Organizational design in public administration

typology of project management offices

Université du Québec à Montréal

Maude Brunet

Université du Québec à Montréal

Abstract

The aim of this paper is twofold. First, it proposes an empirical typology of project management offices (PMOs) based upon four types of projects: (1) engineering and construction, (2) information system and technology, (3) business processes, and (4) new product/service development. Our findings lead us to propose a PmO typology for public administration. PMO types are described according to their strengths or weaknesses in terms of organizational characteristics, project characteristics, structural characteristics, functions, and performance.

Second, the paper aims to enlighten and elucidate the organizational process involved in managing multiple concurrent projects. This is particularly challenging in public administration, where hierarchy and bureaucracy are pervasive. Moreover, public administration is not a single, monolithic entity. Rather, it usually comprises a collection of multiple organizations that differ in status, some of them departments, others agencies. Strong pressures from national auditors general to achieve better results with public funds have created a considerable impetus to search for better approaches to project management.

This paper is based on a survey of 114 entities belonging to 42 departments and agencies within a single public administration. Each of these entities performs activities related to the management of multiple projects. Furthermore, the paper provides a unique internal view of the vastness of such activities within government.

This paper’s main contribution to the academic field is to enrich existing knowledge about organizational design in project management within the specific context of public administration, both in terms of the organization and the process of organizing. Contributions to the professional field range from providing a description of a public administration reality to nurturing the design work of decision-makers through an empirically validated PMO typology. Our findings may help in understanding a situation and open up new avenues for the future in the field of project management organizational design.

Keywords: public administration; organizational design; PMO; PMO typology

1. Introduction

This conference calls for a novel approach to research in the field of project management (Project Management Institute, 2013b). It has been a challenge for authors to address a rather well-known problem related to managing multiple projects while creating some sort of interest and a mystery effect (Alvesson & Sandberg, 2011) and, more importantly, while conducting research with a meaningful impact on organizations (Alvesson, 2013).

This paper aims to advance scientific knowledge about organizational design in the specific context of public administration facing the management of multiple projects. It presents the results of research conducted within a public administration where there is a quest for implementing good practices in project management. Yet, poor project management results have been cited for several years by the auditor general. The media also play a role in demanding a better return on public funds. This is not something new for public administration. Countries like Norway, the UK, and Australia have been criticized for much the same situation (Australia Government, 2012; Samset & Volden, 2013; United Kingdom Government, 2012). This is happening at a time when governments are facing strong pressures to adopt new, transparent public management policies, to open up to the public and to interact with the social media (Osborne, 2000).

In this particular situation, various actions have been taken by the government under study over the last few years that have failed to produce good results. The department responsible for government processes and performance began reflecting on the overall approach to project management. The starting point for this reflection was to determine the current situation. Therefore, a research team was formed and asked to participate in this effort and was mandated to provide a picture of reality showing the functions involved in managing multiple projects. Of course, a public administration is a complex organization or rather, a composite of multiple, complex organizations subject to pressures in the national political system. Therefore, generating a global (and central) understanding of the project management situation is a difficult task. Basic questions remain unanswered, such as: Who is involved in managing multiple projects throughout the governmental apparatus? What are they doing? How sensitive are they to project management performance? In this context, the research presented here was undertaken.

This paper offers only a partial view of the overall reflection on the situation within the government under study at the time of this research. Accordingly, this research supports an overall evidence-based management approach (Rousseau, 2012), calling on academics and professionals to work together toward understanding crucial problems and envisioning contextual solutions. However, we strongly believe that the findings of this research could enlighten similar questions pertaining to any other government. It provides a unique survey of 114 entities in 42 departments and agencies on multiple project management within the same government.

Clearly, this research contributes to a better understanding of project organization and the project organizing process (Van de Ven & Poole, 1995). The methodology adopted in this paper is based on a survey involving statistical analysis. In this sense, findings provide valuable knowledge about the number of entities dealing with multiple projects along with findings useful to decision-makers for organizational purposes, including the development of project management policies.

Some recognition exists that organization design has been neglected as a core concern of organizational theory (Greenwood & Miller, 2010). This is based on the assumption that “a focal interest of organization theory must inevitably be the understanding of how to organize people and resources in order to collectively accomplish desired ends” (Greenwood & Miller, 2010, p. 78). These authors observed some neglect over the past several decades of organizational design among management and organization scholars. They argue that three main reasons explain the lack of interest in organizational design:

  • A shift of empirical and theoretical focus within the field from the organization to the field, population, and community as the unit of analysis
  • The sheer complexity of many of today’s organizations and the potential difficulty involved in data collection methodology requirements
  • The over-riding interest in parts of organizational design to the detriment of the overall organization

Conversely, a clear interest has emerged over the last few decades among scholars in the project management field for work on organizational design such as project business (Artto and Wikstrom, 2005); project management offices (Aubry, Hobbs, Müller, & Blomquist, 2011; Hobbs & Aubry, 2010); project-based and project-oriented firms (Geraldi, 2009; Hobday, 2000); the institutional context (Morris & Geraldi, 2011); and governance (Müller, 2011). Recently, Winch (2014) has proposed an integrative view of organizational design research in project management literature. He proposed a three-domain framework to deal with the tension between the temporary nature of projects and the permanent nature of the parent organization.

Why this interest? One answer probably concerns the well-established proximity of project management to the professional world. Managing multiple concurrent projects poses major challenges in terms of understanding how to organize people and resources in order to collectively accomplish the desired ends.

Moreover, in project management research, these three obstacles mentioned by Greenwood and Miller (2010) seem to be of less importance. First, mainstream research still considers the organization and the problems internal to organization. Second, project management scholars seem fascinated by the complexity of organizational forms. Third, the overall organization is not always the focus, but the networked nature of projects makes research on organizational design in project management more inclusive of multiple views.

Overall, project management research has something to offer to organizational theory on organizational design in overcoming obstacles, as mentioned by Greenwood and Miller (2010). This paper, while highly quantitative and narrow in scope, aims to contribute to this goal of organizational design for project management.

This paper is structured as follows. Following an introduction, the second section provides background information on the naming issue and existent typologies. The third section presents a conceptual framework, and the fourth section describes the methodology used for data collection and analysis. The fifth section contains the results of statistical analysis, and the sixth section presents the findings. Section 7 includes a discussion, followed by the conclusion, which specifies the major limitations of this paper while suggesting avenues for future research.

2. Background

2.1Types of PMO

A substantial effort has been deployed recently to find a PMO typology based on empirical data that would offer an alternative to models found in an array of books written for professionals (Hobbs & Aubry, 2008, 2011; Müller, Glückler, & Aubry, 2013). This task is complicated by the wide variety of PMOs. The approach by variables has not generated strong results yet. The most promising model is based on crossing the following variables (Hobbs & Aubry, 2008): the number of projects covered by the PMO’s mandate and the number of project managers. This suggests four types of PMOs varying in terms of decision-making authority, whether they operate under a matrix type of organization or not, project management maturity, and organizational culture supportiveness. A more advanced statistical analysis creates a new model with six different types of PMO (Hobbs & Aubry, 2011). This refined model reveals significant differences among the project types in the performance area. PMOs with the most projects, the most project managers, and higher levels of decision-making authority perform better that the others.

Using a completely different approach, Müller et al. (2013) suggested a model based upon the PMOs’ relationship roles: serving, controlling, and partnering. Following a qualitative methodology approach, they show the impact of these relationships on innovative performance: slack, innovativeness, and ambidexterity. Serving and controlling PMOs are more likely associated with less operational innovation. On the other hand, the most highly innovative PMOs are shown to include partnership relationships. Most PMOs played the controlling role. Very few had an equilibrium role strong in ambidexterity (March, 1991; O‘Reilly III & Tushman, 2008).

As shown above, the current effort to propose an empirically validated typology for PMOs has not yet produced any convincing results. This may come as no surprise, given that such coordination mechanisms are more likely to be considered an organizational innovation (Hobbs, Aubry, & Thuillier, 2008).

Another way to explore PMO typology is based on project type.

2.2PMO Typology and Project Types

The role of categorization and classification is self-evident. They are important strategies that enable human beings to deal with complexities and numbers. One excellent example is the early Linné plant classification system. Projects have also been an object of classification in an effort to achieve better results, be they operational or strategic. Shenhar and Dvir (2004) have proposed a framework built on differences in project types. Based on contingency theory and innovation literature, they argue that not all projects are the same and they determine project types according to four dimensions: novelty, complexity, technology, and pace. They suggest adapting the project management style to the project type. This approach to project typology was then used to develop a multidimensional model for project success (Shenhar, Dvir, Levy, & Maltz, 2001). Two points must be mentioned regarding this framework: (1) the theoretical approach and (2) the particular study. First, contingency theory has not been extremely useful in explaining the diversity of PMOs. The size of the organization, industries, and regions did not offer any insight in explaining diversity (Hobbs & Aubry, 2010). It seems that a rich context must be understood using a more integrative approach, such as one including historical perspective (Zeitlin, 2008). Second, the goal of this public administration research does not concern projects or project success. Rather, it focuses on entities whose activities relate to multiple projects. This typology might not be as relevant in the context of this specific research.

Another research has been undertaken on projects categorization systems (Crawford, Hobbs, & Turner, 2005). It addresses two main objectives: to develop and to assign appropriate competencies to undertake projects successfully, and to form project portfolios. It also addressed the process of implementing a categorization system within an organization. They found that such a system includes two main components: the purpose of a classification, and the attributes used for its classification. Crawford et al. (2005) did not propose one particular model or framework of categorization. Rather, they “offer a tool for organizations to examine their existing project categorization systems, to better understand how they work, and as a guide to redesign them” (Crawford et al., 2005, p. 45). While this research is of great interest in understanding the underlying features of categorization systems, it is of no help in the specific case of a public context.

The goal of this research clearly deals with organizational design for project management. As such, the project typology of Shenhar and Dvir (2004) is inappropriate for studying entities involving multiple projects within a public administration. The process for developing and implementing categorization systems as suggested by Crawford et al. (2005) is not suitable for this particular research either.

This leads us to the following research question: “Is a PMO typology based on project types helpful in the organizational design of project management in public administration?”

2.3Conceptual Framework

The proposed conceptual model has been developed with the intention of capturing differences between entities dealing with multiple projects based on different types of projects. As illustrated in Figure 1, the global proposition is that these entities have specific characteristics depending on the types of projects they deal with: in terms of their organizational context, their structures, and the functions they perform. Moreover, they may also have different performance results concerning embeddnedness, project management factors, and criteria. The conceptual model that served as the basis for this paper was modeled after Hobbs and Aubry (2010). This model was empirically validated by a worldwide database of more than 500 PMOs, which lends it significant strength. As mentioned recently by Pfeffer (2013), differences must clearly be specified by scholars in research and in teaching to avoid “confusion between what is and what should be or what we would like to be” (p. 277). This descriptive model is not wishful thinking but a product of data. While it may not represent the full truth of a situation, it nonetheless has some significance to the reality.

Conceptual model for the study of types of entities

Figure 1: Conceptual model for the study of types of entities

In this conceptual model, we have adopted the term entity dealing with multiple projects, instead of PMO. PMO is the term largely used in a generic way within academic (e.g., K. Artto, Kulvik, Poskela, & Turkulainen, 2011), normative (e.g., Project Management Institute, 2012), and professional literature (e.g., Kendall & Rollins, 2003), although some exceptions exist, such as portfolio PMO (Unger, Gemünden, & Aubry, 2011) in academia and within a normative, scalable office model (Office of Government Commerce, 2008).

We already know that this generic term covers a multitude of applications (Hobbs & Aubry, 2010). But how well is this PMO term understood outside the field of project management? This question surfaced very early on in this public administration research: the persons to be involved were not necessarily accustomed to the project management jargon or, more specifically, the term PMO.

To return to the etymology of office, we find that the term has Latin roots and is defined as a major administrative unit in some government or a subdivision of some government department. It has primarily been used in public administration (Merriam-Webster, 2007), which is still the case in many governments, such as Post Office or Patent Office. The XIV century word bureau has roughly the same definition, such as Competition Bureau.

The term “entity dealing with multiple projects” has been adopted in this research to avoid the project management jargon term of PMO. It covers a very broad definition of a PMO:

An organizational body or entity assigned various responsibilities related to the centralized and coordinated management of those projects under its domain. The responsibilities of a PMO can range from providing project management support functions to actually being responsible for the direct management of a project (Project Management Institute, 2008, p. 443).

This definition is preferred over the 2012 definition (Project Management Institute, 2013a), which may limit some of the entities included.

The model contains two main components: a description of the organizational entities dealing with multiple projects and the resulting performance of those entities.

3. Methodology

In social science, a quantitative approach is usually associated with a positivist paradigm -where reality is considered to be out there and ready to be observed in a unique and stable state. Following this approach, the usual methods are based on hypothetical-deductive methods. There is a trend among practice scholars in management (including project management) to deviate from this approach where context is evacuated or, at least, to complement it (e.g., Cicmil, Williams, Thomas, & Hodgson, 2006; Drouin, Müller, & Sankaran, 2013).

This paper presents the results of a questionnaire, and as such involves hypothetical-deductive methods. However, this research is not intended to provide prescriptions for project management in a government. Instead, it aims to describe a government reality considered as a very partial element of a much more complex reality. The value of this research is to generate new scientific knowledge to help in understanding this complexity by proposing a categorization based on the types of projects in order to nurture reflection among decision-makers on organizational design for multiple project management. In this sense, the methodology used here is more easily understood when situated within a multidimensional continuum of research projects, between the quantitative and qualitative methods proposed by Teddlie and Tashakkori (2009).Using the three dimensions of these authors, in the concept sphere, the qualitative component relates to the subjective purpose and the aim of this research is to generate understanding (not confirmation). This research is more likely quantitative in the sphere of concrete processes with numeric data and statistical analysis. In the area of inference and explanation, this research leans more on the qualitative side, with more inductive and subjective inferences.

3.1Data Collection Strategy

This research took place in a national government where we had a unique opportunity to provide some input for reflective practitioners on how to significantly enhance the quality of multiple project management undertaken in government departments and agencies. It was managed with the help of professionals and managers in an administrative unit responsible for the management and performance of the government apparatus as a whole.

To meet the needs of the government, the questionnaire developed by Hobbs and Aubry (2007) has been used after being adapted to the specific context of this government setting. This has been done with the participation of community of practices dedicated to public management and with managers and professionals within the government. For example, the question on the position of respondents was modified to adapt to current classes in the government. Other questions were added to capture specific needs, for example, questions on portfolio management. A pretest was conducted with four respondents. Their comments led to some minor adjustments and these answers have not been included in the database. The questionnaire was in French only.

Many targeted respondents were not familiar with project management vocabulary. Different actions have been taken to account for this in our research and ensure confidence in the quality of answers. First, the term project management office or PMO was avoided to facilitate a general understanding among a diversity of respondents in government departments and agencies (DA). The unit of analysis was defined as any entity responsible for managing multiple projects and performing functions related to these projects. This definition is broad enough to include every entity in the research, without excluding relevant ones, as in Hobbs and Aubry (2007). Second, a guide to answering the questionnaire was developed, in which definitions and explanations were provided. Third and lastly, information sessions were organized for all government departments and entities at which the basic concepts where explained and the main sections of the questionnaire were reviewed.

The initial goal of this research was not to create a sample of entities dealing with multiple projects in this government but rather to represent them fully in order to create a census. The method used is not probabilistic, but based on judgment sampling. Participating entities were targeted in the specific aim of studying entities dealing with multiple projects. Invitations to answer the questionnaire were sent by email to the secretary-general of 50 DA. Many entities can coexist in each single DA, but only one respondent was allowed to answer for each entity. The secretary-general was responsible for identifying respondents in his own DA and for following up. The questionnaire was available on a website or in PDF format. Confidentiality and anonymity were ensured by having the completed questionnaires available only to the researchers, as agreed in an ethics certificate issued by the affiliated university.

The questionnaire was released in September 2012 and answers were collected until the end of October. A total of 119 questionnaires were completed, among which 98 valid questionnaires were entered into the database. An official return rate cannot be calculated because the total number of entities dealing with multiple projects in the governmental apparatus is unknown. Exclusions were based on a level of completeness under 60%.

Another 16 questionnaires were added to the data base. These questionnaires were completed during a preliminary phase of the government study using roughly the same questionnaire in the fall of 2011. A comparison of both samples revealed no significant differences, with a p-level < 0.05 (See Table 11 in Appendix). As shown in Table 11, there is no historical bias. The total number of questionnaires (114) adequately represents the entities dealing with multiple projects in the governmental apparatus, although it is not a census. Sample demographic can be found in Table 1.

n Percentage   n Percentage
Position of respondents   Project management maturity level
Senior executive 4 3.6   Initial level 21 18.4
Manager – higher levels 57 50.9   Repeatable level 50 43.9
Manager – lower levels 33 29.5   Defined level 34 29.8
Professional 18 16.1   Managed level 7 6.1
Total 112 100.0   Optimizing level 2 1.8
  Total 114 100.0
Types of deliverables  
Eng. or construction 28 24.8   Internal or external customers
New product/service dev. 11 9.7   Internal 74 71.8
Consultancy services 1 .9   External 29 28.2
IS or IT 44 38.9   Total 103 100.0
Business processes 26 23.0  
Events and cultural act. 3 2.7  
Total 113 100.0   No. of people working on a typical project
  10 or less 63 57.8
No. of employees in department and agency   11 to 25 31 27.9
1 to 100 5 4.4   26 to 50 13 11.7
101 to 500 44 38.6   51 to 75 2 18
501 to 1000 29 25.4   76 to 100 0 0
1001 to 5000 20 17.5   101 to 200 1 .9
More than 5000 16 14.0   More than 200 1 .9
Total 114 100.0   Total 109 100.0

Table 1: Sample demographics

The diversity of the sources of questionnaires is shown in Table 2, which lists DAs. Individual DAs were grouped by logical themes in order to shorten the list and to protect the anonymity of respondents where the number of answers was 5 or less.

Groups of departments and agencies Number of departments or agencies Number of respondents
Administration 4 19
Culture and museums 7 17
Infrastructures 4 14
Health and social services 2 8
Transportation 4 12
Solidarity, education, and research 4 7
Security and justice 4 11
Work and work security 4 10
Seniors and pensions 4 10
Resources, sustainability, and immigration 5 6
Total 42 114

Table 2: List of groups of departments and agencies

3.2Data Analysis Strategy

After having cleaned up the database as explained above, some answers were recoded when appropriate, usually based on the comment field.

Two main types of multivariate methods were used: the first, principle component analysis (PCA) and, the second, nonparametric one-way ANOVA analysis, along with multiples comparisons of means. PCA has been used to determine the underlining structure among three groups of variables; however, previous research has provided models for each of the three groups of variables. Discussion between the current and previous models is provided in the next section on results.

PCA provides summarizations and reduces the number of variables, resulting in values on the dimension of factors which are then included in the second multivariate method (Hair, Black, Babin, & Anderson, 2010, p.98).The three PCA were conducted on Likert scales from 1 to 5. Two types of scales were used:

  • Not important to very important (PCA on functions)
  • Totally disagree to totally agree (PCA on embeddedness and on project performance)

The reliability of the factor data was tested using Cronbach’s alpha: values greater than 0.6 were considered reliable (1951). Factors loading greater than 0.5 were considered significant (Hair et al., 2010). Those factors not meeting these two threshold levels were considered orphans and listed in the respective tables but excluded from further analyses (Aubry et al. 2010).The Kaiser, Meyer, and Olkin method (KMO) was used to measure covariance. All KMO are above 0.7 (Hair et al., 2010).

The second type of multivariate methods included two nonparametric methods:

1. Nonparametric analysis of variance: One-way analysis of variance using the Kruskal and Wallis test (Kerlinger & Lee, 2000). This method was performed to see if differences between the four types of projects are globally significant.

2. Mann-Withney test: Two by two nonparametric comparisons between groups to locate where the significant differences are.

4. Results

In this section, we present the results of the two complementary statistical methods applied: PCA and non-parametric one-way ANOVA, and nonparametric comparison between two groups.

4.1Results from Factor Analysis

Three different factor analyses were performed for the purpose of summarization and reduction: functions, embeddedness, and project performance.

4.1.1 Groups of Functions

Two different sets of questions were asked regarding the 27 functions: the importance and the degree of realization. Several factor analysis strategies were performed using these two sets of answers. In our judgment, the model that better fits the context of this research is based on the question of the degree of realization (instead of importance). However, five functions had to be excluded from this analysis for three reasons: (1) some functions did not load into a factor; (2) some functions loaded into more than one factor; and (3) some functions loaded into a factor making interpretation much more difficult. Ultimately, 22 functions loaded into five factors, each comprising a group of functions, as shown in Table 3.

The first group labeled project performance and portfolio management brings together the monitoring and controlling of a single project within other functions associated with the management of a portfolio of projects. In the public context, it seems that there is no locus for individual project monitoring and control outside portfolio management. This observation is complemented by the exclusion of two orphan functions: implement and operate a project information system, and develop and maintain a project scorecard. Interpretation of this result put into question the quality of the monitoring and control of individual projects in public administration. Alternately, the inclusion of benefits management aligns with a view of portfolio management as a mean to deliver expected benefits provided in a strategic process (Dinsmore & Cooke-Davies, 2006).

The second group of functions, methodologies, and competencies is quite coherent with previous models (e.g., Hobbs & Aubry, 2007), where two different types of activities are loaded together: activities related to standardization, and activities related to the development of project management competencies. The combination of both makes sense, as training people at methodology and standards will certainly lead to better use of these. Noteworthy is that networking and environmental scanning was loaded in this group. It is quite interesting that the public administration includes openness to external views which can incorporate innovation into more conventional approaches to projects. In this particular public setting, there is a strong community of practice dedicated to project management, where practices are constantly discussed between practitioners and where researchers are invited to present their results. One function has been excluded and considered an orphan: provide mentoring for project managers. Mentoring might not be an active way of developing competences in this public setting.

The third group is organizational learning where all functions related to knowledge were loaded with the exception of manage archives of project documentation. This result is surprising, as it does not contribute with the other functions to organizational learning. Archives are associated with explicit knowledge and might not have any value once the project is finished; it might more likely be associated with very technical documentation work. Globally, this group of functions includes lessons learned, which might be the most important reusable knowledge for next projects.

The fourth group, collaboration and communication, comprises relations with upper levels and stakeholders. This is very interesting, as it highlights the role these entities are playing outside project and portfolio management. This can be seen as strength to partnerships and to more active integration (Artto et al., 2011). Interestingly, the function report project status to upper management is part of this group. This can be interpreted to indicate that reporting is more likely a means of communication than project control.

The fifth group is specialized tasks. It includes the manage archives of project documentation function.

Group 1 Project performance and portfolio mgmt. Group 2 Methodologies and competencies Group 3 Organizational Learning Group 4 Collaboration and communication) Group 5 Specialized tasks
Identify, select, and prioritize new projects 0.809 0.032 0.041 0.199 0.202
Manage one or more project portfolios 0.709 0.378 -0.096 0.189 0.152
Coordinate between projects 0.724 0.134 0.090 0.322 0.168
Benefits management 0.711 -0.014 0.437 0.171 -0.038
Allocate resources between projects 0.726 0.242 0.055 0.308 0.020
Monitoring of project 0.658 0.288 0.327 -0.001 0.352
performance
Controlling of project 0.639 0.093 0.405 0.068 0419
performance
Develop and implement a 0.227 0.650 0.329 0.149 0.330
standard methodology
Provide a set of tools without 0.173 0.791 0.001 0.017 0.251
an effort to standardize
Develop competency of 0.128 0.663 0.230 0.135 0.172
personnel, including training
Promote project management 0.087 0.724 0.125 0.406 -0.025
within organization
Networking and environmental 0.122 0.684 0.316 0.087 -0.010
scanning
Conduct project audits 0.200 0.152 0.734 0.301 -0.109
Conduct post-project reviews 0.462 0.089 0.574 -0.028 0.358
Implement and manage 0.080 0.456 0.680 0.111 0.090
database of lessons learned
Implement and manage risk -0.007 0.304 0.724 0.038 0.197
database
Report project status to upper 0.379 0.257 0.006 0.665 0.276
management
Provide advice to upper 0.286 0.288 0.229 0.672 0.026
management
Manage stakeholders 0.272 0171 0.083 0.696 0.470
interfaces
Monitoring and controlling 0.246 -0.015 0.410 0.553 0.357
performance of PMO
Execute specialized tasks for 0.167 0.214 0.034 0.198 0.763
project managers
Manage archives of project documentation 0.293 0.214 0.195 0.318 0.699
KMO 0.868
% Variance 20.030 15.427 13. 100 11.144 10.355
% Cumulative variance 20.030 35.458 48.557 59.701 70.056
Cronbach’s alpha (Q8) 0.901 0.846 0.802 0.842 0.716
Cronbach’s alpha (Q7) 0.874 0.816 0.794 0.798 0.661
Note. Factor loadings >0.50
appear in boldface type.

Table 3: Groups of PMO gunctions

Globally, the factor analysis provides five groups from 22 functions. The five functions that have been excluded from the factor analysis are considered orphans. They play important, but individual functions:

  • Develop and maintain a project scorecard
  • Implement and operate a project information system
  • Provide mentoring for project managers
  • Participate in strategic planning
  • Recruit, select, evaluate, and determine salaries for project managers
4.1.2 Factor Embeddedness

The questionnaire asks for six different aspects of embeddedness. As a reminder, initial research on this construct has produced interesting results on explaining the role of a PMO on the perception of project performance (Hobbs & Aubry, 2011). In searching for a model using the factor analysis method, six variables were loaded together in a single factor labeled embeddedness, as shown in Table 4. In this particular case, the factor analysis method provides a single factor for the reduction of six variables. Together, they help generate an understanding of the different facets of what anchors an entity dealing with multiple projects within the organization.

Factor
Embeddedness
PMO mission is well understood 0.699
PMO mission is aligned with organization’s mission 0.771
PMO work in close collaboration with others in projects 0.678
Expertise is recognized by others 0.718
PMO is supported by executives 0.635
PMO collaborates to the organization’s future vision 0.502
KMO 0.725
% Variance 45205
% Cumulative variance 45.205
Cronbach’s alpha
Note. Factor loadings >0.50 appear in boldface type.
0.753

Table 4: Components of embeddedness

4.1.3 Project Management Performance

Questions were asked on eight variables associated with the evaluation of project management performance. One variable had to be taken out in the process of factor analysis that is upper level satisfaction. The other seven variables loaded in two factors, as shown in Table 5.

Factor 1 Satisfaction with the Project Factor 2 Project Management Performance
Team satisfaction 0.832 0.159
User satisfaction 0.861 0.278
Client satisfaction 0.779 0.336
Stakeholder satisfaction 0.789 0.257
Respect of schedule 0.364 0.814
Respect of cost 0.109 0.924
Sponsor satisfaction 0.447 0.622
KMO 0.846
% Variance 42.957 31.221
% Cumulative variance 42.957 74.178
Cronbach’s alpha 0.880 0.808
      Note. Factor loadings >0.50 appear in boldface type.

Table 5: Factors of project performance

The first factor is labeled satisfaction with the project, and it includes the large variety of actors in and around a project: team, user, client, and more generally, the stakeholders. However, the variable sponsor satisfaction was loaded in the second factor (project management performance). Interpretation of this result suggests that the project sponsor will be satisfied as long as the project management performance, expressed by the more usual indicators of time and cost, is achieved.

4.2 Results from the Comment Fields

In the questionnaire, three questions were asked with free text answers to capture more qualitative interpretations on the impact of the entity:

  • Strengths associated with the entities (142 comments)
  • What should be improved (131 comments)
  • The most important obstacle in the implementation of the entity (82 comments)

A first classification of all comments has been done using a common initial framework for all three questions, as they refer—positively or negatively—to similar themes. Figure 2 illustrates global results. Maturity in project management and in change management is the most cited theme in terms of obstacles and needs for improvement in entities. Respondents complain about the lack of maturity in these terms:

  • Difficulty with the cultural change and the new philosophy of management
  • Change in work organization and resistance to this change
  • Lack of common vision between departments
  • Work in silo

Competencies in project management occupy the second place with the particularity of having the most number of strong points. This is coherent with the results obtained on embeddesness, where expertise is recognized by others.

Obstacles, strengths, and need for improvement in entities

Figure 2: Obstacles, strengths, and need for improvement in entities

5. Findings

The aim of this paper is to propose a typology of entities based on the types of projects they deal with. This requires differentiating among the entities based on their organizational characteristics, structural characteristics, and/or on what they are doing. For all variables, the non-parametric Kruskal-Wallis (K-W) method was used, followed by two by two non-parametric comparisons between the types of projects using the Mann-Whitney (M-W) method. For each of these methods, the usual rules of the p value apply. Table 12 only shows the variables where a significant difference was found with the K-W method. In the interpretation of the results, no difference is also a result and is discussed below when appropriate.

We can easily observe by counting the number of significant differences in each of the six combinations of project types that entities dealing with engineering and construction projects have many different structural characteristics and perform different activities than those dealing with business processes projects (1 versus 3). The least differentiation was found among entities dealing with new product/service development (NPD) projects compared to information system/information technology (IS/IT) projects (2 versus 4) and to business process (BP) projects (3 versus 4).

Each of the following sub-sections emphasizes one aspect of differences among entities dealing with different types of projects. In these sub-sections, a summary table is provided illustrating the major differences using the different sets of rules specified in each particular table. Following Besner and Hobbs (2013), the number of significant differences in pair-wise comparisons are shown in a specific column for each project type (p≤ 0.1). This approach makes it simplest to visualize variables having the best potential to differentiate between types, given six pair-wise comparisons per variable.

5.1Project Types and Organizational Context

Two classes of data are used to describe the organizational context in which entities exist: organizational characteristics and types of projects. For organizational characteristics, three variables display significant differences between types of projects. First, organizational project management maturity shows higher results in engineering and construction projects than the three other types. Significant differences happen with business process and new products and services development. Overall, however, project management maturity levels are almost the same as in previous comparable studies of PMOs (Hobbs & Aubry, 2010).Second, a higher percentage of projects with external clients are based in entities dealing with engineering and construction projects and business process projects. Conversely, a lower percentage of external clients is more likely in entities dealing with IS/IT projects and new product development projects. Differences are significant between these two groups regarding the percentage of external clients. Not surprisingly, IS/IT projects are often performed by internal resources as a service function of organization, contrarily to engineering and construction, where projects are often carried out by contractors and sub-contractors. Public administrations are most likely to develop services (not products) through their mission of serving a population. Expertise in these services is internal to the departments participating in projects. Third, the level of organizational cultural support for project management is significantly lower for business process projects than other types of projects. This could be interpreted to show that business process projects happen in entities newer to the project management approach than engineering and construction and IS/IT projects. Combined, these three variables pertaining to the organizational characteristics ensure good anchorage for differentiating among entities dealing with different types of projects. Table 6 summarizes the results on organizational context.

Organizational Characteristics E & C IS/IT BP NPD
Organizational project management maturity 2.571 2+ 2.341 2.077 1- 2.067 1-
Percentage of external clients 43% 2+ 13% 2- 48% 2+ 13% 2-
Organizational culture support level of project management 3.524 1+ 3.667 2+ 2.579 3- 3.333 1+
Projects characteristics
Scope of project in terms of cost (thousands of CAD) 20 607 3+ 1 964 1+/1- 4,420 1- 500 2-
Scope in terms of duration (months) 20 3+ 14 1- 14 1- 13 1-
Communication in post-delivery activities 25% 2- 48% 1+/1- 42% 1- 80% 3+
Marketing in post-delivery activities 11% 1- 2% 1- 4% 33% 2+
Organizational change in post-delivery activities 14% 3- 50% 1+ 50% 1+ 47% 1+

Table 6: Differences in organizational context between entities eased on types of projects

When considering the characteristics of projects conducted by the entities, the scope of projects both in terms of costs and duration significantly differentiates engineering projects from the three other types, with projects having a longer duration for higher costs. This result is quite obvious; engineering and construction projects usually involve large investments in materials and equipment. The costs of the three other types of projects largely relate to human resources. The scope of projects may include post-delivery activities. Three of these activities involve significant differences in project types. Eighty percent of new product and service projects include communication activities, which significantly differentiate it from the three other types of projects. This result is quite natural as new product and service has to be communicated through the public administration and/or the targeted population.

On the other hand, only 25% of engineering and construction projects include such activities. This could be interpreted as an indication that communication is undertaken by another entity or by the client when external. The same also applies to marketing activities where new product and service development projects differentiate from engineering and construction and IS/IT projects for about the same reason. Very interestingly, engineering and construction projects differentiate from the other three types for not including organizational change management within their scope. The reason for this could relate to the nature of deliverables in the public sector where IS/IT, processes and services projects are more likely to have an impact either on the distribution of services, or internally, in the way the work is done (Crawford, Costello, Pollack, & Bentley, 2003). Accordingly, change management is closely linked to project success (Crawford et al., 2003). About half of these projects include management, which is consistent with previous results in overall sectors (Hobbs & Aubry, 2010).

5.2Project Types and Entity Structural Characteristics

This sub-section explores the differences between entities dealing with different types of projects and their structural characteristics (see Table 7).

Entity Structural Characteristics E & C IS/IT BP NPD
Percent of entities having 5 years and more 48% 2+ 25% 1- 13% 1- 33%
Number of projects simultaneously 161 3+ 43 1- 41 1- 24 1-
Percent of the projects in the entity compared to the overall number of projects in the organization 76% 2+ 68% 2+ 53% 2- 50% 2-
More than 75% of project managers of the organization in the entity 50% 2+ 28% 1+/1- 4% 2- 27%
Profile of personal
Percentage of project managers with master’s degree in project management in the entity 8.4 1- 17,6 2+ 0.2 1- 9.5
Experience in project management of the personal in the entity 7.8 3+ 5.9 1- 4.6 1- 5.1 1-
Practices/tools/methodology
Percentage of home-grown project management methodology 64% 2+ 42% 1+/1- 22% 2- 52%
Role of the entity
Percentage of entities involved in project management 54% 3- 95% 2+ 83% 1+/1- 87% 1+
Percentage of entities involved in monitoring projects 96% 1+ 93% 1+ 78% 2- 100%
Decision-making authority of the entity
Percentage of entities with no authority at all 21% 1- 42% 67% 1+ 44%
Percentage of entities with strong authority 50% 1+ 38% 1+ 0% 3- 22% 1+
Variables NOT included for differentiation
With the exclusion of project managers, number of persons working in the entity Percentage of entities involved in controlling projects

Table 7: Differences in entity structural characteristics based on project types

Considering the percentage of entities having five years or more, of interest in these results pertaining specifically to the public sector is that the entities do not change as frequently, with a mean of nearly 30% for all types among entities having a life span of five years or more. Previous research found 15% for PMOs with five years and more (Hobbs & Aubry, 2010). Recent research has provided some understanding of the reasons why PMOs are changing so frequently and of the change process (Aubry et al., 2011). This result could be interpreted as an indication entities in this research are focused on part of the public administration bureaucratic hierarchy. Political positions aside, this hierarchy is quite stable, even though the governing parties change (Kernaghan, 2005).

However, the variable of entities having five years and more is contributing in differentiation between entities. Entities dealing with engineering and construction projects have greater longevity than the ones dealing with IS/IT and business process projects with highest percentage of entities having five years and more. The number of simultaneous projects significantly differentiates engineering and construction projects from the three others with a high number.

The number of simultaneous projects within the entity distinguishes entities dealing with engineering and construction projects by the high number of projects. This result is to be considered along with the scope of projects, in which engineering and construction projects have highest costs and longer duration.

The percentage of projects and project managers in the entity was used to propose a typology of PMO (Hobbs & Aubry, 2008). In the public sector, this typology was not validated. However, these two variables differentiate entities dealing with different types of project. Entities with engineering and construction projects and IS/IT projects both differ significantly from entities involved with business processes and new product and service development projects. The former are more likely to have more projects than the latter. Having a large number of project managers differentiates entities engaged in engineering and construction projects, where 50% have a large proportion of project managers. Conversely, very few entities with business process projects have such a large number of project managers.

The profile of project managers has been assessed by a high level of education in project management and the extent of their experience in project management. The percentage of project managers having a master’s degree in project management significantly differentiates entities with IS/IT projects as those having the highest percentage. Project manager experience is significantly greater in entities conducting engineering and construction projects. Globally, however, all entities have experienced project management personnel.

In terms of methodology, developing a home-grown methodology differentiates the entities, and those involved in engineering and construction projects are more likely to develop their own. Entities dealing with new product and services projects also post high levels, but do not significantly differ from the others.

Two different roles differentiate the entities: project management and project monitoring. Entities engaged in engineering and construction projects perform the least direct project management compared to the three other types of entities. Conversely, IS/IT entities are more likely to engage in project management compared to engineering and construction projects and business process projects. Interestingly, the role of controlling projects does not differ significantly among the entities. In terms of decision-making authority, entities with less authority are more likely to be engaged in business process projects while those with the highest level of authority are found within engineering and construction projects. It is also worth mentioning that the number of people working in the entities (with the exclusion of project managers) does not differ significantly among the entities.

5.3Project Types and Functions

Conversely, based on the findings of a previous PMO typology study (Hobbs & Aubry, 2011), the functions here seem to differentiate among entities dealing with different sets of projects. All five groups of functions show a significant difference (p ≤ 0.1) for entities dealing with different types of projects.

Upon examining descriptive statistics on the functions, all functions follow about the same distribution pattern: functions in group 4, collaboration and communication, are the most frequently performed by all entities while in group 3, organizational learning is the least frequent. The lowest importance of organizational learning confirms previous results showing a lack of real motivation among organizations in all sectors to invest in knowledge and learning. But what is astonishing about these public sector results is the strength of the collaboration and communication function for all types of projects. This group of functions includes reporting to and advising upper management. In the public sector, it might be the rule of paying careful attention to political relationships, as politicians are major stakeholders. This finding challenged previous results where project performance monitoring and control was typically the major function of a PMO.

Now, turning to a statistical analysis of the differences in what entities do differently in dealing with different types of projects, Table 8 shows that entities involved in new product and service development do not differentiate significantly from the others. Yet, in our interpretation, we emphasize the three other types of projects. Table 8 presents a summary of the significant differences between entities and the types of projects they deal with.

Functions Engineering and construction IS/IT Business processes NPD
Group 1 Project performance and portfolio management 3.593 1+ 3.390 1+ 2.696 2- 3.163
Group 2 Methodologies and competencies 3.438 2+ 3.037 1-/1+ 2.500 2- 3.071
Group 3 Organizational learning 2.723 2+ 2.248 1-/1+ 1.667 2- 2.054 1-
Group 4 Collaboration and communication 4.134 3+ 3.576 1-/2++ 3.167 2- 3.143 2-
Group 5 Specialized tasks 3.446 1 + 3.568 1 + 2.900 2- 3.143

Table 8: Differences between entities based on types of projects

The strongest function differences (all five with p ≤ 0.07) occur between business process projects on the one hand, and both engineering and construction and IS/IT projects on the other. These entities do not perform the functions with the same intensity. Entities dealing with business processes perform significantly lower on all functions compared to engineering and construction projects and IS/IT projects. This is particularly evident for the collaboration and communication function.

Organizational knowledge management is the least frequent function, but still seems to differentiate between project types. In fact, it differentiates between engineering and construction project types which have the highest score, and the three other types. It also shows a significant difference between IS/IT and Business process types, the latter having the lowest score.

Significant differences exist for the methodology and competencies function between engineering and construction type on the one hand and IS/IT and business process types on the other. In this function, IS/IT projects type also differs significantly from business process type. A significant difference is also evident between business process project type and new product and services development. These results are consistent with project management history, given that the engineering and construction industries were among the first sectors to embark on project management practices back in the 1950s (Morris, 2011). More recently, the IS/IT projects within different sectors have shown a strong tendency to adopt project management practices (Rivard & Dupré, 2009). Business process projects, such as IS/IT, cross different sectors. After the business process reengineering era and, more recently, lean projects, it seems that neither methodologies nor competencies are very strong.

5.4Project Types and Performance

Three approaches have been taken to understand performance differences among entities dealing with different types of projects:

  • Project management performance factors
  • Project management success criteria
  • Factor of embeddedness

Table 9 summarizes performance results. Performance in this context is taken as a descriptive, not prescriptive concept.

Performance E& C IS/IT BP NPD
Factors of project
management performance
Satisfaction with the project 3.545 1- 3.921 1- 3.764 1- 4,286 3+
Project management 1 + 1+ 3- 1+
performance 4.188 4.095 3.663 4,232
Most important criteria for
project success
% of entities: Respect of budget 32.1% 1 + 17.1% 1 + 13.6% 0% 2-
% of entities: Satisfaction of 1- 1- 2+
stakeholders 10.7% 12.2% 36.4% 26,7%
Factor of embeddedness
Embeddedness of the entity 4.295 1 + 4.134 1 + 3.812 3- 4,383 1 +
Variables NOT included for
differentiation
Criteria for project performance -
Respect of dates
Criteria for project performance -
Respect of technical specifications
Criteria for project performance -
Quality
Criteria for project performance -
Final clients

Table 9: Summary of performance differences among entities

All three perspectives reveal significant differences in the analysis of the four types of projects. The first performance perspective is based upon the factors of project management performance. Differences in the means are not extremely high, but show significant differences among the types of projects. First, satisfaction with the project differentiates entities with new product/service development from the three others with a higher score on satisfaction with the project. However, developing new products/services in a government setting certainly demands multiple perspectives in a complex environment (Hobday, 2000; Kujala, Putila, & Brady, 2013). Second, project management performance also shows a relatively high score. Significant differences exist among entities with business process projects and the three others, where the lower score applies to the project management performance factor. Results show that these entities assign less importance to factors leading to better project management performance.

The second performance perspective comes from the importance placed by the entities on the project success criteria, with two variables differing significantly among project types. First, respect for budget shows a relatively low percentage with a zero score for new product/services development projects. This result is surprising, but indicates what is perceived as most valued for the success of projects. These entities have significant differences with engineering and construction projects and with IS/IT projects. Entities dealing with new product/service development are more likely to focus their attention on stakeholders as a performance factor than budget. Second, stakeholder satisfaction as the most important criteria for success differs significantly for business process projects compared to engineering and construction projects and IS/IT projects. Overall, in these two latter project types, stakeholder satisfaction does not seem to be highly valued in project management performance and project success.

The third and last perspective is the factor of embeddedness. Globally, this factor shows quite a high level of embeddedness for all types of projects: entities are well embedded in their organization. In this particular context, the entity has a permanent status, being a functional unit not necessarily dedicated uniquely to projects. This context might partly explain why the entity’s mandate is well understood and aligned with the overall mission, and why expertise is recognised. These entities also play a role in the close collaboration with others in projects as well as to the organization’s future vision. However, care should be taken to avoid cognitive bias in this interpretation, or the halo effect. In our study, this bias could have affected the interpretation of the entity in a positive way. Most of the questionnaire respondents are in management positions within the entities. They might overestimate their level of embeddedness. However, statistical analysis shows significant differences between entities. Entities dealing with business process projects post a lower level of embeddedness than the three other ones.

Globally, the three perspectives on performance in relation to entities dealing with multiple projects contribute to the specification of such entities based on the types of projects they deal with. Results for business process projects and new product/service development projects show the most pronounced differences for project management performance and embeddedness factors.

6. Discussion: Proposed PMO Typology Based on Project Types

In this research, we have adopted a very generic approach to the definition of a PMO as an entity dealing with multiple projects. This approach has enabled us to capture the multiple loci of project management activities in a public administration. We have proposed a typology of these entities based on four types of projects. At this stage, we suggest reinstating the term PMO, as it is part of common project management language.

Table 10 presents a synoptic view of the variables from the previous statistical analysis having the potential to differentiate among PMOs dealing with different types of projects. Variables having a stronger power of differentiation, with a maximum of three significant differences, appear in bold.

These empirical results describe for the first time, to our knowledge, the reality of a public administration facing the challenges of managing multiple types of projects. These descriptive results are intended to provide a better understanding of the specific context of public administration. First and foremost, they show the diversity of PMOs dealing with different types of projects. There is no such thing as a unique type of PMO in public administration, as has been shown for different types of organizations (Hobbs & Aubry, 2007). This descriptive typology is of prime importance for this government in setting an organizational design to support the various needs for different project types. For instance, more services to support organizational change could be made available to PMOs dealing with engineering and construction projects because results show very little in project characteristic regarding organizational change in post-delivery activity. Regarding performance, some efforts should be made for business process PMOs where project management performance has less importance as compared with the other three types of projects.

This research provides empirical evidence of the interest in organizational design to counteract the three reasons given by Greewood and Miller (2010) to explain the lack of interest in organizational design. First, the unit of analysis of this research is at the organizational level rather than the population or community level. Second, a government organization is clearly a complex system that faces a number of difficulties in trying to put organizational project management in place. A multi-method methodology would be required to capture such complexity. In this particular case in public administration, a variety of actions have been undertaken, including this quantitative research making only a partial contribution. That said, this piece of research provides a picture of the situation by showing explicitly 1) who is involved in managing multiple projects throughout the government apparatus; 2) what they are doing; and 3) how sensitive they are to project management performance. Third, the focus of this research is on the overall entities in the government that deal with multiple projects. It avoids a narrow view centered on any one part in particular to the detriment of the overall organizational design. In proposing a typology of PMOs in the public sector, this paper’s main contribution to the academic field is to further knowledge about organizational design. It also demonstrates, among other scholars, that project management research can contribute significantly to the management field.

From a methodological point of view, it also illustrates an academic contribution towards elucidating a complex situation. This paper, within the limit of a quantitative methodology, is very much aligned with the reflection undertaken towards the project management scholars to open up our methodological approaches to translational and transformational approaches (e.g., Drouin et al., 2013).

The main contribution of this paper for professionals is two-fold. First, it provides an empirically validated typology of PMOs in the public sector. This typology is based on four project types: engineering and construction, IS/IT, business processes, and new product/service development.

Second, and over and above the strict design of a PMO, it shows how a government can use research as part of its process to answer the difficult question of how to organize for project management in public administration. Given that there are no models that can simply be copied and pasted in any context, work has to be done to construct a local PMO solution. However, such a solution might be only temporary, as previous research has shown (Aubry et al., 2011). To be prepared for change means being aware of the environment in order to anticipate changing needs so that projects can be carried out successfully.

Table 10: PMO typology based on project types

EC IS/IT BP NPD
MORE OF:
Organizational characteristics
Organizational project management maturity % of external clients
Cultural support
Cultural support % of external clients
Project characteristics Scope of projects in terms of cost
Scope of projects in terms of duration
Organizational change in post-delivery activities Organizational change in postdelivery activities Communication in post-delivery activities
Marketing in post-delivery activities
Organizational change in postdelivery activities
Structural characteristics of the entity % of PMOs having +5 years
Number of projects simultaneously
% of projects in the entity
More than 75% of project managers
Experienced personal in project management
% of home grown methodology
% of entities with strong authority
% of PMOs involved in monitoring projects
% of projects in the entity
% of entities with strong authority
% of project managers with a master’s degree
% of PMOs involved in project management
% of PMOs involved in monitoring projects
% of entities with strong authority
% of PMOs involved in project management
Functions Group 1 project performance and portfolio management
Group 2 Methodologies and competencies
Group 3 Organizational learning
Group 4 Collaboration and communication
Group 5 Specialized tasks
Group 1 project performance and portfolio management
Group 4 Collaboration and communication
Group 5 Specialized tasks
Performance Project management performance
Respect of budget
Embeddedness of the entity
Project management performance
Respect of budget
Embeddedness of the entity
Legitimacy
Satisfaction of stakeholders
Satisfaction with the project
Project management performance
Embeddedness of the entity
LESS OF
Organizational characteristics
Organizational project management maturity
Cultural support
Organizational project management maturity
Project characteristics Communication in post-delivery activities
Marketing in post-delivery activities
Organizational change in post-delivery activities
Scope of projects in terms of duration
Marketing in post-delivery activities
Scope of projects in terms of cost
Scope of projects in terms of duration
Communication in post-delivery activities
Scope of projects in terms of cost
Scope of projects in terms of duration
Structural characteristics of PMO
% of project managers with a master degree
% of PMOs involved in project management
% of PMOs having +5 years
Number of projects simultaneously
Experienced personal in project management
% of PMOs having +5 years
Number of projects simultaneously
% of projects in the entity
More than 75% of project managers
% of entities with strong authority
% of project managers with a master’s degree
Experienced personnel in project management
% of home grown methodology
% of PMOs involved in monitoring projects
Number of projects simultaneously
% of projects in the entity
Experienced personal in project management
Functions Group 1 project performance and portfolio management
Group 2 Methodologies and competencies
Group 3 Organizational learning
Group 4 Collaboration and communication
Group 5 Specialized tasks
Group 3 Organizational learning
Group 4 Collaboration and communication
Performance Legitimacy
Satisfaction with the project
Satisfaction of stakeholders
Satisfaction with the project
Satisfaction of stakeholders
Satisfaction with the project
Project management performance
Embeddedness of the entity
Respect of budget
Strengths Competencies in project management Competencies in project management Competencies in project management Competencies in project management
Obstacles Maturity in project management and change management Organizational engagement towards project management Maturity in project management and change management Maturity in project management and change management
To be improved Standardization of project management methods
Maturity in project management and change management
Organizational engagement towards project management
Maturity in project management and change management
Maturity in project management and change management Maturity in project management and change management
Competencies in project management

7. Conclusion

This paper has presented the findings of a quantitative research that took place in a government. It proposed a typology of PMO based upon four project types: engineering and construction, information system and technology, business processes, and new product/service development.

An important limitation of this research is the lack of context due to the quantitative approach. While statistical analysis provides sound differences between project types, it misses rich data to fully interpret the results and provide more generalization to other similar public administrations. In the same vein, having data from a single government imposes limits in terms of generalization. However, we believe that making sense of these findings in any context will help in finding inspiration for a proper solution in terms of organizational design. Moreover, the method employed could be replicated in other governments or industries to complement the results.

This research opens up on several opportunities for future research. First, as mentioned in the introduction, scholars of project management should diffuse more of their work on organizational design through the main stream of organization literature. There is a gap in the literature on management and organization design when time comes to understand new forms of organizing. Project management literature has some answers to offer. Second, the focus of research on organizational design should move to explore the process of organizing for projects, in other words moving from the project-organization to project organizing. Third, this would benefit from the practice theory such as proposed by Nicolini (2013) with a proposed toolkit to tackle together practices and theory. Actually, not that much is known on this process in a practice view.

Acknowledgments

The researchers wish to warmly thank the participants who gave of their time to complete the questionnaire. We would also like to thank the government for its financial support for this research.

Appendix

Table 11: Comparison between the two data collection phases

Engineering and Construction
First Data Collection
Second Data Collection
M-W
p-value
Variables Mean Standard Deviation Mean Standard Deviation Two-tail
No of employees in government departments and agencies 2.835 2.120 1.588 1.891 0.0590
Project management maturity level 2.7 0.8 2.4 0.7 0.5680
Number of people working on a typical project 10 11 10 10 0.3970
What is the duration of a typical project (month) 22.3 3.8 18.0 8.1 0.1740
Percentage of entities delivering projects to external clients 0.5 0.5 0.3 0.5 0.4780
Project management experience of personnel in the entity (year) 7.3 3.6 8.5 2.4 0.5070

Table 12: Non-parametric one-way ANOVA and non-parametric comparisons

Variables 1. Engineering and construction N1=28 Mean 2. IS/IR N2=44 Mean 3. Business process N3=26 Mean 4. NPD N4=15 Mean K-W p-value two-tail M-W p-value two-tail M-W p-value two-tail M-W p-value two-tail M-W p-value two-tail M-W p-value two-tail M-W p-value two-tail
ORGANIZATIONAL CONTEXT Characteristics of the organization Organizational project management maturity 2,571 2,341 2,077 2,067 0,0887 1 VS 2 0,2337 1 VS 3 0,0151 1 VS 4 0,0720 2 VS 3 0,1711 2 VS 4 0,3861 3 VS 4 0,7990
Percentage of entities dealing mostly with external proje 43% 13% 48% 13% 0,0047 0,0057 0,7427 0,0516 0,0032 0,9603 0,0830
Organizational culture support level of project managem 3,524 3 ,6 67 2,579 3,333 0,0123 0,5749 0,0100 0,6570 0,0025 0,4190 0,0760
Characteristics of projects in the entity mandate Scope of project in terms of cost (thousands of CAD) 20 607 1 964 4 420 500 0,0000 0,0000 0,0001 0,0000 0,7159 0,0415 0,2370
Scope in terms of duration (months) 20 14 14 13 0,0001 0,0000 0,0004 0,0022 0,6414 0,4708 0,5310
Percentage of entities including communication within t 25 % 48% 42% 80% 0,0072 0,0555 0,1817 0,0006 0,6624 0,0311 0,0460
Percentage of entities including marketing within the sc 11% 2% 4% 33% 0,0025 0,1301 0,3401 0,0726 0,7046 0,0007 0,1210
Percentage of entities including organizational change i 14% 50% 50% 47% 0,0133 0,0023 0,0052 0,0219 1,0000 0,8250 0,8620
ENTITY STRUCTURAL CHARCTERISTICS Percentage of entities having 5 years and more 48% 25% 13% 33% 0,0462 0,0469 0,0087 0,3584 0,2566 0,5342 0,3000
Number of projects simultenously 161 43 41 24 0,0020 0,0008 0,0050 0,0041 0,9611 0,8319 0,8980
Percentage of the projects in the entity compared to the 76% 68% 53% 50% 0,0222 0,2207 0,0150 0,0330 0,0573 0,0702 0,9640
Pourcentage of entities having more than 75% of projec 50% 28% 4% 27% 0,0040 0,0608 0,0003 0,1440 0,0194 0,9269 0,2460
Percentage of entities with strong authority 50% 38% 0% 22% 0,0048 0,4163 0,0005 0,1568 0,0031 0,3833 0,0414
Profile of personal Percentage of project managers with master degree in 8,4 17,6 0,2 9,5 0,0150 0,0905 0,1234 0,7049 0,0021 0,3062 0,2770
Experience in project management of the personal in th 7,8 5,9 4,6 5,1 0,0102 0,0113 0,0054 0,0163 0,1841 0,4450 0,7270
Practices/tools/methodology Percentage of home-grown project management methc 64% 42% 22% 52% 0,0045 0,0472 0,0007 0,3495 0,0168 0,6061 0,0520
Role of the entity Percentage of entities involved in project management 54% 95% 83% 87% 0,0003 0,0000 0,0302 0,0320 0,0951 0,2688 0,8360
Percentage of entities involved in monitoring projects 96% 93% 78% 100% 0,0547 0,5312 0,0473 0,4642 0,0892 0,2918 0,2730
FUNCTIONS REALIZED IN THE ENTITY Group 1 Project performance and portfolio managemen 3,593 3,390 2,696 3,163 0,0017 0,3485 0,0007 0,1814 0,0006 0,3921 0,1589
Group 2 Methodologies and competencies 3,438 3,037 2,500 3,071 0,0022 0,0529 0,0004 0,2446 0,0078 0,8015 0,0593
Group 3 Organisational Knowledge 2,723 2,248 1,667 2,054 0,0010 0,0412 0,0003 0,0606 0,0032 0,4933 0,2347
Group 4 Collaboration and communication 4,134 3,576 3,167 3,143 0,0002 0,0025 0,0002 0,0013 0,0695 0,0689 0,8435
Group 5 Specialised tasks 3,446 3,568 2,900 3,143 0,0739 0,6391 0,0603 0,3870 0,0132 0,2102 0,5252
PERFORMANCE OF THE ENTITY Factors of project management performance Stakeholders satisfaction 3,545 3,921 3,764 4,286 0,0760 0,1900 0,4723 0,0269 0,6915 0,0683 0,0239
Project management performance 4,188 4,095 3,663 4,232 0,0137 0,7179 0,0265 0,5793 0,0124 0,3036 0,0038
Most important criteria for project success Percentage of entities: Respect of the budget 32% 17% 14% 0% 0,0616 0,1482 0,1322 0,0147 0,7241 0,0900 0,1411
Percentage of entities : Satisfaction of stakeholders 11% 12% 36% 27% 0,0620 0,8514 0,0314 0,1820 0,0250 0,1956 0,5417
Factor of embededdness Embeddedness of the entity 4,295 4,134 3,812 4,383 0,0147 0,2939 0,0052 0,5861 0,0817 0,1558 0,0030
Total number of variables with significant differences 15 22 15 17 7 9

References

Alvesson, M. (2013). Do we have something to say? From re-search to roi-search and back again. Organization, 20(1), 79–90.

Alvesson, M., & Sandberg, J. (2011). Generating research questions through problematization. Academy of Management Review, 36(2), 247–271.

Artto, K., Kulvik, I., Poskela, J., & Turkulainen, V. (2011). The integrative role of the project management office in the front end of innovation. International Journal of Project Management, 29(4), 408–421.

Artto, K. A., & Wikstrom, K. (2005). What is project business? International Journal of Project Management, 23(5), 343–353.

Aubry, M., Hobbs, B., Müller, R., & Blomquist, T. (2011). Identifying the forces driving the frequent changes in PMOs. Newtown Square,PA: Project Management Institute.

Australia Government. (2012). Managing major projects (pp. 76).

Besner, C., & Hobbs, B. (2013). Contextualized project management practice: A cluster analysis of practices and best practices. Project Management Journal, 44(1), 17–34.

Cicmil, S., Williams, T., Thomas, J., & Hodgson, D. (2006). Rethinking project management: Researching the actuality of projects. International Journal of Project Management, 24(8), 675–686.

Crawford, L., Costello, K., Pollack, J., & Bentley, L. (2003). Managing soft change projects in the public sector. International Journal of Project Management, 21(6), 443–448.

Crawford, L., Hobbs, B., & Turner, R. J. (2005). Project categorization systems: Aligning capability with strategy for better results. Newtown Square, PA: Project Management Institute.

Cronbach, L. J. (1951). Coefficient alpha and the internal structure of tests. Psychometrika, 16(3), 297–334.

Dinsmore, P. C., & Cooke-Davies, T. J. (2006). Right projects done right!: From business strategy to successful project implementation. San Francisco, CA: Jossey-Bass.

Drouin, N., Müller, R., & Sankaran, S. (Eds.). (2013). Novel approaches to organizational project management research: Translational and transformational (Vol. 29). Copenhagen: Copenhagen Business School Press.

Geraldi, J. G. (2009). Reconciling order and chaos in multi-project firms. International Journal of Managing Projects in Business, 2(1), 149–158.

Greenwood, R., & Miller, D. (2010). Tackling design anew: Getting back to the heart of organizational theory. Academy of Management Perspectives, 24(4), 78–88.

Hair, J. F., Black, W. C., Babin, B. J., & Anderson, R. E. (2010). Multivariate Data Analysis (7th ed.). Upper Saddle River, NJ: Pearson Prentice Hall.

Hobbs, B., & Aubry, M. (2007). A multi-phase research program investigating project management offices (PMOs): The results of phase 1. Project Management Journal, 38(1), 74–86.

Hobbs, B., & Aubry, M. (2008). An empirically grounded search for a typology of project management offices. Project Management Journal, 39 (Special Issue).

Hobbs, B., & Aubry, M. (2010). The project management office: A quest for understanding. Newtown Square, PA: Project Management Institute.

Hobbs, B., & Aubry, M. (2011, June). A Typology of PMOs Derived Using Cluster Analysis and the Relationship with Performance. Paper presented at the IRNOP, Montreal, Canada.

Hobbs, B., Aubry, M., & Thuillier, D. (2008). The project management office as an organisational innovation. International Journal of Project Management, 26(5, Best Papers Special Issue at 2007 IRNOP Research Conference), 547–555.

Hobday, M. (2000). The project-based organisation: An ideal form for managing complex products and systems? Research Policy, 29(7–8), 871–893.

Kendall, G. I., & Rollins, S. C. (2003). Advanced project portfolio management and the PMO: Multiplying ROI at warp speed. Boca Raton, FL: J. Ross Publishing.

Kerlinger, F. N., & Lee, H. B. (2000). Foundations of behavioral research (4th Ed.). Belmont (CA): Cengage Learning.

Kernaghan, K. (2005). Changing concepts of power and responsibility in the Canadian public service. In B. W. Caroll, D. Siegel. & M. Sproule-Jones (Eds.), Classic readings in Canadian public administration (pp. 166–181). Toronto, CA: Oxford University Press.

Kujala, J., Putila, J., & Brady, T. (2013). Challenges for cost management in complex projects. Paper presented at the IRNOP XI, Oslo, Norway.

March, J. G. (1991). Exploration and exploitation in organizational learning. Organization Science, 2(1), 71–87.

Merriam-Webster, I. (2007). Merriam-Webster’s collegiate dictionary (11th ed.). Springfields, MA: Merriam-Webster.

Morris, P. W. G. (2011). A brief history of project management. In P. W. G. Morris, J. K. Pinto & J. Söderlund (Eds.), The Oxford handbook of project management (pp. 15–36). Oxford, UK: Oxford University Press.

Morris, P. W. G., & Geraldi, J. (2011). Managing the institutional context for projects. Project Management Journal, 42(6), 20–32.

Müller, R. (2011). Project governance. In P. W. G. Morris, J. K. Pinto & J. Söderlund (Eds.), The Oxford handbook of project management (pp. 297–320). Oxford: Oxford University Press.

Müller, R., Glückler, J., & Aubry, M. (2013). A relational typology of project management offices. Project Management Journal, 44(1), 59–76.

Nicolini, D. (2013). Practice theory, work, and organization: An introduction. Oxford: Oxford University Press.

O‘Reilly III, C. A., & Tushman, M. L. (2008). Ambidexterity as a dynamic capability: Resolving the innovator’s dilemma. Research in Organizational Behavior, 28, 185–206.

Office of Government Commerce. (2008). Portfolio, programme and project offices [P3O]. London, UK: The Stationary Office [TSO].

Osborne, S. P. (2000). The new public governance?: Emerging perspectives on the theory and practice of public governance. New York, NY: Routledge.

Pfeffer, J. (2013). You’re still the same: Why theories of power hold over time and across contexts. The Academy of Management Perspectives, 27(4), 269–280.

Project Management Institute. (2008). A guide to the project management body of knowledge (PMBOK® Guide) – Fourth edition. Newtown Square, PA: Author.

Project Management Institute. (2012). Organizational project management maturity model: OPM3 Knowledge Foundation (3rd ed.). Newtown Square, PA: Author.

Project Management Institute. (2013a). A guide to the project management body of knowledge (PMBOK® Guide) – Fifth edition. Newtown Square, PA: Author.

Project Management Institute. (2013b). PMI Reseach and Education Conference 2014: Standing on the Shoulders of Giants: In Search of Theory and Evidence. Retrieved 7 September 2013, http://copevents.pmi.org/rec-2014/home.

Rivard, S., & Dupré, R. (2009). Information systems project management in PMJ: A brief history. Project Management Journal, 40(4), 20–30.

Rousseau, D. M. (2012). Envisioning evidenced-based management. In D. M. Rousseau (Ed.), The Oxford handbook of evidenced-based management (pp. 3–24). Oxford, UK: Oxford University Press.

Samset, K., & Volden, G. H. (2013). Investing for impact: Lessons with the Norwegian state project model and the first investment projects (pp. 1–53). Oslo, Norway: Concept Research Program.

Shenhar, A. J., & Dvir, D. (2004). How projects differ, and what to do about it. In P. W. G. Morris & J. K. Pinto (Eds.), The Wiley guide to managing projects (pp. 1265–1286). Hoboken, NJ: John Wiley & Sons, Inc.

Shenhar, A. J., Dvir, D., Levy, O., & Maltz, A. C. (2001). Project Success: A multidimensional strategic concept. Long Range Planning, 34(6), 699–725.

Teddlie, C., & Tashakkori, A. (2009). Foundations of mixed methods research: Integrating quantitative and qualitative approaches in the social and behavioral sciences. Thousand Oaks,CA: SAGE.

Unger, B., Gemünden, H. G., & Aubry, M. (2011, June). The three roles of a project portfolio management office: The impact on portfolio management execution and success. Paper presented at the IRNOP 2011, Montreal, Canada.

United Kingdom Government. (2012). Assurance for major projects (pp. 44). London: National Audit Office.

Van de Ven, A. H., & Poole, M. S. (1995). Explaining development and change in organizations. Academy of Management Review, 20(3), 510–540.

Winch, G. M. (2014). Three domains of project organising. International Journal of Project Management, 32(5), 721-731.

Zeitlin, J. (2008). The historical alternatives approach. In G. Jones & J. Zeitlin (Eds.), The Oxford handbook of business history (pp. 120–140). Oxford, UK: Oxford University Press.

Monique Aubry, PhD, is director of graduate programs in project management and professor in these programs and in the executive MBA program at the School of Management, Université du Québec à Montréal (UQAM). Her main research interest bears on organizing for projects and organizational design, more specifically on project management offices (PMO). She is member of the Project Management Research Chair (www.pmchair.uqam.ca). In 2012, she received the IPMA Research Award for her research on project management offices. She has joined the editorial team of the Project Management Journal in the “Organizational Side of Project Management and Management of Organizational Projects” department.

Maude Brunet has almost ten years of experience in project management. She is a doctoral student in Management, with a specialization in Project Management, at the School of Management in the University of Quebec in Montreal (ESG-UQAM). Her research interests focus on the governance of major public infrastructure projects. She received her BA from McGill University in International Development Studies and a Master’s in Project Management at ESG -UQAM in 2009. She has worked for the firm GPBL, first as a consultant in project management and as Director of Consultation until December 2012, before pursuing her studies at PhD level.

Maude Brunet has been involved in the Montreal Chapter of the Project Management Institute (PMI), serving as a Board Director for 2012–2013. She is a certified Project Management Professional (PMP®) since 2010. She is also actively involved with the organization GP–Quebec, the association for public projects in Québec.

This material has been reproduced with the permission of the copyright owner. Unauthorized reproduction of this material is strictly prohibited. For permission to reproduce this material, please contact PMI or any listed author.

©2014 Project Management Institute Research and Education Conference

Advertisement

Advertisement

Related Content

Advertisement

Publishing or acceptance of an advertisement is neither a guarantee nor endorsement of the advertiser's product or service. View advertising policy.