic information (e.g.,
gender, age, education, employment) was collected as control variables.
One month after the completion of the field survey, a follow-up questionnaire to assess
Quirky members’ actual contribution was sent to respondents who had voluntarily provided an
email address or link to their profile. The follow-up survey included a question on the number of
product ideas submitted and the number of products influenced (collaborated on) by the
respondents during the last month. A total of 103 Quirky members either responded to the follow-
up survey (78) or provided an answer via their profile (25). The sample size satisfies guidelines
suggested for PLS sample sufficiency (Joseph F Hair et al., 2013).
27
5 DATA ANALYSIS
We used SmartPLS (Ringle, Wende, & Becker, 2015) to validate both the measurement and
structural properties of our research model. In our study PLS analysis is preferred over other
analytical techniques because it simultaneously assesses the psychometric properties of the
measurement items (i.e., the measurement model) and analyzes the direction and strength of the
hypothesized relationships (i.e., the structural model). PLS also facilitates the modeling of
formative constructs and it is recommended for the hierarchical model evaluation used in this
study (Joe F Hair, Ringle, & Sarstedt, 2011; Wetzels et al., 2009). The hypotheses, except H9,
were tested by using the first survey (n = 264). We used the follow-up study (n= 103) to test the
relationship between continuous intention constructs and actual contribution (H9a-c).
5.1 Demographics and Actor Profile Data
Table 2 summarizes the demographics of our sample. More females participated in the survey
than did males. A large majority of respondents were between 26 and 65, and over 70% had at
least some college education. Close to 60% of the respondents were employed outside of their
participation in the SPD. Respondents with less than one month experience with Quirky were
screened from further analysis. More than 76% of the respondents had more than six months’
experience with Quirky and more than 70% visited Quirky at least once a week. A large majority
of respondents had recent experience with ideation (82%), collaboration (100%), and/or
socialization (85%) on Quirky. Over 80% of the respondents had also received monetary credits
for ideation or collaboration (referred to ‘influence’ credits in the Quirky community), an indicator
of active participation in SPD.
28
Table 2. Descriptive Statistics and Respondents’ Profiles
GENDER
Male
Female
Undisclosed
39.8%
52.5%
7.8%
EDUCATION
High school
Some College
College graduate
Post-graduate degree
Other
Undisclosed
Retired
1.4%
24.6%
22.1%
28.3%
15.2%
7.8%
6.1%
AGE
19 or younger
19 - 25
26 - 45
46 - 65
66 and Older
Undisclosed
1.2%
13.9%
43.0%
27.5%
6.6%
7.8%
EMPLOYMENT
Full-time employed
Part-time employed
Self-employed
Unemployed
Undisclosed
45.9%
13.5%
17.6%
4.1%
7.8%
5.2 Evaluation of 1st-Order Measurements
We ran an exploratory factor analysis (EFA) to check the dimensionality of the proposed goal
constructs. We first used Maximum Likelihood with Oblique rotation (direct oblimin) to investigate
the relative importance of each item. Oblique rotation was used to preserve the unique variance
of each measure, achieve more generalizable results, and render a more optimum solution
(Costello & Osborne, 2005; Petter et al., 2007). The result shows that the Kaiser-Meyer-Olkin
measure of sampling adequacy is 0.88 (above the commonly recommended value of 0.6) and
that Bartlett’s test of Sphericity is significant (χ2 = 4936, p = 0.00) indicating that the correlations
between items are sufficiently large for EFA (Joe F Hair & Anderson, 2010). These overall
indicators suggest that factor analysis is suitable for all 28 items. An examination of the
eigenvalues reveals that the first seven components with eigenvalues greater than 1 explain 75
percent of the variance in total. We adopted the seven-component solution because of the
theoretical support, the “leveling off” of eigenvalues on the screen plot after seven factors, and
the insufficient number of primary loadings for the eighth factor and subsequent factors. We
retained all the 28 items for goals as they all have factor loadings higher than 0.4 (±0.30=minimal,
±0.40=important, ±.50=practically significant; Hair & Anderson, 2010). These analyses reveal
seven distinct dimensions underlying SPD goals.
29
The same procedure was followed to assess the dimensionally of perceived participation
affordances constructs. The Kaiser-Meyer-Olkin measure of sampling adequacy is found to be
0.86, above the commonly recommended value of 0.6, and Bartlett’s test of Sphericity is
significant indicating that the correlations between items are sufficiently large for EFA (χ2 = 1940,
p = 0.00). An examination of eigenvalues indicates that the first three components with
eigenvalues greater than 1.0 explain 76 percent of the variance in total, supporting the three-
component solution. We retained all the 12 items for perceived affordances as they have factor
loadings higher than 0.4 (Joe F Hair & Anderson, 2010). These analyses support three distinct
dimensions underlying perceived affordances as we expected.
Next, we evaluated the measurement model by assessing construct reliability (item
reliability and internal consistency), construct factorability, and construct validity (convergent
validity and discrimination validity). As shown in Table 3, all the loadings of measurement items
on their latent constructs, except one, exceed 0.7, indicating acceptable item reliability (Joseph F
Hair et al., 2013). The cross-loadings are also reported in Appendix E. Cronbach’s alpha and
composite reliability of all the constructs are higher than 0.7, indicating good internal consistency
among the items measuring each construct (Joseph F Hair et al., 2013). Three criteria were
adopted to assess convergent validity and discriminant validity: (a) all Average Variance Extracted
(AVE) are higher than 0.50 (Joseph F Hair et al., 2013); (b) the square root of the AVE of each
constructs is larger than the correlations of this construct with the other constructs (Fornell &
Larcker, 2006); and (c) the heterotrait-to-monotrait (HTMT) values are all well below the 0.90
threshold (Joseph F Hair et al., 2013). The results of these tests suggest adequate convergent
and discriminant validity. Lastly, we tested for common method bias using full collinearity
assessment (i.e. vertical and lateral collinearity) (Kock 2015). All the pathological VIFs resulting
from a full collinearity test range from 1.37 to 2.32 (see Appendix F), lower than the 3.3 threshold,
suggesting the absence of common method bias (Kock & Lynn, 2012).
30
Table 3. Psychometric Properties of First-order Constructs
CONSTRUCT
ITEMS
LOADING
AVE
α
CR
MNG
RCN
ALT
ENJ
ENT
LRN
SCL
CII
CIC
CIS
PIA
PCA
PSA
Monetary Gain
(M = 4.82; SD = 1.64)
MNG1
0.91
0.75
0.88
0.92
0.86
MNG2
0.81
MNG3
0.90
MNG4
0.84
Recognition
(M = 4.43; SD = 1.53)
RCN1
0.89
0.79
0.91
0.94
0.41
0.89
RCN2
0.90
RCN3
0.88
RCN4
0.89
Altruism
(M = 4.86; SD = 1.38)
ALT1
0.84
0.63
0.79
0.87
0.20
0.37
0.79
ALT2
0.90
ALT3
0.89
ALT4
0.47
Enjoyment
(M = 5.08; SD = 1.51)
ENJ1
0.95
0.86
0.94
0.96
0.27
0.22
0.56
0.93
ENJ2
0.95
ENJ3
0.95
ENJ4
0.86
Entrepreneurship
(M = 5.20; SD = 1.40)
ENT1
0.85
0.62
0.79
0.87
0.30
0.34
0.33
0.53
0.79
ENT2
0.73
ENT3
0.76
ENT4
0.81
Learning
(M = 5.39; SD = 1.23)
LRN1
0.87
0.69
0.85
0.90
0.23
0.50
0.41
0.43
0.57
0.83
LRN2
0.83
LRN3
0.82
LRN4
0.82
Socialization
(M = 4.81; SD = 1.41)
SCL1
0.90
0.79
0.91
0.94
0.23
0.50
0.59
0.45
0.15
0.33
0.89
SCL2
0.89
SCL3
0.87
SCL4
0.89
Continuous
Intention to
Ideate
(M = 5.27; SD = 1.46)
CII1
0.92
0.75
0.89
0.92
0.48
0.48
0.40
0.46
0.48
0.47
0.31
0.87
CII2
0.87
CII3
0.77
CII4
0.89
Continuous
Intention to
Collaborate
(M = 5.23; SD = 1.37)
CIC1
0.90
0.71
0.86
0.91
0.32
0.35
0.41
0.45
0.46
0.60
0.24
0.75
0.84
CIC2
0.87
CIC3
0.73
CIS4
0.87
Continuous
Intention to
Socialize
(M = 5.06; SD = 1.35)
CIS1
0.91
0.70
0.85
0.90
0.34
0.34
0.65
0.53
0.23
0.36
0.68
0.53
0.52
0.84
CIS2
0.89
CIS3
0.63
CIS4
0.87
Platform Ideation
Affordances
(M = 5.16; SD = 1.33)
PIA1
0.92
0.77
0.90
0.93
0.36
0.36
0.34
0.42
0.43
0.50
0.24
0.64
0.57
0.41
0.88
PIA2
0.93
PIA3
0.82
PIA4
0.84
Platform
Collaboration
Affordances
(M = 5.25; SD = 1.25)
PCA1
0.82
0.69
0.85
0.90
0.25
0.25
0.30
0.37
0.34
0.53
0.26
0.45
0.61
0.39
0.59
0.83
PCA2
0.84
PCA3
0.86
PCA4
0.81
Platform
Socialization
Affordances
(M = 5.01; SD = 1.29 )
PSA1
0.88
0.78
0.91
0.94
0.29
0.29
0.54
0.41
0.22
0.38
0.56
0.39
0.40
0.65
0.45
0.51
0.88
PSA2
0.87
PSA3
0.90
PSA4
0.89
Note: The diagonal elements are the square root of the shared variance between the constructs and their measures.
Monetary Gain (MNG); Recognition (RCN); Altruism (ALT); Enjoyment (ENJ); Entrepreneurship (ENT); Learning (LRN); Socialization
(SCL); Continuous Intention to Ideate (CII); Continuous Intention to Collaborate (CIC); Continuous Intention to Socialize (CIS);
Platform Ideation Affordances (PIA); Platform Collaboration Affordances (PCA); Platform Socialization Affordances (PSA).
31
5.3 Evaluation of 2nd-order Measurements
The evaluation of formative constructs––extrinsic and internalized extrinsic goals––involves an
assessment of the formative indicators’ (predictive) validity and multicollinearity (Joseph F Hair et
al., 2013). Indicator validity, which gauges the strength and significance of the path from the
indicator to the construct, was estimated using the PLS algorithm method with a bootstrapping of
samples to calculate the weight (relative importance) and loading (absolute importance) of each
indicator on its corresponding construct. As Table 4 shows, the weights and loadings of all the
indicators are significant, suggesting satisfactory indicator validity (Joseph F Hair et al., 2013). In
this study, multicollinearity was tested by computing the Variance Inflation Factor (VIF) of each
indicator (Diamantopoulos & Winklhofer, 2001). All computed VIF values are well below the
conservative threshold of 3.3, suggesting that multicollinearity is not a threat to the validity of the
study’s findings (Joseph F Hair et al., 2013). We also conducted additional quality assessment
for higher-order intrinsic goals. Cronbach’s alpha and composite reliability of higher-order intrinsic
goals was 0.76, indicating satisfactory internal consistency (Joseph F Hair et al., 2013). The AVE
of higher-order intrinsic goals was 0.68, higher than 0.50 threshold, and its square root was larger
than the correlations of this construct with the other constructs (Fornell & Larcker, 2006; Joseph
F Hair et al., 2013).
Table 4. Weights and Loadings of the Higher-Order Indicators
CONSTRUCT
INDICATOR*
VIF
LOADINGS
WEIGHTS
Loadings**
t -value
Weights**
t -value
Formative
Extrinsic Goals
(EXG)
MNG
RCN
®
®
EXG
EXG
1.16
1.16
0.90
0.73
16.90
8.83
0.73
0.46
7.77
3.95
Internalized Extrinsic
Goals (IEG)
ENT
LRN
®
®
IEG
IEG
1.50
1.50
0.80
0.95
13.41
31.79
0.38
0.73
3.54
8.20
Reflective
Intrinsic Goals (ING)
ENJ
ALT
SCL
¬
¬
¬
ING
ING
ING
1.56
1.80
1.43
0.87
0.87
0.78
37.11
45.10
22.25
0.41
0.43
0.38
15.84
19.86
14.57
* Monetary Gain (MNG), Recognition (RCN), Entrepreneurship (ENT), Learning (LRN), Altruism (ALT), Enjoyment
(ENJ), Socialization (SCL). ** all loadings and weights are significant at p < 0.00 level
32
5.4 Assessment of the Structural Model
The structural model was tested following the two-step procedure suggested by Wetzels et al.
(2009) for hierarchical models. In the first step, the latent variable scores for the second-order
constructs were obtained using PLS path modeling by specifying a latent variable that represents
all the manifest variables of the underlying first-order latent variables (Wetzels et al., 2009). In the
second step, the hypotheses were tested. The results of structural model analysis are illustrated
in Figure 3 and summarized in Tables 5 and 6. As no control variables (age, employment,
education, and gender) show significant effects on the three types of continuous intention, they
were excluded in further data analysis. As shown in Table 5, our model accounts for 56%, 54%,
and 58% of variance in actors’ continuous intention to ideate, collaborate, and socialize,
respectively. Extrinsic goals significantly influence intention to ideate (β = 0.32, p < 0.001), while
internalized extrinsic goals significantly influence intention to ideate (β = 0.18, p < 0.001) and
intention to collaborate (β = 0.41, p < 0.001). Intrinsic goals exert significant positive impact on
intentions to collaborate (β = 0.10, p < 0.05) and intention to socialize (β = 0.57, p < 0.001). Our
data thus provide support for hypotheses H1 to H5. We did not find support for relationships that
are not hypothesized in exploratory tests of other models.
Comparison of path coefficients using Cohen’s f 2 effect size (Selya, Rose, Dierker,
Hedeker, & Mermelstein, 2012) reveals that: (a) extrinsic goals (f 2 = 0.17) have greater predictive
power on intention to ideate when compared to internalized extrinsic goals (f 2 = 0.05) goals, (b)
internalized extrinsic goals (f 2 = 0.22) are more important in driving intention to collaborate than
are intrinsic goals (f 2 = 0.02), and (c) intrinsic goals (f 2 = 0.48) exert greater influence on intention
to socialize than do any other goals.
We also tested the relationships between continuous intention and actual contribution. The
continuous intention to ideate (β = 0.26, p < 0.05), collaborate (β = 0.37, p < 0.01), and socialize
(β = 0.29, p < 0.05) exert significant impact on actors’ actual contributions, thus corroborating
H9a-c. The three intention constructs account for 71% of variance in actual value creation.
33
Figure 2. Hypothesis Testing Results
Table 5. Results of the Structural Model Assessment
DEPENDENT VARIABLE
HYPOTHESIS
SUPPORT
ß
t
R2
Q2
Continuous intention to ideate (CII)
H1: EXGCII
H2: IEGCII
Supported
Supported
0.32***
0.18***
6.16
2.81
0.56
0.53
Continuous intention to collaborate (CIC)
H3: IEGCIC
H4: INGCIC
Supported
Supported
0.41***
0.10*
5.27
2.00
0.54
0.52
Continuous intention to socialize (CIS)
H5: INGCIS
Supported
0.57***
10.24
0.58
0.57
Actual contribution (AVC)
H9a: CIIAVC
H9b: CIC → AVC
H9c: CISAVC
Supported
Supported
Supported
0.26*
0.37***
0.29*
2.17
4.16
2.19
0.73
0.55
Extrinsic Goals (EXG), Internalized Extrinsic Goals (IEG), Intrinsic Goals (ING)
* p < 0.05; ** p < 0.01; *** p < 0.001; ns = no significant; β = path coefficients; R2 = determination coefficient; Q2 =
predictive relevance (calculated by Blindfolding).
Mediation analysis was conducted to explore the role perceived participation affordances play in
mediating the impact of goals on actorscontinuous intention. In this study, consistent with more
recent practices for testing indirect influence (Rucker, Preacher, Tormala, & Petty, 2011), we
adopted the bootstrapping method as the more rigorous and powerful approach to assess the
Intrinsic
Goals
Extrinsic
Goals
Internalized
Goals
Perceived
Socialization
Affordances
Perceived
Ideation
Affordances
Perceived
Collaboration
Affordances
Continuous
Intention to
Socialize
Continuous
Intention to
Ideate
Continuous
Intention to
Collaborate
Actual
Contribution*
Money
Recognition
Entrepreneurship
Learning
Socialization
Enjoyment
Altruism
PERCEIVED
PARTICIPATION
AFFORDANCES
GOALS * FOLLOW-UP
STUDY
BEHAVIORAL
INTENTION
Mediation Effects:
H6a***: EXG àPIA àCII CI: 0.06 - 0.20
H6b***: IEG àPIA àCII CI: 0.10 - 0.24
Mediation Effects:
H7a***: IEG àPCA àCIC CI: 0.10 - 0.24
H7b*: ING àPCA àCIC CI: 0.01 - 0.11
Mediation Effect:
H8***: ING àPSA àCIS CI: 0.08 - 0.24
H9a: 0.26*
H9c: 0.29*
H9b: 0.37***
H1: 0.32***
H2: 0.17***
H3: 0.41***
H5: 0.57***
H4: 0.10*
(R2= 0.55)
(R2= 0.54)
(R2= 0.58)
(R2= 0.73)
34
mediating role of perceived participation affordances (Preacher & Hayes, 2008). The
bootstrapping method is the preferable approach since the indirect effect is measured directly in
this method, rather than merely inferred to exist through a sequence of tests. Moreover, the
bootstrapping method imposes no assumption of the normality of the dataset, and thus is
recommended for studies with small sample size. In this study, the 95% confidence interval of the
indirect effects is obtained with 5,000 bootstrap resamples. The results of our analysis (see Table
6) reveal that perceived participation affordances significantly carry the influence of the goal
constructs on all the hypothesized intention constructs.
Specifically, the results confirm the role of perceived ideation affordances in mediating the
relationship between extrinsic goals and intention to ideate (H6a: βIndirect = 0.12, CI = 0.06 to 0.20)
and the relationship between internalized extrinsic goals and intention to ideate (H6b: βIndirect =
0.17, CI = 0.10 to 0.24), as none of the bias-corrected 95% confidence intervals contains zero.
Similarly, the results confirm the role of perceived collaboration affordances in mediating the
relationship between internalized extrinsic goals and intention to collaborate (H7a: βIndir ect = 0.17,
CI = 0.10 to 0.24) and the relationship between intrinsic goals and intention to collaborate (H7b:
βIndirect = 0.05, CI = 0.01 to 0.11). Lastly, the results support the role of perceived socialization
affordances in mediating the relationship between intrinsic goals and intention to socialize (H8:
βIndirect = 0.16, CI = 0.08 to 0.24). Therefore, our findings support H6a-b, H7a-b, and H8.
Table 6. Results of the Mediation Assessment
MEDIATOR
HYPOTHESIS
INDIRECT
EFFECT
t-VALUE
CONFIDENCE
INTERVAL
SUPPORTED
Perceived Ideation
Affordances (PIA)
H6a: EXGPIA → CII
H6b: IEGPIA → CII
0.12
0.17
3.40
4.61
0.06 - 0.20
0.10 - 0.24
Supported
Perceived Collaboration
Affordances (PCA)
H7a: IEGPCACIC
H7b: INGPCACIC
0.17
0.05
4.58
2.15
0.100.24
0.010.11
Supported
Perceived Socialization
Affordances (PSA)
H8: INGPSACIS
0.16
4.05
0.080.24
Supported
Extrinsic Goals (EXG), Internalized Extrinsic Goals (IEG), Intrinsic Goals (ING), Continuous Intention to Ideate (CII),
Continuous Intention to Collaborate (CIC), Continuous Intention to Socialize (CIS).
35
6 DISCUSSION
This study theoretically derives and empirically tests a model of actors’ goal-directed behavior on
SPD platforms to inform OI research and practice. Drawing on self-determination theory as well
as literature on OI and affordances, we advance hypotheses on how actors’ goals influence their
ongoing participation in SPD activities and how their perception of actions afforded by the SPD
platform also carries the influence of goals on behavioral intention. We test the hypothesized
relationships between goals, perceived affordances, intentions to participate, and actual
participation in SPD in a field survey of actual SPD participants, an instantiation of OI. This model
provides a nuanced understanding of the distinctive influences of different goals (extrinsic,
internalized extrinsic, intrinsic) that individual innovators bring to the OI context on three essential
categories of OI behaviors (ideation, collaboration, socialization). Distinguishing the indirect
influence of actor goals (via perceived platform affordances of the OI platform) from the direct
impact of actor goals helps explain how the OI platform influences actor’s behavioral intentions.
The study’s model and findings provide a theoretical basis to consider whether and how
actor goals align with the innovation behaviors intended (targeted) by the OI business model.
They also highlight the delicate balancing that OI platform sponsors face to engage participants,
individually and collectively, across the spectrum of OI behaviors through participation rules,
incentives, platform governance and innovation activities that may appeal to different actor goals.
Of note, we found that all three types of goals contribute to an actor’s continuous intention to
participate in SPD, but that each goal type has distinctive implications for the innovation behaviors
the actor is likely to engage in. Interestingly, we found that no one goal category predicts all three
types of innovation behaviors (i.e., ideation, collaboration, and socialization).
OI platforms need to engage individual innovators in generating ideas for new products or
services (ideation). The findings of this study indicate that extrinsic and internalized extrinsic goals
influence actors’ intention to ideate. Consistent with previous research in other OI contexts (Chen
et al., 2012; Salehan et al., 2014), we found extrinsic goals to be the strongest predictor of actors’
36
continuous intention to ideate on the SPD platform. This highlights the importance of the OI
platform rules and processes to attract creative individuals to ideation activities through tangible
rewards (money and recognition), and secondarily through opportunities to learn about innovation.
However, OI platform rules that appeal primarily to extrinsic goals are less likely to elicit
actors’ collaborative or socialization behaviors. Very large external rewards (e.g., cash prizes)
may encourage individuals to work together and compete as a team but discourage them from
working together as an innovation community (Liao & Xu, 2020). Studies of idea competition
platforms demonstrate that competition for extrinsic rewards (particularly cash prizes) may even
suppress collaborative OI behaviors or encourage destructive behaviors (K. J. Boudreau,
Lacetera, & Lakhani, 2011; Faullant & Dolfus, 2017; Füller et al., 2014; Hutter, Füller, Hautz,
Bilgram, & Matzler, 2015). This is a potential problem for OI models like SPD, which attempt to
attract creative individuals with extrinsic rewards while also encouraging collaborative and
community building behaviors. That is, OI platforms that seek to engage creative individuals
beyond initial ideation processes must consider other types of actor goals along with goal-relevant
innovation opportunities.
Collaboration among diverse stakeholders is a core premise of SPD as well as many OI
models (Annosi et al., 2020; Füller et al., 2014; Ungureanu, Cochis, Bertolotti, Mattarelli, &
Scapolan, 2020). As hypothesized, our findings indicate that internalized extrinsic and intrinsic
goals are significant predictors of actors’ continuous intention to collaborate with others on SPD
platforms; internalized extrinsic goals are a stronger predictor of collaboration than are intrinsic
goals. This finding suggests that individuals engage in collaborative activities, such as idea
refinement and product development, primarily to enhance their own innovation capacity and
develop entrepreneurship skills (internalized extrinsic goals), and to a lesser extent, to network
with other members, enjoy the process of collaboration, and help others. An implication of our
study of SPD is that extrinsic rewards (such as influence points) are less likely to engage SPD
actors in collaborative behaviors than would opening a variety of innovation processes (e.g.,
37
refining design, market research) to community members, thus allowing them to learn about
innovation processes, develop their entrepreneurial skills and work together with other creative
individuals on fun and challenging tasks.
SPD platforms, along with some other OI models, seek to stimulate socialization among
participants to build community and to facilitate and encourage continuing participation in ideation
and collaboration. We theorize that socialization behaviors in SPD are driven by intrinsic goals,
and our study’s findings indicate that intrinsic goals do predict actors’ continuous intentions to
socialize with others on SPD platforms. That is, when actors expect opportunities to have fun,
contribute to the community’s innovation success, and build their personal networks on the
platform, they are likely to engage in socialization behaviors on the OI platform. Socialization
behaviors also assist actors to pursue their other goals through ideation or collaboration behaviors
(Yetis Larsson, Di Gangi, & Teigland, 2019). However, an OI platform that too strongly promotes
socializing carries some risk of distracting members from ideation or collaboration tasks (Abhari,
Davidson, & Xiao, 2018); if socialization behaviors veer away from the platform’s core focus on
innovation towards general social interactions, the OI platform’s identity as a serious professional
innovation community may be weakened (cf. Mattson & Davidson, 2018; Ray, Kim, & Morris,
2014).
We theorize and empirically demonstrate that perceived SPD participation affordances
mediate the relationships between goals and continuous intentions to participate, confirming
perceived participation affordances as mechanisms through which distinct types of goals
influence actors’ SPD behaviors. SPD is made possible through the sociotechnical platform that
instantiates the OI process rules and provides the technical features and functions that individual
community members use to participate in innovation activities. However, on multi-featured OI
platforms, actors must seek out and attend to relevant features that afford actions they are
motivated to undertake. We found that both extrinsic and internalized extrinsic goals influenced
how SPD actors perceive ideation affordances, and their perceptions influenced their intentions
38
to continue to ideate. Similarly, internalized extrinsic and intrinsic goals increased actors’
awareness of how the SPD platform affords collaborative behaviors, contributing to their
intentions to collaborate. Intrinsic goals increased actors’ attention to how SPD features could
afford socialization behaviors, increasing their intentions to engage in socializing behaviors.
Finally, our results show that actors’ continuous intention to ideate, collaborate, and
socialize have a significant impact on their actual, measurable contributions to the SPD platform
(R2 = 0.73), providing support for our conceptualization of actor OI behaviors in terms of ideation,
collaboration, and socialization.
6.1 Implications for Research
Our study integrates and extends earlier OI research to develop a comprehensive, theory-based,
and empirically tested framework of goal-oriented OI behavior. Our empirical study is
contextualized for SPD platforms. However, the framework can inform studies of other OI
platforms, for instance to consider potential synergies and conflicts in actor goals and behaviors
in hybrid OI models, such as innovation contests within collaborative open source communities
(Liao & Xu, 2020). It also provides a common foundation for comparative studies across OI
platform types with differing rules, rewards, openness of innovation activities, and community
governance.
Our model has several key advantages over existing approaches. Drawing on OI literature
broadly, we conceptualize actors’ continuous intention to contribute as three distinct constructs
intention to ideate, intention to collaborate, and intention to socialize. In contrast to the single-
construct approach adopted by previous studies of OI participation (e.g. Fernandes & Remelhe,
2016; Zhang et al., 2015), our model provides a finer-grained view of OI behaviors, which allows
for examining how different goal types lead to different OI behaviors. This approach extends
beyond modeling OI as an ideation process by differentiating between ideation (idea submission)
and collaboration (ongoing and iterative interactions to improve ideas) as two distinct goal directed
39
OI behavioral categories. We also account for social aspects of OI to highlight community-building
behaviors that underlie and support project-focused ideation and collaboration. These behavioral
categories are sufficiently general to apply across a range of OI platforms while allowing for
contextualized operational definitions for a specific setting, as we did here for SPD.
Drawing from self-determination theory and prior studies on OI (Acar, 2019; Benbya &
Belbaly, 2010; Füller, 2010; von Krogh et al., 2012), our model extends understanding of actor
goals beyond the extrinsic-intrinsic dichotomy to articulate a parsimonious model of actor goals
for OI comprised of three higher-order goal constructs (extrinsic, internalized extrinsic and intrinsic
goals) with seven lower-order goals. Our specification of higher-order goal categories drew across
the OI literature, and this structure was then operationally defined and tested in the context of
SPD, an OI model that encompasses a wide range of OI activities and processes. These
categories provide a comprehensive framework of OI-relevant goals to study goal-directed
behaviors in a variety of OI models, which would facilitate comparison of how goals influence OI
behaviors under differing business rules and on different OI platforms.
Our research model develops a nuanced explanation of how different goals relate to
various OI behaviors and demonstrates the relative importance of each goal category in
influencing different facets of OI behaviors in the SPD context. Unraveling and contextualizing
these goal-behavior relationships can help researchers investigate why some innovation
behaviors do (or do not) develop on some platforms, based on the platform’s appeal to actors’
goals. We do not suggest there is an optimal or “right” mix of extrinsic, internalized extrinsic, and
intrinsic goals and of ideation, collaboration and socialization behaviors for all OI platforms. SPD
platforms generally appeal across this spectrum and thus provide a research context to examine
multiple goals and behaviors simultaneously. OI platforms dedicated to idea competitions, on the
other hand, might focus primarily on extrinsic goals to motivate ideation behaviors (Hofstetter,
Zhang, & Herrmann, 2018; Mack & Landau, 2020), whereas open source communities generally
prioritize collaborative behaviors aligned with internalized extrinsic goals (Belenzon &
40
Schankerman, 2015; Li et al., 2012). Our model provides a comprehensive framework to assess
what goals-behavior relationships may or may not be present or relevant in different contexts.
An actor’s goals directly influence her intentions to participate in innovation activities (or
not), but she must do so via features and functions on the OI platform that affords her desired
actions. We bring the OI platform to the theoretical foreground as an integral component of goal-
behavior relationships by arguing that perceptions of platform affordances are influenced by
actorsgoals, and that perceived affordances affect their likelihood of engaging in an afforded
action. This approach is consistent with the view of affordances as relational between an actor,
an artifact, and the actions afforded (S. K. Evans, Pearce, Vitak, & Treem, 2017; Fayard & Weeks,
2014; Nagy & Neff, 2015; Volkoff & Strong, 2013). However, our goal-affordances-behavior model
brings nuance to the concept of affordances by suggesting perceptions are formed under the
influence of actors’ goals, and that perception of affordances increases the likelihood that an
afforded action will occur. This role of perceived affordances in mediating the influence of actor
goals on behavioral intentions has not as yet received attention in the IS literature, as quantitative
approaches to specifying affordances and hypothesizing relationships between affordances and
other constructs are still rare (S. K. Evans et al., 2017). Assessing how perceived affordances
mediate goal-behavior relationships will be useful in OI platform design studies. OI platform
designer can consider whether participants perceive designed affordances as meeting their goals
through actions afforded, and thus increase their intentions to participate in OI activities, as
intended by the platform owner. The mediating influence of perceived affordances is also relevant
to study other domains, such as the use of persuasive or “nudgetechnologies on social platforms
intended to change individuals’ behaviors (Fogg, 1999; Piwek, Ellis, Andrews, & Joinson, 2016).
6.2 Implications for Practice
Understanding why actors participate in OI platform activities as they do can inform the design of
business rules and sociotechnical features to better align actor goals and behaviors and to
41
maintain their interest (Bauer, Franke, & Tuertscher, 2016; Bechmann & Lomborg, 2012; Henkel,
Schöberl, & Alexy, 2014; Sorensen & Torfing, 2011). Our study highlights the differential impact
of three goal types on three general categories of OI behavior and suggests that platform
designers consider what innovation behaviors they most want to encourage through the design
of reward systems and openness of innovation activities that appeal to different actor goals.
Rewards such as monetary gain and opportunities for recognition are likely to be more effective
in appealing to actors’ extrinsic goals and engaging them in ideation tasks. Opening innovation
processes and innovation activities to the community presents opportunities for learning and
developing entrepreneurship capacity that appeal to internalized extrinsic goals and thus may be
more effective to encourage collaboration. Highlighting how participation can be fun and increases
the welfare of the SPD community, for instance through success stories, could appeal to actors
intrinsic goals and thus heighten their intentions to socialize. Platform sponsors should also be
aware that actor goals and behaviors may at times be in conflict or incongruent with the OI
sponsor’s intention. Reviewing actual use, or surveying users regularly, could help identify
mismatches. Our research model and construct definitions provide insights on how to do so.
Effective platform design entails affording the actions actors desire through features and
functionalities. While it is important that an OI platform be easy and enjoyable to use (hence
addressing some intrinsic goals), our study suggests that how actors perceive features and
functions enable them to fulfill their own goals is a mechanism to drive OI behavioral intention.
Features that are not clearly relevant to actors’ goals are not only likely to be ignored but may
clutter the user interface and mask desirable affordances. Because different actors may perceive
the actions afforded by various features in different, even unexpected ways, observing users’
behaviors and querying their rationale for their actions (or lack of action) could provide useful
feedback for designers. Our research model outlines a useful framework for such feedback.
42
7 CONCLUSION, LIMITATIONS AND FUTURE RESEARCH
Through Internet and social media technologies, creative individuals around the world can now
engage in innovation through a variety of OI platforms and business models. To realize the full
potential of these creative crowds, OI sponsors need to understand how individuals’ goals
influence the types of innovative behaviors they are eager to participate in and to align actor goals
and innovation opportunities with the OI model through incentives, rules, processes, and
technology affordances (Lifshitz-assaf, 2018; Majchrzak et al., 2020; Randhawa et al., 2016). Our
study contributes to both IS and OI research a comprehensive, theoretically grounded foundation
for examining actor’s goal-directed behavior on individual-level OI platforms. Our research model
highlights general OI behavioral categories and offers a nuanced, systematic way to examine the
relative influence of diverse actor goals on these OI behaviors. This is particularly important when
OI platforms engage participants broadly in innovation, because diverse goals motivate
individuals’ participation in different innovation behaviors (Bogers et al., 2017; Kohler &
Chesbrough, 2020). We advance understanding of technology affordances for OI (cf. Nambisan
et al., 2017) by theorizing the mediating influence of perceived platform participation affordances
in goals-behavior relationships. Our study of social product development also builds knowledge
of a novel open source model and complements studies of OI in new product development (R. D.
Evans et al., 2018; Hidayanti, Herman, & Farida, 2018).
In this paper, we empirically examined OI goals and behaviors in a field survey of active
participants on an SPD platform, an approach that strengthens the external and ecological validity
of the study. Our empirical findings are limited to the context studied, but our theoretical model is
analytically general and could inform empirical studies of other OI settings (Lee & Baskerville,
2003). Research methods such as case studies, in-depth interviews, or user-experience
experiments can help assess the robustness of our model for other OI platforms. Longitudinal
studies in which researchers follow actors’ behavior in an OI platform could help assess how the
43
(re)design of business models and platform features influence the goals participants bring to the
platform, their perception of afforded actions, and activities they participate in.
This study lays a foundation for comparative studies of alternative OI models, which could
build knowledge of OI actors’ goal-directed behaviors more generally. Cross-sectional studies of
actor behavior in competing platforms or business models could then highlight the implications of
different rewards, rules, structures, and processes across OI platforms. Beyond the goals-
behavioral links, future studies of OI could consider approaches to social engagement, extent of
actor involvement, flexibility of roles, use of social technology, and diversity of activities performed
by actors on OI platforms. Such studies could inform the design of the mix of technological
capabilities for OI platforms and innovation activities directed at different sets of actor goals
(Ardolino, Saccani, Adrodegari, & Perona, 2020; Belenzon & Schankerman, 2015; Nambisan et
al., 2017). Finally, future research could also consider the influence of individual characteristics
such personalities, priorities, risk tolerance, and experience on actors’ behaviors relative to goals.
REFERENCES
Aarts, H., & Elliot, A. J. (2012). Goal-directed Behavior. New York: Psychology Press.
Abhari, K., & Davidson, E. J. (2016). Creative Co-production: The Adaption of an Open Innovation Model in Creative
Industries. In Information Systems and Management in Media and Entertainment Industries (pp. 119130).
Springer.
Abhari, K., Davidson, E. J., & Xiao, B. (2017). Co-Innovation Platform Affordances: Developing a Conceptual Model
and Measurement Instrument. Industrial Management & Data Systems, 117(5), 873895.
Abhari, K., Davidson, E. J., & Xiao, B. (2018). A Risk Worth Taking? The Effects of Risk and Prior Experience on Co-
Innovation Participation. Internet Research, 28(3), 804828.
Abhari, K., Davidson, E. J., & Xiao, B. (2019). Collaborative innovation in the sharing economy: Profiling social product
development actors through classification modeling. Internet Research, 29(5), 10141039.
Abhari, K., Davidson, E. J., & Xiao, B. (2020). Modeling Social Product Development Process, Technology, and
Governance. IEEE Transactions on Engineering Management, 114.
Acar, O. A. (2019). Motivations and solution appropriateness in crowdsourcing challenges for innovation. Research
Policy, 48(8), 113.
Allen, B. J., Chandrasekaran, D., & Basuroy, S. (2018). Design Crowdsourcing: The Impact on New Product
Performance of Sourcing Design Solutions from the Crowd. Journal of Marketing, 82(3), 106123.
Anderson, C., & Robey, D. (2017). Affordance potency: Explaining the actualization of technology affordances.
Information and Organization, 27(2), 100115.
Annosi, M. C., Marzi, G., Ciampi, F., & Rialti, R. (2020). An Ambidextrous Approach to Practice-Based Innovation for
Social Product Development: Lessons From A Dutch Company. IEEE Transactions on Engineering Management,
1–12.
Antikainen, M. J., Mäkipää, M., Ahonen, M., & M. (2010). Motivating and supporting collaboration in open innovation.
European Journal of Innovation Management, 13(1), 100119.
Antikainen, M. J., & Vaataja, H. K. (2010). Rewarding in open innovation communities--how to motivate members.
International Journal of Entrepreneurship and Innovation Management, 11(4), 440456.
Ardolino, M., Saccani, N., Adrodegari, F., & Perona, M. (2020). A business model framework to characterize digital
multisided platforms. Journal of Open Innovation: Technology, Market, and Complexity, 6(1), 10.
44
Austin, J. T., & Vancouver, J. B. (2005). Goal constructs in psychology: Structure, process, and content. Psychological
Bulletin, 120(3), 338375.
Ayer, A. J. (1972). Phenomenalism. In Philosophical Essays (pp. 125166). London: Palgrave Macmillan UK.
Barkin, E. (2003). Relative Phenomenalism Toward a More Plausible Theory of Mind. Journal of Consciousness Studies,
10(8), 313.
Battistella, C., & Nonino, F. (2012). What drives collective innovation? Exploring the system of drivers for motivations
in open innovation, Web-based platforms. Information Research, 17(1), 133.
Battistella, C., & Nonino, F. (2013). Exploring the impact of motivations on the attraction of innovation roles in open
innovation web-based platforms. Production Planning & Control, 24(23), 226245.
Bauer, J., Franke, N., & Tuertscher, P. (2016). Intellectual property norms in online communities: How user-organized
intellectual property regulation supports innovation. Information Systems Research, 27(4), 724750.
Bechmann, A., & Lomborg, S. (2012). Mapping actor roles in social media: Different perspectives on value creation in
theories of user participation. New Media & Society, 15(5), 765781.
Belenzon, S., & Schankerman, M. (2015). Motivation and sorting of human capital in open innovation. Strategic
Management Journal, 36(6), 795820.
Benbya, H., & Belbaly, N. (2010). Understanding Developers’ Motives in Open Source Projects: A Multi-Theoretical
Framework. Communications of the AIS, 27(30), 589610.
Bhattacherjee, A., & Premkumar, G. (2001). Understanding information systems continuance: an expectation-
confirmation model. MIS Quarterly, 25(3), 351370.
Bhimani, H., Mention, A. L., & Barlatier, P. J. (2019). Social media and innovation: A systematic literature review and
future research directions. Technological Forecasting and Social Change, 144, 251269.
Bogers, M., Chesbrough, H., & Moedas, C. (2018). Open innovation: Research, practices, and policies. California
Management Review, 60(2), 516.
Bogers, M., Zobel, A.-K. K., Afuah, A., Almirall, E., Brunswicker, S., Dahlander, L., … Ter Wal, A. L. J. J. (2017). The
open innovation research landscape: established perspectives and emerging themes across different levels of
analysis. Industry and Innovation, 24(1), 840.
Boudreau, K. J., Lacetera, N., & Lakhani, K. R. (2011). Incentives and problem uncertainty in innovation contests: An
empirical analysis. Management Science, 57(5), 843863.
Boudreau, K., & Lakhani, K. (2009). How to Manage Outside Innovation. MIT Sloan Management Review, 50(4), 69
76.
Brandstatter, V., Lengfelder, A., & Gollwitzer, P. M. (2001). Implementation Intentions and Efficient Action Initiation,
81(5), 946960.
Bretschneider, U., Rajagopalan, B., & Leimeister, J. M. (2012). Idea Generation in Virtual Communities for Innovation:
The Influence of Participants’ Motivation on Idea Quality. In 45th Annual Hawaii International Conference on
System Sciences (HICSS) (pp. 34673479). Maui.
Bstieler, L., & Hemmert, M. (2015). The effectiveness of relational and contractual governance in new product
development collaborations: Evidence from Korea. Technovation, 111.
Camacho, N., Nam, H., Kannan, P. K., & Stremersch, S. (2019). Tournaments to crowdsource innovation: The role of
moderator feedback and participation intensity. Journal of Marketing, 83(2), 138157.
Cañal-Bruland, R., & van der Kamp, J. (2009). Action goals influence action-specific perception. Psychonomic Bulletin
and Review, 16(6), 11001105.
Carillo, K., Huff, S., & Chawner, B. (2017). What makes a good contributor? Understanding contributor behavior within
large Free/Open Source Software projects A socialization perspective. Journal of Strategic Information Systems,
26(4), 322359.
Cenfetelli, R. T., & Bassellier, G. (2009). Interpretation of Formative Measurement in Information Systems Research.
MIS Quarterly, 33(4), 689707.
Chen, I. Y. L. (2007). The factors influencing members’ continuance intentions in professional virtual communities a
longitudinal study. Journal of Information Science, 33(4), 451467.
Chen, L., Marsden, J. R., & Zhang, Z. (2012). Theory and Analysis of Company-Sponsored Value Co-Creation. Journal
of Management Information Systems, 29(2), 141172.
Ciani, K. D., Sheldon, K. M., Hilpert, J. C., & Easter, M. A. (2011). The British Psychological Society Antecedents and
trajectories of achievement goals: A self-determination theory perspective. British Journal of Educational
Psychology, 81, 223243.
Coelho, D. A., Nunes, F., & Vieira, F. L. (2016). The impact of crowdsourcing in product development: an exploratory
study of Quirky based on the perspective of participants. International Journal of Design Creativity and Innovation,
6(12), 115.
Conner, M., & Armitage, C. J. (1998). Extending the Theory of Planned Behavior: A Review and Avenues for Further
Research. Journal of Applied Social Psychology, 28(15), 14291464.
Cooke, D. J., & Michie, C. (2001). Refining the construct of psychopathy: Towards a hierarchical model. Psychological
Assessment, 13(2), 171188.
Corral de Zubielqui, G., Fryges, H., & Jones, J. (2019). Social media, open innovation & HRM: Implications for
performance. Technological Forecasting and Social Change, 144(July 2017), 334347.
45
Costello, A. B., & Osborne, J. W. (2005). Best Practices in Exploratory Factor Analysis. Practical Assessment, Research
& Evaluation, 10(7), 19.
Deci, E. L., Olafsen, A. H., & Ryan, R. M. (2017). Self-Determination Theory in Work Organizations: The State of a
Science. Ssrn, 4(April), 1943.
Deci, E. L., & Ryan, R. M. (1980). The empirical exploration of intrinsic motivational processes. In L. Berkowitz (Ed.),
Advances in Experimental Social Psychology (pp. 3980). New York: Academic Press.
Deci, E. L., & Ryan, R. M. (1985). The general causality orientations scale: Self-determination in personality. Journal
of Research in Personality, 19(2), 109134.
Deci, E. L., & Ryan, R. M. (2000). The “what” and “why” of goal pursuits: Human needs and the self-termination of
behavior. Psychological Inquiry, 11(4), 227268.
Diamantopoulos, A., & Winklhofer, H. M. (2001). Index Construction with Formative Indicators: An Alternative to Scale
Development. Journal of Marketing Research, 38(2), 269277.
Dingler, A., & Enkel, E. (2016). Socialization and innovation: Insights from collaboration across industry boundaries.
Technological Forecasting and Social Change, 109, 5060.
Downes, P. E., Kristof-Brown, A. L., Judge, T. A., Todd, •, Darnold, C., & Darnold, T. C. (2017). Motivational
Mechanisms of Self-Concordance Theory: Goal-Specific Efficacy and PersonOrganization Fit. Journal of
Business and Psychology, 32(2), 197215.
Evans, R. D., Gao, J. X., Martin, N., & Simmonds, C. (2015). Integrating Social Knowledge and Collaboration Tools
into Dispersed Product Development. International Journal of Advanced Corporate Learning, 8(2), 20.
Evans, R. D., Gao, J. X., Martin, N., & Simmonds, C. (2018). A new paradigm for virtual knowledge sharing in product
development based on emergent social software platforms. Journal of Engineering Manufacture, 232(13), 2297
2308.
Evans, S. K., Pearce, K. E., Vitak, J., & Treem, J. W. (2017). Explicating Affordances: A Conceptual Framework for
Understanding Affordances in Communication Research. Journal of Computer-Mediated Communication, 22(1),
3552.
Faullant, R., & Dolfus, G. (2017). Everything community? Destructive processes in communities of crowdsourcing
competitions. Business Process Management Journal, 23(6), 11081128.
Fayard, A. L., & Weeks, J. (2014). Affordances for Practice. Information and Organization, 24(4), 236249.
Fernandes, T., & Remelhe, P. (2016). How to engage customers in co-creation: customers’ motivations for collaborative
innovation. Journal of Strategic Marketing, 24(34), 311326.
Fogg, B. J. (1999). Persuasive Technologies. Communications of the ACM, 42(5), 2629.
Forbes, H., & Schaefer, D. (2017). Social Product Development: The Democratization of Design, Manufacture and
Innovation. Procedia CIRP, 60, 404409.
Forbes, H., Schaefer, D., Panchal, J., & Han, J. (2019). A Design Framework for Social Product Development. IEEE
Transactions on Engineering Management, PP, 112.
Fornell, C., & Larcker, D. F. (2006). Structural Equation Models with Unobservable Variables and Measurement Error:
Algebra and Statistics. Journal of Marketing Research, 18(3), 382388.
Frey, K., Lüthje, C., & Haag, S. (2011). Whom Should Firms Attract to Open Innovation Platforms? The Role of
Knowledge Diversity and Motivation. Long Range Planning, 44(56), 397420.
Füller, J. (2006). Why consumers engage in virtual new product developments initiated by producers. Advances in
Consumer Research, 33(1), 639646.
Füller, J. (2010). Refining Virtual Co-Creation from a Consumer Perspective. California Management Review, 52(2),
98122.
Füller, J., Hutter, K., Hautz, J., & Matzler, K. (2014). User