Project Management Institute

Modern virtual project management

the effects of a centralized and decentralized project management office

Introduction

Project management today is an art that combines the skills and knowledge of the project manager with the tools and techniques within the project management profession to deliver a product within the specifications required of the sponsor (Project Management Institute [PMI], 2004). Project management has been viewed as a means to track and organize a project. Research indicates that for a company to succeed there must be a view to the future while understanding the company's past (Brown & Eisenhardt, 1997). Brown and Eisenhardt (1997) found that the successful companies were not overly structured but that they allowed chaos and flexibility and promoted independent thinking. As Brown and Eisenhardt (1997) suggested, if project management is to survive as a discipline, the community must be flexible in incorporating technologies, instituting flexible environments such as virtual project management, and assessing continually new ways to measure a project's success.

Statement of Problem

One of the many challenges within the project management community is how and where to structure the project management organization (PMO) to optimize the value of virtual project managers. In fact, Hobbs and Aubrey's (2007) study has found that PMOs longevity is short. This may indicate that organizations leadership has not truly understood the merits or that leadership is too fluid in an organization for a PMO to become an integral part of the system. Some organizations choose to include the project managers as part of discrete sales or solution teams, with a small corporate group overseeing the project management methods, training, and other miscellaneous responsibilities. This example is characteristic of a decentralized PMO (Ormand, Bruner, Birkemo, Hinderliter-Smith, & Veitch, 2000; Hales, 1999; Kerzner, 1998). Other businesses and organizations choose to centralize the project managers in one organization (PMI, 2004). These project managers receive direction and guidance from an overarching centralized PMO.

Centralized project managers report to a centralized PMO. A centralized PMO is defined as an organization to which project managers report and from which they receive direction, guidance, and oversight. The centralized or projectized PMO is responsible for processes, procedures, systems, and tools (PMI, 2004).

Decentralized project managers report to a decentralized PMO. A decentralized PMO is defined as a corporate group that oversees the project management methodology, training, and other miscellaneous responsibilities, but project managers do not report to this organization directly or in a matrix environment (Ormand et al., 2000; Hales, 1999; Kerzner, 1998). The project managers normally report to a functional or solution chain-of-command (PMI, 2004) common in a matrix organization.

Significance of the Study

One of the many challenges within the project management community is how to structure the PMO for virtual project managers. The project manager in a decentralized organization, which may be referred to as a matrix organization, does not have an overarching organization that provides administrative and functional guidance and support nor a structure for the virtual environment (Ormand et al., 2000; Hales, 1999). In the decentralized organization, the project manager may have at least two supervisors, which can create a difficult situation for the project manager. The project manager may decide upon an unpopular, strategic, and tactical direction that moves the project contrary to the functional supervisor's desires. However, the decision may be justifiable for the administrative supervisor and for the company (Kahai, Snyder, & Carr, 2001/2002).

Research Questions

  1. Are there reported differences in the training received by project managers working on virtual projects in a centralized PMO versus a decentralized PMO?
  2. Are there reported differences in the use of standardized processes by project managers working on virtual projects in a centralized PMO versus a decentralized PMO?
  3. Are there reported differences in levels of electronic communication and collaboration by project managers working on virtual projects in a centralized PMO versus a decentralized PMO?
  4. Are there reported differences in leader behaviors perceived by project managers working on virtual projects in a centralized PMO versus a decentralized PMO?

Hypotheses

Hypothesis One – Project Manager Training

H1: There will be no reported differences in training received by project managers working on virtual projects in a centralized PMO versus a decentralized PMO.

Hypothesis Two – Use of Standardized Processes

H2: There will be no reported differences in the use of standardized processes by project managers working on virtual projects in a centralized PMO versus a decentralized PMO.

Hypothesis Three – Levels of Electronic communication and collaboration

H3: There will be no reported differences in levels of electronic communication and collaboration by project managers working on virtual projects in a centralized PMO versus a decentralized PMO.

Hypothesis Four – Leader Behaviors

H4: There will be no reported differences in leader behaviors perceived by project managers working on virtual projects in a centralized PMO versus and decentralized PMO.

Definitions

For the purposes of this study, the following terms are defined:

  1. Virtual projects: These projects consist of more than 50% of the project team members not being resident in the same physical location, but are not necessarily dispersed over different time zones. Team members depend on technology to communicate, rarely or never meet face-to-face more than once every two weeks as a project team, and they make decisions about the project (Kanawattanachai & Yoo, 2007; Kelley, 2001; Townsend & DeMarie, 1998; Maznevski & Chudoba, 2000).
  2. Centralized project management organization: The organizational structure is designed such that the project managers, project coordinators, and other personnel performing project activities report to an administrative chain of command within the PMO. The project personnel are assigned to projects by the administrative chain of command. The centralized PMO is responsible for project management training, project management organizational processes, and technology used and implemented for project managers (Milosevic, Inman, & Ozbay, 2001; Toney, 2002b). In addition, this organization is responsible for evaluating the project personnel's performance and compensation (Milosevic et al., 2001; Toney, 2002b).
  3. Decentralized project management organization: This small corporate or business unit organization is responsible for maintaining project management methods and/or training and best practices. This type of PMO does not have a central decision-making authority. Authority may be delegated or collaborative, depending on the project (Ormand et al., 2000; Hales, 1999; Kerzner, 1998). Wren (1972) describes the decentralized organization as a matrix or project organization. Therefore, a definition for the decentralized PMO is “an organizational body or entity assigned various responsibilities related to the centralized and coordinated management of those projects under its domain” (PMI, 2004, p. 369).
  4. Project manager: The project manager is the “person assigned by the performing organization to achieve the project objectives” (PMI, 2004, p. 369).
  5. Project: A project is “a temporary endeavor undertaken to create a unique product, service, or result” (PMI, 2004, p. 368).

Literature Review

Modern Virtual Projects

Modern virtual projects are when more than 50% of the project team members are not resident in the same physical location, but are not necessarily dispersed over different time zones. The team depends on technology to communicate, rarely or never meets face-to-face more than once every two weeks as a project team, and team members themselves make decisions about the project (Kelley, 2001; Townsend & DeMarie, 1998; Maznevski & Chudoba, 2000). It is not unusual for one or two of the team members to rely on technology; however, it is more exceptional to have most members using technology to communicate with each other, the customer, and the project manager in order to accomplish objectives (Reinsch, 1999).

Training and learning are major factors in the virtual environment's success (Townsend & DeMarie, 1998; Duarte & Snyder, (2001). A virtual team appears to be more successful when training is conducted on communication skills and communication technology (Townsend & DeMarie, 1998). Townsend & DeMarie's (1998) studies indicate that technology training should occur more often for virtual teams than for traditional teams, since technology is the mainstay for communication and is evolving at a fast pace. Duarte and Snyder (2001) find that a successful virtual project team distributes “learnings within the team and beyond to the wider organization” (p. 127).

The seamless integration of the technology within the organization and among the federation members (Boudreau, Loch, Robey, & Straud, 1998) allows local projects to have the support of a worldwide virtual organization, and the client does not realize that the product is a result of several companies or organizations. A well-run virtual organization should be able to function with very little regard to geographical distance and time barriers.

Centralized Project Management Organizations

Jiang, Klein, and Chen's (2001) research provides further evidence that the project manager's success in centralized or decentralized organizations requires senior management's support and active participation. The research concludes that the project manager must be involved early in the proposal or business case development to enhance an information technology (IT) project's success. The centralized PMO would be a hybrid of a group 3 and group 4 from Hobbs and Aubrey's (2007) study. This PMO would include the oversight of projects, programs, and resources, and for the project standards and processes within the organization.

Traditionally, information systems (IS) organizations have been centralized. However, a study conducted by Kahai et al. (2001/2002) notes a trend in Fortune 1000 companies in the United States. IS departments are transitioning to a hybrid of decentralized and centralized management (Kahai et al., 2001/2002). Resources are becoming decentralized while the decision authority is still centralized (Kahai et al., 2001/2002). Finally, the study by Kahai et al. (2001/2002) found that each company surveyed struggles for the control of the resources and how best to unitize the resources.

Orwig and Brennan's (2000) studies found that companies such as professional services companies that were project based had better project management success over those companies that were decentralized and were not able to replicate project management methodologies. Quality standards such as benchmarking, statistical controls, and flowcharting are key to the survival of centralized or project-based organizations (Orwig & Brennan, 2000).

Decentralized Organizations

A decentralized PMO may be characterized as a small corporate or business unit organization that is responsible for maintaining project management methods and/or training and best practices. This type of PMO does not have a central decision-making authority. Authority may be delegated or collaborative depending on the project (Ormand et al., 2000; Hales, 1999; Kerzner, 1998). Wren (1972) describes the decentralized organization as a matrix or project organization. Therefore, a definition for the decentralized PMO is “an organizational structure in which the project manager shares responsibility with the functional managers for assigning priorities and for directing the work of individuals assigned to the project” (PMI, 2004, p. 368). The decentralized PMO may also be any organization responsible for PM functions.

The decentralized project management organization may have little authority over projects. Benchmark data (Toney, 1999) implies that decentralized project organizations hinder the attainment of project goals since the project manager has to borrow resources and reports to two bosses. Toney's (1999) studies indicate that a best practice for the project organization is to “build partnerships with and gain support from senior executives” (p. 20).

Quresh, Liu, and Vogel (2006); Townsend and DeMarie (1998); and Duarte and Snyder's (2001) research indicates that there appears to be a correlation between the virtual teams' successes and communication technology training. Reinsch (1999) and Sethi, Smith, and Park (2001) indicate training in team communication methods also enhances a team's success. The survey evaluates the training and development project managers receive.

Virtual teams normally do not have the ability to assess each other's working habits visually. Toney's (1999) benchmark data indicate and Frame (1995) emphasizes that a virtual project succeeds more often when team members have confidence in the leader and in each other. Survey participants will answer questions regarding their perception of the team leader and the members' competence on the project team.

Organizational processes are the foundation of many project management methodologies. The literature research provides conflicting results as to the necessity of detailed project management processes for project teams. There is no research found that specifically addressed virtual project teams and the merits of organizational processes. The literature appears to favor more organizational processes in a centralized organization versus a decentralized one, although there is literature that provides conflicting results.

Technology continues to provide the ability for a project team to communicate without the necessity to be colocated and face-to-face (Guss, 1998; Reinsch, 1999; Toney, 1999). This allows organizations and companies to seek to reduce overhead expenditures and to promote the viability of virtual organizations. Communication technology is also another component of the survey.

Methods

Two of the Project Management Institute's (PMI) components were approached to distribute the survey via e-mail to its membership and to post the survey on its website. The confidentiality and purpose statements were included in the survey when posted on the Web site or distributed via e-mail. The estimated project management population was 1,500 project managers. The project management components were a local chapter of PMI and a PMI Specific Interest Group.

The project managers in the PMI components represented a cross section of many industries that use project management skills. Examples of these industries include banking, IT, automotive, healthcare, pharmaceuticals, construction, government, independent project management contractors, and finance. The project managers may be new or experienced, and may be certified by PMI.

This research was an organizational investigation studying training development, standard organizational processes, technology, and leadership. A survey was used to collect quantitative data. The project management survey was sent electronically to all known project managers within the defined population.

The demographics of the project managers, the types of virtual projects, and the organization have been established to understand the study population. Chi-square analysis was done on the demographic data to understand the fitness for use. The collected data should also identify opportunities where future research may be needed.

The survey was designed to collect quantitative data including team operations and effectiveness on a virtual project team. The project managers were asked to provide information about one virtual project they participated on within the last year. The project managers were asked to provide data about the projects, including cost and schedule baseline and the range of deviation from the baseline.

A Likert-type survey was developed for this research. The survey was derived from existing project management and virtual teams research questionnaires (Duarte & Snyder, 2001; Lurey, 1998). The survey consisted of 22 Likert-type questions/statements and 12 yes/no, multiple-choice questions, and fill-in-the-blank statements.

This study was generalized, to the degree possible, for the project management industry. It was not targeted for any one specific industry, but to provide overarching data regarding virtual projects' cost and schedule results in centralized versus decentralized PMOs. The resultant data cannot be generalized for traditional face-to-face project management environments or specific project virtual environments. This study should provide data for project management leadership in how to structure the PMO.

Presentation and Analysis of Data

A total of 73 project managers replied to the survey. This number represented 4.8% of the membership of the two project management societies. Thirteen surveys were rejected for one or more of the following reasons:

  • The project manager was not a virtual project manager;
  • The survey was not completed properly; and/or
  • The survey was incomplete.

This research was an organizational investigation studying training development, standard organizational processes, technology, and leadership. A validated survey was used to collect quantitative data (Duarte & Snyder, 2001). Demographic data was added to the survey, including gender, experience, metrics used on the project, and certification. The project management survey was sent electronically to all known project managers in one PM society, while the other society chose to post the survey on its website.

A chi-square analysis was used to evaluate the demographic data as it related to a centralized and decentralized project manager. For the purposes of this study, levels of significance of 0.010 or greater were considered significant. Correlations and significance of relationships among statements in each category were also evaluated.

The demographics included gender, project management certification, and years of experience. The statistical analysis revealed no statistical significance between centralized and decentralized project managers. The project's dollar value and the role of the project manager were also statistically insignificant.

The demographic data according to the statistics did not significantly affect the centralized and decentralized project manager sample. Therefore, all detected differences are attributable to the hypotheses and not to the demographic differences.

Each of the four hypotheses' data is presented in the following manner.

  • A summary table with significant statements and correlation (the summary tables has a list of the statements),
  • Those questions with significant chi-square values, by centralized (C) and decentralized (D),
  • The correlation of centralized versus decentralized project managers.

Research Question #1 and Hypothesis One - Project Management Training

The statistical analysis indicated that there were reported differences in the training received by project managers working on virtual projects in a centralized PMO versus a decentralized PMO. To further clarify hypothesis one, Table 1 presents the significant statements and correlations between statements. Based on statistical data shown in Table 2, the hypothesis (H1) for training was rejected. The research question was positive since there were reported differences in the training received. The centralized virtual project managers perceived that they had access to cultural training and there was adequate access for all to databases that had learning opportunities (statement 6).

Table 1: Hypothesis one summary table

Hypothesis #1 Significance Correlations Centralized Correlations Decentralized
H1: There will be no reported differences in training received by project managers working on virtual projects in a centralized PMO versus a decentralized PMO.
Survey
Training and Development Statements (Duarte & Snyder, 2001)
1. There is good access to technical training. Positive correlation to statement 4 (LOS 0.001) Positive correlation to statement 4 (LOS 0.001)
2. I took advantage of the available training.
3. The training increased my project management skill set.
4. There is access to training in working across cultures. Yes, stronger for project managers in centralized PMOs Positive correlation to statement 6 (LOS 0.005)
5. There are methods available for continual and just-in-time learning, such as Web-based training. Positive correlation to statement 6 (LOS 0.001)
6. There are mechanisms, such as lessons-learned databases, for sharing across boundaries. Yes, stronger for project managers in centralized PMOs

Table 2: Chi-square data for training

Chi-Square 0.005 0.010 0.050
X2 calc 88.858
df 10
X2 crit 25.188 23.209 18.307
0.005 0.010 0.050
Statement 4 25.788
Statement 6 32.166

Access to training in working across cultures and to lessons-learned databases for sharing were the two statements that were responded to favorably by centralized versus decentralized project managers (see Tables 2 and 3). The statistical data for the virtual project managers in a centralized organization consistently reported better access to crosscultural training and access to lessons-learned databases, over the decentralized virtual project managers.

Table 3: Proportions centralized/decentralized

Measure and Variable S4 S6
C D C D
Agree/Strongly Agree 44.0 24.2 56.0 30.3
Disagree/Strongly Disagree 32.0 54.5 36.0 51.5

As indicated in Table 3, these two statements (statements 4 and 6) also had a positive correlation among the centralized virtual project managers. This correlation did not exist for the decentralized virtual project managers.

This data supported the hypothesis that training, especially crosscultural and databases with lessons-learned, was more readily available in the centralized organization (see Tables 1 and 3).

Access to training in working across the cultures (statement 4) also had a positive correlation to the technical training provided (statement 1) (see Tables 1 and 4). The level of significance was the same for both the centralized and decentralized project managers. This was the only correlation that resulted for the significant questions for the decentralized project managers (see Table 4). The research question positively correlates to the data reported by the centralized virtual project managers. Differences were seen in the training received by the virtual project managers in a centralized versus a decentralized PMO.

Centralized project managers showed a positive correlation between the availability of continual learning and just-in-time learning (statement 5) and the availability of lessons-learned databases (statement 6). This correlation was only seen with the centralized project management respondents (see Tables 1 and 4). The statistical data suggested that a centralized virtual project manager had better access to cultural training. Statistically, the centralized virtual project managers reported positively to having access to databases which appeared to allow for continual and just-in-time learning.

Table 4: Correlation Levels of Significance

Centralized PMs
Statements          S1        S4        S5
S4 0.001
S6 0.005 0.001
Decentralized PMs
Statements          S1        S4        S5     
S4 0.001

Research Question #2 and Hypothesis Two - Use of Standardized Processes

Centralized project managers reported differences in standardized processes (see Table 5). To clarify hypothesis two further, Table 5 presents the significant statement and correlations between statements. Based on this data shown in Table 6, the hypothesis (H2) for standardized processes was rejected. The data suggested that standardized processes and agreed upon soft processes were correlated positively to the centralized project management community. An affirmative answer was appropriate for the research question since there was a difference in one statement in hypothesis two.

Table 5: Hypothesis two summary table

Hypothesis #2 Significance Correlations Centralized Correlations Decentralized
H2: There will be no reported differences in the use of standardized processes by project managers working on virtual projects in a centralized PMO versus a decentralized PMO.
Survey Section 2
Standardized Processes (Duarte & Snyder, 2001)
7. There are standard and agreed-on technical team processes used throughout the organization and with partners. Positive correlation to statement 8 (LOS 0.001) Positive correlation to statement 8 (LOS 0.001)
8. There are standard and agreed-on “soft” team processes used throughout the organization and with partners. Yes, stronger for project managers in centralized PMOs
9. Adaptation of processes is encouraged when necessary. Positive correlation to statement 8 (LOS 0.005) Positive correlation to statement 8 (LOS 0.010)
10. The culture supports shared ways of doing business across teams and partners. Positive correlation to statement 8 (LOS 0.010)

Table 6: Chi-square data for standardized processes

Chi-Square 0.005 0.010 0.050
X2 calc 55.053
df 6
X2 crit 18.548 16.812 12.592
Statement 8 42.323

Centralized virtual project managers responded positively to having standard and agreed-on “soft” team processes that were used throughout the organization and with partners (statement 8) versus the decentralized virtual project managers (see Tables 6 and 7). The reported data suggested that centralized virtual project managers had consistent written direction to refer to and a culture that communicated acceptable nonwritten policy.

Table 7: Proportions centralized/decentralized

Measure and Variable S8
  C D
Agree/Strongly Agree 64.0 33.3
Disagree/Strongly Disagree 28.0 51.5

For both the centralized and decentralized virtual project managers, standard and soft team processes (statement 8) were positively correlated with standard and agreed upon technical team processes (statement 7) (see Tables 5 and 8). Additionally, with both virtual project manager groups there was a positive correlation between the standard and soft team processes (statement 8) and the encouragement to adapt processes when necessary (statement 9). This correlation was stronger with the centralized project managers (see Tables 5 and 8). The correlation appeared to imply that technical team processes need to be present for the virtual project manager to have the fullest use of the processes. Additionally, both communities emerged as being willing to adapt processes, as necessary, although it was a stronger correlation in the centralized environment.

The decentralized project managers also had a positive correlation to statement 10. A culture that supported sharing ways of doing business across teams and partners (statement 10) was a positive correlation to having standard and soft team processes (statement 8). The correlation suggested that the decentralized project managers may have a greater need to share successful ways of doing business among the community, since there was no central authority.

Table 8: Correlation levels of significance

Centralized PMs Statement   S7 S9 S10
Statement   S7 S9 S10
S8   0.001 0.005  
Decentralized PMs        
Statement S7    S9        S10  
S8 0.001 0.010 0.010

Research Question #3 and Hypothesis Three - Electronic Communication and Collaboration Technology

The centralized virtual project managers reported a difference in the levels of electronic communication and collaboration. Table 9 is made available to present the significant statements and correlations between statements. Based on this data shown in Tables 9 and 10, the null hypothesis (H03) for technology was rejected. Statement 14 was agreed to almost equally by the centralized and decentralized project managers (see Table 7). More centralized project managers responded neutrally, whereas the decentralized project managers' responses were skewed toward disagreeing with statement, 14 (see Table 10). As shown in Table 9, the answer to the research question was yes, because statement 14 was different between the two environments.

Table 9: Hypothesis three summary table

Hypothesis #3 Significance Correlations Centralized Correlations Decentralized
H3: There will be no reported difference in levels of electronic communication and collaboration by project managers working on virtual projects in a centralized PMO versus a decentralized PMO.
Survey Section 3
Electronic Communication and Collaboration (Duarte & Snyder, 2001)
11. There are consistent standards for electronic communication and collaboration tools across the organization. Positive correlation to statement 14 (LOS 0.010)
12. There are ample resources to buy and support state-of-the-art electronic communication and collaboration technology.
13. People from all functional areas have equal access to and are skilled in using electronic communication and collaboration technology. Positive correlation to statement 14 (LOS 0.001) Positive correlation to statement 14 (LOS 0.001)
14. People from all geographic areas have equal access to and are skilled in using electronic communication and collaboration technology. Yes, stronger for project managers in centralized PMOs

Table 10: Chi-square data for electronic communication

Chi-Square 0.005 0.010 0.050
X2 calc 190.39
df 6
X2 crit 18.548 16.812 12.592
Statement 14 165.02

The centralized and decentralized project managers agreed about equally (see Table 7) that personnel in all geographic locations had the same access and were skilled in using technology to communicate and collaborate (statement 14). However, the decentralized project managers responded more negatively than the centralized, while the centralized respondents were more neutral. The variation between the two groups was that over half of the decentralized respondents did not agree that the community had equal access to the technology needed to conduct business. In the centralized community, the respondents were positive or neutral which may indicate that adequate technology was provided to the project management community.

Table 11: Proportions centralized/decentralized

Measure and Variable S14
C D
Agree/Strongly Agree 40.0 39.4
Disagree/Strongly Disagree 24.0 54.5

The centralized and decentralized project managers' responses demonstrated a positive correlation between personnel in all geographic locations (statement 14) and in all functional areas (statement 13). Both had the same access and skill in using technology to communicate and collaborate (see Table 12). The correlation suggested that the various companies of the respondents might have a standard technology which employees were expected to adopt.

As shown in Tables 9 and 12, the decentralized project managers' responses also indicated a positive correlation between having consistent standards for communicating and collaborating electronically (statement 11) and personnel in all geographic locations having equal access and skills in using the communication and collaboration technology (statement 14). The centralized community did not have this correlation. The lack of a correlation may have indicated that the centralized virtual project managers had different levels of technology for the various geographic locations, depending on the need of the area.

Table 12: Correlation levels of significance

Centralized PMs Statement S11 S13
S14 0.001
Decentralized
PMs Statement             S11       S13     
S14 0.010 0.001

Research Question #4 and Hypothesis Four - Project Management Leader Behavior

The centralized virtual project managers reported differences in all four statements (statements 15 through 18). Table 13 is provided to further clarify the significant statements and correlations between statements. Based on this data, shown in Tables 13 and 14, the null hypothesis (H04) for leader behaviors was rejected. This section was more absolute than the previous three hypotheses, since all statements in this section were supported by the statistical data. Since all statements were positive for the centralized PMO, the research question was yes, because there were differences in the responses.

Table 13: Hypothesis four summary table

Hypothesis #4 Significance Correlations Centralized Correlations Decentralized
H4: There will be no reported difference in leader behaviors perceived by project managers working on virtual projects in a centralized PMO versus a decentralized PMO.
Survey Section 4
Leader behaviors (Duarte & Snyder, 2001)
15. Leaders set high expectations for virtual team performance. Yes, stronger for project managers in centralized PMOs Positive correlation to statements 16 (LOS 0.001) and 18 (LOS 0.005) Positive correlation to statements 16 (LOS 0.010) and 18 (LOS 0.001)
16. Leaders help gain the support of customers and other stakeholders. Yes, stronger for project managers in centralized PMOs Positive correlation to statements 15 (LOS 0.001) and 17 (LOS 0.001) Positive correlation to statement 15 (LOS 0.010)
17. Leaders allocate resources for the training and technology associated with virtual teams. Yes, stronger for project managers in centralized PMOs Positive correlation to statement 16 (LOS 0.001) Positive correlation to statement 18 (LOS 0.001)
18. Leaders model behaviors such as working across boundaries and using technology effectively. Yes, stronger for project managers in centralized PMOs Positive correlation to statement 15 (LOS 0.005) Positive correlation to statement 15 (LOS 0.001) and 17 (LOS 0.001)

Table 14: Chi-square data for leader behavior

Chi-Square 0.005 0.010 0.050
X2 calc 115.86
df 6
X2 crit 18.548 16.812 12.592
Statement 15 18.758
Statement 16 25.711
Statement 17 48.575
Statement 18 22.813

Statement 15, as shown by Tables 13 and 14, was agreed to more so by centralized project managers versus decentralized. Proportionally more decentralized project managers disagreed with statement 15 than did the centralized project managers. Leaders in the centralized community, according to the data, established higher expectations (statement 15) than in the decentralized group.

As shown by Tables 13–15, the centralized project managers responded positively to statement 16. Leaders in the centralized group were more likely to gain the support of customers and stakeholders (statement 16), according to the respondents. The lack of agreement within the decentralized virtual project managers may have resulted due to the differing goals of the various functional managers, which may have resulted in a confusing message being portrayed to the project team.

The decentralized group disagreed that the leaders allocated the necessary resources for training and technology for the virtual teams (statement 17), whereas the centralized community respondents agreed with this statement (see Tables 13 and 15). This data suggested that the centralized PMO understood the needs of the virtual project management team and ensured that the resources needed were represented properly to allocate the appropriate funds.

The decentralized PMO community, among the respondents, indicated that there was not a central group that advocated the needs of the virtual project team.

Statement 18, similar to statement 17, was positive for the centralized project managers (see Tables 13 and 15). The perception among those virtual project managers in the centralized PMO provided evidence that the leaders of the organization modeled the expected behavior with other organizations and with the ability to use the technology (statement 18). In the centralized PMO, the focus was on the project managers' needs and the leaders projected the expected behavior. Since there are normally various leaders on a project in a decentralized organization, the model behavior may not be consistent with what may be needed with the project management community and/or the virtual teams' needs.

Table 15: Proportions centralized/decentralized

Measure and Variable S15 S16 S17 S18
C D C D C D C D
Agree/Strongly Agree 80.0 63.6 72.0 51.5 44.0 27.3 60.0 39.4
Disagree/Strongly Disagree 8.0 27.3 16.0 12.1 24.0 57.6 20.0 42.4

Both the centralized and decentralized project managers responded in such a manner that there were three positive correlations between statements in each group (see Table 16). The centralized project managers' responses demonstrated positive correlations between statements 15 (leaders set high expectations) and 16 (leaders gain support), between statements 16 (leaders gain support) and 17 (leaders allocated resources), and between statements 15 (leaders set high expectations) and 18 (leaders modeled expected behavior). The decentralized project managers' positive correlation between leaders setting high expectations (statement 15) and leaders gaining support among other organizations and stakeholders (statement 16) was not as strong as with the centralized project managers. The suggested implication in the centralized PMO was that the leaders had a stronger understanding of the needs and what was required for the success of the virtual project team.

The virtual project managers in the decentralized organization had a stronger positive correlation between leaders setting high expectations (statement 15) and leaders modeling the expected behavior (statement 18) than the centralized project managers. The implication with this correlation may be related to centralized PMO leaders being more rounded leaders; the correlation was present but not as strong as with the decentralized. Within the decentralized community, the project managers may more clearly see the distinction since the leaders were not centralized but were scattered throughout the organization.

The positive correlation between leaders that allocated the needed resources for training and technology (statement 17) and leaders that modeled the expected behavior (statement 18) was unique to the decentralized project managers. As with the previous correlation, the decentralized project managers may have been more aware of this relationship, since there was not a centralized group that provided the needed resources. In the centralized PMO, the resources were allocated, and there was not a clear distinction between the leader's behavior and the needed resources.

Table 16: Correlation levels of significance

Centralized PMs
Statements               S15       S16          S17
S16 0.001
S17 0.001
S18 0.005
Decentralized PMs
Statements               S15        S16          S17
S16 0.010
S18 0.001 0.001

Summary

This paper reports a portion of a larger study. The chi-square analysis revealed that there was no statistical difference among the demographic data. Since the statistical analysis rendered the demographic data insignificant, all differences detected for hypotheses one through four were attributed to the differences between centralized and decentralized PMOs. For hypotheses one through four, the hypotheses were rejected based on the chi-square statistical analysis. Significant variables existed within each section as they related to each hypothesis. The leadership area leader behavior (hypothesis four) had the strongest significance. Leader behavior had variability with each of the four statements. The data for hypotheses and research questions one through four indicated centralized virtual project managers were more positive in the following areas than their counterparts in a decentralized PMO:

  • Technical and just-in-time training,
  • Standard and agreed-on soft team processes,
  • Electronic communication and collaboration technology was available and team members were skilled in its use, and
  • Leaders established high expectations, assisted in gaining support for various stakeholders, provided the necessary training and technology resources, and modeled expected behavior.

Each of the four hypotheses had positive correlations between statements for both the centralized and decentralized project managers. Leader behavior (hypothesis four) had similar correlations between the two groups of project managers.

Hypothesis two (standard processes) and hypothesis three (electronic communication/collaboration) each had one statement that was significant. Standard and soft processes electronic communication and collaboration technology were the two relevant statements. For hypothesis two, there were three positive correlations and for hypothesis three there were two. Within hypothesis two, both the centralized and decentralized virtual project managers and the standard and soft team processes were positively correlated with standard and agreed upon technical team processes. Additionally, with both virtual project manager groups there was a positive correlation between the standard and soft team processes and the encouragement to adapt processes when necessary. This correlation was stronger with the centralized project managers (see Tables 5 and 8).

For hypothesis three, the centralized and decentralized project managers' responses demonstrated a positive correlation between personnel in all geographic locations and in all functional areas. Both had the same access and skill in using technology to communicate and collaborate (see Table 12).

Training (H1) had two significant statements: access to training in working across cultures (statement 4) and mechanisms for sharing learning across boundaries (statement 6). There were three positive correlations with the significant questions among the centralized and only one within the decentralized group. The two significant statements (statements 4 and 6) also had a positive correlation among the centralized virtual project managers. This correlation did not exist for the decentralized virtual project managers.

Availability of continual learning and just-in-time learning (statement 5) was positive to the availability of lessons-learned databases (statement 6). This correlation was only seen with the centralized project management respondents (see Tables 1 and 4). The statistical data suggested that a centralized virtual project manager had better access to cultural training. The centralized virtual project managers statistically reported positively to having access to databases that appeared to allow for continual and just-in-time learning. Availability of continual learning and just-in-time learning (statement 5) positively correlated to the availability of lessons-learned databases (statement 6). This correlation was only seen with the centralized project management respondents (see Tables 1 and 4).

Summary and Recommendations

Project Management Training (H1) Recommendations

Duarte and Snyder's (2001) research found that providing the latest technology to virtual teams was not enough. There needed to be an equal amount of attention focused on technology and cultural awareness training for the virtual team (Thomas & Bostrom, 2008; Duarte & Snyder, 2001). The findings from this study indicated that the centralized project managers had better access to cultural training (statement 4) and on-the-job training, such as lessons-learned databases (statement 6). Of note, the centralized project managers indicated that technology was available and team members were skilled in using the technology (hypothesis 3). When these two hypotheses were coupled, it appeared that the centralized PMO's leadership ensured that the proper training and technology were provided to the teams.

It is recommended that organizations with virtual project managers, whether a centralized or decentralized PMO exists, ensure that all team members are kept current on the technology being used, as well as, training for working with different cultures (Kanawattanachai & Yoo, 2007; Khazanchi, & Zigurs, 2005). The centralized organizations are statistically better at providing the virtual project managers with databases that allowed sharing and learning. Duarte and Stuart (2001) recommend establishing “shared lessons, databases, knowledge repositories, and chat rooms” (p. 17) to enhance virtual teams' learning opportunities. This recommendation is also supported by Toney (2002a), who states, “The best practices project organization has a personalized development and training program based on identification of skills and competencies needed by the individual or group” (p. 241).

Standard Organizational and Team Processes (H2) Recommendations

Statement 8, “There are standard and agreed-on soft team processes used throughout the organization and with partners” (Duarte & Snyder, 2001, p. 13), was statistically positive for the centralized project managers. Based on the one statement, it was difficult to say whether all areas of standardized processes were better in the centralized versus the decentralized PMO. Numerous research studies demonstrated that standardized processes help the efficiency of projects and organizations (Duarte & Snyder, 2001; Toney, 2002a). A project normally had goals that include completion within budget and on schedule (PMI, 2004).

The recommendation is that standard processes within the organization are adapted for the virtual organization, as necessary. In addition, there should be agreed upon soft processes for the virtual environment, including items such as conflict-resolution and communication (Qureshi, Liu, Vogel, 2006; Duarte & Snyder, 2001). The project manager and the team members must also be competent and understand the importance behind the standardization.

Adding to the importance of standardized processes were the results of hypothesis five, team leader and team member competency. This result was positively correlated to the centralized project management organization. Toney (2002a) states, “Project teams are more efficient when utilizing a repeatable and predictable approach” (p. 183). It, therefore, may be extrapolated that one of the reasons that team leaders and team members are more competent and experienced in a centralized PMO is due to the repeatable and standard processes.

Electronic Communication and Collaboration (H3) Recommendation

Duarte and Snyder (2001) emphasized and a study by Quresh, Liu, and Vogel (2006) supported that the electronic collaboration and communication technology needed to meet the needs of the team. Toney's (2002b) benchmark data supported Duarte and Snyder (2001). The technology needed to be customizable to meet the needs of the team and did not need to be the latest technology (Quresh, Liu, & Vogel, 2006; Toney, 2002b). There was one statement for electronic communication and collaboration that was positive for the centralized project manager and stated that individuals from all geographic areas have equal access to and are skilled in using the communication and collaboration technology (statement 14) (Duarte & Snyder, 2001).

It is difficult to come to any specific conclusions based on the one statement. However, it appears from this study that the centralized virtual project managers were more adept at using the technology offered. Whether the virtual project manager is centralized or decentralized, the main way that work is conducted is via technology that is provided. Therefore, the proper technology and appropriate technology training are required to provide the team the tools to succeed.

Leader Behaviors (H4) Recommendations

Of the five hypotheses, the organizational leadership section statistically favored the centralized PMO. Duarte and Snyder (2001) emphasized, “working across time and distance and with organizational partners is not just a temporary fad but a new way of doing business” (p. 20). The centralized project management organization's leadership statistically understood this new way of doing business. The leadership established high expectations of the virtual team but ensured that partners in and out of the organization supported the team. In addition, the leadership ensured that the virtual team has the resources and training to accomplish the work. The leadership also was a model of the desired behavior when working across time and distance and organizational boundaries. The project manager establishes,models, and enforces the rules of communications expected by all team members (Malhotra, Majchrzak, & Rosen, 2007).

Virtual leadership required a different skill set than the traditional face-to-face leadership. Trust was an integral aspect of a virtual team and studies indicated that without this trust a virtual team was more likely to fail (Malhotra, Majchrzak, & Rosen, 2007; Cascio, 2000; Hage & Powers, 1992; Kezsbom, 2000). Trust enhances communication and this is essential in a virtual environment and, in fact, alternate channels of communications should be encouraged. In traditional face-to-face projects, this may be seen as noise and not on task focused but for the virtual team it encourages trust and enhances communication (Qureshi, Liu, Vogel, 2006). The organization's leadership must make a concerted effort to demonstrate to the whole organization the successes of a virtual team (Duarte & Snyder, 2001). The leaders of the organization also support the virtual team by providing the needed resources (Duarte & Snyder, 2001).

In a centralized project management organization, the focus is on the project manager and the project. When project managers are spread through the organization the management has to provide administrative and leadership support to many different types of employees with differing needs. It is recommended that the organizational leadership ensure that a small overarching PMO be established to review the training, technology, processes, and competencies for the virtual project managers. This provides the virtual project teams with the understanding that there is trust and that the organization values the work and the results.

Recommendations for Further Research

Based on the conclusions and findings of this study and the limitations encountered, the following is recommended for further study:

  1. Similar studies of best-in-class organizations in different industries with virtual projects need to be studied. This would enhance the body of knowledge within the project management industry for virtual projects within specific industries.
  2. Further studies need to be done with larger study groups to determine if demographics truly are insignificant to the hypotheses posited in this study.
  3. A relatively small sample of project managers was used. The Project Management Institute (PMI) has over 95,000 members. This study should be replicated and conducted in conjunction with the PMI Research Foundation to reach a much broader audience. The study would enhance the quality of the findings.

This study provides evidence that centralized project management offices provide better support services for the virtual project manager. This information needs to be further studied to understand the reasons and what is being used in its place.

References

Boudreau, M., Loch, K., Robey, D., & Straud, D. (1998). Going global: Using information technology to advance the competitiveness of the virtual transnational organization. Academy of Management Executive, 12(4), 120.

Brown, S., & Eisenhardt, K. (1997). The art of continuous change: Linking complexity theory. Administrative Science Quarterly, 42(1), 1.

Cascio, W. (2000). Managing a virtual workplace. The Academy of management Executive, 14(3), 81.

Duarte, D., & Snyder, N. (2001). Mastering virtual teams (2nd ed.). San Francisco: Jossey-Bass.

Frame, J. (1995). Managing projects in organizations. San Francisco: Jossey-Bass.

Guss, C. (1998). Virtual project management: Tools and the trade. Project Management Journal, 29(1), 22.

Hage, J., & Powers, C. (1992). The post industrial lives: Roles and relationships in the 21st century. New York: Sage.

Hales, C. (1999). Leading horses to water? The impact of decentralization on managerial behaviour. Journal of Management Studies, 36(6), 831.

Hobbs, B., & Aubrey, M. (2007). A multi-phase research program investigating project management offices(PMOs): The results of phase1. Project Management Journal, 38(1), 74-86.

Jiang, J., Klein, G., & Chen, H. (2001). The relative influence of IS project implementation policies and project leadership on eventual outcomes. Project Management Journal, 32(3), 49.

Kahai, P., Snyder, C., & Carr, H. (2001/2002). Decentralizing the IS function: Resources and locus of decision. Computer Information Systems, 42(2), 44.

Kelley, E. (2001). Keys to effective virtual global teams. Academy of Management Executives, 15(2), 132.

Kanawattanachai, P., & Yoo, Y. (2007). The impact of knowledge coordination on virtual team performance over time. MIS Quarterly, 31(4), 783-808.

Kerzner, H. (1998). In search of excellence in project management. New York: Van Nostrand Reinhold.

Kezsbom, D. (2000). Creating teamwork in virtual teams. Cost Engineering, 42(10), 30.

Khazanchi, D., & Zigurs, I. (2005). Patterns of effective management of virtual projects: An exploratory study. Newtown Square, PA: Project Management Institute.

Lurey, J. (1998). A study of best practices in designing and supporting effective virtual teams. ProQuest Digital Dissertations (UMI No. 9831029).

Malhotra, A., Majchrzak, A., & Rosen, B. (2007). Leading virtual teams. Academy of Management Perspectives, 21(1), 60-70.

Maznevski, M., & Chudoba, K. (2000). Bridging space over time: Global virtual team dynamics and effectiveness. Organization Science, 11(5), 473.

Milosevic, D., Inman, L., & Ozbay, A. (2001). Impact of project management standardization on project effectiveness. Engineering Management Journal, 13(4), 9–16.

Ormand, J., Bruner, J., Birkemo, L., Hinderliter-Smith, J., & Veitch, J. (2000). A centralized global automation group in a decentralized organization. Journal of Automated Methods & Management Chemistry, 22(6), 195.

Orwig, R., & Brennan, L. (2000). An integrated view of project and quality management for project-based organizations. The International Journal of Quality & Reliability Management, 17(4/5), 351.

Project Management Institute. (2004). A guide to the project management body of knowledge (PMBOK® Guide) – Third Edition. Newtown Square, PA: Project Management Institute.

Qureshi, S., Liu, M., & Vogel, D. (2006). The effect of electronic collaboration in distributed project management. Group Decision & Negotiation, 15(1),55-75.

Reinsch, L. (1999). Selected communications variables and telecommuting workers. The Journal of Business Communications, 36(3), 247.

Sethi, R., Smith, D., & Park, C. (2001). Cross-functional product development teams, creativity, and the innovativeness of new consumer products. Journal of Marketing Research, 38(1), 73.

Thomas, D. & Bostrom, R. (2008). Building trust and cooperation through technology adaptation in virtual teams: Empirical field evidence. Information Systems Management, 25(1), 45-56.

Toney, F. (1999). Competencies of superior project managers in large functional organizations (Vol. 1). New York: Marcel Dekker.

Toney, F. (2002a). The superior project organization: Global competency standards and best practices. New York: Marcel Dekker.

Toney, F. (2002b). The bottom line: Assuring project financial success. Unpublished.

Townsend, A., & DeMarie, S. (1998). Keys to effective virtual global teams. Academy of Management Executive, 12(3), 17.

Wren, D. (1972). The evolution of management thought. New York: The Ronald Press.

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.

© 2008 Project Management Institute

Advertisement

Advertisement

Related Content

  • PM Network

    Leveraging Locals

    PM Network interviews Veronica Lum Kok Yin, PMP, Director of Project Management, Global Delivery of Technology Solutions, QuintilesIMS, Singapore

  • PM Network

    Balance the scales

    By Mustafa, Abid Abid Mustafa describes two key ways of striking the right team balance for your project management office (PMO) team: A balance of “half-full” and “half-empty” thinkers can improve the accuracy of…

  • PM Network

    Positive reinforcement

    By Kretschmer, David | Garg, Saurabh | Rosenfeld, Gary | Gikovski, Emiliyan There are negative stakeholders in every project, and it's a project manager's responsibility to take negative stakeholders into account and neutralize their influence on the project affairs. This…

  • Virtual project management office

    By Guimarães, Eduardo Barbosa Gomes | Bôto, Elizeu Fonsêca Over the last decade, new technologies have enabled countless organizations to easily and effectively expand their operations worldwide. As a result, these organizations have needed to operate in…

Advertisement