JPMA-01765; No of Pages 14

Available online at www.sciencedirect.com

ScienceDirect
International Journal of Project Management xx (2015) xxx – xxx
www.elsevier.com/locate/ijproman

Project management and its effects on project success:

Cross-country and cross-industry comparisons
Marly Monteiro de Carvalho ⁎, Leandro Alves Patah, Diógenes de Souza Bido
Production Engineering Department, Polytechnic School University of São Paulo, São Paulo, Brazil

Received 11 April 2014; received in revised form 1 April 2015; accepted 14 April 2015

Abstract

This study aims to investigate the effects of project management (PM) on project success under the parameters of scheduling, cost, and margins.
We adopt a contingency approach that evaluates the complexity of the project, according to 4 categories, the effect of industry sector and countries.
The methodological approach involved a longitudinal field survey in 3 countries (Argentina, Brazil, and Chile) with business units from 10
different industries over a 3-year period, and data from a total of 1387 projects were analyzed. Structural equation modeling was used to test the
research hypotheses. The results show a significant and positive relationship between the response variable schedule with PM enablers and project
management efforts in training and capabilities development. Project complexity has a significant effect on 2 aspects of project success: margin and
schedule. Both cross-country and cross-industry analyses show a significant explanatory effect.
© 2015 Elsevier Ltd. APM and IPMA. All rights reserved.

Keywords: Project management; Performance; Success; Contingency approach; Project management maturity; Project complexity

1. Introduction PM methodologies (Ala-Risku and Kärkkäinen, 2006; Besner

and Hobbs, 2013; Chou and Yang, 2012; Hong et al., 2011).
Studies conducted over the last decade have aimed to This occurs because internal and external contexts can affect
analyze project success based on a variety of dimensions PM (Papke-Shields et al., 2010).
(Belout and Gauvreau, 2004; Besner and Hobbs, 2006; Bizan, On the one hand, some studies try to show the relationship
2003; Dvir et al., 2003; Gray, 2001; Kendra and Taplin, 2004; between PM maturity and project success. The core thesis in the
Lipovetsky, 2005; Raz et al., 2002; Repiso et al., 2007). This studies is that companies that expend efforts and resources to
interest is associated with the increasing efforts (and resources) develop PM and to expand their PM capabilities demonstrate
that companies are expending to implement project management better performance in their projects. However, the evidence for
(PM). However, PM remains a challenge because many projects that thesis is limited and inconclusive (Grant and Pennypacker,
have failed, as evidenced by several studies (Buchanan, 2008; 2006a, 2006b; Ibbs and Kwak, 2000a, 2000b; Jiang et al.,
Dai and Wells, 2004; The Standish Group International, 2009; 2004; Jugdev et al., 2002; Mullay, 2006; Thomas and Mullaly,
White and Fortune, 2002). 2007; Yazici, 2009).
Methods and techniques have been developed and encapsu- On the other hand, executives seek evidence that their PM
lated in bodies of knowledge by institutes and professional PM efforts are both effective and producing expected results. Thus,
associations (IPMA, 2006; PMI, 2013). However, empirical from the project performer organization perspective the
studies highlight the challenges associated with implementing existence of a positive relation between the organizational
efforts into improving project management and project success
⁎ Corresponding author at: Av. Prof. Almeida Prado, trav 2, n 128, 05508-900 is critical to sustaining these efforts.
São Paulo/SP, Brazil. There remains a gap in the literature with respect to
E-mail address: [email protected] (M.M. Carvalho). understanding the relationship between PM and project success

http://dx.doi.org/10.1016/j.ijproman.2015.04.004
0263-7863/00/© 2015 Elsevier Ltd. APM and IPMA. All rights reserved.

Please cite this article as: M.M. Carvalho, et al., 2015. Project management and its effects on project success: Cross-country and cross-industry comparisons, Int. J. Proj.
Manag. http://dx.doi.org/10.1016/j.ijproman.2015.04.004
2 M.M. Carvalho et al. / International Journal of Project Management xx (2015) xxx–xxx

(Aubry and Hobbs, 2010; Ika, 2009; Thomas and Mullaly, success criteria: efficiency (related to the iron triangle),
2008). Further, Besner and Hobbs (2013) suggest that maturity effectiveness, impact of the project on society, relevance to
and competence have a diverse, complex and intertwining real needs and priorities in society, and sustainability, which
relationship with project success, which should be studied more relates to the project effects on the future. Shenhar and Dvir
thoroughly. The literature continues to lack a broader empirical (2007) propose five slightly different dimensions of success:
basis (Grant and Pennypacker, 2006a, 2006b). project efficiency, impact on the customer, impact on the team,
This study aims to contribute to filling the research gap, business and direct success, and preparation for the future. For
evaluating both the relationship between PM training efforts Carvalho and Rabechini Junior (2015), there is also a
and project results and the relationship between PM context and sustainability dimension, but it relates to the impact of the
project results. PM contexts involve developing and using PM project on social and environmental aspects, more aligned with
practices and methodology, along with providing organization- the current triple bottom line literature.
al and administrative support. The methodological approach The distinction between the success of project management
involved a longitudinal study, combining quantitative and and that of its product/service is also an important issue in the
qualitative strategies. literature of project success emphasized by several authors
This article is divided into 5 sections. Section 2 summarizes (Barclay and Osei-Bryson, 2010, Carvalho and Rabechini
the theoretical framework. The methodological approach is Junior, 2015; Cooke-Davies, 2002, Pinto and Slevin, 1988,
presented in Section 3. Section 4 presents and discusses the Shenhar and Dvir, 2007).
results. Finally, Section 5 concludes and provides recommen-
dations for future studies. 2.2. Project management and success

2. Literature review A systematic PM consists of methods, toolkits and models.

It can be viewed as the sequential application of structured
2.1. Project success processes for the purpose of institutionalizing standardized
practices. Using a well-structured and well-implemented
Project success has been the target of fruitful discussions in approach, capabilities can be stored and transferred over time,
the project management literature (Carvalho and Rabechini space and context. Additionally, PM can make organizations
Junior, 2015) that reveals the social and political contextuali- less vulnerable to the loss of tacit knowledge stored in
zation of performance in project management (Sage et al., individual memories (Ibert, 2004).
2014). According to Carvalho et al. (2003), maturity models meet
The traditional view of project success is associated with these needs because they systematize project methods, tool
fulfilling time, cost and quality objectives (the iron triangle). packages and methodologies, proposing a continuous improve-
Financial criteria have been used to measure project perfor- ment model to manage the change between an organization's
mance, including economic return and cost/benefit analyses initial and desired statuses. In general, maturity models involve
(Archer and Ghasemzadeh, 1999) and profits (Shenhar and structuring managerial processes and the key areas in which
Dvir, 2007; Thomas et al., 2002). Another way to evaluate the the capabilities and practices to be developed—and the
benefits of PM is to analyze the margins of a company's key performance indicators—are grouped. The models may
ongoing projects (Patah and Carvalho, 2007). The most-often be structured according to proficiency levels, processes or
utilized project performance metrics are those related to domains by analyzing their repeatability and continuous
obtaining the initially planned schedule and cost values at the improvement.
end of the project (Gray, 2001; Katz and Allen, 1985; Larson Maturity models assume that organizations' knowledge and
and Gobeli, 1989; Ling, 2004; White and Fortune, 2002), in experience can be translated into procedures, roadmaps,
which—as expected—there is a consensus on the financial routines and databases, which leads to the configuration of a
issues involved (Archer and Ghasemzadeh, 1999; Patah and “collective brain” (Gareis and Huemann, 2000).
Carvalho, 2007; Thomas et al., 2002). There are several maturity models, most of which have been
However, various studies have investigated new dimensions proposed by consulting companies (Foti, 2002). Although most
of project success (Carvalho and Rabechini Junior, 2015; of the models in the PM area have been created over the last
Samset, 1998; Shenhar and Dvir, 2007; Barber, 2004; Ika, two decades, several maturity models already exist, such as the
2009; Jugdev and Muller, 2005). There seems to be no simple Capability Maturity Model (CMM) and Capability Maturity
definition for this construct, once it may be measured Model Integration (CMMI) (SEI, 2006), the Project Manage-
differently in different types of projects, from different ment Process Maturity Model (Ibbs and Kwak, 2000), the
perspectives, at different stages, and in absolute or relative Project Management Maturity Model (Kerzner, 2001), the
terms (Samset, 1998). Organizational Project Management Maturity Model (PMI,
It is a multidimensional construct (Carvalho and Rabechini 2008) and the PM Competence Model (Gareis and Huemann,
Junior, 2015; Samset, 1998; Shenhar and Dvir, 2007) and 2000).
different stakeholder groups have their own perceptions of When companies adopt systematic PM, they assume that
project success (Chou and Yang, 2012; Davis, 2014, Toor and increased PM maturity will generate better project performance.
Ogunlana, 2010; de Vries, 2009). Samset (1998) explores five However, the empirical results of this assumption are not yet

Please cite this article as: M.M. Carvalho, et al., 2015. Project management and its effects on project success: Cross-country and cross-industry comparisons, Int. J. Proj.
Manag. http://dx.doi.org/10.1016/j.ijproman.2015.04.004
 M.M. Carvalho et al. / International Journal of Project Management xx (2015) xxx–xxx 3

conclusive (Yazici, 2009). Studies on maturity models 2.3. Project management skills and success
implementation in the information technology (IT) industry,
specially related to the CMM and CMMI models, have Besner and Hobbs (2012) observe that the PMBoK guide
concluded that these models tend to develop higher quality seems to emphasize certain practices over others. The BoKs
software, a faster development cycle, higher productivity highlight hard skills, stressing the need for documentation,
(Dion, 1993), better organizational performance (Hersleb, measurement and control of a project during its life cycle.
1996) and increased project success (Jiang et al., 2004). In However, Carvalho (2014) highlights the importance of
their study, Jiang et al. (2004) also conclude that project soft skills in PM, especially related to communication and
success is only significant beyond CMM Level 3. stakeholders' management and skills. Jugdev et al. (2007)
In a cross-sector study in the USA, Yazici (2009) concludes corroborate and highlight the relationship between PM and a
that organizational performance is associated with PM maturity firm's capability based on the resource-based view.
but also that the latter influences organizational performance Sage et al. (2014) suggest that in project management,
when combined with organizational culture. Besner and Hobbs failure is often assumed to be an evidence of deficient
(2012, 2013) corroborate this finding and suggest that there are management, and thus, a problem that can be overcome by
different relationships among maturity, competency and project better management. However, the managerial BoKs and
success. According to Ghoddousi and Amini (2011)), among maturity models discussed previously focus mainly on hard
large construction companies in Iran, only those with high skills. Thus, more effort should be put into the soft side
maturity levels have been able to win contracts for international because, for implementing projects successfully, it is necessary
projects. to combine both hard and soft skills (Soderlund and Maylor,
In general, maturity models are strongly grounded in the 2012), considering organizational and behavioral issues such as
bodies of knowledge (BoKs) of PM associations and institutes. the support and commitment of middle and top management,
In the models, BoKs are a reference for management process training and careers in PM. Accordingly, efforts in PM maturity
groups, knowledge areas, and tools and practices, including the are important and must be combined with an understanding of
Project Management Body of Knowledge (PMBoK) (PMI, the manager and project team's cultural orientations and skills.
2013) and the ICB Competence Baseline (IPMA, 2006). Project manager is a critical success factor (CSF) in a project
Chou and Yang (2012) analyze the impact of the PMBoK (Archibald, 2003), providing direction, goals, motivational
guide on project performance, customer satisfaction and project support and assistance in resolving interpersonal and organiza-
success in the civil construction industry in Taiwan and tional issues (Rauniar and Rawski, 2012). Other authors
conclude that it is convenient to adopt PMBoK practices in corroborate and include training and education issues as key
this industry. Similarly, McHugh and Hogan (2011) present factors in PM (Cooke-Davies, 2002; Dai and Wells, 2004; Takey
the reasons to adopt an internationally recognized PM and Carvalho, 2015). Plaza and Rohlf (2008) go further and
methodology, as follows: the guarantee that the organization suggest that the choice of a training strategy has a significant
is employing what are considered to be best practices; external impact on project cost performance whereas Hong et al. (2011)
customers' requirements for using a recognized methodology; emphasize the impact of learning and knowledge sharing on
assistance in external recruitment and the availability of training; performance. For McHugh and Hogan (2011), training on
and support to the methodology. Conversely, Sanchez-Losado international BoKs supports PM methodology.
(2012) opposes the PMBoK guide and Lean Thinking in the According to Czuchry and Yasin (2003), effective executive
context of construction projects by arguing that projects with a knowhow exerts influence on the 3 modes (decisional, critical
particular level of uncertainty cannot be managed exclusively skills and technical) of a project life cycle. Technical know-how
through PMBoK but must incorporate aspects of Lean Thinking includes the importance of applying PM skills that are well
in a way that makes both methods compatible. documented in the PMBoK.
Some studies have tried to understand the impact of the In the IT context, Nfuka and Rusu (2011) highlight
implementation of these PM standards (Besner and Hobbs, both training and attracting, besides retaining competitive
2012, 2013; Chou and Yang, 2012; McHugh and Hogan, 2011) IT leadership and professionals as critical. For Ali and
and suggest a relationship between PM and success. Other Kidd (2013), one of the top-ranking CSFs for configuration
studies investigate the impact of the adoption of PM practices management professionals is certification and training in that
on performance (Besner and Hobbs, 2012, 2013; Chou and area. It might be relevant to note that the same CSFs—training
Yang, 2012; McHugh and Hogan, 2011), and all of the in PM methods and Project Management Professional (PMP)
referenced studies are based on the structure of the PMBoK, certification—hold top rankings in the PM context.
fourth edition (PMI, 2008). Thus, to assess whether organizational efforts in PM training
In determining PM context, we consider both PM areas, improve performance, we established the following hypothesis:
as suggested by the BoKs, and organizational enablers, as
suggested by the maturity models. Therefore, the following H2. Efforts in PM training are positively associated with
hypothesis is established: project success.

H1. The context of PM is positively associated with project The conceptual model and research hypotheses are shown in
success. Fig. 1.

Please cite this article as: M.M. Carvalho, et al., 2015. Project management and its effects on project success: Cross-country and cross-industry comparisons, Int. J. Proj.
Manag. http://dx.doi.org/10.1016/j.ijproman.2015.04.004
4 M.M. Carvalho et al. / International Journal of Project Management xx (2015) xxx–xxx

Due to the complex nature of interdependencies between the identify that the levels of perceived risk vary significantly in
independent construct context of PM and the efforts in PM three countries, once it is higher in New Zealand, as compared
training with the dependent construct project success, the to lower levels in Israel and Japan.
choice of control variables represented a particularly important More recently, Chou et al. (2013) focus on construction
part of the research model (see Fig. 1). In the research models, engineering projects and try to identify cross-country (Taiwan,
it was control for effects for country-specific, sector-specific Indonesia, and Vietnam) similarities and differences consider-
factors, and project complexity, while relying on a limited ing the contribution of the Project Management Body of
number of control variables. It is important to highlight that the Knowledge (PMBOK) to success. Empirical data obtained
model is designed from the project performer organizations in a cross-country comparison suggest modified models for
point of view, i.e., the organization that is investing resources in Taiwan, Indonesia, and Vietnam. Just 7 (20%) of the PMBOK
PM. techniques/tools/skills (TTSs) were common in the three
The first control variable is the national business environ- studied countries, namely: product analysis, alternatives
ment (country variable). More attention for cross-country identification, control charts, risk data quality analysis,
analysis has been given in studies related to international procurement negotiations, procurement performance reviews
development (ID) projects (Ahsan and Gunawan, 2010; Prasad and meeting quality requirement. The highest project success
et al., 2013) and suggests that cultural issues of the host country index (PSI) values were obtained for project risk management
should be considered, including aspects related to projects (PRM) in Taiwan, project human resource management
being more susceptible to political corruption (Khang and Moe, (PHRM) in Indonesia, and project procurement management
2008). Prasad et al. (2013) propose a framework that considers (PPM) in Vietnam. Le-Hoai et al. (2008) also study Vietnam
the characteristics of the external environment such as construction industry and point out the five most important
infrastructure, stakeholder variance, and flexibility and auton- causes of delay and cost overruns, and compare it with other
omy. Some studies focus more on the impact of culture countries and identify that the results are more similar to
(Chevrier, 2003) on international projects than on the impact of developing economies.
the stage of PM methodology development in particular These studies suggest the contextual variable of national
countries. However, as Ahsan and Gunawan (2010) suggest, business environment (country variable) is relevant. This
ID projects are different from traditional business projects, and expected homogeneity within a country such as culture and
make traditional project management tools in the developed PM externalities allowed us to perform an analysis controlling
world less appropriate. effects per country and thus, the following hypothesis is
Zwikael et al. (2005) identify differences in project proposed:
management style between the Japanese and the Israeli,
considering the nine knowledge areas of PMBoK third edition H3a. The national business environment significantly influ-
(the fifth edition considers 10 knowledge areas, PMI, 2013). ences project success.
The results reveal significant cultural differences between the
two countries. In Israel, project managers are more focused on The second control variable is the effect across industries
performing “Scope” and “Time”, while in Japan they focus on (Carvalho and Rabechini Junior, 2015; Ibbs and Kwak, 2000a,
“Communications” and “Cost”. Zwikael and Ahn (2011) also 2000b; Pennypacker and Grant, 2003; Raz et al., 2002; Zwikael

Fig. 1. Conceptual model and hypotheses. Note: To simplify the understanding of the model, some arrows were not shown. For example, there are six arrows at the
H1(+) position (going from the two dimensions of “PM context” to the three dimensions of “Project success”, there are three arrows at the H2(+) position, etc.

Please cite this article as: M.M. Carvalho, et al., 2015. Project management and its effects on project success: Cross-country and cross-industry comparisons, Int. J. Proj.
Manag. http://dx.doi.org/10.1016/j.ijproman.2015.04.004
 M.M. Carvalho et al. / International Journal of Project Management xx (2015) xxx–xxx 5

and Ahn, 2011). These studies have come up with controversial implementation processes in organizations that allow for
findings. Although some studies (Raz et al., 2002; Zwikael and temporal analysis and access to project data. As Thomas and
Ahn, 2011) have found significant effect of industries, other Mullaly (2008) suggest, it is difficult to obtain various types of
studies (Carvalho and Rabechini Junior, 2015; Pennypacker information from most organizations; thus, key aspects of the
and Grant, 2003; Ibbs and Kwak, 2000a, 2000b; Pennypacker study include access to company information, along with the
and Grant, 2003) were statistically inconclusive. Zwikael and company's willingness to participate in the study, and allow
Ahn (2011) identify that the level of project risk varies access to project values, project performance data, and audit
significantly across seven industries, in which engineering data related to PM methodology in its business units.
and construction projects have the highest levels of perceived The company's PM program is systemized by an organiza-
risk, while projects in the service industry have the lowest. tional matrix based on the best practices of BoKs, customized
Carvalho and Rabechini Junior (2015) study the moderate to the company's context. The implementation guide, which is
effect of industry on risk project management and on success, in its 4th edition, contains 60 best practices divided into 12
but the results show no significant effect of industries. In modules. The guide, which was released in 2001, is a worldwide
the studies on project management maturity (Grant and program for implementing a singular PM methodology.
Pennypacker, 2006a, 2006b; Ibbs and Kwak, 2000a, 2000b; Within the Southern Common Market, the company operates
Pennypacker and Grant, 2003), the sector does not seem to be in Argentina, Chile and Brazil; since 2003, the company's
statistically significant. This previous discussion on the Brazilian operation has been responsible for program implemen-
industry contingent variable suggests the following hypothesis: tation. There are 14 different business units in the region that
participate in the implementation program. The evolution of the
H3b. The industrial sector significantly influences project
business units within the company's PM program is evaluated by
success.
a central audit department. Therefore, this study analyses are
based on independent audits, not on the perceptions of people
Finally, various studies pointed out the relevance of
involved in PM.
analyzing the effect of project complexity on PM research.
The contingent effect of complexity has evolved in prior
3.1. Data collection
decades, influenced by Shenhar and Dvir's research (Shenhar,
2001; Shenhar and Dvir, 1996; Shenhar et al., 2002) that
Several sources of evidence were used in this study;
proposes the Diamond model, which includes four
however, three documents were decisive for the sample
dimensions—novelty, complexity, technology and stage—.
selection within each business unit and for project analysis:
Various authors argue that a project complexity influences its
standard sheets, cockpit and project status report. The standard
performance (Cleland and Ireland, 2006; Cooke-Davies, 2002;
sheets, provided by all business units, contain a list of each unit
Crawford et al., 2004; Larson and Gobeli, 1989; Schwalbe,
primary projects, results and periods and provide a general
2007; White and Fortune, 2002). Besner and Hobbs (2012,
overview of their projects. The cockpit is a consolidated report
2013) reinforce the influence of the variable project type both
with a general overview of each business unit, based on the
on adopting the practices and on the results. The last control
standard sheets report. Project status report is a detailed report
variable previously pointed out in this section as relevant
containing the project scope, categorization, primary financial
suggests the following hypothesis:
data, an analysis based on SWOT (strengths, weaknesses,
H3c. Project complexity interferes significantly with project opportunities and threats) specific to the project, the project
success. primary risks, as well as the opportunities and good practices
generated by the project. In addition, the authors participated in
3. Research methods monthly workshops conducted in Brazil with representatives of
all of the business units that implemented the program along the
The methodological approach involved a longitudinal field two years of data collection for this study. The workshops
survey conducted in the Southern Common Market business involved evaluations, mapping key processes, discussing the
units of a multinational organization in Argentina, Brazil, and requirements to be implemented by the business units,
Chile over a 3-year period. A total of 60% of the gross sales of presentations of the best practices found, follow-up reports,
the studied company come from projects, and there is a high audit results and information exchanges.
rate of product innovation. The company has analyzed eight The authors analyzed the number of projects per business
different business groups, finding that those groups represented unit to determine and to establish a valid sample. The
almost 45 billion Euros worldwide; this finding demonstrates availability of the data shown in Table 1 was analyzed for
the importance to the company of an effective PM system. The each project selected. As shown in the following section, 1387
company operates in different segments of the economy, valid project samples were obtained.
including manufacturing, components, safety, maintenance, Due to the scarcity of data and complex nature of
oil and gas, energy, healthcare, telecommunications, IT, and interdependencies between PM maturity and success, and PM
transportation. The company sells customized solutions to its Skills and success, the choice of control variables represented
customers. The selection of this research approach was guided a particularly important part of the model design. As all
by the need for a deeper analysis of the PM methodology business units have been operating for 10 years with the same

Please cite this article as: M.M. Carvalho, et al., 2015. Project management and its effects on project success: Cross-country and cross-industry comparisons, Int. J. Proj.
Manag. http://dx.doi.org/10.1016/j.ijproman.2015.04.004
6 M.M. Carvalho et al. / International Journal of Project Management xx (2015) xxx–xxx

Table 1
Project success measurement.
Cost variation Schedule variation Margin variation
Relative variation of budget Relative variation of the increase of Relative variation of the project
increases, which is measured as the the project schedule measured as the margin measured as the difference
difference in monetary value difference between the original between the final and the originally
between the original budget planned schedule in days and the forecasted margins, divided by the
forecasted and the actual final final schedule divided by the original margin. (%)
project budget and divided by the original schedule. (%)
original budget. (%)
Mean 9.51% 4.66% − 2.68%
Standard deviation 47.07% 7.42% 84.4%
Project performance Number of projects % Number of projects % Number of projects %
Negative 180 12.98 412 29.70 367 26.46
On target 667 48.09 1 0.07 505 36.41
Positive 540 38.93 974 70.23 515 37.13
Total 1387 100 1387 100 1387 100

organizational PM maturity model, this homogeneity allowed The variations were defined such that an increase in the
us to perform an analysis controlling effects for country- indicator designated a higher degree of success, as shown in
specific and sector-specific factors, while relying on a limited Eqs. (1) and (2).
number of control variables.
Thus, from the project performer organization perspective,
Cost variation ¼ budget−actual budget ð1Þ
the existence of a positive relation between the organizational
efforts to improve of project management and project success is
critical to sustain these efforts. (+) ➔ budget N actual ➔ the amount spent was lower
than planned ➔ higher project success
3.2. Operationalization of the dependent variables (−) ➔ budget b actual ➔ the amount spent was higher
than planned ➔ lower project success
As mentioned before, several studies have attempted to
final−forecasted
analyze how to measure project success (Belout and Gauvreau, Magin variation ¼ ð2Þ
2004; Berssaneti and Carvalho, 2015; Besner and Hobbs, 2006; forecasted
Bizan, 2003; Carvalho and Rabechini Junior, 2015; Dvir et al.,
2003; Gray, 2001; Kendra and Taplin, 2004; Lipovetsky et al., (+) ➔ budget b actual ➔ the amount earned was higher
2005; Raz et al., 2002; Repiso et al., 2007; Samset, 1998; than planned ➔ higher project success
Shenhar and Dvir, 2007). Thus, project success can be (−) ➔ budget N actual ➔ the amount earned was lower
operationalized through operational and strategic performance than planned ➔ lower project success
indicators from different dimensions. Because of the lack of
consensus in defining project success concept and the
quantitative approach of this research method, the traditional As all subsidiaries have been operating for 15 years or more
dimension of project efficiency was adopted. Moreover, this prior to the observation period; the subsidiary age was deemed
research is designed from the project performer organization irrelevant. Subsidiaries have also exhibited homogeneity with
perspective. Although it represents just part of the project respect to the product portfolio, brand and high-level pricing
success discussion, it is still considered central to the strategy (allowing a modest degree of discretion for individual
measurement of project success (Papke-Shields et al., 2010; client negotiations).
Berssaneti and Carvalho, 2015). This study focuses on the
efficiency dimension also because data on other dimensions 3.3. Operationalization of the independent variables
were not available in a systemized and documented fashion.
The reports concern on cost, schedule, and margins apply the To analyze the intensity of use of PM methods in the business
same measurement methodology across the distinctive business units, the central audit report evaluated 11 variables of the
units studied. However, for customer satisfaction measurement, worldwide program guidelines on a scale of 0% to 100%. The
the business units employ different methodologies that could modules addressed different aspects of PM maturity and
not be compared. As all the selected projects were accepted methodology using the consolidated areas of the BoKs. Thus,
without legal issues by the project client, we could neutralize two latent variables relating to PM maturity and PM methodology
the effect of this potential success dimension. Table 1 presents were created, called PM enablers and PM areas, respectively (see
project success measures—including project descriptions and Table 2). The construct of PM enablers was modeled as a
project units—and distribution according to a project perfor- formative latent variable (Jarvis et al., 2003), whereas PM areas
mance in those measures. were designed as first-order reflective because they are designed

Please cite this article as: M.M. Carvalho, et al., 2015. Project management and its effects on project success: Cross-country and cross-industry comparisons, Int. J. Proj.
Manag. http://dx.doi.org/10.1016/j.ijproman.2015.04.004
 M.M. Carvalho et al. / International Journal of Project Management xx (2015) xxx–xxx 7

Table 2
Measurement of the use of project management methods.
Variables Description Unit
PM enablers PM processes roles • Definition of the business unit's processes, establishing the main 0 to 100%
benchmarks of project and documents
PM web portal • Creation of a portal with easy to access and to obtain information, to 0 to 100%
exchange project management information and best practices
Benchmarking and implementation status • Management of the information transfer processes and program 1 to 100%
implementation in the other areas of the organization
PM assessment • Self-evaluation of project management in the company using a 2 to 100%
standard tool for all the business units
Small project • Analysis of the requirements and recommendations applicable to 3 to 100%
the management of small projects
PM areas Contract mgt • Management of the project contract, from the pre-sale until… (missing text) 4 to 100%
People mgt • Management of the project team from the career definition and team stimulus 5 to 100%
to the implementation of system of goals and incentives for results achieved
Quality mgt • Project quality management 6 to 100%
Knowledge mgt • Knowledge Management about project management 7 to 100%
Procurement Mgt • Project-related acquisitions management 8 to 100%
Project Control • Project control, both financial and technical 9 to 100%

Mean standard deviation

PM methodology* 77.18% 3.29%
PM training budget** 238,609.00 270,344,00
Number of the PMPs** 7 1.16
Notes: *Extracted from central audit department. **Extracted from PMO data system (in Real$).

as parts of a whole, as suggested in the main BoKs. Table 2 Project distribution according to complexity categories is
presents the variables of both PM enablers and PM areas, also shown in Table 3. To use this variable in the model
including their descriptions and units of measurement. estimation, the variable was re-coded from 1 to 4, and the
The effort put into PM training and capability development highest score (4) was attributed to the most complex projects
was measured using 2 indicators. First, the financial value in (A). These data were obtained from the Latin America PMO
Brazilian currency (R$) of the investment in PM training at all data system.
available levels (basic, intermediate and advanced) of the The nominal variables (country and industry) were opera-
business unit. Second, the number of project managers in the tionalized as dummy variables, consistent with Falk and Miller
business unit responsible for leading the project that are (1992). Therefore, a value equal to 1 was attributed to the
PMP-certified, which is an internationally recognized certifica- projects that belonged to the same category and 0 to all the
tion given by the PMI to project managers. Table 2 also presents other categories, as detailed in Table 4.
statistics related to these latent variables. For confidentiality
purposes, only a summary of these data is presented.
3.4.1 . Data analysis
The data were analyzed according to their frequency
3.4. Operationalization of the control variables distribution, descriptive statistics and bivariate analyses
(cross-tables and correlations), and the full model was
Within the company, projects are classified into 4 categories evaluated using partial least squares path modeling (PLS-PM).
according to the system proposed in the worldwide program PLS-PM was considered the most adequate method for various
guidelines. The projects are evaluated according to four criteria reasons. First, it was possible to incorporate nominal variables
that are unfolded into variables, as shown in Table 3. into the structural model, as was the case for the country and
Based on the analysis of the projects according to all of the industry control variables. Second, it was possible to incorpo-
criteria presented in Table 3, projects are classified into four rate variables measured by formative indicators, as was the case
categories of complexity. Each category has variable scoring for the PM enablers and the control variables. Finally, it
from specific minimum to maximum, and the total project depended neither on the normality of the variables (as was the
complexity score is the sum of all the variables (see Table 3). case of LISREL) nor on the normality of the residuals because
The maximum score that a project can achieve is 1360. Each the significance probabilities were estimated by bootstrap
project is classified, according to its score, into four categories (Henseler et al., 2009; Tenenhaus et al., 2005).
from A to D. Category A groups projects with scores higher The significances were estimated by bootstrap directly on the
than 1000 points. Categories B, C and D group projects from SmartPLS software, using 1000 resamplings (Tenenhaus et al.,
500 to 999, 60 to 499 and 0 to 59 points, respectively. 2005).

Please cite this article as: M.M. Carvalho, et al., 2015. Project management and its effects on project success: Cross-country and cross-industry comparisons, Int. J. Proj.
Manag. http://dx.doi.org/10.1016/j.ijproman.2015.04.004
8 M.M. Carvalho et al. / International Journal of Project Management xx (2015) xxx–xxx

Table 3
Criteria and variables adopted by the company to categorize projects.
Criteria Variables Unit
Financial – Financial volume; 1–340
– Percentage value of the estimated risks;
– Sales margin of the project; and
– Percentage value of investment in research and development or engineering for the project.

Contractual – Contractual position of the company in the project; 1–340

– Number of external partners contractually associated;
– Internal partners of the company; and
– Degree of the relationship with the customer.
Technical complexity – Clarity of the definition of the product or scope of the project and 1–340
– Need for a new technological development.
Organizational considerations – Type of project (supply, system or turn-key); 1–340
– Contractual complexity;
– Strategic significance of the project for the company; and
– Strategic relevance for the customer.

Category Score Value used in the structural research model Number of Projects %
A 1000 to 1360 4 209 15.07
B 500 to 999 3 547 39.44
C 60 to 499 2 598 43.11
D 0 to 59 1 33 2.38
Total 1387 100

4. Results Regarding the project complexity control variable, the

sample was composed of predominantly type C (598, or
4.1. Demographics 43.1%) and B (547, or 39.4%) projects, contributing 82.5% of
the total. Therefore, the extremes—type A projects of high
We obtained 1387 valid samples of projects in the following complexity (209) and type D projects of low complexity
10 industries: energy (713), health care (218), manufacturing (33)—were the minority.
(152), transportation (105), IT (82), telecommunications (43),
components (37), maintenance (18), safety (13) and oil and gas 4.2. Evaluation of the measuring model
(6). Therefore, more than half of the sample was from the
energy industry (51.4%). 4.2.1. Constructs measured with a single indicator
The sample was predominantly composed of projects Cost variations, schedule variations, margin variations and
developed in Brazil (823 projects, or 59.3%), followed by complexity were measured with only one indicator. Thus, they
Argentina (294) and Chile (270). cannot be interpreted as latent variables but as the indicator

Table 4
Industry and country categories coded by formative dummy formative indicators.
Industry category D1 D3 D4 D5 D6 D7 D7 D9 D10 n %
Components 1 0 0 0 0 0 0 0 0 37 3%
Energy 0 0 0 0 0 0 0 0 0 713 51%
Manufacturing Industry 0 1 0 0 0 0 0 0 0 152 11%
Maintenance 0 0 1 0 0 0 0 0 0 18 1%
Health care 0 0 0 1 0 0 0 0 0 218 16%
Oil and gas 0 0 0 0 1 0 0 0 0 6 0%
Safety 0 0 0 0 0 1 0 0 0 13 1%
Information technology 0 0 0 0 0 0 1 0 0 82 6%
Telecommunications 0 0 0 0 0 0 0 1 0 43 3%
Transportation 0 0 0 0 0 0 0 0 1 105 8%

Country category D_arg D_chi n %

Argentina 1 0 294 21%
Brazil 0 0 823 59%
Chile 0 1 270 19%
Note: Any category might be used as a reference (Falk and Miller, 1992), we chose the industry category energy and country category Brazil because it is the category
with the highest numbers of projects.

Please cite this article as: M.M. Carvalho, et al., 2015. Project management and its effects on project success: Cross-country and cross-industry comparisons, Int. J. Proj.
Manag. http://dx.doi.org/10.1016/j.ijproman.2015.04.004
 M.M. Carvalho et al. / International Journal of Project Management xx (2015) xxx–xxx 9

Table 5
Loading factor.
Reflective indicators Y1 Y2 Y3 M1 PMEf PMA T Statistics P value
Y1—Cost variation 1.000 − 0.231 0.018 0.059 − 0.062 − 0.033 – –
Y2—Schedule variation − 0.231 1.000 0.090 − 0.160 0.234 0.163 – –
Y3—Margin variation 0.018 0.090 1.000 − 0.124 0.051 − 0.019 – –
M1—Coplexity 0.059 − 0.160 − 0.124 1.000 − 0.307 − 0.102 – –
PMEf1—Investments in training − 0.037 0.195 0.054 − 0.292 0.992 0.041 846.7 0.000
PMEf2—Number of PMPs − 0.079 0.260 0.047 − 0.314 0.996 0.086 2711.4 0.000
PMA1—Contract mgt − 0.012 0.077 − 0.005 − 0.082 − 0.012 0.884 30.4 0.000
PMA2—Knowledge mgt − 0.023 0.151 − 0.010 − 0.155 0.116 0.900 41.4 0.000
PMA3—Quality mgt − 0.034 0.106 0.014 − 0.058 − 0.047 0.633 7.3 0.000
PMA4—Project control − 0.030 0.141 − 0.045 − 0.013 0.091 0.685 9.3 0.000
PMA5—Procurement mgt Note 1
PMA6—People mgt Note 1
Source: Data from this study.
Note 1: Item removed from the measurement model because of low loading factor.

itself. Evaluation of convergent validity and reliability was 4.2.3. Constructs measured by multiple formative indicators
performed for the cases in which multiple reflective indicators For the control variables (country and industry), each
were used; therefore, that evaluation was not performed for category was coded as a dummy variable, which was later
these 4 constructs. modeled as a formative indicator (Falk and Miller, 1992). In
these cases, there was no evaluation of validity and reliability
4.2.2. Constructs measured by multiple reflective indicators because correlation among the formative indicators was neither
necessary nor desired. Table 7 presents the outer weight for all
PM efforts in training and PM areas show loading factors
formative indicators.
that are significant (p b 0.001) and higher than 0.6 (Table 5),
For the control variables, content validity is considered
which result in an Average Variance Extracted (AVE) of 0.99
adequate because all of the relevant categories are represented
and 0.62; this is much higher than the recommended minimum
value of 0.5. There is convergent validity for these two latent in the model estimation (see Fig. 1).
variables with reflective indicators; the Cronbach's alpha is PM enablers are the third construct related to formative
higher than 0.75, the composite reliability is higher than 0.82 modeling. It was assumed that the construct was defined by its
for both constructs (Chin and Newsted, 1999; Henseler et al., five indicators.
2009; Tenenhaus et al., 2005).
4.3. Evaluation of the structural model: hypothesis testing
Discriminant validity was evaluated in two ways—at the
indicator level and at the latent variable level. In the first case,
The project success was operationalized as three comple-
Table 5 shows that the indicators have loading factors for their
respective constructs that are higher than in any other latent mentary variables (evaluating different dimensions of success).
variable. In the second case, Table 6 shows that for all the latent The correlations were analyzed to evaluate the possibility of
variables, the square root of the AVE is higher than the modeling success as a latent variable, showing that these
correlation among them (Henseler et al., 2009; Tenenhaus et al., variables have very low correlations (see Table 6), which
means that it is not feasible to model success as a reflective
2005), which confirms discriminant validity.
latent variable (Henseler et al., 2009; Mackenzie et al., 2005;
Tenenhaus et al., 2005). Furthermore, modeling success as a
Table 6
Correlation matrix between the latent variables (n = 1387).
formative latent variable would hinder the interpretation of the
results with outer weights that are positive, negative or equal to
Construct Y1 Y2 Y3 M1 PMEf PMA PMEn 0. Therefore, we decided to evaluate each project success
Y1-Cost 1 indicator separately (see Fig. 1).
Y2-Schedule −0.231 1 The structural model tested on the SmartPLS 2.0.M3 software
Y3-Margin 0.018 0.090 1 contained all of the relationships shown in Fig. 1; however, to
M1-Complexity 0.059 −0.160 −0.124 1 present the results more clearly than in the figure with all of the
PMEf-PM Efforts −0.062 0.234 0.051 −0.307 0.994 estimated coefficients and arrows, we decided to split the figure
PMA-PM areas −0.033 0.163 −0.019 −0.102 0.067 0.784 into three tables, one for each project success indicator.
PMEn-PM Enablers −0.029 0.166 −0.021 0.085 −0.013 0.591 (Note2) In addition, we opted to test four versions of the structural
Note 1: The numbers in the diagonal contains the AVE square root. and all model to evaluate the increase of R2 from the model with the
values higher than |0.08| are significant at the 1% level. control variables (Models 1 to 3 in Table 8) to the full model
Note 2: Values higher than |0.08| are significant at the 1% level. (Model 4 in Table 8).

Please cite this article as: M.M. Carvalho, et al., 2015. Project management and its effects on project success: Cross-country and cross-industry comparisons, Int. J. Proj.
Manag. http://dx.doi.org/10.1016/j.ijproman.2015.04.004
10 M.M. Carvalho et al. / International Journal of Project Management xx (2015) xxx–xxx

Table 7
Outer weights.
Formative indicators PMEn PM enablers Country Industry T statistics P value (Note 2)
PMEn1_PM assessment 0.442 1.63 0.1035
PMEn2_PM web portal 0.276 1.80 0.0727
PMEn3_Processes roles 1.224 4.71 0.0000
PMEn4_PM benchmarking − 0.351 1.47 0.1413
PMEn5_PM small projects 0.336 1.32 0.1884
D_Argentina 1.032 101.7 0.0000
D_Brazil (Note 1)
D_Chile 0.331 3.64 0.0003
D1—Components 1.000 1.10 0.2697
D2—Energy (Note 1)
D3—Manufacturing industry 1.002 4.24 0.0000
D4—Maintenance 1.003 1.05 0.2940
D5—Health care 1.004 7.75 0.0000
D6—Oil and gas 1.005 0.94 0.3471
D7—Safety 1.006 2.95 0.0033
D8—Information technology 1.007 16.44 0.0000
D9—Telecommunications 1.008 0.88 0.3800
D10—Transportation 1.009 11.14 0.0000
Note 1: Item represents the reference category (see Tables 5 and 6).
Note 2: For formative indicators, the outer weights and their significance probabilities are not interpretable when there is multicollinearity among them; nevertheless,
the structural coefficients are not biased.

Table 8
Cost, schedule and margin variation as dependents variables.
Dependent variable Independent variable Hypothesis Path coefficient Standard error T statistics P value R2 Adjusted R2
Cost Model 1 Country 0.154 0.026 5.9 0.0000 2.4% 2.4%
Model 2 Industry 0.244 0.040 6.1 0.0000 6.0% 6.0%
Model 3 Country 0.092 0.026 3.5 0.0005
Industry 0.218 0.038 5.7 0.0000 6.7% 6.6%
Model 4 PM Enablers H1a′ x −0.004 0.029 0.1 0.8926
PM Areas H1b′ x 0.001 0.018 0.0 0.9609
PM Efforts H2′ x 0.034 0.022 1.6 0.1151
Country H3a′ v 0.102 0.035 2.9 0.0035 7.2% 6.8%
Industry H3b′ v 0.222 0.039 5.7 0.0000
Complexity H3c′ v 0.076 0.023 3.3 0.0009
Schedule Model 1 Country − 0.322 0.027 11.8 0.0000 10.3% 10.3%

Model 2 Industry − 0.535 0.023 23.1 0.0000 28.6% 28.6%

Model 3 Country −0.183 0.030 6.2 0.0000
Industry −0.482 0.026 18.8 0.0000 31.7% 31.6%
Model 4 PM Enablers H1a′ v 0.101 0.039 2.6 0.0092
PM Areas H1b′ x 0.045 0.028 1.6 0.1129
PM Efforts H2″ v 0.079 0.022 3.7 0.0003
Country H3a″ v −0.106 0.027 4.0 0.0001 36.8% 36.5%
Industry H3b″ v −0.496 0.026 19.0 0.0000
Complexity H3c″ v − 0.161 0.024 6.8 0.0000
Margin Model 1 Country 0.044 0.042 1.04 0.2974 0.2% 0.2%
Model 2 Industry 0.015 0.052 0.29 0.7733 0.0% 0.0%

Model 3 Country 0.043 0.029 1.49 0.1375

Industry 0.003 0.029 0.09 0.9246 0.2% 0.0%
Model 4 PM Enablers H1a‴ x 0.021 0.028 0.7 0.4566
PM Areas H1b‴ x − 0.033 0.026 1.3 0.2084
PM Efforts H2‴ x 0.054 0.029 1.8 0.0693
Country H3a‴ v 0.077 0.037 2.1 0.0369 2.1% 1.6%
Industry H3b‴ x −0.008 0.030 0.3 0.7951
Complexity H3c‴ v −0.118 0.039 3.0 0.0026
Note: The model was estimated using the SmartPLS 2.0.M3 software (Ringle et al., 2005).

Please cite this article as: M.M. Carvalho, et al., 2015. Project management and its effects on project success: Cross-country and cross-industry comparisons, Int. J. Proj.
Manag. http://dx.doi.org/10.1016/j.ijproman.2015.04.004
 M.M. Carvalho et al. / International Journal of Project Management xx (2015) xxx–xxx 11

Table 8 shows that the control variables (country H3a′ and success index value in Brazil (p-value = 0.000), and margin
industry H3b′) explain most of the variance of the cost variation success index value also in Argentina (p-value = 0.000).
(6.6%). The increase in R2 is only 0.2% from Model 3 to Model Finally, the cross-country comparison also suggests different
4, indicating that although complexity (H3c′) has a statistically degree of PM methodology implementation and investments in
significant coefficient, its contribution to explaining cost is Argentina, Brazil and Chile. The highest average number of
indeed negligible (Hair, 2005), once the increase in R2 is only PMP by project was obtained in Brazil (p-value = 0,000);
0.2%. similarly, the degree of PM methodology and investments in
Table 8 shows that the control variables (country and Brazil is higher than in the other two studied countries.
industry) explain most of the variance of the schedule variation Our study shows that the country control variable explains a
(31.6%). In Model 4, besides the control variables (country and significant amount of the effect on performance. In this sense,
industry), three others are shown to have significant coeffi- our study reinforces the difficulty in explaining the project
cients, which are PM enablers (H1a″), PM efforts (H2″) and success variable as suggested by Sage et al. (2014). The country
complexity (H3c″), but these improve the explanation of the control variable can combine several factors, including a
schedule variance only by 5%. specific PM environment such as the availability of skilled
The third project success indicator is margin variation, and PM professionals, the presence of PM professional associations
once again, the predictor variables show low predictive power and the presence of world-class companies in the PM area, but
(see Table 8), indicating that its contribution to explaining also economic and political issues among other factors with a
margin is only 1.6%. Of all the predictor variables, only high explanatory effect. Considering the three Latin American
country (H3a‴), complexity (H3c‴) and PM Efforts (H2″) have countries studied, Brazil has a leading role in the PM area
a significant coefficient. because it has the most chapters of the main associations (PMI
and IPMA), and has more than 15,000 PMPs (Carneiro, 2014),
which together demonstrate the interest in developing PM
5. Discussion methodology in that country, as corroborated by our empirical
data.
Table 8 summarizes the findings by describing the Similar analysis was performed for cross-sector comparison;
significant impact of the national environments (Brazil, Chile however, just on the degree of PM methodology were
and Argentina), project complexity and industries on all of the significant differences found across sectors. The highest the
project performance measures investigated. Our study indicates degree of PM methodology was obtained in the Telecommu-
that all of the contingent variables have a significant relation nications sector (p-value = 0.004). We designed distinctive
with all of the performance measures studied, but the effect models that considered the three performance measures studied.
magnitude is not the same. The schedule model is affected by PM variables. Both PM
The findings confirm the descriptions in the literature enablers (that relates to PM maturity) and PM efforts in training
highlighting the impact of complexity (Cleland and Ireland, have a positive significant impact on schedule performance. It
2006; Cooke-Davies, 2002; Shenhar, 2001; Shenhar and Dvir, is interesting to note that PM areas do not show a significant
1996; Shenhar et al., 2002; White and Fortune, 2002) and impact on any project performance measure. These two
industry sectors on project performance (Patterson et al., 1999; variables relate more to the soft side of PM that involves
Raz et al., 2002). However, the findings add to the literature in managing stakeholders' roles, individual skills and capabilities
two primary ways. First, the current literature poorly explores and the notion of project ecology (Grabher, 2004). Most PM
whether the national environment (Brazil, Chile and Argentina) approaches have disregarded the soft side of PM (Sharma and
has a strong impact on project performance. Second, the Gupta, 2012; Soderlund and Maylor, 2012) in favor of focusing
schedule performance measure makes more sense for these on the hard side, characterized in our study as the PM areas
contingent variables than for the other performance measures, variable.
considering their effect on performance. Although much effort from professional associations,
Several empirical data suggest significant differences across companies and scholars has resulted in the consolidation of
the three studied countries. First, the cross-country comparison the discipline of PM over time, that effort continues to show
suggests modified project complexity profile for Argentina, weak results. There are several implications for practice. The
Brazil and Chile, because significant differences between absence of effect of the PM areas variable with respect to the
project proportions among the four complexity levels were three performance measures studied is consistent with
found (using chi-square test, with p-value = 0000). The highest Ahlemann et al. (2013), who suggest that the prescriptive
proportion of complex projects (level 4 class A) were obtained characteristics of PM approaches lead to numerous problems,
in Chile (56%). such as non-acceptance in practice, limited effectiveness and
Second, the cross-country comparison also suggests ambiguous application scenarios.
modified project success profile for Argentina, Brazil and The cost and margin models are not affected by PM
Chile, because significant differences between medians in the variables. The lack of impact of these variables naturally leads
three success indicators applied were found (using Mood to questions about the way PM is implemented in organiza-
median test). The highest project cost success index values tions, and we remain far from a definitive explanation. The
were obtained in Argentina (p-value = 0.000), schedule issue of PM implementation may be partially explained by

Please cite this article as: M.M. Carvalho, et al., 2015. Project management and its effects on project success: Cross-country and cross-industry comparisons, Int. J. Proj.
Manag. http://dx.doi.org/10.1016/j.ijproman.2015.04.004
12 M.M. Carvalho et al. / International Journal of Project Management xx (2015) xxx–xxx

the research of Jiang et al. (2004) on maturity, which the References

results suggest that tangible results are only perceived
beyond a certain degree of maturity (CMM Level 3). Ahlemann, F., El Arbi, F., Kaiser, M.G., Heck, A., 2013. A process framework
for theoretically grounded prescriptive research in the project management
Moreover, it is necessary to invest more time and effort in
field. Int. J. Proj. Manag. 31, 43–56.
the soft side (Soderlund and Maylor, 2012) and to conduct Ahsan, K., Gunawan, I., 2010. Analysis of cost and schedule performance of
further research. international development projects. Int. J. Proj. Manag. 28 (1), 68–78.
The limitations of the present research lie in the nature of the Ala-Risku, T., Kärkkäinen, M., 2006. Material delivery problems in construction
research methodology. This study was developed in Latin projects: a possible solution. Int. J. Prod. Econ. 104 (1), 19–29.
Ali, U., Kidd, C., 2013. Critical success factors for configuration management
America, which means that there are some biases that make it
implementation. Ind. Manag. Data Syst. 113 (2), 250–264.
difficult to generalize the study conclusions to developed Archer, N.P., Ghasemzadeh, F., 1999. An integrated framework for project
countries. Nevertheless, the large sample of more than 1000 portfolio selection. Int. J. Proj. Manag. 17 (4), 207–216.
projects in different sectors and countries offers important Archibald, R.D., 2003. Managing High-Technology Programs and Projects.
insights for further research and creates a basis for comparison John Wiley, New York.
Aubry, M., Hobbs, B., 2010. A fresh look at the contribution of project
with other countries. Moreover, the results pointed out the
management to organizational performance. Proj. Manag. J. 42 (1), 3–16.
importance of cross-countries and cross-sector analysis to better Barber, E., 2004. Benchmarking the management of projects: a review of
understand the variables that can moderate the relation between current thinking. Int. J. Proj. Manag. 22 (4), 301–307.
project management and success. Barclay, C., Osei-Bryson, K.-M., 2010. Project performance development
framework: an approach for developing performance criteria & measures for
information systems (IS) projects. Int. J. Prod. Econ. 124, 272–292.
Belout, A., Gauvreau, C., 2004. Factors influencing project success: the impact
6. Conclusions of human resource management. Int. J. Proj. Manag. 22 (1), 1–11.
Berssaneti, F.T., Carvalho, M.M., 2015. Identification of variables that impact
This paper explored three research hypotheses. The research project success in Brazilian companies. Int. J. Proj. Manag. 33 (3), 638–649.
hypotheses related to contingent approach are validated for all Besner, C., Hobbs, B., 2006. The perceived value and potential contribution of
project management practices to project success. Proj. Manag. J. 37 (3),
performance measures. The study shows that national environ-
37–48.
ment plays a key role in project performance, with an increased Besner, C., Hobbs, B., 2012. An empirical identification of project management
effect on performance in the country (Brazil) where PM toolsets and a comparison among project types. Proj. Manag. J. 43 (5),
methodology is at a more developed stage (as compared to 24–46.
Chile and Argentina) with respect to several aspects, such as Besner, C., Hobbs, B., 2013. Contextualized project management practice: a
cluster analysis of practices and best practices. Proj. Manag. J. 44 (1), 17–34.
PM associations, certified professionals and regulations. This
Bizan, O., 2003. The determinants of success of R&D projects: evidence from
effect is poorly addressed in the current PM literature, and American-Israeli research alliances. Res. Policy 32 (9), 1619–1640.
further research is needed for an in-depth understanding of how Buchanan, J., 2008. Measuring Up. PM Network, Project Management Institute.
this variable impacts performance. The effects of complexity Carneiro, M., 2014. Uma entrevista com Margareth Carneiro: Primeira latino-
and industry sectors confirm recent streams of PM literature. americana no Board do PMI. Rev. Mundo PM 10 (55), 24–27.
Carvalho, M.M., 2014. An investigation of the role of communication in IT
The research hypotheses related to PM enablers, PM efforts
projects. Int. J. Oper. Prod. Manag. 34 (1), 36–64.
in training and capabilities development and PM areas still Carvalho, M.M., Rabechini Junior, R., 2015. Impact of risk management on
show weak effects on performance, except for the schedule project performance: the importance of soft skills. Int. J. Prod. Res. 53 (2),
success indicator. The PM enablers and PM efforts in training 321–340.
stand out and reinforce the importance of the soft side of PM. Carvalho, M.M., Laurindo, F.J.B., Pessôa, M.S.P., 2003. Information technology
project management to achieve efficiency in Brazilian companies. In: Kamel,
This paper provides new contributions to the current
S. (Ed.), Managing Globally with Information Technology, pp. 260–271
literature in two ways: it provides an understanding of the (Hershey).
effect of PM enablers and PM efforts on project performance Chevrier, S., 2003. Cross-cultural management in multinational project groups.
and it provides an understanding of the contingent effect of J. World Bus. 38 (2), 141–149.
national business environments. Chin, W.W., Newsted, P.R., 1999. Structural Equation Modeling Analysis With
Small Samples Using Partial Least Squares. In: Hoyle, R.H. (Ed.), Sage
Publications, Thousand Oaks, CA, pp. 307–341.
Chou, J.S., Yang, J.G., 2012. Project management knowledge and effects on
Conflict of interest construction project outcomes: an empirical study. Proj. Manag. J. 43 (5), 47–67.
Chou, J.-S., Irawan, N., Pham, A.-D., 2013. Project management knowledge of
None. construction professionals: cross-country study of effects on project
success. J. Constr. Eng. Manag. 139 (11) (04013015-1-04013015-15).
Cleland, D.I., Ireland, L.R., 2006. Project Management Strategic Design and
Implementation. 5th ed. McGraw Hill, New York, NY.
Acknowledgments Cooke-Davies, T.J., 2002. The “real” success factors on projects. Int. J. Proj.
Manag. 20 (3), 185–190.
The authors gratefully acknowledge the financial support Crawford, L., Hobbs, B., Turner, R., 2004. Project categorization systems and
their use in organizations: an empirical study. In: Slevin, D.P., Cleland, D.I.,
of the Brazilian research funding agencies CNPq (National
Pinto, J.K. (Eds.), Innovations Project Management Research (Pennsylvania:
Council for Scientific and Technological Development, Newton Square).
311293/2013-1) and FAPESP (São Paulo Research Founda- Czuchry, A.J., Yasin, M.M., 2003. Managing the project management process.
tion, 2012/10117-0). Ind. Manag. Data Syst. 103 (1), 39–46.

Please cite this article as: M.M. Carvalho, et al., 2015. Project management and its effects on project success: Cross-country and cross-industry comparisons, Int. J. Proj.
Manag. http://dx.doi.org/10.1016/j.ijproman.2015.04.004
 M.M. Carvalho et al. / International Journal of Project Management xx (2015) xxx–xxx 13

Dai, C.X., Wells, W.G., 2004. An exploration of project management office Kerzner, H., 2001. Project Management—a Systems Approach to Planning,
features and their relationship to project performance. Int. J. Proj. Manag. 22 Scheduling, and Controlling. John Wiley & Sons, New York.
(7), 523–532. Khang, D.B., Moe, T.L., 2008. Success criteria and factors for international
Davis, K., 2014. Different stakeholder groups and their perceptions of project development projects: a life-cycle-based framework. Proj. Manag. J. 39 (1),
success. Int. J. Proj. Manag. 32 (2), 189–201. 72–84.
de Vries, J., 2009. Assessing inventory projects from a stakeholder perspective: Larson, E., Gobeli, D., 1989. Significance of project management structure on
results of an empirical study. Int. J. Prod. Econ. 118, 136–145. development success. IEEE Trans. Eng. Manag. 36 (2), 119–125.
Dion, R., 1993. Process improvement and the corporate balance sheet. IEEE Le-Hoai, L., Lee, Y.D., Lee, J.Y., 2008. Delay and cost overruns in vietnam
Softw. 10 (4), 28–35. large construction projects: a comparison with other selected countries. KSCE
Dvir, D., Raz, T., Shenhar, A., 2003. An empirical analysis of the relationship J. Civ. Eng. 12 (6), 367–377.
between project planning and project success. Int. J. Proj. Manag. 21 (2), 89–95. Ling, F.Y.Y., 2004. How project managers can better control the performance
Falk, R.F., Miller, N.B., 1992. A Primer for Soft Modeling. The University of of design-build projects. Int. J. Proj. Manag. 22 (6), 477–488.
Akron Press, Ohio. Lipovetsky, S., Tishler, A., Dvir, D., Shenhar, A., 2005. The relative importance of
Foti, R., 2002. Maturity noun 21st century. Synonym: survival. PMNework, project success dimensions. R&D Manag. 27 (2), 97–106.
Upper Darby, PA 21(9), pp. 14–20. Mackenzie, S.B., Podsakoff, P.M., Jarvis, C.B., 2005. The problem of
Gareis, R., Huemann, M., 2000. Project management competences in the measurement model misspecification in behavioral and organizational
project-oriented organization. In: Turner, J.R., Simister, S.J. (Eds.), research and some recommended solutions. J. Appl. Psychol. 90 (4), 710–730.
Published in the Gower Handbook of Project Management. Aldershot, McHugh, O., Hogan, M., 2011. Investigating the rationale for adopting an
Gower, pp. 709–721. internationally-recognised project management methodology in Ireland: the
Ghoddousi, P., Amini, Z., 2011. Hosseini, MR.,2011. A survey on the maturity view of the project manager. Int. J. Proj. Manag. 29 (5), 637–646.
state of Iranian grade one construction companies utilizing OPM3 maturity Mullaly, M., 2006. Longitudinal analysis of project management maturity. Proj.
model. Technics Technologies Education. Management 6 (1), 69–77. Manag. J. 36 (3), 62–73.
Grabher, G., 2004. Learning in projects, remembering in networks? Commu- Nfuka, E.N., Rusu, L., 2011. The effect of critical success factors on IT
nality, sociality and connectivity in project ecologies. European Urban and governance performance. Ind. Manag. Data Syst. 111 (9), 1418–1448.
Regional Studies 11, 103–123. Papke-Shields, K.E., Beise, C., Quan, J., 2010. Do project managers practice
Grant, K.P., Pennypacker, J.S., 2006. Project management maturity: an what they preach, and does it matter to project success? Int. J. Proj. Manag.
assessment of project management capabilities among and between selected 28 (7), 650–662.
industries. IEEE Trans. Eng. Manag. 53 (1), 59–68. Patah, L.A., Carvalho, M.M., 2007. Quantifying the Value of Project
Gray, R., 2001. Organizational climate and project success. Int. J. Proj. Manag. Management: the Actual Situation in the IT Market in Brazil. EUROMA
19 (2), 103–109. Conference Proceedings, Ankara.
Hair, J.F., Babin, B., Money, A.H., Samouel, P., 2005. Fundamentos de Patterson, F.D., Neailey, K., Kewkey, D., 1999. Managing the Risks within
Métodos de Pesquisa em Administração. Bookman, Porto Alegre. Automotive Manufacturing. Risk Manag. 1, 9–23.
Henseler, J., Ringle, C.M., Sinkovics, R.R., 2009. The use of partial least Pennypacker, J.S., Grant, K.P., 2003. Project management maturity: an industry
squares path modeling in international marketing. Adv. Int. Mark. 20 (1), benchmark. Proj. Manag. J. 4–9 (March).
277–319. Pinto, J.K., Slevin, D.P., 1988. Project Success: Definitions and Measurement
Herbsleb, J.D., Goldenson, D.R., 1996. A system survey of CMM experience and Techniques. Proj. Manag. J. 19 (1), 67–72.
results. Proceedings of International Conference on Software, Engineering 18, Plaza, M., Rohlf, K., 2008. Learning and performance in ERP implementation
pp. 323–330. projects: a learning-curve model for analyzing and managing consulting
Hong, P., Doll, W.J., Revilla, E., Nahm, A.Y., 2011. Knowledge sharing and costs. Int. J. Prod. Econ. 115 (1), 72–85.
strategic fit in integrated product development projects: an empirical study. PMI Project Management Institute, 2009. PMI Today June 2009. Project
Int. J. Prod. Econ. 132 (2), 186–196. Management Institute, Four Campus Boulevard, Newtown Square.
Ibbs, C.W., Kwak, Y.H., 2000. Assessing project management maturity. Proj. PMI Project Management Institute, 2013. A Guide to the Project Management
Manag. J. 31 (1), 32–43. Body of Knowledge (PMBoK Guide). 5th ed. Project Management
Ibert, O., 2004. Projects and firms as discordant complements: organisational Institute, Four Campus Boulevard, Newtown Square.
learning in the Munich software ecology. Res. Policy 33 (10), Prasad, S., Tata, J., Herlache, L., McCarthy, E., 2013. Developmental
1529–1546. project management in emerging countries. Oper. Manag. Res. 6 (1-2),
Ika, L.A., 2009. Project success as a topic in project management journals. 53–73.
Project Management Journal 40 (4), 06–19 (Four Campus Boulevard: Rauniar, R., Rawski, G., 2012. Organizational structuring and project team
Project Management Institute). structuring in integrated product development project. Int. J. Prod. Econ.
IPMA, International Project Management Association, 2006. ICB—IPMA 135 (2), 939–952.
Competency Baseline. IPMA, Nijkerk. Raz, T., Shenhar, A.J., Dvir, D., 2002. Risk management, project success, and
Jarvis, C.B., Mackenzie, S.B., Podsakoff, P.M., 2003. A critical review of technological uncertainty. R&D Manag. 32 (2), 101–109.
construct indicators and measurement model misspecification in marketing Repiso, L.R., Setchi, R., Salmeron, J., 2007. Modelling IT projects success:
and consumer research. J. Consum. Res. 30 (2), 199–218. emerging methodologies reviewed. Technovation 27 (10), 582–594.
Jiang, J.J., Klein, G., Hwang, H., Huang, J., Hung, S., 2004. An exploration of Ringle, C.M., Wende, S., Will, A., 2005. SmartPLS 2.0 M3 (Beta). University
the relationship between software development process maturity and project of Hamburg, Germany.
performance. Inf. Manag. 41 (3), 279–288. Sage, D., Dainty, A., Brookes, N., 2014. A critical argument in favor of
Jugdev, K., Muller, R., 2005. A retrospective look at our evolving understanding theoretical pluralism: project failure and the many and varied limitations of
of project success. Proj. Manag. J. 36 (4), 19–31. project management. Int. J. Proj. Manag. 32 (4), 544–555.
Jugdev, K., Thomas, J., 2002. Project management maturity models: the silver Samset, K., 1998. Project management in a high-uncertainty situation:
bullets of the competitive advantage? Proj. Manag. J. 33 (4), 4–14. Uncertainty, risk and project management in international development
Jugdev, K., Mathur, G., Fung, T.S., 2007. Project management assets and their projects. Ph.D. Thesis. Norwegian University of Science and Technology.
relationship with the project management capability of the firm. Int. J. Proj. Faculty of Civil and Environmental Engineering. Department of Building
Manag. 25 (6), 560–568. and Construction Engineering, Trondheim.
Katz, R., Allen, T.J., 1985. Project performance and the locus of influence in the Sanchez-Losado, J.M., 2012. Project management models: lean thought project
R&D matrix. Acad. Manag. J. 28 (1), 67–87. management. Dyna 87 (2), 214–221.
Kendra, K., Taplin, L., 2004. Project success: a cultural framework. Proj. Schwalbe, K., 2007. Information Technology Project Management. Thomson
Manag. J. 35 (1), 30–45. Course Technology, Boston.

Please cite this article as: M.M. Carvalho, et al., 2015. Project management and its effects on project success: Cross-country and cross-industry comparisons, Int. J. Proj.
Manag. http://dx.doi.org/10.1016/j.ijproman.2015.04.004
14 M.M. Carvalho et al. / International Journal of Project Management xx (2015) xxx–xxx

SEI, Software Engineering Institute, 2006. CMMI-DEV The capability maturity The Standish Group International, 2009. CHAOS Summary 2009. Disponível
model for development. University, Carnegie Mellon. em. http://www.standishgroup.com/. Acesso em:28 dez.
Sharma, A., Gupta, A., 2012. Impact of organisational climate and Thomas, J., Mullaly, M., 2007. Understanding the value of project
demographics on project specific risks in context to Indian software management: first steps on an international investigation in search of
industry. Int. J. Proj. Manag. 30, 176–187. value. Proj. Manag. J. 38 (3), 74–89.
Shenhar, A.J., 2001. One size does not fit all projects: exploring classical Thomas, J., Mullaly, M., 2008. Researching the Value of Project Management.
contingency domains. Manag. Sci. 47 (3), 394–414. PMI, Pennsylvania, Newtown Square.
Shenhar, A., Dvir, D., 1996. Toward a typological theory of project Thomas, J., Delisle, C.L., Jugdev, K., 2002. Selling Project Management to
management. Res. Policy 25 (4), 607–632. Senior Executives. Project Management Institute Inc, Newtown Square.
Shenhar, A., Dvir, D., 2007. Reinventing Project Management: the Diamond Toor, S., Ogunlana, S.O., 2010. Beyond the 'iron triangle': Stakeholder
Approach to Successful Growth and Innovation. Harvard Business School perception of key performance indicators (KPIs) for large-scale public
Press, Boston. sector development projects Int. J. Proj. Manag. 28 (3), 228–236.
Shenhar, A.J., Tishler, A., Dvir, D., Lipovetsky, S., Lechler, T., 2002. Refining White, D., Fortune, J., 2002. Current practice in project management—an
the search for project success factors: a multivariate, typological approach. empirical study. Int. J. Proj. Manag. 20 (2), 1–11.
R and D Manag. 32 (2), 111–126. Yazici, H.J., 2009. The role of project management maturity and organizational
Soderlund, J., Maylor, H., 2012. Project management scholarship:Relevance, culture in perceived performance. Proj. Manag. J. 40 (3), 14–33.
impact and five integrative challenges for business and management Zwikael, O., Ahn, M., 2011. The effectiveness of risk management: an analysis
schools. Int. J. Proj. Manag. 30 (6), 686–696. of project risk planning across industries and countries. Risk Anal. 31 (1),
Takey, S.M., Carvalho, M.M., 2015. Competency mapping in project management: 25–37.
an action research study in an engineering company. Int. J. Proj. Manag. 33 (4), Zwikael, O., Shimizu, K., Globerson, S., 2005. Cultural differences in project
784–796. management processes: a field study. Int. J. Proj. Manag. 23 (6), 454–462.
Tenenhaus, M., Esposito Vinzi, V., Chatelin, Y.-M., Lauro, C., 2005. PLS path
modeling. Comput. Stat. Data Anal. 48 (1), 159–205.

Please cite this article as: M.M. Carvalho, et al., 2015. Project management and its effects on project success: Cross-country and cross-industry comparisons, Int. J. Proj.
Manag. http://dx.doi.org/10.1016/j.ijproman.2015.04.004