A GENERIC METHOD FOR BEST PRACTICE REFERENCE

MODEL APPLICATION

Stefanie Looso

Frankfurt School of Finance and Management, Frankfurt, Germany

Keywords: Method Engineering, Design Science Research, Best Practice, Reference Models, IT Governance.

Abstract: The perceived importance of the topic IT governance increased in the last decade. Best practice reference

models (like ITIL, COBIT, or CMMI) promise support for diverse challenges IT departments are confronted

with. Therefore, the interest in best practice reference models grows and more and more companies apply

BPRM to support their IT governance. But there is limited knowledge about how BPRM are applied and

there is no structured method to support the application and lift the full potential of BPRM. Therefore, this

paper presents the construction and evaluation of a generic method for the application of BPRM. Following

the language-based approach of method engineering, elements of methods will be derived and formally

described. The criteria of design science research presented by Hevner et al., 2004 will be applied to the

evaluation of the constructed method. Intention of this research is to reduce the inefficiencies caused by the

inconsistent use of best practice reference models.

1 INTRODUCTION

As a central instrument for the design of corporate

information systems within the field of information

systems research, information models have

traditionally been used for decades. Literature on

this subject suggests the concept of reference

modeling for an improvement in the development of

enterprise-specific models (see Hars, 1994; Becker,

1995; Frank; Scheer, Seel, & Georg, 2002; Goeken,

2002; Becker & Knackstedt, 2002; Loos & Fettke,

2005 among others).

A reference model is defined as a generic

conceptual model which is useful when developing

an individual model of a specific class. It formally

presents state-of-the-art knowledge and best practice

knowledge and is considered as an example for a

corporate model (Fettke & Loos, 2003, Rosemann &

van der Aalst, 2007) . Precisely the mentioned best

practice knowledge is contained in the models of IT

governance focused on herein. (Co-) produced by

practitioners these models contain profound and

consolidated knowledge based on experience in the

field of IT governance and tend to become quasi-

standards (PWC, 2006).

Thus, the part of the definition concerning best

practice knowledge is fulfilled by the models of IT

governance. However these models are conceived as

structured compilations of best practices rather than

semiformal conceptual models. Therefore, the part

of the definition which states reference models are

conceptual models is not fulfilled by all models of

IT governance.

By metamodeling these models they could be

described more formally (Goeken & Alter, 2009). A

meta model includes the inner structure of the best

practice knowledge and it is a first step to model

best practices as conceptual models. This

conceptualization makes some research findings of

reference model application utilizable. However, in

order to avoid misleading terms and misconceptions,

reference models of IT governance will be referred

to as best practice reference model (BPRM) in this

paper.

Those BPRM have reached a certain degree of

commonness in practice. Their application is still

growing, but seems to be inconsistent. The study "IT

Governance in Practice - Insight from leading

CIO’s" quotes one participant on the application: „I

use frameworks and standards for inspiration, and

we use what we think is useful and relevant for our

organization“(PWC, 2006, p.18). Other companies

use BPRM even more holistic and with a higher

degree of obligation. The missing methodical

support for their application results in several forms

of inconsistent application of one BPRM.

149

Looso S. (2010).

A GENERIC METHOD FOR BEST PRACTICE REFERENCE MODEL APPLICATION.

In Proceedings of the 12th International Conference on Enterprise Information Systems - Information Systems Analysis and Speciﬁcation, pages

149-158

DOI: 10.5220/0002870701490158

 SciTePress

Additionally to that one-model case the

simultaneous use of several BPRM increases in

enterprises (PWC, 2006). Simultaneous means that

an IT department for instance uses CMMI for the

development of new systems and COBIT to provide

IT governance. In this multi-model case the

inconsistent and simultaneous application can lead

to problems (Alter & Goeken, 2009, Siviy, Kirwan,

Marino, & Morley, 2008a and b). For example,

various different, sometimes contradicting,

languages are encountered which complicate

cooperative work spanning several divisions.

Furthermore, multiple BPRM produce overlaps and

contradictions among each other. This leads to

redundancies and further inefficiencies. Those

inefficiencies are clearly opposed to the nature of

reference models since the construction of

corporation-specific models based on customized

models or model components promises positive

effects on effectiveness and efficiency see Fettke &

Loos, 2002 (p.9), Goeken, 2002 (p.1) or Becker,

Delfmann, & Knackstedt, 2004 (p.1). Therefore this

paper presents a method for the methodical and

structured application of IT governance BPRM.

2 RESEARCH DESIGN

This paper presents a part of broader a research

project. This research project is designed as follows.

To broaden the understanding of BPRM application

the research design includes explorative expert

interviews in addition to an extend literature review

as a first step. Based on that knowledge, a method

has been constructed for the one-model case. The

generic method has been evaluated and its usability

has been tested by applying it to a specific BPRM

(COBIT). After that the methods is extend to the

multi-model case. Goal of the research project is to

support effective and efficient use of several BPRM.

This paper presents the generic method for the one-

model case. The illustrated method is a result of a

research process using the knowledge base of IS

research and the business needs concerning the topic

IT governance and BPRM application. Figure 1

shows the position of the research project in the well

known conceptual framework of Hevner, March,

Park, & Sudha, 2004.

According their framework the knowledge base

“provides the raw materials from and through which

IS research is accomplished. The knowledge base is

composed of foundations and methodologies. Prior

IS research and results from reference disciplines

provide foundational theories, frameworks,

instruments, constructs, models, methods, and

instantiations used in the develop/build phase of a

research study“ (p.80). Following this definition the

knowledge base for this research includes research

on reference models and modeling, method

engineering and on IT governance but also on

research methods like interview techniques.

The environment defines the problem space

which includes the phenomena of interest. „In it are

the goals, tasks, problems, and opportunities that

define business needs as they are perceived by

people within the organization“(p. 79).

Figure 1: Research project (according to Hevner et al.,

2004).

For research on the BPRM of IT governance this

environment is composed of IT employees, IT

organization, IT goals and processes and the existing

best practices.

The research findings could be distinguished in a

generic method and several specific methods,

derived from the generic method. The constructions

process of the generic method follows a rigor

research design by using the existing and proofed

knowledge of IS research. The various specific

methods represent relevant IS artifacts which

provide support for practical problems. Following

Hevner et al., 2004 these practical applications of

the presented generic method follow the design

science paradigm. The latter is “fundamentally a

problem solving paradigm. It seeks to create

innovations that define the ideas, practices, technical

capabilities, and products through which the

analysis, design, implementation, management, and

use of information systems can be effectively and

efficiently accomplished“(p.78). That means the

practical application of the specific method is an

application for the IT governance environment

whereas the generic method is a contribution to the

knowledge base. The generic method for the one-

model case is presented in this paper. Following the

argumentation of Hevner et al., 2004 using the

knowledge base to support the IT governance

environment leads to a rigor research design.

ICEIS 2010 - 12th International Conference on Enterprise Information Systems

150

3 METHOD CONSTRUCTION

3.1 Preliminaries: Method Engineering

Methods describe a systematic approach to the

solving of problems. A problem is defined as a

discrepancy between actual and desired state

(Becker, Knackstedt, Pfeiffer, & Janiesch, 2007).

Focused on the creation of methods, the research

area of method engineering is a commonly accepted

and frequently debated concept in construction-

oriented IS research. Brinkkemper, 1996 defines:

“method engineering is the engineering discipline to

design, construct and adapt methods, techniques and

tools for the development of information

systems“(p.276).

Since it is commonly accepted that no universal

method exists (see Brooks, 1987, Fitzgerald, Russo,

& O'Kane, 2003 among others), tendencies such as

"domain-specific method engineering“ (Kelly,

Rossi, & Tolvanen, 2005) and "situational method

engineering” (Brinkkemper, 1996, Brinkkemper,

Saeki, & Harmsen, 1999); (Harmsen, 1997) have

developed. Basically, there are two tendencies

within this area. Some approaches of method

construction emphasize aspects of the construction

process and project management (Kaschek, 1999).

In contrast, the approach of a language-based

construction of method elements focuses on the

artifacts created. In recent years the latter approach

has been focused on in the field of methods

engineering (Brinkkemper, 1996, Ralyté & Rolland,

2001, Karlsson & Wistrand, 2006).

A method is primarily taken as a tupel of a type

of exercise and a number of rules (Becker,

Knackstedt, Holten, Hansmann, & Neumann, 2001),

p.5. According to Zelewski, 1999 however, not all

potential tupels are a method per se. In order to

qualify as a method, an observation of certain

qualitative requirements has to be assured. For

instance, all elements and their relations to one

another need to have an unambiguous interpretation

for the support of which a fairly unified

documentation of methods should be used. A

language-based reconstruction of method elements

and formalized documentation thereof is thus

required (Becker et al., 2001, p.6).

The St. Gallen description model of method

engineering includes a schematic composition of the

elements: meta model, result, activity, technique,

tool, and role. According to its language-based

interpretation, the description of these elements

offers a systematic development of a method (Heym,

1993, Gutzwiller, 1994, Becker, 1998).

Unlike the St. Gallen model, the method

presented herein presumes a relation between

activity and technique for the application of BPRM.

Two of the required method elements, namely role

and tool, will not be addressed in this paper. The

paper does not primarily deal with personnel related

and psychological aspects of the element role. It

neither deals with the specific technical aspects of

tools. Thus, the generic elements result, activity, and

technique will be included in the construction of the

method in accordance to the language-based

approach. Following the description of the generic

method element types, the relations between these

elements will be exemplified on a detailed level by

presenting instantiations of the generic types.

3.2 Method Element: Result Type

Results of the suggested method are several different

models. These belong to certain result types which

can be divided by two dimensions. The first division

is between two abstract levels, the meta level and the

model level. The second dimension distinguishes

between reference level and corporate level. Figure

2 depicts a metamodel of the method element result

type. The relationships shown in Figure 2 represent

possible transformations between several result

types. Their dynamic aspect will be described in

detail in section 3.4.

Meta Level

Model Level

Reference Layer

Specific Company

Layer

Best Practice

Reference Model

Best Practice

RM Subset

Explicit

Model

Best Practice

Reference Meta

Model

Company Specific Best

Practice Reference Meta

Model

Company Model

is applied to

has Subset is adjusted to

is adjusted to

has high level

abstraction

is adjusted to

Figure 2: Meta classification of result types of the generic

method.

Hence, result type best practice reference model is

defined as a model on model and reference level. A

possible instantiation of this type would be the

BPRM COBIT 4.1. Result type best practice

reference model subset has partly been adjusted to

corporate-specific conditions, and is thus a result

type of the company layer. An exemplifying

instantiation is a COBIT 4.1 subset which contains

PO processes exclusively. The explicit model shows

the externalized consensus of individual subjects in

A GENERIC METHOD FOR BEST PRACTICE REFERENCE MODEL APPLICATION

151

relation to corporate reality (Becker, Niehaves, &

Knackstedt, 2004). The explicit model is more

closely adjusted to corporate conditions than is the

BPRM subset. Both are instantiations of the (meta)

type company model. Additionally, the method is

familiar with various forms of company models

apart from the aforementioned result types. All of

them are associated with the lower right section of

the matrix. Building a company model it could be

necessary to change the meta model the model is

based on. These changes of the best practice

reference meta model result in a company-specific

best practice reference meta model. These two

abstract result types complete the result types used

for the presented method.

3.3 Method Element: Technique Type

Technique is defined as “a procedure, possibly with

a prescribed notation, to perform a development

activity” (Brinkkemper, 1996), p.276. Structured

interviews or questionnaires are common examples

for techniques used in methods. Techniques used for

the application of best practice reference models

support activities transforming models to other

models. Therefore, techniques used in this method

are defined as adaptation mechanisms transforming

a reference model step by step to a company’s

model. These techniques could be derived from

available research on reference model application.

Conclusions from research on reference models can

be included especially if the methods themselves are

formally represented by models. The method

presented herein contains several models as result

types (shown in section 2.2). Thus, the concepts of

reference model application provide important

information about the design of techniques within

the presented method.

Becker et al., 2004 provide two types of

adaptation mechanisms: The mechanisms of

generative adaptation describe all modes of a

reference model’s configuration, given the existence

of rules which determine how to adjust the reference

model depending on mechanisms of configuration.

These rules should be included in the reference

model. As mentioned, best practice reference models

of IT governance are conceived as structured

compilations of best practice rather than conceptual

models. Therefore, BPRM do not usually contain

explicit rules for model configuration.

Apart from configuration, Becker et al., 2004

describe four mechanisms of non-generative

adaptation: what generally characterizes

mechanisms of non-generative adaptation is “that

while they support the creation of specific model

variants, they leave room for variety to be filled by

the user of the reference model” (Becker et al., 2007

p.1). As this matches the situation in the area of

BPRM, the four non-generative adaptation

mechanisms will be concisely described and

integrated into the method as technique types.

Ad 1) Aggregation requires the reference model

to be divided into its components which are

recomposed by aggregation for new solutions.

Combinations can be limited by defined joints.

Ad 2) Instantiation ultimately describes the

existence of deliberately vague formulations or

blank spaces as placeholders to be specified by

users. In order to develop a BPRM into an explicit

model system, placeholders have to be filled in a

corporation-specific way. A BPRM is more freely

and individually adaptable trough instantiation than

trough aggregation.

Ad 3 and 4) analogy construction and

specialization are very free forms of adaptation in

which prescriptions for adjustment are mostly

absent. However, Becker et al., 2004 stress that even

these free adaptation mechanisms should give details

for the user about which model elements are suitable

for specialization and analogy (p.259).

The use of adaptation mechanisms of reference

models within the area of method construction has

been accomplished several times. For instance

Harmsen, 1997 or Brinkkemper, 1996 use the

mechanism aggregation, whereas Baskerville &

Stage or Patel, de Cesare, Iacovelli, & Merico, 2004

use specialization. A broad overview is given from

Becker et al., 2007, p. 5, table 1. The herein

presented method includes the following types of

technique: aggregation, instantiation, specialization,

and analogy construction (their degrees of

prescription about adjustment in descending order).

3.4 Method Element: Activity Type

The method for reference model application

described by Fettke & Loos, 2002 includes two

phases. The phase reuse follows after the phase

construction of the model. Reuse is divided into four

sub phases, which adapt the model to the corporate-

specific situation. These sub phases represent

activities according to the language-based approach

of method engineering. A possibility to distinguish

these activities is presented by Schütte, 1998.

Firstly, compositional activities means that

individual parts of a model are erased, altered, or

added in order to improve a reference model’s fit.

Secondly, generic adaptation activities means

explicitly described rules of adaptation. These rules

are defined explicit within the model to be observed

for adjustment of the reference model. Thus,

ICEIS 2010 - 12th International Conference on Enterprise Information Systems

152

activities needed to be taken before applying a

reference model to a corporate-specific model.

Generic adaptation activities are not usually

employed since most BPRM do not contain rules for

adaptation. Furthermore compositional adaptation

activities need to be divided into more parts. For the

presented method, there is a difference between the

two compositional adaptation activities “choice of

model components”, i.e. alteration of the model (e.g.

by erasing certain parts Gammelgard, Lindstrom, &

Simonsson, 2006) and “adaptation to corporate

conditions” (e.g. corporation-specific indices).

The generic method considers this by the activity

types subset selection and adjustment. If a BPRM is

not entirely used, it is limited by selecting a BPRM

subset to the part relevant for a corporation. The

reason for this decision does not necessarily based

within the model itself but can be entirely strategic

(Bowen, Cheung, & Rohde, 2007). Hence activity

type subset selection is take place before activity

type adjustment. During the subsequent adaptation,

the chosen subset is continuously adjusted to the

corporation. The activity type application completes

the generic activity types. In the following, all three

types of activity will be described in detail jointly

with their proposed techniques. The order of the

activity types in the method is depicted in Figure 3.

Figure 3: Event-driven-chain the proposed method.

Activity 1. Subset Selection

By selecting a model subset the BPRM is

transformed resulting in a BPRM subset. The

process of selection itself with its internal

organizational and communicative aspects is not

addressed in this paper. However, possible kinds of

this subset are interesting for this research. The

criterion to classify subsets is completeness. Two

cases occur in the first place, complete and partial

use. The former makes the result type subset

obsolete as the BPRM and the subset are identical.

But if some parts are selected while others are not,

the following applies to the contents described

within the model: BPRM Subset <= BPRM.

For detailed specification, further classification

criteria are required. These criteria can be derived by

abstraction of BPRM into a best practice reference

meta model. A meta model created by semantic

abstraction can show possible sub divisions of the

case “partial application” by means of content and

structure (Alter & Goeken, 2009). A model’s

structure is defined by its meta model components.

A Limitation of the applied meta model components

typical for a reduction of a model’s range, for

instance a subset which only contains the meta

model component COBIT control objectives

(Simonsson & Johnson, 2008, De Haes & van

Grembergen, 2008). It turns out that the structure of

a model is changed. Still, chosen model components

need to be aggregated into BPRM subset

the form of

aggregation being defined by internal relations

within the meta model. A coherent subset does not,

for instance, allow the use of metrics of the COBIT

processes unless the goals of the process are used as

well. This is because the component goal links

process with metrics. Thus, the meta model shows

options of aggregation.

The second case to be regarded is defined by

reduction of the model’s profundity. Accordingly,

all meta model components are employed in respect

to the COBIT model but not all model components.

Thus the content is reduced. These subsets leave the

model’s structure unchanged (Gammelgard et al.,

2006). The reduction of model components results in

different problems than the reduction of meta model

components does. This is due to interconnections of

content, such as predecessor-successor-relations,

which can cause successors to be left without any

input or the output of a process to remain unused

even though links within the meta model are intact.

Hence, model components have to be aggregated on

a model level as well. This aggregation of model

components (e.g., several COBIT processes) can be

supported by the connection among meta model

A GENERIC METHOD FOR BEST PRACTICE REFERENCE MODEL APPLICATION

153

Instantiations

Model Level

Meta Level

Goal

me asured

Met ri csupportsProcess

Deliver Projects on

Ti m e an d on Bu dg et

M eet in g Q u ali ty

Standards

me asured

Percent of Projects

M eet in g Sta keh ol de r

E x pe c ta tio ns

supports

Re fe ren ce Pr o cess

PO 10

Deliver Projects on

Ti m e an d on Bu dg et

M eet in g Q u ali ty

Standards

measured

Pe rc ent of P roj ects Mee tin g 90 % o f IS O 90 00

R equ ir em en ts

supports

Co mpan y’ s P roce ss

PO 10 Manage

Projects

Perc ent of Pr oj ect s on Ti me ( 1,2 5-fold of

Ta rg et Du ra tio n)

Figure 4: Instantiation of a COBIT metric.

components. These convenient options of

aggregation can be derived from the meta model.

For the COBIT meta model, the meta model

component result proves to be an initial point for

interconnection of model component process.

It turns out that connections occur on a meta

model level when transforming generic BPRM into

specific BPRM sections that can be used for

aggregation. With its components, the meta model of

a BPRM offers possibilities for aggregation of both

meta model components and model components.

Hence, the quality of meta models in best practice

reference models is crucial for this research project

(Alter & Goeken, 2009, Goeken & Alter, 2009).

Activity 2. Adjustment to Corporate Conditions

Once the relevant BPRM subset has been selected,

the next step is the transformation into the explicit

model particular to one specific BPRM and one

specific corporation. During instantiation the user

specifies those model sections which formerly

remained deliberately vague. However, it usually

remains unclear for BPRM which model sections

have remained vague on purpose and require

instantiation.

Order and design of the model component

“metric of process x” in the COBIT model allow the

assumption that metric is a components which

requires instantiation. Exemplary in character, the

metrics of a COBIT process should be completed

with individual metrics. Along with the mechanism

of adaptation in the present example, Figure 4

depicts the instantiation of the metric placeholder for

a number of corporate-specific metrics.

Other mechanisms are applied during the

development of the explicit model as well.

Supporting the instantiation both specialization and

analogy construction should be primarily used in the

following third step. This is due to the relation

between the BPRM and the explicit model. Here, the

explicit model is taken as an altered part of the

BPRM, which should basically remain recognizable

in this intermediate result. Control by IT auditors is

thus facilitated in case of COBIT. This can change

due to the more variable mechanisms of model

adaptation, which is why too much room for

variation in adaptation mechanisms should be

avoided in this activity.

Activity 3. Application to the Corporate Model

Depending on the BPRM the corporate model can

consist of either the process model of IT processes

or a smaller part such as a model of IT project

management. Specialization and analogy

construction are important mechanisms during this

phase since BPRM of IT governance usually specify

what to do rather than how to do it. Those challenges

of establishing have to be fulfilled by means of

analogy construction in which the explicit model

serves as a state-to-be and to inspire ideas.

„Analogies can be drawn from any aspect of the

reference model which can be indicated by the

annotation of analogy construction advices“(Becker

et al., 2007, p.3).

4 METHOD EVALUATION

In a second step the method obtained in a design-

oriented research process for the application of

BPRM is to be evaluated. The evaluation can be

carried out in two ways (Hevner et al., 2004).

Firstly, the method itself can be focused on;

secondly, the process of constructing the method

plays a role as well. Although Hevner's guidelines

were meant to enhance the probability for a

publication of design science research, they are also

used for support of a systematic evaluation of

research (Arnott & Pervan, 2008, Zelewski, 2007).

ICEIS 2010 - 12th International Conference on Enterprise Information Systems

154

According to these criteria both artifact and research

process will be examined in the following (Hevner

et al., 2004).

1. The first guideline states that construction-

oriented research is supposed to create an innovative

artifact for the accomplishment of one task and to

solve an existing problem. While the application of

best practice reference models has not been

supported by scientifically developed methods, the

systematic design of the application process is a

manifest problem in corporate practice, particularly

in a multi-model case (Cater-Steel, Tan, &

Toleman, 2006, Siviy et al., 2008a and b). Thus,

construction of a method applicable to several

BPRM solves an existing problem.

2. The relevance of the scientific problem

characterizes the importance of the problem for

scientific practice. This importance might result

from specific sentences of relevance or the obvious

notion of problems within the respective decisive

constituent community (Zelewski, 2007). IT

governance is the relevant field of research for this

artifact. Various results provide a certain evidence

for an effect of applying BPRM on the achievement

of a corporation's goals. Simonsson & Johnson,

2008 prove the effect of governance maturity on IT

within a corporation. Governance maturity itself is

raised by use of BPRM. Wagner, 2006 presents a

case study in which a specific BPRM (ITIL) has a

positive effect of IT on corporate success. Studies by

Tuttle & Vandervelde, 2007 are based on the

positive influence of the use of COBIT on risk

management performance. Heier, Borgman, &

Maistry, 2007 prove the positive association of

governance software, governance processes, and IT

value contribution by empirical research. In another

case study they conclude that the performance of

governance processes is a critical factor for the

success of corporate goals (Heier, Borgman, &

Hoffbauer, 2008). An empirical study proves a

measurable connection between IT governance

software and a corporation's IT value contribution

(Heier, Borgmann, & Mileos, 2009). Based on this

connection, the case study by Larsen, Pedersen, &

Anderson, 2006 analyses 17 tools of IT governance,

including the BPRM of IT governance dealt with

herein. A Delphi study by De Haes & van

Grembergen, 2008 presents the ten most important

COBIT processes for achievement of IT goals and,

indirectly, corporate goals. As the author of this

paper points out, this implies that certain goals are

probably achieved by the application of COBIT or

its individual processes. Lunardi, Becker, &

Macada, 2009 examine the financial influence of „IT

governance mechanisms’ adaptation“ within

Brazilian companies. They distinguish between two

cases, adaptation by means of BPRM (ITIL, COBIT)

or without BPRM. The procedure of this adaptation

is not described. Debreceny & Gray, 2009 present a

case study on the effect of the BPRM COBIT on

maturity of processes and on IT capability. A survey

by de Espindola, Rodrigo Santos, Luciano, & Audy,

2009 proves that successful adaptation of BPRM

affects various corporate goals.

To sum up, literature frequently states that the

application of BPRM has, if only indirectly, positive

effects on the achievement of a company’s goals. It

has been shown that the “constituent community”

uses BPRM without dealing with their application in

great detail. The scientific and practical relevance is

proved but scientific work on the application is

almost entirely lacking. The method presented

herein fill parts of this gap by developing a method

for the application of BPRM.

3. The evaluation of research results includes

usefulness, quality, and effectiveness Results are to

be evaluated by strictly scientific evaluation

methods as Hevner et al. 2004 emphasize. They

suggest a number of methods for detection of

misbehavior of the artifact, prove of usefulness, etc.

The evaluation of the generic method includes two

steps due to the characteristics of the artifact. As the

method described has a generic character, it is partly

evaluated in a first step by application to a specific

case, in which its general usability is proven. The

specific methods then have to be evaluated

individually. The degrees of coverage and separation

of the elemental types in generic constructions have

been examined in a research project. It is

questionable whether all method elements required

in a specific method are available in the generic

model and whether they can unambiguously be

matched to model component types. If this is not the

case, the generic model needs to be adjusted.

4. Construction-oriented research, as any

research, has to contribute to the progress within a

field of research. Three characteristics are generally

employed for this: novelty, general validity, and

relevance. In addition, it is recommended for design-

oriented research to create an artifact providing

significance for an unsolved problem of scientific

community or a new application of existing

knowledge.

Novelty. Currently, there is no complete method for

the application of BPRM in IT governance. Existing

methods for application of reference models have

partly been included in the artifact. Apart from this,

experience from the area of BPRM and their

application in corporate practice have been

integrated and theoretically dealt with. Also, the

A GENERIC METHOD FOR BEST PRACTICE REFERENCE MODEL APPLICATION

155

artifact is a novelty as it recombines existing

knowledge from model theory and research on

method construction.

General Validity. The developed artifact is

generally valid. The method can be used to support

any BPRM. The degree of abstractness corresponds

with the differences between existing and future

BPRM of IT governance. If, for example, the

method is to be applied to a BPRM which does not

require subset selection this step can be omitted.

This is the case if a desired certificate is linked to a

complete use of a model.

Relevance. The relevance of the artifact for the field

of IT governance is demonstrated in the presented

research papers mentioned in guideline 2. The

described method allows a systematic application of

BPRM in scientific practice.

Significant Use for an Unsolved Problem. The

method presents a systematic procedure for the

application of BPRM for a corporation’s support.

The procedure is directly useful to solve company’s

problems. It also provides an indirect use as a basis

for the application of multiple BPRM for the multi-

model case.

5. This guideline prescribes a rigorous use of

scientific methods when constructing and evaluating

the artifact. According to Hevner et al., the scientist's

experience is required for „skilled selection of

appropriate techniques” (Hevner et al., 2004, p.80).

Also, they stress that scientific rigor is based on

consequent definitions, consistence within research,

and formal representation of the topic. Scientific

methods for evaluation have already been described.

The process of construction meets the criteria of

language-based method engineering. The artifact

components have been selected by means of the St.

Gallen approach of method engineering. Design and

composition of components have been obtained by

application and aggregation of available knowledge

taken from a broader literature review and structured

interviews with practitioners.

6. This guideline provides that “search for an

effective artifact requires utilizing available means

to reach desired ends while satisfying laws in the

problem environment“ (Hevner et al., 2004, p.81).

They describe means as a set of actions and

resources available to construct a solution. Thus, the

research process should follow a generate-test-cycle.

After creating a design alternative it has to be tested

against requirements and constraints. The

construction of the presented method follows this

cycle. The method is and will be tested and adjusted

several time during the research project.

7. This guideline deals with the communication

of research. It is crucial to present results adequate

to the appealed audience. It is also important that the

audience „understand [s] the processes by which the

artifact was constructed and evaluated“ (Hevner et

al., 2004, p.90). By publishing and presenting the

results in academia and practice results can broaden

the knowledge base for further research efforts and

construct solutions for practitioners.

5 CONCLUSIONS

By using the knowledge base of IS research and

business needs concerning BPRM application, we

have constructed and partly evaluated a generic

method for best BPRM application. This generic

method is on the one hand an addition to the

knowledge base of IS research and on the other hand

a possibility to support the application of BPRM in

practice. For this practical use the generic method

has to be instantiated to a specific method.

The paper follows the language-based method

engineering approach by presenting formally

described static method element types and their

instantiations. Furthermore the presentation includes

dynamic aspects by describing processes and

procedures concerning the transition between

various instantiations of the method element types.

This method aims to be an addition to the knowledge

base of IT governance research and a practical

solution for challenges IT departments are

confronted with.

In an ongoing research process a specific method

for the BPRM COBIT is derived from the presented

generic method. This specific method supports the

methodical application of the COBIT BPRM. These

research findings were also used to develop a

governance tool, based on semantic software which

supports the application of COBIT. Furthermore, the

method provides a sound basis for the construction

of methods for applying several BPRM, i.e. in the

multi-model case.

REFERENCES

Alter, S., & Goeken, M. (2009). Konzeptionelle

Metamodelle von IT-Governance-Referenzmodellen

als Basis der Kombination und Integration in einer

Multi-Model-Umgebung. In Hansen, H.R.,

Karagiannis, D., Fill, H.G. (Ed.), Business Services:

Konzepte, Technologien, Anwendungen.: 9.

Internationale Tagung Wirtschaftsinformatik, Band 1 .

Wien.

ICEIS 2010 - 12th International Conference on Enterprise Information Systems

156

Arnott, D., & Pervan, G. (2008). Eight key issues for the

decision support systems discipline. Decision support

systems, 44(3), 657–672.

Baskerville, R. L., & Stage, J. Accommodating Emergent

Work Practices: Ethnographic Choice of Method

Fragements: IFIP TC8/WG8.2 Working Conference

on Realigning Research and Practice in IS

Development: The Social and Organisational

Perspective, Boise, ID, 2001, pp. 12.28.

Becker, J. (1995). Strukturanalogien in

Informationsmodellen – Ihre Definition, ihr Nutzen

und ihr Einfluß auf die Bildung der Grundsätze

ordnungsmäßiger Modellierung (GoM). In W. König

(Ed.), Wirtschaftsinformatik '95 –

Wettbewerbsfähigkeit, Innovation, Wirtschaftlichkeit

(pp. 133–150). Heidelberg.

Becker, J., Delfmann, P., & Knackstedt, R. (2004).

Konstruktion von Referenzmodellierungssprachen:

Ein Ordnungsrahmen zur Spezifikation von

Adaptionsmechanismen für Informationsmodelle.

Wirtschaftsinformatik, 46(4), 251–264.

Becker, J., & Knackstedt, R. (Eds.) (2002).

Referenzmodellierung 2002 Methoden – Modelle –

Erfahrungen (Vol. 90). Arbeitsberichte des Instituts

für Wirtschaftsinformatik WWU, 90. Münster.

Becker, J., Knackstedt, R., Holten, R., Hansmann, H., &

Neumann, S. (2001). Konstruktion von Methodiken:

Vorschläge für eine begriffliche Grundlegung und

domänenspezifische Anwendungsbeispiele

(Arbeitsberichte des Instituts für

Wirtschaftsinformatik No. 77). Münster: Westfälische

Wilhelms-Universität Münster, from

http://www.wi.uni-muenster.de/inst/arbber/ab77.pdf.

Becker, J., Knackstedt, R., Pfeiffer, D., & Janiesch, C.

(2007). Configurative Method Engineering: On the

Applicability of Reference Modeling Mechanisms in

Method Engineering. In Proceedings of the AMCIS

2007 .

Becker, J., Niehaves, B., & Knackstedt, R. (2004).

Bezugsrahmen zur epistemologischen Positionierung

der Referenzmodellierung. In J. Becker (Ed.),

Referenzmodellierung: Grundlagen, Techniken und

domänenbezogene Anwendung, mit 6 Tabellen (pp. 1–

17). Heidelberg: Physica-Verlag.

Becker, M. (1998). Umsetzung betrieblicher Prozesse:

Methode, Fallbeispiele, Workflow-Technologie.

Dissertation, Universität St.Gallen, St.Gallen, from

Universität St.Gallen: .

Bowen, P. L., Cheung, M.-Y. Decca, & Rohde, F. H.

(2007). Enhancing IT governance practices: A model

and case study of an organization’s effort.

International Journal of Accounting Information

Systems, 8, 191–221.

Brinkkemper, S. (1996). Method engineering: engineering

of information systems development methods and

tools. Information & Software Technology, 38(4),

275–280, from http://www.sciencedirect.com/science/

article/B6V0B-3VTJK0G-

Y/2/4211ed8ea0fdafa25ee5b90754e89ea1.

Brinkkemper, S., Saeki, M., & Harmsen, F. (1999). Meta-

modelling based assembly techniques for situational

method engineering. Information Systems, 24(3), 209–

228, from http://www.sciencedirect.com/

science/article/B6V0G-3WXWT1F-

3/2/a253952d59a29fb748286910bcb97b8a.

Brooks, F. P. (1987). Essence and Accidents of Software

Engineering. IEEE Computer, 20(4), 10–19.

Cater-Steel, A., Tan, W. G., & Toleman, M. (2006).

Challenge of adopting multiple process improvement

frameworks. In Proceedings of the 14th European

Conference on Information Systems .

de Espindola, Rodrigo Santos, Luciano, E. Mezzomo, &

Audy, J. L. N. (2009). An Overview of the Adoption

of IT Governance Models and Software Process

Quality Instruments at Brazil - Preliminary Results of

a Survey. In HICSS 2009 - Proceedings of the 42th

Hawaii International Conference on System Sciences .

Debreceny, R., & Gray, G. L. (2009). IT Governance and

Process Maturity: A Field Study. In HICSS 2009 -

Proceedings of the 42th Hawaii International

Conference on System Sciences .

Fettke, P., & Loos, P. (2002). Methoden zur

Wiederverwendung von Referenzmodellen –

Übersicht und Taxonomie. In J. Becker & R.

Knackstedt (Eds.), Arbeitsberichte des Instituts für

Wirtschaftsinformatik WWU: Vol. 90.

Referenzmodellierung 2002 Methoden – Modelle –

Erfahrungen (pp. 9–30). Münster.

Fettke, P., & Loos, P. (2003). Classification of reference

models—a methodology and its application.

Information systems and ebusiness management, 1(1),

35–53.

Fitzgerald, B., Russo, N. L., & O'Kane, T. (2003).

Software Development: Method Tailoring at

Motorola, Communications of the ACM, 46(4), 65–70.

Frank, U. Conceptual Modelling as the Core of the

Information Systems Discipline – Perspectives and

Epistemological Challenges. In W. D. Haseman & D.

L. Nazareth (Eds.). Proceedings of the Fifth Americas

Conference on Information Systems (AMCIS 1999),

August 13-15, 1999. Milwaukee, Wisconsin

(pp. 695.697).

Gammelgard, M., Lindstrom, A., & Simonsson, M.

(2006). A reference model for IT management

responsibilities. In . EDOCW ’06: Proceedings of the

10th IEEE on International Enterprise Distributed

Object Computing Conference Workshops (pp. 26-26).

Washington, DC, USA: IEEE Computer Society.

Goeken, M. (2002). Grundlagen und Ansätze einer

Referenzmodellierung für Führungs informations

systeme: Fachbericht Nr.03.

Goeken, M., & Alter, S. (2009). Towards Conceptual

Metamodeling of IT Governance Frameworks: –

Approach – Use- Benefits. In . HICSS 2009 -

Proceedings of the 42th Hawaii International

Conference on System Sciences .

Gutzwiller, T. (1994). Das CC RIM-Referenzmodell für

den Entwurf von betrieblichen,

transaktionsorientierten Informations systemen.

Heidelberg: Physica-Verlag.

Haes, S. de, & van Grembergen, W. (2008). An

exploratory Study into the design of an IT Governance

Minimum Baseline through Delphi Research.

A GENERIC METHOD FOR BEST PRACTICE REFERENCE MODEL APPLICATION

157

Communications of the Association for Information

Systems, 22(Article 24), 443a.

Harmsen, A. F. (1997). "Situational Method Engineering.

Utrecht.

Hars, A. (1994). Referenzdatenmodelle – Grundlagen

effizienter Datenmodellierung. Wiesbaden:

Dissertation 1993 Saarbrücken.

Heier, H., Borgman, H., & Hoffbauer, T. (2008). Making

the most of IT governance software: understanding

implementation processes. In HICSS 2008 -

Proceedings of the 41st Annual Hawaii International

Conference on System Sciences, Washington, DC,

USA: IEEE Computer Society.

Heier, H., Borgman, H., & Maistry, M. (2007). Examining

the relationship between IT governance software and

business value of IT: evidzence from four case studies.

In, HICSS 2007: Proceedings of the 40th Annual

Hawaii International Conference on System Sciences .

Washington, DC, USA: IEEE Computer Society.

Heier, H., Borgmann, H., & Mileos, C. (2009). Examining

the Relationship between IT Governance Software,

Processes, and Business Value: A Quantitative

Research Approach. In . HICSS 2009 - Proceedings of

the 42th Hawaii International Conference on System

Sciences .

Hevner, A. R., March, S. T., Park, J., & Sudha, R. (2004).

Design Science In Information Systems Research.

MIS Quarterly, 28(1), 75–106.

Heym, M. (1993). Methoden-Engineering - Spezifikation

und Integration von Entwicklungsmethoden für

Informationssysteme. Disseration, Universität

St.Gallen, Hallstadt, from Universität St.Gallen: .

Karlsson, F., & Wistrand, K. (2006). Combining method

engineering with activity theory: theoretical grounding

of the method component concept. European Journal

of Information Systems, 15(1), 82–90.

Kaschek, R. (Ed.). 1999. Entwicklungsmethoden für

Informationssysteme und deren Anwendung: EMISA

'99, Fachtagung der Gesellschaft für Informatik e. V.

(GI). Stuttgart, Leipzig.

Kelly, S., Rossi, M., & Tolvanen, J. P. (2005). What is

Needed in a MetaCASE Environment? Enterprise

Modelling and Information Systems Architectures,

1(1), 25–35.

Larsen, M. H., Pedersen, M. K., & Anderson, K. V.

(2006). IT Governance: Reviewing 17 IT Governance

Tools and Analysing the Case of Novozymes A/S. In

HICSS 2006: Proceedings of the 39th Annual Hawaii

International Conference on System Sciences,

Washington, DC, USA: IEEE Computer Society.

Loos, P., & Fettke, P. (2005). Referenzmodellierung:

Entwicklungsstand und Perspektiven, Information

Management & Consulting, 20(Sonderausgabe).

Lunardi, G. Lerch, Becker, J. Luiz, & Macada, A. C.G.

(2009). The Financial Impact of IT Governance

Mechanisms' Adoption: An Empirical Analysis with

Brazilian Firms. In HICSS 2009 - Proceedings of the

42th Hawaii International Conference on System

Sciences .

Patel, C., Cesare, S. de, Iacovelli, N., & Merico, A.

(2004). A Framework for Method Tailoring: A Case

Study. In . 2nd OOPSLA Workshop on Method

Engineering for Object-Oriented and Component-

Based Development . Vancouver.

PWC (2006). IT Governance in Practice - Insight from

leading CIOs. PricewaterhouseCoopers Johannesburg,

South Africa, IT Governance Institute Rolling

Meadows, USA.

Ralyté, J., & Rolland, C. (2001). An Assembly Process

Model for Method Engineering. In Lecture notes in

computer science: 13th International Conference on

Advanced Information Systems Engineering (CAiSE

2001) (pp. 267–283).

Rosemann, M., & van der Aalst, W. M. P. (2007). A

configurable reference modelling language.

Information Systems, 32(1), 1–23.

Scheer, A.-W., Seel, C., & Georg, W. (2002).

Entwicklungsstand in der Referenzmodellierung.

Industrie Management, 18(1), 9–12.

Schütte, R. (1998). Grundsätze ordnungsmäßiger

Referenzmodellierung: Konstruktion konfigurations-

und anpassungsorientierter Modelle (Vol. 233). Neue

betriebswirtschaftliche Forschung, 233. Wiesbaden:

Gabler.

Simonsson, M., & Johnson, P. (2008). The IT

Organization Modeling and Assessment Tool:

Correlating IT Governance Maturity with the Effect of

IT. In HICSS 2008 - Proceedings of the 41st Annual

Hawaii International Conference on System Sciences

(p. 431). Washington, DC, USA: IEEE Computer

Society.

Siviy, J., Kirwan, P., Marino, L., & Morley, J. (2008a).

Maximizing your Process Improvement ROI through

Harmonization (White Paper SEI). Software

Engineering Institute.

Siviy, J., Kirwan, P., Marino, L., & Morley, J. (2008b).

The Value of Harmonizing Multiple Improvement

Technologies: A Process Improvement Professional’s

View (White Paper SEI). Software Engineering

Institute.

Tuttle, B., & Vandervelde, S. D. (2007). An empirical

examination of COBIT as an internal control

framework for information technology. International

Journal of Accounting Information Systems, 8, 240–

263.

Wagner, H. Theo (2006). Managing the impact of IT on

Firm Success: The Link between the resources-based

view and the IT Infrastructure Library. In . HICSS

2006: Proceedings of the 39th Annual Hawaii

International Conference on System Sciences

Washington, DC, USA: IEEE Computer Society.

Zelewski, S. (1999). Grundlagen. In H. Corsten & M. Reiß

(Eds.), Betriebswirtschaftslehre (pp. 1–125). Wien.

Zelewski, S. (2007). Kann Wissenschaftstheorie behilflich

für die Publikationspraxis sein?: Eine kritische

Auseinandersetzung mit den Giudelines vn Hevner et

al. In F. Lehner & S. Zelewski (Eds.),

Wissenschaftstheoretische Fundierung und

wissenschaftliche Orientierung der

Wirtschaftsinformatik (pp. 71–120). Berlin: Gito.

ICEIS 2010 - 12th International Conference on Enterprise Information Systems

158