NO PAIN, NO GAIN!
On the role of both system and organization in the success of an information system
No pain, no gain!
As each organization, the municipality of Groningen needs to keep a close eye on its financial position. This is one of the main tasks of the Management Service. It acts as a hub and gathers financial information from the eight divisions on topics like policies, budgets, money spent, etc. Within IAS the responsibility to supply this information to the Management Service lies with Finance & Control (F&C). As shown in the figure below, IAS is split up in three main departments, namely Public Services, Resources and Control and Facility Services. F&C is placed under Facility Services but nonetheless has the responsibility to report to the Management Service on behalf of all the departments within IAS. General Manager Resources and
Control Facility Services Public Services
Communication
Finance and
Control Facility Affairs Human Resources Information Management Public Affairs Civil Affairs Taxes Legal Protection Commercial Affairs Central ICT Organization Financial Services
Research problem
A part of the vision of IAS is to make a profit while sustaining the quality of the service (Gemeente Groningen, 2010). This implies that each decision always has a financial component that needs to be taken into account. The head of F&C has the impression that this is insufficiently the case at the moment. He would like to see that more financial information gets used by the heads of departments in their decision process to come to financially sound decisions so that their budgets aren’t overdrawn. To accomplish this they need relevant financial information. F&C tries to deliver this information through Cognos but the IS is barely (if at all) in use. Because of the costs and time that have already gone into implementing Cognos, the head of F&C wants to try one last time to make it a success. Therefore, the research question is: How can the head of F&C introduce Cognos successfully so that it delivers relevant information to other department heads?Research method
To discover the answer to the research questions mentioned above, several sources of data were explored. The graphical representation of the research method that addressed these sources is shown below. Figure 4. Research methodInterviews Pilot interview
Research Flow of information Shannon and Weaver (1949) Technical level Semantic level Effectiveness or influence level
Mason (1978) Production Product Receipt Influence on recipient Influence on system Delone and McLean (1992) System quality Information quality Use User satisfaction Individual impact Organizational impact Table 1. D&M model in relation to Mason (1978) and Shannon and Weaver (1949) After reviewing many studies that identify success measures in each of the aspects, DeLone and McLean argue that “no single measure is intrinsically better than another; so the choice of a success variable is often a function of the objective of the study, the organizational context, the aspect of the information system which is addressed by the study, the independent variables under investigation, the research method, and the level of analysis” (p. 80). However, they think it is unlikely that only one measure explains IS success. Therefore, they integrate the identified aspects of IS success into one model, which is shown in the figure below. Information quality
System quality Use User satisfaction
Individual impact Organizational impact Figure 5. Visual representation of D&M model As can be seen in the figure, the model adopts a process approach with interdependent causal relations between the factors. DeLone and McLean believe that this model allows for “a reasonably coherent organization of at least a large sample of the previous literature, while, at the same time, providing a logic as to how these categories interact” (p. 87).They acknowledge that the model needs further development and validation; a task that has been well received by the IS research field.
Updated D&M model
diagnostic tool for managers, including IS managers [and that it] can be a useful tool in IS service evaluation systems” (p.150). The four‐dimension Servqual is therefore a reliable tool to measure the added service quality aspect of the updated D&M model. The second adjustment to the model comes from the realization that the impact of an IS might reach further than assumed in the original model. Delone and McLean (2003) conclude that “there is a continuum of ever‐increasing entities, from individuals to national economic accounts, which could be affected by IS activity” (p.19). Although the original model could be expanded by all these entities, DeLone and McLean choose to group this continuum into one measure, namely net benefits. These two modifications result in the updated D&M model that is shown in the figure below. Figure 6. Visual representation of updated D&M model As discussed in the previous section, DeLone and McLean already discovered that information seems to flow through a series of stages. The modifications to the original model can again be classified into these stages, which is shown in the table below. Research Flow of information Shannon and Weaver (1949) Technical level Semantic level Effectiveness or influence level
Table 2. Updated D&M model in relation to D&M model, Mason (1978) and Shannon and Weaver (1949)
The existing models
The models that were selected to be included in the research have one essential similarity: intention and/or usage is employed as the key dependent variable. Usage is seen as the ultimate goal for acceptance and intention to use is established as a solid predictor of usage. The selection consisted of eight models, each with between two and seven determinants of acceptance. The table below (adapted from Venkatesh, Morris, Davis, & Davis, 2003) shows these models and their constructs.the models (except for MM and SCT) and individual factors increase further. Now that the models have been compared, Venkatesh et. al. (2003) go on to formulate the Unified Theory of Acceptance and Use of Technology.
Unified theory of acceptance and use of technology
As can be seen in the figures above, seven of the constructs appear significant. Only four of them are selected by the authors as suspected to have a “significant role as direct determinants of user acceptance and usage behavior: performance expectancy, effort expectancy, social influence, and facilitating conditions” (p. 447). These are new labels to show that they are uncoupled of any former models and their meaning will be explained below. Also, the effect of the previously mentioned key moderators on the relationship between the constructs and user acceptance and user behavior is discussed. With these four constructs and the key moderators, Venkatesh et al. (2003) formed the Unified Theory of Acceptance and Use of Technology, which is presented graphically in the figure below. Figure 9. Visual representation of Unified Theory of Acceptance and Use of Technology The constructs that weren’t selected for UTAUT are attitude towards using technology, self efficacy and anxiety. Based on research, Venkatesh et. al. (2003) conclude that self‐efficacy and anxiety are only indirect determinants of intention because they are mediated by effort expectancy. They hypothesize “self‐efficacy and anxiety to behave similarly, that is, to be distinct from effort expectancy and to have no direct effect on intention above effort expectancy” (p. 455). The results fully support this hypothesis. Attitude towards using technology is used in four of the previously mentioned models to represent the individual’s pleasure associated with the use of a technology. In some of the models this construct is a significant predictor of behavioral intention and in others it is not. Venkatesh et. al. (2003) suggest that the discriminating factor between research that finds this construct significant or not, is the “omission of the other key predictors (specifically, performance and effort expectancies)” (p. 455). Because these key predictors are embedded in UTAUT they Performance expectancy Effort expectancy Social influence Facilitating conditions
Behavioral intentions Use behavior
Dependant variable Independent variable Result
Behavioral intention Performance expectancy Significant, with the effect stronger for men and younger workers
Behavioral intention Effort expectancy Significant, with the effect stronger for women, older workers, and those with limited experience
Behavioral intention Social influence Non‐significant without inclusion of moderators. Significant with the inclusion of moderators, with the effect stronger for women, older workers, under conditions of mandatory user, and limited experience
Usage Facilitating conditions Significant only in conjunction with moderators age and experience, with the effect stronger for older workers with increasing experience
Usage Behavioral intention Direct effect.
Pilot study
Function Gender Interview
IAS, staff F&C (sFEA) Male Yes
SAW, department head Small Business Owners (hDSBO) Male Yes
SAW, staff Information Management (sIM1) Male Yes
SAW, staff Information Management (sIM2) Male Yes
Total 4
Figure 11. Subjects in pilot study
Main Study
Department head of (IAS) Gender Interview
Procurement Male No
Facility Affairs (hFA) Male Yes
Communication (hCEM) Female Yes
IAS Male No
Team leader Application Management (tAM) Male Yes
Information Management (hIM) Female Yes
F&C (hF&C) Male Yes
Taxes (hT) Male Yes
Legal Affairs (hLP) Male Yes
Civil Affairs (hCA) Male Yes
Human Resources (hHR) Male Yes
Quality and Auditing (hQA) Male Yes
Commercial Affairs (hCA) Male Yes
Public Affairs (hPA) Male Yes
Facility Services (hFS) Male Yes
Central ICT Organization (hCIO) Male Yes
Total 14
improved is the ability to keep a close eye on the budgets and spending in projects. The general impression is that the current information supply doesn’t give information that can be used to manage the department. To provide overview of the issues mentioned by the two groups – department heads and users –concerning the performance expectancy of Cognos, the results are summarized in the table below.
Group Positive Negative
The results presented at the service quality section earlier in this chapter are therefore not repeated but should be considered as complimentary for this section. The degree to which an individual believes that an organizational and technical infrastructure exists to support use of the system. Cognos isn’t in use at IAS but the department head that did ask for it said: “[the head of F&C] knows that I’m a supporter of Cognos. I asked for Cognos a long time ago but it doesn’t get enough momentum. This year I suggested making Cognos available to Taxes and to let Martin [advisor from F&C] build that for me. I also think that it’s the role of F&C to help me with this. If the need is too high and nothing gets done I’ll eventually do it myself” (hT). A similar situation unfolded at Public Affairs, where F&C wanted to combine efforts. However, “it’s quiet on that front now. I also don’t know what F&C is now doing with Cognos and the developments regarding Cognos” (hPA) a department head remarked. Another department head said: “If F&C wants us to use [Cognos], I think they have to take on a more pro‐active role” (hCA). When asked what would make it easier to use Cognos, an interviewee said: “you should show me the features and qualities of the application. I also want to know if its user friendly. I have to know what kind of information it has to offer” (hCEM). The application should be ‘sold’ to the department heads, showing which information they could get which they couldn’t with MIS.
Refreezing
Once the moving stage of Schein’s (1988) Planned Change Theory has passed, the achieved changes should be anchored in the employees’ behavior; the organization needs to be refrozen. However, since the interviews showed that Cognos was barely introduced, let alone completely implemented, it is clear that the moving stage has not yet been completed. Gathering data on the refreezing stage was therefore seen as time spent inefficient.Summary
In this section the influence of the will to change in the organization on the implementation of Cognos is examined using the Planned Change Theory of Schein (1988), combined with the Unified Theory of Acceptance and Use of Technology (Venkatesh et al, 2003). The results are summarized in the table below.Phase Sub phase Findings
Unfreezing ‐ Budget is for +/‐ 80% static
Need for clear link between activities and expenses