The technological affordances in a business context
A qualitative systematic literature review
Niels IJntema S3445887 24 June 2019
University of Groningen Faculty of Economics & Business
Msc Business Administration: Change Management Supervisor: Dr. H. Bruns
Co-assessor: Dr. I. Maris-de Bresser Words: 12452
2
Abstract
Purpose: Although technology is developing rapidly, the last literature review about technological affordances is published five years ago. The purpose of this paper is to map the current landscape of the literature about technological affordances in the business context.
Methods: This study used a qualitative, systematic literature review and contains peer-reviewed articles from 2014 to 2019. In total, 60 articles are reviewed.
Findings: Most articles agree that planned affordances are expected to occur but differ in their potency. Whether affordances are actualized, depends on the user’s perception and user’s goal, where gamification can mediate perception. Actualized affordances lead to positive or negative outcomes.
Outcomes can create, strengthen or constrain other affordances. Contextual factors influence the whole process.
Conclusion: Different variables of the affordance theory are often confused by scholars; future research should work on clarity of the variables. Also, scholars strongly focussed on social media as technology. In the future, scholars should pay attention to a more equal distribution of the type of technology. Furthermore, the focus is mainly on the ‘technological side’ of affordances. The ‘human side’ is underexposed yet; in the future, scholars should investigate this side in more depth. Lastly, the use theory is intermingled with the affordance theory. Researchers should investigate the influence of the use theory on the affordance theory, to what extent both theories overlap, or even complement each other.
3
Table of content
Introduction ... 4
Methodology ... 6
Findings ... 6
Planned and Potential Affordances ... 7
Actualized Affordances ... 9
Outcomes ... 12
Contextual factors... 14
Discussion ... 19
Key Concepts ... 19
Findings in Light of Existing Literature ... 19
Limitations ... 22
Recommendations for Future Research ... 22
References ... 25
Appendix 1 Codes ... 32
Appendix 2 Main statements ... 33
Appendix 3 Distribution of papers ... 35
Appendix 4 Table of papers ... 39
4
Introduction
The concept of affordances refers to what we perceive a technology afford us to do, or how it can hinder us in reaching our goals (Markus and Silver, 2008). Gibson (1986), who had an ecologic psychologist perspective, came up with the affordance theory. He said that everyone can perceive the features of an object (e.g. its roundness) but the usage of those features, what they afford (e.g. rolling on), is relative to the user. The concept of affordances is about the relation between objects or technologies which have various properties – also called ‘features’ – and the users of the objects, who have personal needs, goals and characteristics (Majchrzak and Markus, 2012; Markus and Silver, 2008; Nambisan, Majchrzak, Lyytinen, & Song, 2017). This concept is useful in the field of business, for example to explain why and how the same technology has different outcomes or to predict outcomes (Nambisan et al., 2017).
Leonardi (2013) confirmed this: “often, new technologies are implemented and fail to bring about the kinds of changes to work that were envisioned by their champions” (p.772). It is important to review the existing literature; it provides an insight in the gaps of the literature that need to be filled. Therefore, this literature review maps the literature about technological affordances in the business field.
Leonardi (2011) emphasized the relational aspect of affordances by arguing that when users, with their current set of technologies and routines, were not able to achieve their desired goals, they changed the routines or the technologies. So, users evaluate technology on how they perceive the technology and how it fits their goals (Leonardi, 2011). Markus and Silver (2008) called this perception the ‘perceived ease of use’, and the ‘perceived relative advantage’, and Michaels and Carello (1981) wrote an example of this: “we would say that humans do not perceive chairs, pencils, and doughnuts;
they perceive for example places to sit, objects with which to write, and things to eat” (p. 42).
In 2013, Leonardi made in his article a distinction between the concepts of individualized -, collective-, and shared affordance. Individualized affordance is explained as an affordance that a person enacts while using a feature, while that affordance is not common by the fellow users; which means that this person can do something that the fellow users cannot. Collective affordance is created collectively by certain group members, which allows the whole group to accomplish something that otherwise was not possible; this is in some cases the result of pooled individual affordances. Shared affordance is explained as an affordance that is shared by the group as a whole. On the other hand, Strong et al. (2014) looked at affordances and goals at the individual and organizational level.
An alternative term for technological affordances that is often used in the literature, is socio- materiality. In the literature, this term is explained the same as technological affordances, which is for example shown in the article of Faulkner and Runde (2012), who defined socio-materiality as:
“technological objects are shaped by the activities of humans, that technological objects in turn shape human activities” (p. 64). In this literature review, I stick to the term ‘affordances’
5 Fayard and Weeks (2014) wrote the last literature review that provided a comprehensive view of technological affordances. Their literature review is my starting point. Fayard and Weeks (2014) performed a narrative literature review of the existing literature. They found that scholars take either a functional (also called dispositional or realist) or a relational perspective on affordances. A functional perspective means that human actors and technology are distinct; they are noticeable and when perceived, directly linked to practice. So, if arguing in extremes, researchers need to develop very detailed descriptions of technological features, assuming that technology provide certain affordances and not others so affordances (this makes affordances isolated). A relational perspective explains that affordance is about the relation between human and the technology. Because both perspectives have their drawbacks – the functional perspective does not take misperceived or not-perceived affordance into account, while the relational perspective is not clear in how to recognize an affordance and requires a higher level of abstraction – Fayard and Weeks (2014) stated that scholars should integrate these perspectives.
Beside the two perspectives, Fayard and Weeks (2014) found that the practical aspect is missing in literature about affordances. They argue that researchers mainly focus on affordance of technological features (i.e. affordance of email), while they should focus more on the practice enacted through the technology (i.e. for collaboration). Lastly, Fayards and Weeks (2014) argued that scholars should combine the concept of affordance with middle-ground theories; whereas affordance is about the physical and social construction of technology, middle-ground theories (i.e. habitus) can add the explanation of symbolic and social structures.
As explained in the beginning of this chapter, it is important to explore the gaps in the literature that need to be filled. Although technology is developing rapidly, the last literature review about technological affordances was written in 2014. So, it is time to perform a new literature review. The purpose of this thesis is to map the current landscape of the literature about technological affordances in the business context. I aim to answer the following research question:
“What is in the literature about technological affordances?”
In order to answer the research question, I performed a systematic, qualitative literature review. The last article Fayard and Weeks (2014) reviewed, was published in February 2014. Therefore, I reviewed the articles that appeared after their time span; which means a timespan of five years (February 2014 till February 2019). The contribution of this literature review is a comprehensive way of the concept of technological affordances. It maps the divergent and convergent areas in the literature, and explains the gaps in the research in the field of technological affordances which provide subjects or areas that needs future research.
6
Findings
As explained in the methodology section, I discovered themes in the literature; planned affordances, potential affordances, actualized affordances, outcomes and contextual factors. Appendix 3 shows the distribution of papers in terms of type of technology, industry, research design and journal. Figure 2 shows how the variables of the affordance theory are related. This is rather a process model rather than a variance model; the boxes to the left are necessary but not sufficient precursors of boxes to the right.
The arrows in the figure display that an element has influence on the element where the arrow is pointed at.
The model starts with planned affordances. These are the affordances that are predicted to occur, considering the developed features of the technology and the expected abilities of the user.
However, the potential affordances are determined by the actual implemented features and the actual abilities of the user. Whether an affordance is actualized or not depends not only on its potential; but on the perception and goals of users too. Gamification can mediate the user’s perception. Actualized affordances lead to positive or negative outcomes which enable or constrain new or other affordances.
The model is divided into boxes, which is discussed – in order of numbering – in the next sections.
Since the model is complex and contains a lot of arrows and lines, the contextual factors – that influence the whole process – are not displayed in the model but discussed in the last section of this chapter.
After reviewing all articles, I found out that a part of the relations I discovered were similar to the relations Anderson and Robey (2017) showed in their model. Therefore, the first part of my model is similar to the model of Anderson and Robey (2017). The variables that match their model are the planned affordances and potential affordances, influenced by the technology’s features and human abilities, and the actualized affordances, influenced by the user’s goals and perceptions. So, I found the relations Anderson and Robey (2017) to be confirmed in the literature, and expand their model with extra variables and relations I explored in the articles, which are gamification as a mediating factor of user’s perception, the affordance outcomes and the feedback loop.
7
Planned and Potential Affordances
The planned affordances are the expected affordances which are expected to occur, considering the designed features and the expected abilities of the users (Anderson & Robey, 2017). Not many scholars made a distinction between the planned and potential affordances, although affordances of a technology differ in their potency to help users reach their goals; like pharmaceutical drugs will differ in their potency for treating diseases. Therefore, not all planned affordances have potency, as
concluded by Jarzabkowski and Kaplan (2015): “Yet, many of us have long suspected that tools do not operate as they are ostensibly designed to do” (p.554). The concept of affordance potency is defined as: “the strength of the relationship between the abilities of the individual and the features of the system at the time of actualization, conditioned by the characteristics of the work environment”
(Anderson & Robey, p.103); in which the strength of the relationship means the likelihood of actualizing an affordance by the user. This means that although technology designers try to create affordances, the potency of affordances is always based on the implemented system features and the actual abilities of the individual (Anderson & Robey, 2017).
Figure 1 Concept of affordances. A part is adapted from “Affordance potency: Explaining the actualization of technology affordances” by C.
Anderson and D. Robey, 2017, Information and Organization, 27, p. 112. Copyright 2017 by Elsevier Ltd.
8 So, planned and potential affordances both depend on the technological features and user’s abilities. Potential affordances are considered in the expected context, whereas potential affordances are considered in the actual context. Therefore, the literature about planned affordances mainly wrote about the design of features, whereas the literature about potential affordances wrote about the features that were implemented. There is little about the expected or actual user’s abilities in the literature;
scholars who discussed this topic mainly emphasized the importance of understanding the user in order to align the design of features with goals and needs of the user.
Technological features are determinant for the affordance potency, because they can enable the user’s action (Ge & Gretzel, 2018). The features of a technology are interpreted differently in the literature. Some scholars consider features as visible characteristics of a technology, such the on/off button of a technology (Suh & Wagner, 2017). Other researchers give more insight in the capabilities of a technology in their way of describing the features, such as the mentioned rapid elasticity and abstraction for cloud computing technology (Krancher, Luther, & Jost (2018). The fact that some researchers make a clear distinction between ‘use’ and ‘affordance’, whereas others do not, can explain this difference (Leidner, Gonzalez and Koch, 2018). Sheer and Rice (2017), for example, made a clear distinction between use and affordances. They mentioned multiple usage types of mobile instant messaging, such as texting, transmission of images, create chat rooms, and label them
‘features’, whereas many other scholars consider those as affordances. Features of a technology can also constrain the user’s action. Sheer and Rice (2017) found for example the constraint that users of a mobile phone need to know the phone number of the other person. Ge and Gretzel (2018) found a constraint of social media in the character limitation, which is a barrier to provide details.
Leidner et al. (2018) criticize scholars who do not sufficiently distinguish between features, use and action. For example, Treem and Leonardi (2012) describes four affordances of social media;
visibility, editability, persistence and association, while these are according to Leidner et al. (2018) not affordances but technology attributes. Also, Majchrzak and Markus (2013) talked about the affordance
‘metavoicing’, while this is according to Leidner et al. (2018) just a direct use of the technology’s features, and therefore not an affordance. Leidner et al. (2018) took the isolated position that
affordances occur when actions are possible because technology is used. For example, “riding the train is the direct use of the object whereas working, sleeping, meditating, or conversing are not use of the train itself, but affordances made possible by the train ride” (p.120). Many scholars would say that the outcome in this example is that a person arrives at work, but Leidner et al. (2018) state that the arrival is just the outcome of riding the train; they think that the outcome in this example is that a person is able to complete more work in a day, or arrives more relaxed. For the technology Enterprise Social Media (ESM), they describe twelve affordances which all are actions, such as building relationships, helping peers, etc. (Leidner et al., 2018). So, Leidner et al. (2018) contradict the whole concept of affordances; the discussion contains recommendations for future research about this topic.
9 Some scholars disagree with the fact that most scholars take a single technology into
consideration; they think that it is more valuable if multiple technologies – that are used in combination while doing a task/job – are taken into consideration. Azad, Salamoun, Greenhill, &
Wood-Harper (2016), who interviewed 36 employees of a consultancy company, considered combined technology and came to the conclusion: “the smartphone that is email-enabled offers asynchronous connectivity affordance far beyond SMS” (p.18). So, two technologies that are combined, can lead to new affordances; or to a greater change that a certain desired affordance arise. Also, a combination of tool X (smartphone) with tool Y (e-mail), can have a higher possibility of leading to a desired
affordance than a combination of tool X (smartphone) with tool Z (SMS). So, the focus on a single technology limits the possibility to discover affordances that are created by combinations of technologies.
The potential of an affordance is higher in case of an alignment between the design of the technology and the goals and needs of the intended user. Many scholars argued that understanding the intended user during the design phase, is crucial to reach this alignment. In this way, the technology can be adapted to its user; technologies therefore often have different versions for different types of users, such as regular Facebook and Facebook Business. Other scholars went deeper than the types of users and considered the behaviour of users also. Choi and Im (2015), who performed an experiment among 423 young adults, are scholars who tried to understand their user, and found: “minimizing cognitive burden is a natural human behavior" (p.128). So, by ‘figuring out’ the user of the
technology, designers can consider requirements – in this case ease of use – in the design phase, which can increase the potential of the affordance.
Although most scholars agree that the abilities of the user influence process of affordances, they wrote scarcely about the expected or actual user’s abilities. Some scholars gave examples in the introduction of their paper – for example, the affordance ‘communication’ by a mobile phone do only have potency when the user is capable of using a mobile phone (Krancher et al., 2018) – but did not discuss it in their own research. However, some scholars mentioned that the user’s ability can change by convenient information (Bae, Jun, & Hough, 2016; Krancher et al., 2018). Krancher et al. (2018) interviewed 48 users of a cloud computing technology, distributed over 16 teams in the manufacturing industry, and argued; “However, our informants reported that they had to learn PaaS- based
development principles before they could ‘appropriately’ [12, p. 89] actualize the affordance” (p.796).
This means that user’s abilities are better aligned with the technology when convenient information about the technology is provided, for example in the form of training.
Actualized Affordances
The actualization of an affordance is defined as the “engagement of a user with an affordance in pursuit of specific goals as made possible by the affordance” (Strong et al. 2009, p. 6). Not all
10 potential affordances are actualized. So, the same technology can have different actualized
affordances, and different technologies can have the same actualized affordance. Factors that influence whether an potential affordance is actualized or not are the user’s perception, the user’s goals and the context (Leidner et al., 2018; Azad et al., 2016; Lehrer, Wieneke, vom Brocke, Jung, & Seidel, 2018;
Christianson, 2018; Petrakaki, Klecun, & Cornford, 2016; Van Osch & Steinfield, 2018; Kim, 2018;
Anderson and Robey, 2017).
User’s goals. The goals of the user influence whether a potential affordance is actualized or not. Some scholars state that if there is a disconnection between the affordance and user’s goals, the affordance will not be actualized (Anderson and Robey, 2017). Petrakaki et al. (2016), who performed semi structured interviews among users of an electronic patient record in a healthcare organization, confirmed this by saying: “technological affordances of change are realised when health- care professionals’ perceptions of technology fit in with their sense of (professional) self” (p.220); which means that users compare their goals with the affordances of a technology. The goals of the user are, beside other things, related to their work routine; in case of a new system, a user will often search for the features that fits their work routine. Some scholars argue that an affordance can lead to different routines or work practices in organizations (Azad et al., 2016; Robinson, Chan, & Lau, 2016).
Findikoglu and Watson-Manheim (2016), who observed and interviewed 18 users of an Electronic Health Record (EHR) in a healthcare organization, argued that this phenomenon occurs in the case of a misfit between the affordances and the goals and gave an example:
On the contrary, when the goals are misaligned, users perceive system features as constraints and show divergent EHR usage behaviors. As exemplified in the findings section, the
physicians’ aim was to record the patients’ anamnesis according to medical norms. However, the Ministry’s data structure did not support this goal. Instead, it was designed to capture only a narrower set of patient data. When faced with this constraint, some physicians attempted to create a limited discretionary workaround, such as taking notes in text boxes despite the fact that these notes could not be filtered or analyzed systematically afterwards. This finding supports prior research arguing that users reconcile their goals with the materiality of the technology and develop new routines of behaviors or imbrications (Leonardi, 2011) (p.397).
This example shows that users change their current work routine to match the goal and technology.
However, new routines are not always aligned with the goals of the organization. The chance of workarounds by the user increases, which can mean that work is not done as intended by the organization. Moreover, it is possible that because of workarounds, a certain application of the technology – that is in the organization considered as important – is ignored by the user (McKechnie
& Beatty, 2015).
User’s perception. Many scholars agree that an affordance potency need to be perceived by the user in order to be actualized (Kim, 2018; Petrakaki et al., 2016; Anderson & Robey, 2017; Van Osch & Steinfield, 2016; Brotheridge, Neufeld, & Dyck, 2015; Sheer and Rice, 2017; Huotari and
11 Hamari, 2017). Some scholars call it ‘awareness’ of the affordance (Suh & Wagner, 2017).
Gamification, signifiers and contextual factors mediate whether a user perceive the affordance or not, which is explained in the text below.
Some scholars in the literature found that gamification - adding gamified elements to the technology (Huotari & Hamari, 2017) – serves to increase the perception of an affordance. The motivational, or even called ‘hedonic’ value of gamification impacts the behavior of the user, which increases the chance of a perceived affordance (Suh & Wagner, 2017; Huotari & Hamari, 2017;
Sigala, 2018; Suh, Cheung, Ahuja, & Wagner, 2017). Sigala (2018), who wrote a conceptual paper about the technology-driven transformation of customer relationship management (CRM) into social CRM, described an example of gamification. Customers of airline Lufthansa were, beside many other rewards, rewarded with an ‘early bird badge’ if they checked in before 6 a.m. The extrinsic reward (the bagde) as well as the intrinsic rewards (increase self-esteem by sharing badge on social media), worked motivating for the customers to use the technology. However, Suh and Wagner (2017) found that perceptions of gamification features significantly differ; some actors saw the features of
gamification as a tool for progress tracking or goal setting, (rewardability; visibility of achievement) whereas others considered them as an opportunity for competition. The result was that some
gamification systems failed to engage users, whereas others successfully engaged users. So, adding gamification to a technology can encourage but also constrain affordance perception.
Beside gamification, Piccoli, Lui and Grün (2017) performed a mixed method study among users of a customer service system in the service industry and found that signifiers can provide guidance to the user, which increase the chance on the user’s perception and described how signifiers can be implemented in a system. They described:
[…] the redesigned learning subprocess on the MyPage site offered a menu of options, enabled real-time collection and storage of preferences, allowing the customer to check that they had been recorded. It is these features of the personalized page that serve as signifiers to ensure that customers perceive the service personalization affordance. Without those features, as in the traditional approach, guests may not be aware of the specific possibilities to
customize their experience (p.354).
So, the guidance provided by signifiers can have value, especially in the learning phase of a technology.
Level of affordances. The actualization of affordances can take place at different levels.
Scholars in the literature mainly considered the affordances at individual level (individual affordances lead to individual outcomes). However, a group of scholars take multiple levels into consideration, most of them acknowledged the levels ‘individual’, ‘collective’ and ‘shared’ affordances. These levels were created by Leonardi (2013), who studied how affordances were used between and within groups, while an organizational change is going on. The three levels were distinguished based on task
interdependence. An individualized affordance is explained as an affordance that a person enacts while
12 using a feature, while that affordance is not actualized by the fellow users; which means that this person can do something that the fellow users cannot. Collective affordance is created collectively by members of a group, which allows the whole group to accomplish something that otherwise was not possible; this is in some cases the result of pooled individual affordances. Shared affordance is explained as an affordance that is shared by the whole group: the sum of the individuals’ results is the result of the group (Leonardi, 2013).
One paper, who acknowledged the levels ‘individual’, ‘collective’ and ‘shared’, took the isolated position that another level of affordances can be identified: the connective level, which can be actualized by actors who complement each other (Vaast, Safadi, Lapointe, & Negoita, 2017). The scholars of this paper used a mixed-methods design, and defined three different but interdependent roles in users of Enterprise Social Media: “without amplifiers’ activity, advocates’ and supporters’
content would not have been further circulated. Without advocates and supporters’ activity, amplifiers would not have been able to circulate new content” (Vaast et al., 2017, p.1198). So, they suggest that although the users had different roles and different patterns of feature use, they were still
interdependent. The difference of collective- and connective level of affordances, is that the interdependence between users is at the collective level minimal – users complete their tasks independently, the team-results are the sum of the individuals’ results - while there is mutual dependence at the connective level (Vaast et al., 2017).
Some scholars who take other levels into consideration, argued that individual affordances can become or strength affordances at other levels (Van Osch & Steinfield, 2016; Plesner & Gulbrandsen, 2015; Aten & Thomas, 2016; Robinson et al., 2016). This phenomenon is also called ‘cross-level theorizing’, which is defined as “the relationship between independent and dependent variables at different levels” (Rousseau, 1985, p.20; Chatterjee et al., 2017). There are different thoughts in the literature why a transition between levels occurs. Firstly, some scholars argued that combined individual affordances can lead to affordances at other levels as a result of the degree of task interdependence (Aten & Thomas, 2016; Rehm & Goel, 2017; Verhulst & Rutkowski, 2018).
Secondly, it is explained by the occurrence that an outcome of a certain affordance can lead to a new affordance, which can be at another level (Burton-jones & Volkoff, 2017; Krancher et al., 2018;
Leidner et al., 2018; Suh and Wagner; 2017). This is described in more depth in the end of the section Outcomes.
Outcomes
Scholars agree that an actualized affordance results in a certain outcome. The outcomes described in the literature were very divergent. Some scholars described the outcomes of affordances generally (such as ‘standardized data’, ‘reduced efforts’), whereas other scholars considered outcomes more organization-specific (such as ‘appropriate clinical decision made’ for an health care institution)
13 (Burton-jones & Volkoff, 2017; Krancher et al., 2018). Some articles saw a certain behaviour of the user as the outcome of an affordance. Leonardi (2017), for example, surveyed 60 users of social media in a firm that provide financial services, and found that the affordances of social media affected the user’s knowledge sharing contribution by taking certain barriers away. This means that the outcome of an affordance also can be a (change of) behaviour.
Another way in which the outcomes were different, is that they can be at individual, group or organizational level (Plesner & Gulbrandsen, 2015). Outcomes can be both: positive and negative.
McKechnie and Beatty (2015), for example, acknowledged the possibility of negative outcomes. They conducted 22 interviews among users who shared the personal calendar with their colleagues.
Participants said about the notifications of the calendar: “it makes us ‘lazy’ because we don't have to remember anything [Female, age 33]. Further it can lead to procrastination since ‘no actions are taken until you get prompted to do so’ [Male, age 37]”. Anders (2016), who observed users of Team communication platforms, argued that the affordances of that technology – more social collaboration and information sharing – can beside positive outcomes also lead to cognitive overload of users. So, affordances of a technology can lead to negative outcomes.
The variety in the ways how the outcomes are described, is also noticed by Leidner et al.
(2018), who stated that use, affordances and outcomes of technology has become muddled; “the literature on affordance has been inconsistent in carefully distinguishing the outcome of affordance actualization from the affordance itself” (p.119). This means that affordances of a technology are by researchers often considered as the outcomes; for example, Strong et al. (2014) saw ‘standardized data’ as an outcome, whereas Leidner et al. (2018) argue that this is an affordance. Leidner et al.
(2018) defined several outcomes for Enterprise Social Media, which are for example cultural understanding, productivity enhancement, stress, etc. Although it has not been proven whether the position of Leidner et al. (2018) is right or wrong, it is visible in the literature that there is no consistency when something is an outcome and when it is not; which gives reason to doubt the strength of the affordance literature. The discussion section describes how this could be improved in future research.
Feedback loop. A group of scholars found that the outcomes of some affordances can contribute to the strength or occurrence of other (new) affordances; the feedback loop of affordances (Burton-jones
& Volkoff, 2017; Krancher et al., 2018; Leidner et al., 2018; Suh and Wagner; 2017; Chatterjee et al., 2015; Alam & Campbell, 2017). Some scholars who noticed that an affordance occurred due to another (previous) affordance, made a distinction between first-order (or basic level) affordances and second-order (or higher level) affordances. The latter (i.e. helping peers) can be actualized after the first-order affordances (i.e. interacting with peers) are actualized. However, this does not mean that first-order affordances are simpler or easier to activate than second-order affordances (Leidner et al., 2018; Suh and Wagner, 2017; Krancher et al., 2018).
14 Leinder et al. (2018), who focussed on Enterprise Social Media, and performed a case
study in a financial service company, described a second-order affordance that occurred as a result of a first-order affordance:
In our case, new hires meeting senior managers as a result of their participation in OnBoard events not only made the new hires more comfortable around their superiors, but also led to new affordances for senior managers, who recognized the potential insights new hires could provide into new product and service ideas and who therefore began soliciting feedback from new hires (p.131).
So, this example shows how the outcome of an affordance results in a new affordance. The new affordance does not always arise by the same individual; the outcome of an affordance can also lead to new affordances for different set of actors (Erhardt, Gibbs, Martin-Rios, & Sherblom, 2016; Leidner et al., 2018). However, the outcome of an affordance can also constrain another affordance (Paroutis, Franco & Papadopoulos, 2015).
Contextual factors
Around half of the papers paid attention to the contextual factors that influence the process of affordances. Some scholars mainly came up with contextual factors for a specific part of the affordance process. However, the fact that a certain contextual factors are described for each phase, does not mean that they do not have influence on other phases. In some cases, scholars considered a particular phase and mentioned contextual factors for that phase, but did not investigate whether those contextual factors impact other phases too. Scholars mainly described contextual factors for the actualization, perception and outcomes of affordances.
Factors as the support, governance, structure and financial situation of the organization were mentioned by multiple scholars (Jacobson, 2016; Krancher et al., 2018; Aversa, Cabantous, &
Haefliger, 2018; Bettany & Kerrane, 2016; Cardon, 2016; Alam & Campbell, 2017). The influence of the past is also determining whether an affordance is actualized or not: “The past was used as a resource as actors drew on it to make sense of the potential uses of a new material technology of memory (computers)” (Blagoev, Felten, & Kahn, 2018, p.1776). Jarzabkowski & Kaplan (2015) confirmed this, by their findings that users prefer familiarity. This occurrence of an actualized affordance is more likely if the tool is already embedded in organizational practice.
Various contextual factors influence the affordance outcome; therefore, the same actualized affordances can lead to different outcomes. Leidner et al. (2018) also found a contextual factor that explains why outcomes of the same affordance can differ. They surveyed 100 users of ESM in a financial service company and found that the most active users gained a greater network and more resources as result of the affordances. Therefore, they defined ‘human effort’ as contextual factor for the outcome. These examples show that the outcome of an actualized affordance is context dependent.
15 However, there are also some examples in the literature that different actualized affordances can still result in the same outcome. For example, Tafesse (2016) found in the context of social media that affordances for perceptual experience (defined as the activation of someone’s hearing, seeing), affordances for social experience (defined as consumers participation) and affordances for epistemic experience (defined as the consumer’s cognition of relevant brand information) all lead to consumer engagement.
There are multiple contextual factors that influence the user’s perception of the affordance.
These contextual factors are mainly about the user as a person, such as the ethics, professional values, privacy concerns, motivation and personality traits – is the user openminded or not? – of the user (Bae, et al., 2016; Petrakaki et al., 2016; Mettler & Winter, 2016; Lai, Hill & Ma, 2015). Petrakaki et al.
(2016) mentioned the strength of their hierarchical position; users lower in hierarchy are more likely to just do what they are told. Lastly, the frequency of use was also mentioned as a contextual factor. The more the technology is used (by the same user), the higher the change that a user will perceive a certain affordance (Sheer & Rice, 2017). One paper paid attention to the contextual factors that influence the existence of a feedback loop. Krancher et al. (2018) interviewed 48 users of a cloud computing technology, distributed over 16 teams in the manufacturing industry, and argued; “The degree to which teams actualized higher-level affordances depended on the work environment
characteristic of self-contained tasks” (p.789). However, this is not confirmed or contradicted by other papers, so future research needs to investigate this contextual factor.
Although most contextual factors are about the user and organization (micro and meso level), some papers paid attention to the culture of the country (macro level). Chen, Lee and Hwang (2018), who performed experiments in a service organization with employees from different countries, considered the response of those employees on waiting screens, and argued:
That is, people from Taiwan (i.e., a strong uncertainty avoidance society) experience a significantly larger difference of perceived uncertainty when they view a progress bar with high time affordance than a progress indicator with low time affordance compared to those from U.S. (i.e., a weak uncertainty avoidance society). (p.569).
So, the alignment between technology and the user’s needs or goals differ per culture.
Although scholars came up with multiple contextual factors that influence the process of affordances, they are often detailed, specific to the type of technology and industry and therefore not generalizable. One of the pie charts in appendix 3 shows that a third of the studies considered a
network technology; which was in most cases social media. This reinforces the idea that the contextual factors are not generalizable. Besides contextual factors, another type of affordance was found in the literature that influenced the outcome of a technological affordance; spatial affordances (Erhardt et al., 2016; Sandström et al., 2016). With a multi-method case analysis in an insurance company that focused on e-mailing, Erhardt et al. (2016) found that affordances of space (face-to-face meetings), had a positive influence on the outcomes of technological affordances (team learning).
16
Methodology
A literature review means that existing studies in a specific field are gathered, critically appraised, combined, analysed and in the end summarized (Liberati et al., 2009; Wolfswinkel, Furtmueller, &
Wilderom, 2013; Zurynski, 2014). This literature review covers academic, peer reviewed articles over the last five years; February 2014 till February 2019, since the timeframe of the most recent literature review ran until January 2014 (Fayard & Weeks, 2014). The nature of the review is systematic, which entails according to Tranfield, Denyer and Smart (2003) a trail of the decisions, procedures and conclusions of the reviewer. For systematic reviews, literature is used as data to be coded, analysed and synthesised in order to reach overall conclusions (Ridley, 2012). To this end, I followed methods outlined in the book “The Literature Review” by Ridley (2012) and the article “Using grounded theory as a method for rigorously reviewing literature” by Wolfswinkel et al. (2013). I went through five stages: define, search, select, analyse and present (Wolfswinkel et al. 2013).
Define and search. I used the database Web of Science, a comprehensive database with extensive search filters. To include all derivations of this term, I searched for “affordanc*”. I chose to select within Web of Science the categories ‘Business’ and ‘Management’, since I write my thesis in the context of a Business Administration master. I searched for the mentioned search terms in the topic of articles, within the time span of February 2014 until February 2019. This resulted in 3431 articles.
After I selected the categories ‘business’ and ‘management’, 121 articles were left. All of these articles were written in English, so I did not have to exclude articles based on language.
Select. I read the title and abstract and scanned the text of each article to decide if the concept of affordance was central in the article. I excluded articles in which the concept of affordances was just used to introduce a certain subject or theory, or was just used in the theory building. I double-checked the centrality by exploring whether the concept of affordances was a considerable part of the study or research question, and the paper therefore contributes to the affordance literature. I excluded 37 articles in which the concept of affordances was not central. Then, I checked whether the articles address technological affordances rather than other types, such as the affordance of power. 24 articles were not about technological affordances, and therefore excluded. This resulted in a list of 60 articles for the literature review. I kept track of the literature in a Microsoft Excel table (see Appendix 4). The search strategy which I described in the define-, search- and select stage, is summarized in Figure 1.
17
Figure 2 Search strategy
Analyse. In this stage, I carefully read and coded each article. I performed three ways of coding.
Firstly, I performed open coding; for this way of coding, I tried to explore themes and categories while reading. In their literature review about affordances. In their literature review, Fayard and Weeks (2014) did not write around categories and themes, but were rather ‘storytelling’ while briefly mention different variables of the affordance theory. However, other literature – before 2014 – describes themes as the affordance potential, -perception, -actualization and the effect of affordances. While reading and coding the articles, I noticed various themes that were for many articles similar, but slightly differed from the themes of previous literature; planned affordances, potential affordances, actualized affordances, outcomes and contextual factors. Secondly, I performed axial coding; this way of coding explored sub-categories (and their properties) and created hierarchy of importance between codes. For example, the sub-categories of the theme ‘affordance potential’ are human abilities and technological features. Finally, I performed selective coding; this way of coding explored the
relationships within (sub-)categories. An example is that I started with the code ‘human factors’. After reading, I realized that this code was not comprehensive enough and divided it into the codes: human skills, human capabilities, human perception and human goals. Later on, I realized that the codes
‘human skills’ and ‘human capabilities’ could be combined into the code ‘human abilities’, and that the codes ‘human goals’ and ‘human perception’ determines the ‘affordance actualization’.
18 For each article, I performed those three ways of coding iteratively; when new insights or extra information about a concept or the relationships between concepts arose, I recoded the articles. When no new insights occur after all the coding, I reached the theoretical saturation. Appendix 1 shows the open and axial ways of coding, which are the main categories and sub- categories. The model of figure 2 in the findings section, shows the relationships between codes and therefore the selective way of coding.
Present. In this last stage, I structured and represent the content that is gained from the previous stages. I put my set of 60 articles in the program Mendeley which enabled me to mark the text and add notes where I wrote down the codes. I kept track of what I read; I grouped the articles based on the codes, described the purpose, type of technology, research design and type of industry in a table of Microsoft Excel (see appendix 4). I also wrote down the main statements – an example is shown in appendix 2– and brief summaries of the articles and kept interesting citations apart. I wrote the findings section while reading and after I reached the theoretical saturation, I read my tables, summaries, statements and citations again in order to complete the findings section. I organized my findings section based on the main- and sub-categories; the model in figure 2 shows the connections I found between them. In order to write the discussion section, I summarized the key concepts of the findings, compared the findings with the existing literature, wrote my conclusions about the
developments, described the limitation of my study and wrote recommendations for future research, based on the previous sections.
19
Discussion
Key Concepts
The aim of this thesis is to map the current landscape of the literature about technological affordances in the business context. Technologies have certain planned affordances which are expected to occur, based on the developed features and the expected abilities of the users. The potential affordances can differ significantly from planned affordances because of variations in the users’ abilities and in the way features are implemented in the actual context. Whether the potential affordances is actualized, depends on the users’ goals and the users’ perception, where gamification can mediate the perception.
Actualized affordances lead to outcomes which can strengthen or constrain other affordances, or lead to the occurrence of new affordances. The whole process is influenced by contextual factors.
Findings in Light of Existing Literature
As mentioned in the Introduction, the literature review of Fayard and Weeks (2014), the starting point of this literature review, made a distinction between dispositional (functional) and relational
affordances. They concluded with a suggestion that researchers should take an integrated view. Over the last five years, the dispositional and relational perspectives are acknowledged by a few scholars, however; they mostly explained both views but did not describe which perspective they chose. None of the papers acknowledged or mentioned the integrative view. However, it seems that scholars use aspects of both, the dispositional and relational view, and in that way take an integrative perspective which was recommended by Fayard and weeks (2014). If researchers would take the dispositional perspective, they would only describe which possible affordances a certain technology have, which actions the technology affords. If researchers would take the relational perspective, they would only describe what actions the users perceive in a technology. Instead, researchers assume that a certain technology affords certain actions (dispositional perspective), acknowledging that the abilities, goals and perception (relational perspective) of the user also determines whether the technology affords a certain action or not. So, researchers combine aspects of both perspectives and therefore take an integrative perspective.
In conclusion, although scholars take the integrative view Fayard and Weeks (2014) recommended, it seems that they are not aware of it since none of them acknowledged this
perspective. Moreover, the fact that a few scholars explained the perspectives briefly instead of taking a position and do not describe how they combine the views, points out that scholars do have little knowledge about the dispositional, relational or integrative view. By not taking a position, they leave the perspective(s) they chose to the interpretation, which can cause confusion and incorrect
20 interpretation of the reader. Moreover, it is striking that scholars focussed on the dispositional aspects rather than relational aspects by combining the perspectives. It seems easier for the scholars to describe the technology’s features and the possible actions they can afford than the – influence of – goals, abilities and perception of the users. This is a disturbing development; it is possible that scholars choose the ‘easy’ parts while combining the dispositional and relational view, which can cause the
‘hard’ parts – in this case the human side – to be ignored. However, it is also a possibility that the functional aspects are closer to the reality than the relational aspect. Recommendations for future research on this topic is described later this chapter.
Besides the distinction between the dispositional, relational and integrative perspective, Fayard and Weeks (2014) also concluded that the focus of affordances should be on the practice enacted trough technology (i.e. collaboration, communication) rather than the features or the
environment’s characteristics. However, in the literature is still much variety and conflicting ways in how researchers describe affordances. What is an affordance for the one scholar, is according to another scholar just a feature. It is a possibility that this variety is due to a lack of clearly defined variables of the affordance concept, which was argued by some scholars. Otherwise, suppose that a part of the scholars follows the advice of Fayard and Weeks (2014) to consider affordances more as a practice, it is possible that the affordance theory is in a ‘transition’ towards the more practice-based view. In that case, a part of the scholars will take the practice-based view, while the other part of the scholars still focusses on the features; which result in variety and conflicting ways in how affordances are defined.
Fayard and Weeks (2014) are in favour of combining the concept of affordances with middle- ground theories, and suggested the habitus theory: “Together, affordances and habitus shape the possibilities for action that show up, either consciously or unconsciously, for the actor” (p.247). Over the last five years, this recommendation was neglected by the scholars. Only a few scholars took a middle-ground theory – also habitus – into consideration. However, whether they did it as a result of the recommendation is not clear, because they did not refer to Fayard and Weeks (2014).
Over the last five years, there are developments in the literature that are not discussed by Fayard and Weeks (2014). Multiple scholars considered gamification as a mediating factor of
affordance perception. Also, although most scholars stick to the first-order affordances, some scholars acknowledge the existence of the feedback loop, and begin to search for patterns how and when the outcome of an affordance create or strengthen another one. Considering the impact of other type of affordances – for this study mainly spatial and motivational affordances – on technological
affordances is a new, but rare, development in the literature. A part of the by scholars discussed contextual factors, might come from other types of affordances. Fayard and Weeks (2014)
acknowledged the impact of the context but did not make a distinction on what specific variable or phase – i.e. actualization, outcome – the contextual factor has influence on. However, the indication of scholars which phase is influenced by a contextual factor does not mean that other phases are not
21 influenced by the contextual factor; it is possible that the scholars took only one phase into
consideration or confuse the different phases.
As described in the Introduction, two scholars made a distinction between levels of
affordances: individual, group and organizational (Strong et al., 2014), or individual, collective and shared (Leonardi, 2013). Although these papers are reviewed by Fayard and Weeks (2014), they did not discuss the different levels of affordances. Over the last five years, scholars increasingly
acknowledged the distinction of affordance levels; the levels described by Leonardi (2013) are described multiple times. None of them mentioned the levels of Strong et al. (2014). So, the levels of Leonardi (2013) are widely accepted in the literature. Although scholars in general seem to hesitate in applying the levels to their results, some of them tried to assign certain affordances to the related levels. One scholar described a new level of affordances as an addition to the individual, collective and shared level; the connective level of affordances. However, the general focus of the literature is on affordances at individual level.
Another development is the possibility for subjectivity in the affordance literature: scholars sometimes seem to confuse elements and phases of the affordance theory and interpret them in the way that suits their research. The definitions and examples researchers gave of features, affordances and outcomes are different, and sometimes contradicting. Additionally, scholars intermingle the theory of use with the theory of affordances. Aspects that are related to the theory of use are visible in the theory of affordances; scholars used terms in the affordance concept that are central elements in the theory of use. Examples of those elements are the ease of use and usefulness of technology, which predict user intention according to the Technology Acceptance Model (TAM) and Theory of Planned Behaviour (TPB) model (Mathieson, 1991). Scholars have conflicting opinions about the value of the present elements of the use theory in the affordance theory; future research should find out whether this intermingling adds value or is a decay for the affordance theory. The section ‘Recommendations for future research’ describes how this future research should look like.
Although the concept of affordances is about the relationship between human and material, the literature of technological affordances mainly focusses on technology and less on the human. The discussed factors of the user – abilities, goals and perception – are not comprehensive in explaining the drive of the user. Scholars – and therefore the model – does not take for example personality, mood, workload and demographic characteristics into account. I think these are important aspects to consider because I can imagine that a person with a high workload is not that into ‘exploring’ the technology, which impacts the affordance actualization. Also, it seems logical to me that elderly persons have a lower affordance perception. My suggestion is to add a box to the model that is about the user’s characteristics.
This study contributes to the development of affordance theory by providing a comprehensive overview of the technological affordance literature over the last five years. Since the last literature review of Fayard and Weeks (2014) did not provide a model, the relation between elements in the
22 affordance theory are not all clear in their review. Therefore, my findings also extent the literature by providing a model which shows the relation between the variables of the affordance theory. As explained in the beginning of the findings section, a part of this model is equal to the model of
Anderson and Robey (2017). The model contains elements of both, the relational and the dispositional view. Lastly, this study contributes to the development of technological affordance theory by
recommendations for future research, which are described at the end of this section.
Limitations
I think my focus on ‘technological’ affordance limited me in mapping the ‘human side’ of the affordance theory. I guess that literature in the psychological field, will have interesting things to say about the human factors. For example, consider the user’s goals, which impacts the actualization of an affordance. Scholars mainly wrote about the importance of alignment between the user’s goals and the technology. However, there are much – psychological – elements might impact the goals of a user;
such as norms, values and group pressure. So, for this example, I think it would have been valuable to consider literature that contains such psychological elements and explains how human goals arise.
However, only one paper was from a psychological journal, which I think is due to the focus on
‘technological’ affordances.
Recommendations for Future Research
In appendix 3, pie charts show the distribution of the research over the last five years, in terms of technology, industry and nature of research. A third of the papers took a network technology – which mainly consist of (enterprise) social media – into consideration. The other two-third types of
technology are shredded into small groups. The distribution of research design is not remarkable; there are not very dominant groups on this topic. Many researchers performed their research in the service industry. The focus on social media and service industry impacts the results by making it less generalizable for all types of technology. A recommendation for future research is to focus on other types of technology and other types of industries. Modern types of technology – blockchain – can add value to the affordance literature by providing new insights.
Based on unfilled gaps or conflicting results in the literature, I found several points for future research. First of all, scholars wrote a lot about contextual factors that researchers found to have influence on the process of affordances. However, these contextual factors are varied, seem a bit random and not generalizable. Research on contingency factors for the affordance process would be valuable. Whereas researchers now describe characteristics of their context that could have an
influence, there should be certain generalizable factors which have a positive or negative relation with the concept of affordances. So, for example; the strength of leadership is positively related to the user’s perception. It would make the actualization of affordance more predictive, and the theory more
23 prescriptive. Quantitative research can be a way to reach this; it measures the strength of the
relationships. A remarkable fact is that the described contextual factors are almost all on the micro and meso level. So, it would be interesting to take more macro-level aspects into account while
investigating contingency factors. For example, the influence of law on the affordance process.
Although scholars emphasize the importance of understanding the user in order to increase affordance potency by aligning technology and user, not much is written about the type of users. It would be interesting to investigate what user characteristics – besides abilities – impacts an affordance; demographic aspects, such as gender, age, etcetera. So, scholars should for example investigate the relation between age and affordance perception, gender and goals, and other
demographic characteristics in relation with elements of the affordance theory. Also, as mentioned in the limitations, adding psychological literature into the affordance research would be interesting to explore patterns in human factors by learning more about the characteristics of the user. For example, gamification can enhance affordance perception because of the motivational value. Psychological literature can add value by exploring the motivation of humans.
Most of the researchers consider a single type of technology; the fact that a technology is often used in combination with other technologies is most of the times not taken into account. This is also an interesting cause for research, since the combination of technologies can on itself lead to new
affordances too. An example of a combination is social media and big data. Besides considering a single type of technology, researchers also performed their research mostly in a single setting. It is to perform more research about aspects of affordances in multiple industries. It might reveal remarkable similarities or differences, patterns, or other valuable information. Also, almost two-thirds of the papers performed their research in the service industry or health care industry; so, the research need a better distribution of industries. It would add value to find contingency factors which are needed in the affordance literature (as I explained in the previous paragraph).
The studies in the field of affordances are often short-term oriented in the way that researchers mainly consider the basic-order (or: first order) affordances of a technology. They do often not look at the effects of the affordance outcomes, while those outcomes can create, stimulate or constrain one or multiple (new) affordances So, firstly, it is interesting to consider the process of affordances rather as a
‘circle’, whereby the higher-order affordances are taken into account too. Secondly, the literature does not provide clear conditions whether a higher-order affordance occurs or not. Longitudinal case studies would add value at this point. Although this kind of study is difficult to do and to publish;
examining circle of basic-level affordances and higher-level affordances (feedback loop) requires time.
Another topic that requires time and asks for longitudinal case studies is the different levels of affordances; individual, shared and collective. The different levels of affordances are explained in the literature, and several scholars assign certain affordances to the related level. However, there is not much in the literature about the transition between the different levels. More research on how often
