Reaching full potential: the actualization of
shared affordances
Theory development Demy van Leusden Rijksuniversiteit Groningen Faculty of Economics and Business
Postbus 72 9700AB Groningen Supervisor: Hille Bruns d.van.leusden@student.rug.nl
2
SUMMARY
3
INTRODUCTION
In their study (2013), Seidel, Recker & Vom Brocke show how information systems could support the ‘green transformation’ of a global software solutions provider. They concluded that the transformation to a more sustainable organization was facilitated by the realization of functional affordances. In this specific organization, one of the material properties of the information system was that it offered the possibility to control the number of prints employees made. This material property led to the emergence of a so-called functional affordance: the ability of the organization to perform output management by controlling the amount of paper that was used for printing. After this affordance was successfully realized, work practices became more sustainable and paper consumption was reduced by 28% per employee (Seidel, Recker & Vom Brocke, 2013). The realization of the output management affordance, combined with the realization of three other affordances, facilitated organizational learning and more sustainable work practices. This in turn helped reaching the overarching organizational goal - becoming more sustainable.
Markus & Silver (2008) define affordances as the relationship between an individual user or user-group and a technological object that indicates what the user can possibly do with that technological object - given the capabilities and goals of the user. Affordances are the potential - not actual - uses of a technological object that an individual perceives; the perspective of the user is important. They are possibilities for action that emerge from a relation between information technology (IT), often considered in terms of IT features, and the systems that are present in organizations (Zammuto, Griffith, Majchrzak, Dougherty & Faraj, 2007). However, in order to benefit from these potential uses of affordances, they need to be actualized. Affordance actualization can be defined as a series of actions taken by organizational actors in order to take advantage of perceived affordances to support organizational goals by achieving certain outcomes (Strong, Johnson, Tulu, Trudel, Volkoff, Pelletier, Bar-On & Garber, 2014).
4 itself. In addition, the review concludes by advocating the need to focus research attention on what outcomes emerge and what goals are achieved by means of the actualization process.
Strong and colleagues (2014) analyzed this actualization process for both individualized- and collective level affordances. Their study also concludes that future studies should look into the difficulties that individuals experience while actualizing affordances. Another level of affordances - shared affordances - exists in situations in which the work of specific organizational groups is characterized by a high level of reciprocal interdependence (Leonardi, 2013). This means that in order to achieve coordination of work activities, members of these groups have to use technology features in a similar way. Although Strong and colleagues (2014) theorized the actualization process for individualized and collective-level affordances, the process of affordance actualization for shared affordances remains, to the best of my knowledge, unexplored.
It is important to fill this gap because affordance actualization substantiates the behavior within organizations that is needed to achieve the desired organizational goal (Pozzi et al., 2014). Thebenefit of opening the ´black box´ of shared affordance actualization is that it will give organizations a better view on how they can stimulate this process when organizational groups depend on each other for the coordination of work. Insight in the actualization process of shared affordances will also add to the literature stream of affordance theory in the IS discipline in general and, more specific, to affordance actualization. If the process of shared affordance actualization becomes clear, future studies can focus their efforts on theorizing about how this process can be stimulated. Therefore this study will try to fill this gap by answering the following research question: how does the actualization of shared affordances
evolve? The research question was studied by means of a qualitative, exploratory study based on
5
LITERATURE REVIEW
This section starts with a review of the literature concerning affordances in general and theory related to the actualization of affordances. Subsequently, the different levels of affordances as specified by Leonardi (2013) will be explained, as this article focusses on shared affordances. Since no theory about the actualization of shared affordances is available yet, this section ends with a review of the literature surrounding the actualization of the other two levels of affordances.
Affordances
The actual concept of affordances originates in ecological psychology. James Gibson (1986) explained that individuals only interact with an object when they perceive the functionality of the object. He defined an affordance as something that is offered or provided to someone or something by a certain object. Affordances are relational; the capabilities of the technology combined with the choices that people make about how they use those capabilities explain what effect a technology will have (Leonardi, 2011). A relational view on affordances in the context of information systems emphasizes that individuals’ goals shape their perspective on the features of a technology (Markus & Silver, 2008). People have varying goals, which is why they perceive different action possibilities in the material properties of technologies (Hutchby, 2001). Leonardi (2011) explains that the material properties of technologies afford different action possibilities, depending on the context in which the technology is used. The material properties of an information system are the same for every person, but the affordances of the system depend on how an individual perceives these material properties.
6 the possibility of performing output management with the new system existed, whether employees perceived and acted on it or not.
The second step is the perception of affordances, which is a process of recognition (Greeno, 1994). Organizational actors first need to perceive an affordance, before they can exploit its potential (Pozzi et al., 2014). In order to perform output management when utilizing a new system, organizational actors have to perceive that possibility, before they can decide whether or not to act on it. The framework continues with a third step, affordance actualization. When organizational actors actualize affordances they take steps that support the realization of advantages that affordances have to offer, which in turn should help reach organizational goals (Strong et al., 2014). Lastly, the completion of all the steps leads to an affordance effect. In the example of Seidel, Recker & Vom Brocke (2013) the actualization of the possibility to perform output management created a more sustainable organization. When an organizational actor actualizes an affordance this leads to results, which can be either organizational change, the development of new IT features or appropriate conditions for additional affordances (Pozzi et al., 2014). As the actualization process is the main focus of this study, this will be discussed in detail in the next section.
Affordance Actualization
Many studies assume that affordances exist and can be simply utilized. However, these studies do not explain how and why affordances emerge and which factors influence the actualization of affordances by users of an object (Bernhard, Recker & Burton-Jones, 2013). Therefore, Bernhard and colleagues (2013) argue that the emergence, perception and actualization of affordances should be acknowledged as different concepts. This opinion is shared by Strong and colleagues (2014), who argue for a clear distinction between affordances as potentials for action and actualization as a process in which individuals realize the potentials that affordances offer. Actualization of an affordance leads to certain consequences which can be both intended effects by the user and/or initial developer of the artifact, as well as unintended effects (Markus & Silver, 2008). The decision to actualize an affordance by a user does not necessarily align with the intended use of an object (Orlikowsi, 1992).
7 important. Lastly, the information about an affordance that is available to an individual user affects the extent to which the individual perceives the affordance. Strong and colleagues (2014) add three other factors that influence the actualization of affordances by individuals: the abilities and preferences of the individual, features of the system itself and characteristics of the work environment.
Anderson & Robey (2017) studied the difficulties that individuals experience while actualizing affordances and introduce the concept of ‘affordance potency’. They define the concept as the strength of the relationship between the system features and individual abilities during actualization, conditioned by work-environment characteristics. The strength of the relationship points to the ease of actualizing the affordance for a specific individual. An affordance with a weak potency requires more energy to actualize than a similar affordance that holds a strong potency. The article argues that a higher energy requirement will lead to more difficulty and a situation in which it is less likely that the individual will actualize an affordance, compared to an affordance with strong potency. The affordance potency is unique for individuals, as it is partly determined by individual abilities and the fact that it can change when the environment, system features or individual abilities change. This is why affordance potency, together with the perception of the performance and individual goals, will be a necessary condition for affordance actualization (Anderson & Robey, 2017). This relationship between system features and individual characteristics that together determine affordance potency reflects the relational perspective on affordances as explained by Markus & Silver (2008).
Affordance Levels
Affordance actualization is often perceived as an individual-level process in which each organizational actor takes goal-oriented actions and experiences the actualization process alone (Pozzi et al., 2014). The work of Strong and colleagues (2014) discussed earlier is an exception, as it looks at the actualization of affordances at an organizational level. They conclude that the aggregation of individual-level actualization processes forms organizational-level actualization. Leonardi (2013) however explained that members of a group do not always enact affordances in the same way: organizational actors can use technology features differently within and between groups. Therefore, he suggests that affordances should be perceived in three levels: individualized affordance, collective affordance and shared affordance.
8 available to every group member; the group member that enacts the affordance might be capable of doing something that other group members cannot. A collective affordance, on the other hand, is created by a collective effort of group members; it allows the group to do something that it could not have done without enacting the collective affordance. This type of affordance arises when individuals carry out different tasks and when there is pooled interdependence - group members have their own tasks that together combine into a final output (Thompson, 1967). It is not necessary for members of a group to use the technology in the same way, because different patterns of use afford the different capabilities that suit the specific kind of work that group members are responsible for. These distinct capabilities are essential ingredients in order for the group to be able to complete its work (Oborn, Barrett & Davidson, 2011). Next to the individual-level and group-level, affordance encompasses a third level to which this study is focused.
This third level is an affordance that members of a work group share. It differs from collective affordance, because shared affordances are common when work groups experience a high level of reciprocal interdependence. All members of the group use technology features in a similar manner, whereas in collective affordance, members show differential feature use that they need for finishing their non-interdependent individual tasks. Work group members who experience reciprocal interdependence complete similar tasks in a project and work together closely in order to achieve coordinated output. The technology use affords the same capabilities to individual group members. Members of such groups have to use features similarly in order to achieve a successful work outcome and performance at both individual and group level. Enacting a shared affordance creates a new resource for group members that allows for coordination of work and the achievement of both individual and group goals (Leonardi, 2013).
Actualization of individual and collective affordances
9 concrete outcomes’ in turn form a source of feedback that help adjust individual actions to support affordance actualization.
Strong and colleagues (2014) introduce and define the concept of immediate concrete outcomes as a specific expected outcome for the actualization process. An outcome can be for example coordination or standardization; the outcomes are seen as supportive of the realization of organizational goals (in the case of this EHR, higher quality of care and/or more efficient use of available resources). An affordance determines the potential function of immediate concrete outcomes. However, how organizational actors perceive these outcomes varies across actors. The structure of an outcome emerges during the actualization process and differs across individuals or groups. In the model developed by Strong and colleagues (2014) an immediate concrete outcome serves as an intermediate step between actualization actions and the ultimate goals from the overall organization. The actualization process consists of a ‘’set of individual journeys occurring in a dynamic organizational context as users learn, the electronic health record (EHR) system evolves and managers intervene and a multi-level process in which the journeys of many individuals interact and aggregate to form an organizational journey (the organizational change process)’’ (Strong et al., 2014, p.72).
The study from Strong and colleagues (2014) explains the individual actualization process as follows. Individuals are perceived to be goal-directed actors who start thinking about the outcomes that they should be able to achieve by making use of the new system. They start this process of thinking in terms of goals and affordances even before they start using the system itself. Next, these actors think about what actions they should take and if those actions will produce outcomes that contribute to goal achievement: they think about the actualization process. In Strong’s study, expectations of these goal-oriented actors were future-goal-oriented and driven by their goal of delivering high-quality care. In the process of using the EHR system, the individual users experienced both supporting and restricting factors that did not influence all actors similarly. How actors choose to handle these supporting and restricting factors depends on their individual abilities and preferences, the features of the EHR system and the properties of their specific work environment.
10 extent and alignment of actualization. The consistency of actualization points to how well individual actions work together to actualize organizational affordances. If, for example, some parts of the organization in the case would strive for standardization, whilst another part would strive for personalization of work styles, the organizational-level immediate concrete outcome with regard to standardization would be lower. If individual actualization outcomes reinforce each other and are compatible, they are consistent. This does not require for individual actions to be the exact same. The extent of actualization assesses how well the actualization process achieves the intended organizational-level outcomes. Even when consistency in individual actions is high, the extent to which they actualize organizational immediate concrete outcomes can be low. Lastly, the alignment of actualization comprises how well individual actions support the achievement of organizational goals. For example, some organizational actions in the case took actions to enhance the quality of care they provided. However, those quality oriented actions did not enhance the organizational substitutability affordance that should help the organization in using resources more efficiently. Together, this set of concepts might provide an instrument adequate for evaluating the consistency, extent and alignment of shared affordance actualization in this study.
11
METHODS
Background Information on Organization
The research site for this study is a large social services organization in the healthcare sector. The healthcare sector is appropriate to study the process of shared affordance actualization because new information technology (IT) implementations are common (Berwick, 2002). Next to that, the research site for this study shows the implementation of a new information system in stages. The various locations of the organization are in different stages of the implementation process, which makes it an ideal site to study the evolvement of the actualization process. It allows for a comparison of early and late implementation stages. Studying the actualization process in different stages of the implementation process is appropriate, since the research question aims to explore the evolvement of a process. In order to preserve anonymity, the organization will be referred to as ‘Alpha’. Alpha is a Dutch organization, an association in which multiple healthcare providers join forces. These providers employ about 30,000 people at different locations in the Netherlands.
Around 2014, Alpha concluded that their current information system was inadequate for the implementation of a new work process that was developed at the time. Next to that, the old system found itself at the end of its life and Alpha needed a system that would integrate all of its locations, as the different locations used a variation of systems before. The new system was also intended to meet increasingly strict privacy requirements with regard to storage of client data. I will refer to this information system as ‘Smart’. After a pilot introduction in late 2017 Alpha has started implementing Smart in its various locations in the course of 2018. This study will be conducted at multiple locations from Alpha that are in different stages of the implementation process.
Data Collection
12 and further explanation. This semi-structured approach improves both reliability and objectivity in data collection (Yin, 2009).
Data collection took place between April and June 2019. The interviewees are Alpha employees, mainly social workers and administrative employees, as these groups form the main users of Smart. Also one of the main new functionalities of Smart, the possibility for people to assign send to each other within the system instead of sending emails, depends on agreement on system usage between these two groups. This dependency is the reason why it is important to interview both groups. I also interviewed one implementation-coordinator and one application manager that were both involved in the implementation of Smart. The interviews were conducted at five different locations, of which two found themselves early in the implementation process of Smart (six weeks after implementation). The other three locations worked with Smart for one till one-and-a-half year and found themselves in a late stage of the implementation process. The respondents were collected by means of an email request by location managers, which means that all respondents signed up for participation in the study voluntarily. An important note here is that many interviewees were ‘key users’, which means they were involved in the arrangement of Smart prior to implementation and that they play an advising role after the implementation. Key users also participate in Smart-meetings in which they discuss issues regarding the use of the system that appear during their daily work activities.
13 on using new functionalities in a similar way and what led to this decision, if the respondent felt like co-workers used specific affordances in different ways than themselves and if dependency-relationships between employees changed due to the actualization process. Appendix 2 contains the interview protocol that was used in the interview with the implementation coordinator. In this interview I focused on how the implementation coordinator experienced the implementation of Smart, what difficulties arose and how it could be that the use of the system varied widely per location.
The first of the two following tables contains a description of the position of interviewees, the duration of the interviews and where the interviewee works. The table follows the order in which I conducted the interviews. The names of the interviewees are fake in order to preserve their anonymity. The second table shows for how long the employees at various locations have worked with Smart at the time the interviews were conducted. Their fake name also indicates whether they find themselves in an either early or late stage of the implementation process. Only the last three locations provided enough information useful for explaining the actualization process in the findings section. Of these three locations, one found itself in a late stage and two in an early stage. This is why the table makes no difference between the locations that are in a late implementation stage, but the two locations in an early stage are called EarlyOne and EarlyTwo. The two early locations differ in their actualization of the identified shared affordance, which is why these locations cannot be analyzed as one identical group.
TABLE 1 Descriptive Information about Interviewees
Name of interviewee Position of interviewee Location Interview duration in
minutes
Henk Social worker Late 32.42
Evelien Administrative employee Late 27.57
Kevin Application manager Late 15.55
Teddy Administrative employee Late 26.14
Susan Administrative employee Late 45.57
Mary Administrative employee EarlyOne 46.39
Janet Social worker EarlyOne 38.03
Lelia Social worker EarlyOne 26.18
Bob Social worker EarlyTwo 46.35
Megan Administrative employee EarlyTwo 37.37
Barbara Social worker EarlyTwo 37.35
Harry Social worker EarlyTwo 23.53
Wendy Social worker Late 25.32
Debby Implementation coordinator Other 45.53
14
TABLE 2 Descriptive Information about Locations
Data Analysis
I recorded and transcribed the interviews to improve data quality and reliability (Yin, 2009). All interviews were processed anonymously; I changed the names of the interviewees and locations in the findings section and interview transcripts to fake names. I uploaded the interview-transcripts in ATLAS.ti, a qualitative data analysis program. Then I analyzed the data using a coding procedure, in order to discover important patterns by constantly comparing and theorizing the data. The coding procedure consisted of three steps (Saldaña, 2015). First, I applied open coding, a process during which codes were attached to the interview transcripts. I did not interview the respondents with much existing theoretical concepts but with a general interest in understanding the actualization process in the organization. As little theory about affordance actualization exists and this research is the first to study shared affordance actualization, the only deductive open codes I used were ‘shared affordance’, ‘agreement on work practices’ and ‘similar system usage’ because the study of Leonardi (2013) explained that work groups have to agree on similar system use in the case of shared affordances.
In the open coding process I learned that the actualization process of the identified shared affordance could be understood from how employees from Alpha handled three issues related to the use of a new task system within Smart: 1) how social workers used the system for sending tasks to administrative employees; 2) how administrative employees used it for sending tasks to social workers and 3) how both groups handled rush jobs. I compared the interviews by comparing the data from different locations with regard to these three issues to understand the actualization process. A few examples of inductive open codes related to the issue that administrative employees could not distinguish between tasks that were rush jobs and tasks that were no rush jobs are ‘not clear if task is rush job’, ‘social workers discuss rush job in key user meeting’ and ‘administrative employees tell social workers that they want to see whether a task is a rush job’.
Location Amount of time
working with Smart (months)
Late 12-18
EarlyOne 1,5
EarlyTwo 1,5
15 In the second step of the coding process, axial coding, I grouped similar open codes into categories if they shared characteristics. For example, I grouped open codes ‘not clear if task is rush job’ and ‘administrative employees tell social workers that they want to see whether a task is a rush job’ into the first order category ‘administrative employees cannot distinguish between rush and no rush’. Hence, I also grouped first order categories into second order themes. For example, I grouped ‘administrative employees cannot distinguish between rush and no rush’, ‘administrative employees get both emails and tasks’ into the second order theme ‘administrative employees recognize opportunity or problem’, based on the fact that in both cases the administrative employees recognize a problem. Lastly I applied selective coding to identify core categories and to establish relationships between them, which led to three main theoretical dimensions: intra-group problem or opportunity recognition, intra-group agreement and inter-group agreement.
17
FINDINGS
This section will first explain work practices for the two main employee-groups in Alpha: social workers and administrative employees. It is important to explain the work activities of these two work groups because the shared affordance that forms the focus of this study depends on the agreement on system use between these two groups. An understanding of their reciprocal interdependence supports the understanding of this findings section. Subsequently, an explanation of the identified shared affordance will be given as well as a description of the actualization process as identified in the data. The actualization process is described by looking at three issues with regard to actualizing the identified affordance that came up in the data analysis and comparing how different locations handled these issues.
Job Description of Work Groups
Within Alpha two main employee groups can be distinguished: social workers and administrative employees.
Social workers. Social workers are the main employees of the healthcare providers from Alpha, they provide care to Dutch people. They develop reports and action plans in Smart.
Administrative employees. Administrative employees act as support staff for the social workers. They receive reports and letters from social workers and process them by means of an editing process, for example by checking for spelling errors and missing information. Sometimes reports also contain unnecessary or double information that needs to be shortened or summarized. Administrative employees also make sure that the documents that leave Alpha are readable and understandable for the recipient. After the editing process, they send the finished reports and letters to the appropriate recipients, such as related institutions, children or families. This work group also processes incoming letters and other correspondence by uploading the documents in Smart and/or archiving them in a paper file. In the case of address changes or other personal information changes, they change this in Smart to ensure correctness of personal information of clients. Administrative employees spend most of their day working with the Smart system.
Shared Affordance: the Integration of Information and Communication
18 affordance can be categorized as a shared affordance because there is reciprocal interdependence between these work groups and agreement on their work activities is needed in order to achieve successful output (Leonardi, 2013). This shared affordance can be identified as the ‘integration of information and communication’.
Within the old system, social workers and administrative employees communicated either by e-mail, phone or face-to-face. Social workers would send documents and requests for editing to the administrative group and ask them to send or edit documents by one of those means. Reversely, administrative employees would email, call or talk to social workers about adjustments that they had to make. However, this way of communicating leads to a situation in which information and documents are scattered over different locations: it can either be located in the mailbox of employees, in their personal to-do lists (paper or digital) or simply in their head. This situation is undesirable, as many requests that social workers have for administrative employees and the other way around have deadlines. For example, some documents have to be delivered at court before a specific date. As information within this social services organization is often very personal and important for clients, there was a need to integrate this information and communication in one location. This would at the same time provide a way to support automatic file creation within the new system.
To this end, Smart contains a new functionality for social workers and administrative employees to send each other tasks within the system, in order to keep all communication and information (mostly in documents) within one location. The task system was used primarily for communication from social workers to administrative employees at the time of the interviews. To send out a task, social workers create a task for either administrative employees or colleagues and attach documents and/or a mailing list - a list of people to whom the recipient needs to send the attached documents. When the receiver opens the task he or she immediately opens the right document in Smart that is in need of adjustment or reading.
19 system, information is still not integrated in one location and the possibility of missing requests for editing by administrative employees or social workers remains present. Social workers and administrative employees need to agree on their use of the task system in order to produce desirable work outcomes. This reciprocal interdependence between administrative employees and social workers indicates that this affordance is a shared affordance.
Actualization Process
This section will analyze the data with regard to the actualization processes for the identified shared affordance within Alpha. It is structured based on three main issues related to the task system actualization that differs between locations. These issues are how social workers use the system for to send tasks to administrative employees, how administrative employees use it for sending tasks to social workers and how both groups handle rush jobs. Three main themes emerged from the data that together form the steps of a shared affordance actualization process.
Double tasks: sending both emails and tasks. Social workers can send tasks to administrative employees instead of emails with requests for editing or sending documents. When analyzing the data concerning the actualization of the integration of information and communication, I found that social workers from the different locations used this possibility to replace emails with tasks to various extents. Table 3 provides quotes that show how different locations deal with this issue of double tasks.
TABLE 3 Excerpts for issue: social workers sending tasks to administrative employees
Late EarlyOne EarlyTwo Intra-group
problem/ opportunity recognition
A: ‘’So, no we have both and a lot of task in the regular mail and a lot of tasks in Smart, sometimes just to be sure they both send an email and a task in Smart’’. – 1
A: ‘’It is just that both the secretariat as the social workers both are like: ‘Oh, but do I have to use the task system within Smart or…’ ‘’. –
A: ‘’[…]. And when I started doing it, I saw that the document was already in Smart. So she could have just sent me a task with the note: ‘Please add document from […]’. Because I could have just retrieved it from the documents. She did not have to email that, so in that case the email was redundant’’. – 1 S: ‘’Yes, I think it works really nicely. It takes a bit of searching for
20 A: Administrative employee, S: Social worker. The number behind a quote indicates the number of the quote and corresponds with referrals to quotes in the text.
At first, either administrative employees or social workers had to recognize a problem or opportunity with the current way of using the task system this way. For example at location Late administrative
2
A: ‘’Just after the
implementation it really was just constructing tasks. But the social workers did not have it clear that you could make the task from a document already’’. – 3
the colleagues that we can do it like this. It is also not entirely a habit to do it like this’’. – 2 S: ‘’Well, the old way often was sending an email. But then I found out: ‘Yes, the possibility in Smart is there, so we should use it’ ´´. – 3
A: ‘’Because, if they would not do that, then you would have some in the mailbox and others in Smart and then it would become unclear of course. And we do not like unclarity’’. – 2
Intra-group agreement
A: ‘’So now we are really punishing the social workers like: ‘Everything that can be done within Smart has to be done in Smart’. [...]. And if they still send an email, we more and more send it back’’. – 4
A: ‘’Now you see more and more that social workers really make tasks from the document, which is more efficient’’. – 5
A: ‘’Well, that is a matter of teaching that to everyone very well. Then you should, what is needed then is that in such a situation I go to her or send an email like: ’Well, next time…’. And often we do that’’. – 4
S: ‘’I think partly it is also a matter of recognition. And maybe they have to… it has been quite a change. So maybe we should address each other more about that. What other possibilities there are’’. – 5
S: ‘’It was briefly discussed in the training, but eventually a colleague of mine pointed it out to me’’. – 6
S: ‘’Well, they have a fixed structure that they have to adhere to. So the moment you send an email, you immediately get it back like: ‘No, you have to make a task’. Always actually’’. – 3
Inter-group agreement
A: ‘’[...]. So I look at the tasks and I think: ‘Well, I don’t see a task’. So I went to him and he forgot to make a task. So actually they have to do that, but in principle everyone does. And if they don’t, I tell them how they should do it. Because that is the goal in the end’’. – 7
21 employees Susan explains (Quote 1 and 2) that the administrative employees receive tasks both in their task tray and in their mailbox and that both work groups are not sure about when to use the task system or when to send an email. She explains why this is a problem:
So we both have a lot of tasks in the regular mail and a lot of tasks in Smart. Sometimes workers send us an email and a task in Smart, just to be sure. Which leads to a situation in which one administrative employee is sometimes working on a task received via e-mail, and the other one saying: ‘Stop, this is not working, are you also working on that? I am as well’.
Susan explained that although the task system provides them with multiple advantages, like less search-work and higher efficiency, the task system was not used in an optimal way due to social search-workers still sending emails to them, instead of or in combination with tasks. After recognizing this problem with the use of the task system, the administrative employees at location Late agreed that they would more and more send emails back to social workers to stimulate them to send tasks (Quote 4). Although the administrative employees at this location agreed on how to handle the issue, there was no agreement between them and social workers yet. This indicates that there was no actualization of the integration of information of communication with regard to this issue at location Late, since work groups did not agree on similar system use yet and social workers would still send both tasks and emails to administrative employees instead of integrating information by using the task system in all cases.
22 Mary notes how she and the other administrative employees walk to social workers if they do not know how to send a task and explain to them how they should do it. In this short, personal contact the social worker and administrative employee will then agree on how to send the task. By means of this agreement on system use between the two work groups all information and communication is integrated in Smart and the integration affordance is actualized.
At EarlyTwo, administrative employee Megan explained how she and the other administrative employees recognized a problem with social workers sending both tasks and emails to them (Quote 2). She explains how this leads to unclarity for the secretariat, because administrative employees have to check both their mailbox and task tray. Sometimes, requests for editing were both in the mail box and in the task tray. Social worker Bob recognizes a related opportunity in sending only tasks instead of both tasks and emails (Quote 1). He explains how he thinks this will lead to fewer mistakes. Later, he describes how administrative employees agreed on how to deal with this issue in that they are very consistent in sending emails back to social workers (Quote 3). Megan later confirms this:
Well, that is a matter of, a matter of teaching *laughs*. We are still teaching. It is actually going quite well. There are always people who try, but, we are actually really consequent in that. Because everyone would always just send emails to the secretariat, since we have a team wise mailbox. And ehm… If they send an email about a person or a case, we send it back: ‘That email, could you make a task or construct a mailing list?’.
23 at this location would still send both tasks and emails to administrative employees, preventing the integration of information and communication in Smart.
Reverse tasks: tasks from administration to social workers. Besides social workers sending out tasks to administrative employees, this process can also be used in reverse by administrative employees sending tasks to social workers. Table 4 provides quotes that show how different locations deal with the issue of reverse tasks.
TABLE 4 Excerpts for Issue: Administrative Employees Sending Tasks to Social Workers
Late EarlyOne EarlyTwo Intra-group
problem/ opportunity recognition
A: ‘’It is useful, but now we have to get used to it like: ‘Oh do we have to do this via Smart now, does everyone check their inboxes in Smart?’. The secretariat does, but the social workers not always so well’’. – 1
A: ‘’So even after a year it asks for: ‘Hey, does everyone uses it?´ ‘’. – 2
A: ‘’And actually in the ideal world, if you know: ‘Everyone has a look and is continuously aware of what comes in in Smart’, then we don’t have to send the email’’. – 3
A: ‘’Yes, I think it is interesting. Because I have thought about that, like: ‘Should we maybe do more with these tasks? Shouldn’t we use that even more?’ ’’. – 1
A: ‘’But you know what’s important then? Then the workers also have to learn to check their task tray often. Because of course I do that all day; that is where I receive my work. So I constantly check: ’Did something come in?’. And workers don’t do that I think. And you don’t get a notification right?’’. – 2 A: ‘’But that is my ‘fear’, that the workers will not see it. Because they oversee an email quickly already and an email even comes in on their phone, you know?’’. – 3 S: ‘’I sure think that, and
somewhere I am like: ’The more we can do at one place and that it just becomes really clear to us what is expected from us, the better’. So there definitely are advantages’’. – 4
S: ‘’It would be nice if they for example, well I don’t know what they could send us. But if they see something is missing in a file, that they point that through us via a task’’. – 5
A: ‘’Oh by the way, what is new since this week, I totally forgot about that, is that now you have… [...]. Now you can, since this week, that’s true. Because I talked to a colleague about that this week, I said: ‘Well, that is useful’. Because now we can see whether the task has been handled yes or no and what is done with it’’. – 1 S: ‘’I think I had to send something again or put something in a document again. But I, I did not see it until very long. Because, there is a list with the tasks you have, but that is something which, I don’t look at it since it is not in my system’’. – 2
S: ’Well, I found out about that really late. And I thought, you know, I also have to get used to that. I am like: ‘Why don’t you just walk over to me?’ ‘’. – 3
Intra-group agreement
A: ‘’Well… they [social workers] know that if it is urgent, it’s also in the email. So when the secretariat
A: ‘’Oh yes, we have spoken about that already. But, it is all a lot of course right now. In the beginning everything is new. So of course it is
24 A = administrative employee, S = social worker. The number behind a quote indicates the number of the quote and corresponds with referrals to quotes in the text.
At location Late, Susan explained that administrative employees were aware of the fact that they could send tasks to social workers, however, that both social workers and administrative employees sometimes doubted between using the task system and sending an email (Quote 1, 2 and 3). She perceived this situation, in which social workers are not aware of the fact they can receive tasks from administrative employees, to be a problem. Because they are not aware, administrative employees cannot be sure that the tasks they send will be noticed by the other work groups. In Quote 4, Susan explains how she and her colleagues agreed that they would stop sending both emails and tasks when something is not urgent. They agreed on this work practice to make social workers more aware of the possibility to receive tasks. However, at this location there is no agreement between these two work groups yet on how to handle the issue. Especially in case of urgent situations, administrative employees still send both an email and a task to social workers. There is no integration of information and communication in Smart and the integration affordance is not actualized.
At EarlyOne, social worker Janet recognized an opportunity in how administrative employees could use the task system to send tasks to social workers (Quote 4 and 5). She explains how the use of this possibility by administrative employees will lead to a better integration of information and communication and more clarity with regard to what is expected from social workers. Mary, administrative employee at this location, notices that she and her colleagues agreed on how to handle the issue (Quote 6). However, they feel like they should wait with the use of this possibility until all employees at this location have a better understanding of the system. Mary felt like some more practicing time with the system is needed before new possibilities can be taken in. Similar to location Late, there is no agreement between the two work groups on how to handle this issue. This prevents the integration of all information and communication in Smart and the actualization of the shared integration affordance.
would stop sending an email at such moment, and we do that more often if it’s not urgent… [...]. But we can also see that they, that they haven’t read messages yet’’. – 4
good to first just… and I think that later you have think: ‘Okay, we have everything under control now. What is more, what else can we do? What else can we make easier?’ ’’. – 6
25 At EarlyTwo, social worker Bob recognized a problem related to this issue (Quote 2). He explained that he once received a task from the administrative employees with a request for sending something over again, but only noticed the task by accident very late. He was not aware of the possibility to receive a task from the administrative employees, which is why it took him so long to notice. Colleague Barbara reports a similar incident in Quote 3. She later explains that she now uses to walk to the administration every now and then to ask if everything is alright or whether she needs to adjust things. She also notices:
‘’I don’t know if everyone is alert to that, that tasks can be sent back. [...]. So, there is a risk that things will not get done’’. Also Barbara was not aware that she did not complete the task until she accidentally
noticed it in a very late stage. Megan, administrative employee at EarlyTwo, explained how she and her colleagues recently recognized an opportunity related to this problem (Quote 1). Administrative employees now have the possibility to see whether social workers have opened the tasks they receive, so if they are aware of them. In Quote 4 she explains how she and the other members of her work group agreed that they will start using this possibility to handle the issue, although they did not know how exactly. Also no agreement between the two work groups existed yet, which prevents actualization of the integration affordance with regard to this issue for location EarlyTwo.
Rush jobs. Another area in which it is important that social workers and administrative employees agree on similar use of the task function are rush jobs from social workers that need to be handled quickly by administrative employees. For administrative employees it was often not clear if tasks were rush jobs or if they were not. Table 5 provides quotes that show how different locations deal with this issue.
TABLE 5 Excerpts for Issue: Rush Jobs
Late EarlyOne EarlyTwo Intra-group
problem/ opportunity recognition
A: ‘’We were like: ‘Hey, that is not useful at all’. Because, we have a task list and
sometimes it contains ten, but sometimes also sixty tasks. We don’t see at a glance whether something is a rush job. We only see ‘letter’, we want to be able to see ‘rush letter’ ‘’. – 1
S: ‘’’That the secretariat indicated that they could not distinguish between tasks and see at a glance what tasks are urgent and what tasks can wait a bit’’. – 1 S: ‘’It would be nice if you would just get the option: rush. That you can just opt for: ‘yes’ or ‘rush’ ‘’. – 2
A: ‘’Every now and then we find out about something that we are like: ‘Well, if we do it like this it is more useful’. – 1
Intra-group agreement
A: ‘’And then we are like: ’Hey, that is not useful at all’ ’’. – 2
S: ‘’We did discuss that lately. Let’s say I want to send a task to the secretariat now, then it is
26 A: Administrative employee, S: Social worker. The number behind a quote indicates the number of the quote and corresponds with referrals to quotes in the text.
At Late, the administrative employees recognized a problem in the task system related to rush jobs. Susan (Quote 1) explained how it was not clear to them if something was a rush job or not, which is not useful given the large amount of tasks they receive. In Quote 2 and 3 she explains how she and her colleagues agreed on how to handle the issue. They discussed it with the location manager, after which the inconvenience was adjusted in a new release of the system by adding an option for social workers to choose ‘rush job’ when making a task. After the new option for social workers to indicate ‘rush job’ was released, social workers and administrative employees agreed that from that moment social workers
A: Regularly we say: ‘Let’s send it to [role] and see whether they can adjust it in the new release’ ’’. – 3
important that we indicate whether it is a rush job in the task description’’. – 3
S: ‘’We did agree lately that we indicate whether a task is a rush job’’. – 4
get those two weeks before they have to be in court if everything goes well. So that we still have time to make things right. So they are also in the general task list. [...]. But that is quite unclear, so now we put a last expiration date behind it, that is the only thing that we do that for, behind an extension. [...]’’. – 2 A: ‘’We just do that with each other: what makes sense and what are we going to do, what is useful and what works best’’. – 3 A: ‘’Well, they could of course give an end date by themselves. But we don’t want that because we put an end date behind extensions by ourselves’’. – 4
Inter-group agreement
A: ‘’It could be adjusted in the system, that now they can choose: do I want to define it myself or pre-defined’’. – 4
S: ‘’I think a few weeks ago… Yes I think two weeks ago we discussed it in the key user meeting’’. – 5 S: ‘’And then I also fed it back to the team’’. – 6
27 would use this to solve the ‘rush job’ issue and actualize the integration affordance (Quote 4). At EarlyOne and EarlyTwo, the issue concerning rush jobs was handled differently.
At EarlyOne, social worker Janet explained that the administrative employee recognized a problem in that they could not distinguish between tasks that are rush jobs and tasks that are not (Quote 1). In quote 3 and 4 she describes how she and her work group members agreed on how they would indicate that something is a rush job to administrative employees. Janet explained that after discussing the issue, she and her team agreed on reporting a task as a rush job by adding their own description to the task (this is optional). In quote 5 she explains that after the agreement on how to handle the issue within the work group of social workers, she discussed the issue in a key user meeting. In this meeting key users from the administrative employees were present and both work groups agreed on how to handle the issue, by indicating ‘rush job’ in the task description. Another social worker from EarlyOne, Lelia, confirms this practice: ‘’Ehm, we cannot opt for a ‘rush’ option. But we agreed a while ago that we now
indicate whether something is a rush job in the description’’. At EarlyOne both social worker interviewees
are key user within different teams and both of them reported the same practice with regard to the rush job issue. After the recognition of a problem by administrative employees and agreement on how to handle the issue within the work group of social workers, there was agreement between social workers and administrative employees: the integration affordance could be actualized.
At location EarlyTwo, administrative employee Megan reported the recognition of a problem related to the issue of rush jobs (Quote 1). Some of the tasks that social workers send to the administrative employees have to be out before a certain date, but this is not visible to them in the general task list. She explained that this was one example of things the administrative employees thought could be handled better. Within their work group, they agreed on how to handle this problem (Quote 2). They agreed that they put a last expiration date behind the concerning tasks for the following reason: ‘’So that we know
right away: ‘Oh that one has to leave the 23th and the other one the 25th and that one has to be out that specific week’. That is how we do it’’. The fact that at this location the administrative employees agreed
on how to handle this problem had consequences for how they dealt with the rush job issue.
28 6). Contrary to how Late and EarlyOne tried to deal with this rush job issue, via the task system itself, at EarlyTwo the administrative employees preferred a different approach as they had to account for another agreement on system usage they had already incorporated in their work routines. However, also at this location a similar actualization process is found with regard to the rush job issue. There is a problem recognized within a work group, in this case the administrative employees, after people within a work group agree on how to handle the issue. Next, they agree on how to handle the issue with the other work group to actualize the integration affordance.
Actualization process for the integration of information and communication. The data analysis shows the actualization process for the shared affordance ‘integration of information and communication’ by analyzing how Alpha employees handled three different issues. Three main themes with regard to the actualization processes of the three issues emerged from the data: the recognition of a problem or opportunity within one of the two work groups, the agreement on how to handle that issue within either the social workers or administrative employees work group and eventually agreement between these two work groups to allow for actualization of the integration of information and communication. The data reveals a three-step actualization process in which members of a work group first recognize a problem with the current use of the task system or an opportunity to optimize its use. In the second step the members of the work group agree on how they want to deal with the issue. As it pertains the actualization of shared affordances, in the third step agreement between work groups takes place which completes the actualization of the shared affordance. The process model in Figure 2 shows the three steps and if they occur at either intra- or inter group level.
30
DISCUSSION
Within the relatively new theoretical field of affordance actualization, this paper is the first to explore the actualization process of shared affordances. Although prior research by Strong and colleagues (2014) explored the actualization process for individualized and collective affordances, the question of how shared affordances are actualized remained unanswered. Within this study in a large social services organization two main employee groups were studied in their attempt to actualize a shared affordance that was offered by a new information system. The process of actualizing the ‘integration of information and communication’ affordance was identified to follow three steps. First, members of a work group recognize a problem or opportunity within their current way of using the system. In the second step, members within a work group together agree on how to handle the recognized problem or opportunity. Lastly, there is agreement across work groups that experience reciprocal interdependence on how to handle the identified problem or opportunity in their system usage to achieve similar system usage. This process model for shared affordance actualization is an important contribution to the literature stream of affordance actualization and affordances in general.
This study supports the relational view on affordances as proposed by Markus & Silver (2008). The perspective of work groups within the organization in this study influenced the problems and opportunities they recognized within current system usage. The goals from administrative employees for example, supporting social workers by processing their requests in time, caused them to perceive a problem with the rush job issue. Pozzi and colleagues (2014) explained that affordances first need to exist and be perceived by organizational actors before they can be actualized to lead to an affordance effect in their framework, which is supported by this study. The implementation of Smart caused new affordances to exist, for example caused by the task system within Smart that allows for integration of information and communication. Next, the social workers and administrative employees needed to perceive this affordance before they became capable of recognizing problems or opportunities with current system usage. Next, they actualized the shared affordance which led to an affordance-effect: integration of information and communication within Smart. This study confirms the framework as constructed by Pozzi and colleagues (2014) for affordances in general for a specific level of affordances: shared affordances.
31 order to achieve this coordinated output. He explains: ‘’In work of this type, individuals who share similar patterns of technology use are afforded the same capabilities and, consequently, can continue to coordinate their work’’ (Leonardi, 2013, p. 752). In this sense, it could be inferred that he perceives this state of coordinated output to be the ‘end state’, when the affordance is actualized. However, within this study, the process in which work groups agreed on similar system use can be perceived to be a process of coordination as well. An example of this process in this study is when social workers and administrative employees together agreed on how to handle the rush job issue during a key user meeting. This process in which the agreement is achieved could be looked at as a form of coordination, despite the fact that during this process no coordination of output is reached yet. Therefore, it could be that coordination plays a role in the (shared) affordance actualization process, instead of being the ‘end goal’ of affordance actualization as Leonardi (2013) points to. However, this potential role of coordination in affordance actualization remains unexplored in the affordance literature; future studies should try to explore if coordination is important during (shared) affordance actualization. If this role of coordination exists, existing theories about coordination can offer insights that are valuable for affordance actualization literature.
The study by Strong and colleagues (2014) indicated that the affordance actualization process is iterative, that system related actions by individual organizational actors generate certain outcomes that in turn provide feedback that adjusts subsequent individual actions. This study supports the argument that the affordance actualization process is iterative for shared affordances in specific. However, in the case of shared affordances, individual system related actions do not provide the feedback that adjusts future actions, but work group actions. An example that illustrates this argument can be found in how the rush job issue was handled. The fact that social workers sent tasks without indicating ‘rush job’ in the task description, led to feedback from the side of administrative employees. They indicated that this system usage was a problem to social workers, which caused adjustments in system related actions of social workers: they started indicating rush job in the task description. This illustrates that the shared affordance actualization process is iterative, which adds to this similar argument by Strong and colleagues (2014) for the case of individualized and collective affordances.
32 how for ‘consistency of actualization’, individual actions do not have to be the same in order for them to work together and actualize organizational affordances. However, in case of shared affordances, it is important that individuals within a work group agree on similar system usage. Therefore, consistency of actualization is not appropriate for evaluating shared affordance actualization, unless the consistency of work group actions instead of individual actions in actualizing organizational affordances is evaluated. Within this study, it could be evaluated how well the actions of both work groups were consistent in actualizing the integration of information and communication affordance.
The second concept, extent of actualization, evaluates how well the actualization process achieves organizational level outcomes. In this study, the organization wanted to enhance integration of information and communication. The extent to which the actualization of the shared integration affordance supports this can be evaluated. The last concept, alignment of actualization, evaluates how well individual actions support organizational goals. Because in the case of shared affordances it is important that individuals within a work group agree on similar system usage, so similar individual actions, this concept should be adjusted to become appropriate for evaluating shared affordance actualization. Instead of evaluating how well individual actions support organizational goals, there could be an evaluation of how well work group actions support organizational goals. Within this study, it could be evaluated how well work group actions supported the integration of communication and information. To conclude, the instrument of Strong and colleagues (2014) can be made appropriate for evaluating shared affordance actualization as well, if for consistency and alignment the individual actions are replaced with work group actions. Future research should study the applicability of this instrument for evaluating shared affordance actualization.
33 Lastly, the organization might want to enhance agreement making opportunities for the involved work groups, for example by rewarding efforts that support this process by work groups.
Research Limitations and Further Research
This study has certain limitations. First, the study focused on one organization in one industry and one technology, which limits the generalizability of results. Further research should look into the applicability of the identified process model within other contexts to see whether the model still holds. In this research, the use of the technology depended predominantly on similar system use by two main work groups. However, in case of a different technology, in which there is a high dependency between more than two work groups or a different type of shared affordance, future studies should look into the applicability of this process model.
A second limitation is that most of the respondents were ‘key user’, which means that they were more involved in the pre-implementation, implementation and post-implementation process and that they were often better informed about the new system before implementation. A better understanding of the system might help these respondents actualize the integration affordance more easily. However, when I asked respondents (both key user and not key users) if they felt like they had an advantage in their use of the system due to their key user membership, they explained that this would only be true for a very limited amount of time after the implementation. They indicated that after some weeks, most of the other (not key user) users were up to speed and that their ‘knowledge advantage’ had faded.
Conclusion
35
REFERENCES
Anderson, C., & Robey, D. (2017). Affordance potency: explaining the actualization of technology affordances. Information and organization, 27, 100-115.
Bernhard, E., Recker, J., & Burton-Jones, A. (2013). Understanding the actualization of affordances: a study in the process modeling context. International conference on information sytems (ICIS 2013), 15-18.
Berwick, D.M. (2002). A user’s manual for the IOM’s ‘Quality chasm’ report. Health affairs, 21(3), 80-90. Burton-Jones, A., & Gallivan, M.J. (2007). Toward a deeper understanding of system usage in organizations: a multi-level perspective. MIS Quarterly, 31(4), 657-679.
Edmondson, A.C., & McManus, S.E. (2007). Methodological fit in management field research. The
academy of management review, 32(4), 1155-1179.
Gibson, J. J. (1986). The Ecological Approach to Visual Perception. Hillsdale, NJ: Lawrence Erlbaum Associates.
Greeno, J.G. (1994). Gibson’s affordances. Psychological review, 101(2), 336-342. Hutchby, I. (2001). Technologies, texts and affordances. Sociology, 35(2), 441-45.
Leonardi, P.M. (2011). When flexible routines meet flexible technologies: affordance, constraint and the imbrication of human and material agencies. MIS Quarterly, 35(1), 147-167.
Leonardi, P.M. (2013). When does technology use enable network change in organizations? A comparative study of feature use and shared affordances. MIS Quarterly, 37(3), 749-775.
Markus, M.L., & Robey, D. (1988). Information technology and organizational change: causal structure in theory and research. Management science, 34(5), 583-598.
Markus, M. L., & Silver, M. S. (2008). A foundation for the study of IT effects: a new look at DeSanctis and Poole’s concepts of structural features and spirit. Journal of the Association for Information Systems,
9(10/11), 609-632.
Oborn, E., Barrett, M., & Davidson, E. (2011). Unity in diversity: electronic patient record use in multidisciplinary practice. Information systems research, 22(3), 547-564.
Orlikowski, W.J. (1992). The duality of technology: rethinking the concept of technology in organizations.
Organization science, 3(3), 398-427.
36 Saldaña, J. (2015). The coding manual for qualitative researchers. Thousand Oaks, California: Sage Publications Ltd.
Seidel, S., Recker, J., & Vom Brocke, J. (2013). Sensemaking and sustainable practicing: functional affordances of information systems in green transformations. MIS Quarterly, 37(4), 1275-1299.
Strong, D. M., Johnson, S. A., Tulu, B., Trudel, J., Volkoff, O., Pelletier, L. R., Bar-On, I., & Garber, L. (2014). A theory of organization-EHR sffordance actualization. Journal of the association for information systems,
15(2), 53–85.
Thompson, J.D. (1967). Organizations in action: social science bases of administrative theory. New York: McGraw-Hill.
Yin, R.K. (2009). Case study research – design and methods. Thousand Oaks, California: Sage publications Ltd.
37
APPENDIX 1 – FINAL INTERVIEW PROTOCOL
Bedankt dat u tijd wilde vrijmaken voor een interview voor dit onderzoek. Dit interview zal gaan over de verandering naar het nieuwe informatie-systeem Smart. Dit interview zal ongeveer 45 minuten in beslag nemen. De resultaten worden geanonimiseerd, wat betekent dat de door u gegeven informatie in het artikel niet naar u persoonlijk kan worden teruggeleid.
Algemeen
Wat zijn uw verantwoordelijkheden/wat is uw functie? Waar bent u de meeste tijd aan kwijt op één dag? Hoeveel tijd per dag met het systeem?
Verandering in informatiesysteem
Wat is het doel van Smart? / Waarom is Smart geïntroduceerd?
Wat vindt u persoonlijk van het nieuwe systeem, ten opzichte van het oude systeem? Invloed op werkzaamheden
Hoe heeft het nieuwe systeem uw werkzaamheden veranderd? - Kunt u voorbeelden geven?
Wat zijn belangrijke nieuwe functies of mogelijkheden van het systeem? - Kunt u voorbeelden geven?
Wat is voor u de belangrijkste nieuwe functie of mogelijkheid van het systeem? - Waarom? Kunt u voorbeelden geven?
Takensysteem/functie
Hoe maakte u vlak na de implementatie gebruik van de nieuwe takenfunctie in Smart? - Waarom? Waar werd dit door beïnvloed? Kunt u voorbeelden geven?
Hoe maakt u nu gebruik van de takenfunctie in Smart?
- Waarom? Waar werd dit door beïnvloed? Kunt u voorbeelden geven? Maken uw collega’s op een andere manier gebruik van deze functie?
