project evaluation, weight assignment,...

International Journal of Construction Engineering and Management 2013, 2(3): 70-84 DOI: 10.5923/j.ijcem.20130203.05

Quantitative Weighting for Evaluation Indexes of Construction Project Success by Application of Structural

Equation Modeling

Thu Anh Nguyen1,*, Visuth Chovichien1, Shin-ei Takano2

1Department of Civil Engineering, Chulalongkorn University, Bangkok, 10330, Thailand 2Department of Civil Engineering, Hokkaido University, Sapporo, 060-8628, Japan

Abstract Although there is no universal defin ition of construction project success, no one can deny the importance of having a methodology to evaluate the success level of a p roject. It is necessary to develop a complete framework. Consequently, this paper aims to describe an evaluation index system for construction project success. It includes the list of indicators and criteria representative off project success and their important weight. The list of these indicators and criteria is the result of both academic and practical point of v iew. Quantitative important weight and ranking of these indicators are achieved by the application of factor analysis and structural equation modeling. The orig inal indicators were established from a review of literature and were used in preliminary survey. The informat ion from twenty-eight projects and sixty-five respondents were collected in this preliminary survey. Results from statistical analysis indicated that seven criteria could be rejected. The final index system contained 11 indicators and 46 criteria which were used in the main survey. In the main survey, 266 completed questionnaires were used for analysing. The resulting model was an index system which included ten indicators, which involved forty-six criteria to evaluate project success. The model also described their important weight for use in evaluation. It is anticipated that this model can serve as a means to evaluate the current status for construction industry in developing countries, that it would be, fair and objective, yet easy to understand and apply. In practice, it is hoped that this model will contribute to the improvement of project success rate and benefit all parties. Evaluating pro ject success is a useful tool for the construction industry to manage, control, and improve policies, and to anticipate future project success.

Keywords Project Evaluation, Weight Assignment, Construction Project Evaluation Indexes, Project Success

1. Introduction Along with the steady development of recent decades, the

construction industry faces several problems that cause serious damage and loss of men and material. Many people have studied to solve these problems, but construction project “success” has not improved significantly (stated by Hatush and Skitmore[1]). The construction sector still faces problems related to time and cost overruns, diminished quality and safety, and serious claims and lit igation.

In order to overcome these problems, the first mission is to understand how well a pro ject is performed, and how much the end result satisfied the initial objectives. It means that a project success measurement is necessary. A project manager cannot manage, control, or improve if he cannot measure a pro ject’s success. This is a difficult concept and

* Corresponding author: [email protected] (Thu Anh Nguyen) Published online at http://journal.sapub.org/ijcem Copyright © 2013 Scientific & Academic Publishing. All Rights Reserved

has been studied a long time by many researchers. Although there is no universal defin ition of pro ject success, no one can deny the importance of evaluating pro ject success, particularly in construction. Project success is the foundation for managing and controlling current pro ject, and for planning and orienting future project.

In fact, it is difficu lt to evaluate pro ject success in the construction field, especially in developing countries. The reasons are several. Customarily, project part icipants (owners, contractors, and consultants or project managers) have never evaluated a project upon complet ion. Until now, there has been no reliable tool to perform this evaluation. An appropriate model to evaluate project success is necessary to develop a past performance database.

From literature review, there are a great number of researchers interested in studying the factors which influence project success and in criteria to measure project success. In order to reduce misunderstanding, a d istinction should be made between project success factors and project success criteria. According to Oxford Advanced Learner’s Dict ionary, criterion means “a standard or princip le by

International Journal of Construction Engineering and Management 2013, 2(3): 70-84 71

which something is judged, or with the help of which a decision is made”; whereas a factor is “one or several things that cause or influence something”. So, the concept of “project success criteria” and “project success factor” are totally different but sometimes misunderstood. From this definit ion, a set of criteria for project success forms the basis of judging a project’s success. It includes a set of standards or principles to judge the project. On the other hand, project success factors are a set of several things that cause or influence the project outcomes. They contribute to the success or failure of pro ject. Up to this time, the majority of the studies conducted have focused on project success factors. These published articles include[2-12].

It is important to stress that, the concept used in this research is project success criteria. This research will not focus on what factors influence or contribute to project success or failure; it completely concentrates on the principle or standard by which the project is judged. There are several studies that focused on the area of p roject success measurement. However, these models for measuring project success contain some problems.

First, measuring project success models depends on the evaluator’s perception[13], bias and sensibility cannot be avoided, so the model may not be suitable for use as a database for contractor bidding information. To develop a contractor performance database, which will be used as criteria in bidding project, especially for public pro jects, a fair, straightforward, unbiased evaluation project success tool is required. If a contractor is required to evaluate a project based on their perception, they may make a b iased evaluation because the results may influence their business. Therefore, it is necessary to develop a quantitative evaluation project success model.

Secondly, each model was developed based on one party’s point of view[14]. A project should satisfy the requirement of all stakeholders (i.e., owner, contractor and consultant or project manager), so project success should be evaluated from their standpoints to avoid bias. Stakeholders use different indexes to evaluate the project. These are also appropriate to provide d ifferent informat ion to evaluate project success. Therefore, a model measuring project success model should allow stakeholders to evaluate the project independently and to combine their evaluations to achieve the final p roject success evaluation.

Thirdly, some quantitative evaluation models are d ifficult to put into practice in developing countries. For example, in order to evaluate contractor safety performance, it has been suggested that OSHA assessment be used, or using Environmental Impact Assessment to evaluate environment. Therefore, a feasible evaluation of pro ject success should be studied to be practiced in developing countries. It should consider carefu lly which indexes should be used and how to evaluate them based on the real information of completed projects in quantitative way.

Previous researches lacked methodology to combine the evaluation of all indexes. They provided the methods to evaluate each index separately. However, they did not

suggest the method to combine all indexes in a final p roject success score. The relative weight of each index also was rarely studied. For these reasons, this paper seeks to identify the project success indexes and sub-indexes from a long list of potential indexes which were suggested by authors in previous research. It includes the list of index and sub-index representative for project success and their important weight.

The following part of th is paper contains: literature review on the construction project success measurement and methods of weight assignment, preliminary framework of project success measurement system, research methodology, factor analysis, structural equation modeling result, and conclusion.

2. Review on the Construction Project Success Measurement

2.1. Review on the Construction Project Success Measurement

Project success is a difficult concept because of project’s complexity and dynamics. Until now, there has been no universal defin ition of pro ject success accepted. Defin ition of project success may vary depending on different points of view, from different persons, industries, and countries[2]. Even in a project, success or lack thereof is different among participants, scope of services, project size, and time-dependent[15]. However, accord ing to Parfitt and Sanvido[2], project success definition is based on the basic concept of overall ach ievement of objectives.

The problem of whether the project success can be measured or not has been addressed by many researchers. According to de Wit[16], measuring success is complex because it depends on the stakeholders’ point of v iew and it is time dependent. One party may acknowledge project as successful, but another may take the opposite view. A project maybe successful today but may fail tomorrow. de Wit[16] believed that it is an illusion to measure the p roject success objectively. However, he pointed out that it is possible and valuable to evaluate a project at its post-completion stage. He also provided evidence, the Project Management Institute conference held in Montreal in 1986, to demonstrate the possibility of success measurement. Th is conference discussed the importance of a measurement index system to evaluate project success. It reviewed the earlier versions of papers related to “measuring success”, implied a message that project success is possible to determine.

This section will consider previous research about measurement of project success in order to develop a measurement system of pro ject success indexes.

Many researches created a solid foundation for this study when they described the whole p icture of project success measurement indexes. They were de Wit[6], Songer et al.[17], Liu et al.[21], Crane et al.[20], Tukel et al.[18], White et al.[17], Bryde et al.[15], Ahadzie et al.[14], and

72 Thu Anh Nguyen et al.: Quantitative Weighting for Evaluation Indexes of Construction Project Success by Application of Structural Equation Modeling

Al-Tmeemy et al.[13]. They collected the indexes from previous researches or industry and then asked the perception of respondents. Most of them were based on the important scale to evaluate the important level of each. These researches provided a valuable reference for this research.

Project goals were the most appropriate criteria for project success assessment. They were based on the level of these objectives being met. In almost all previous researches, triangle project objectives, which included cost, time, and quality, were the main components in the evaluation system. de Wit[16] discussed the results from a pilot study about construction project success at Texas University. He suggested an evaluation indexes system which contained six criteria. These were budget and schedule performance, functionality of project, satisfaction of client, contractor, and project manager.

Another list of six success criteria was developed from Songer et al.[17]. Similar to de Wit[16], Songer et al.[17] also stressed the importance of budget and schedule achievement to evaluate project success. They were measured by the variat ion of budget and schedule between initial plan and practice. Songer et al.[17] mentioned quality of project by adding criteria about specifications and quality of workmanship to the model. He also focused on the satisfaction of users compared with their expectations, and the aggravation in the project. Th is indexes system was compatible with construction industry at that period.

In ten years, from 1990 to 2000, more than twenty studies were carried out to establish project success criteria[13]. They were separated into objective measures and subjective measures. Related to objective measures, four criteria in most of all research were Cost, Time, Health and Safety, and Quality. Other five measures were Techn ical performance/-Meeting specification, Functionality, Productivity, and Profitability, rarely appeared. In the subjective measures group, only one criterion, stakeholders’ satisfaction, was predominant in almost all studies. Seven other criteria were only mentioned in one or two studies. They are Expectation/ Aspiration, Dispute/Conflict management, Claim management, Professional image, Aesthetics, Educational/social/ professional aspects, and Environmental sustainability.

A group of researches concentrated on exploring the important weight and methodology to combine all indexes. They were Griffith et al.[18]; Chua et al.[19]; Shawn et al.[20]; Menches and Hanna[14]; and Shahrzad Khosravi [21]. Although some limitations made them difficult to apply in developing countries, these studies were very important in developing this research framework.

A success indexes equation was developed by Griffith et

al.[18]. Their equation considered four main criteria with their careful definit ion. The first criterion was Budget Achievement, which kept the highest proportion, weighted at 33% in evaluating project success. It was measured by percent of deviation between authorization budget and completion. The second criterion was Schedule Achievement. It was weighted at 27% in project evaluation and was measured by the difference between the authorized schedule and schedule of actual completion. Two other criteria were Design Capacity and Plant Utilizat ion. They were weighted at 12% and 28%, respectively, and were measured by authorization and actual attainment after six months of operation. Their relat ive weights were calculated by summing up all responses in important scale. This framework was developed specifically for facility projects. Therefore, it required more indicators and modifications to apply in construction building.

After two years, another group of researchers, Shawn et al.[20], developed a Construction Project Success Survey (CPSS) instrument. Their instrument included classic objective measures such as cost, schedule, quality, performance, safety, and operating environment. They used the seven point Likert system to assess each criterion. In their instrument, respondents’ perceptions about the importance of each issue was calculated. The instrument included thirty two issues related to six groups of criteria as mentioned above with the seven scale of answering. It made the instrument difficult and confusing for respondents. The result was still subjective because it depended on the perceptions of respondents.

A quantitative measurement method of successful performance was developed by Menches and Hanna[14]. They provided a quantitative methodology to measure the success from the qualitative evaluation. This method was the nearest basis for conducting the project success frame in this research. In the end, six factors were selected for the measurement. They were Project profit, Schedule achievement, Amount of time to perform the project, Communicat ion among project participants, Cost achievement, and Change in work time. This method was suitable from a contractor’s point of view. From the owner’s standpoint, these criteria were not enough to cover their entire objective to evaluate project success. However, this research provided an effective method to convert a qualitative parameter to quantitative and the concept of the probability of successful performance.

The summary list of indexes and sub-indexes from previous researches are described in Table 1 below. The table below also explained the evaluation methodology that previous researchers suggested for each index.


Table 1. Summaries List of Indicators and Methodology to Evaluate Project Success

Researchers List of Indicators and Evaluation Method Weighting Method

Tabish and Jha [11] Overall success : Nine-point scale Anti-corruption norms: Nine-point scale Financial norms: Nine-point scale

Not mentioned

Shahrzad Khosravi[21]

Time Performance Cost Performance Quality Performance HSE Client Satisfaction

Mean Rank method from 0 (not important) to 10

(very important)

Al-Tmeemy et al.[22]

Quality Targets Schedule Budget achievement Satisfaction of customer Functionality Meeting specification Profit achievement Market development Reputation Competitive Improvement

Not mentioned

Ahadzie et al.[23]

Project Cost Project Duration Project Quality Customer Satisfaction Environmental impact

Not mentioned

Menches and Hanna [14]

Profit (0.583) Schedule achievement (0.117): Percent t ime variation over/underrun Realistic schedule (0.033): How realistic: 1-5 Communication (0.133): Rate how good: 1-5 Achieved budget cost (0.083): Exceed or not: Y/N Work hours (0.05): Percent change in work hours

Summing up all responses for six

variables

Bryde and Robinson [24]

Project Cost (*) Project Duration (*) Technical specification Customer Satisfaction Stakeholders Satisfaction (*)

Not mentioned

Chan and Chan [25]

Time: Construction duration, Construction speed, Schedule variation Project cost: unit Profit: net present value Safety: Accident rate, EIA or ISO 14000 Environmental performance: Number of complaints Quality: Seven-point scale Functionality: Seven-point scale Satisfaction: Seven-point scale

Not mentioned

Shawn et al.[20]

Cost: Seven-point scale Schedule: Seven-point scale Quality: Seven-point scale Performance: Seven-point scale Safety: Seven-point scale Operating Environment: Seven-point scale

Relative important weight assigned by

respondents

Chan et al.[13]

Time: T ime overrun, Construction duration, Construction speed Cost: Unit cost, Cost overrun Health and Safety: Accident rate per 1,000 Profitability: Total net revenue over total costs Quality Technical Performance Functionality Productivity Satisfaction Environmental Sustainability

Not mentioned

White and Fortune [26] (General Project)

Project Cost Project Duration Meets client’s requirements Organizational objectives Business benefits

Not mentioned


Researchers List of Indicators and Evaluation Method Weighting Method

Quality and Safety requirement

Tukel and Rom [27] (General Project)

Project Cost Project Duration Technical specification Customer Satisfaction Rework

Not mentioned

Chua et al.[19] Achieve budget target (0.314) Achieve schedule target (0.360) Achieve quality target (0.325)

AHP Technique

Lim and Mohamed [28]

Time Cost Quality Performance Safety Satisfaction

Not mentioned

Crane et al.[29]

Cost Schedule Safety Quality Litigation

Not mentioned

Griffith et al.[18] (Facility projects)

Budget achievement (0.33): Percent deviation Schedule achievement (0.27): Percent deviation Plant utilization (0.12): Percent of planned utilization and actual attainted after 6 months Design capacity (0.28): Percent of planned utilization and actual attainted after 6 months

Weighting by summing up all responses for four

variables

Liu and Walker [30]

Project goals (1st level): Time, budget, functionality/ quality/ technical specification, safety, environmental sustainability. Satisfaction of the claimant (2nd level) Perception and awareness of different claimant.

Not mentioned

Shenhar and Levy [15] (General Project)

Budget and Schedule: Seven-point scale Customer Satisfaction Business benefits Potential Competition: extend market, new products, and new technology.

Not mentioned

Songer et al.[17] Budget variation, Schedule variation, Conformity to expectations

Not mentioned

3. Development of a Preliminary Survey 3.1. Preliminary Survey

This section describes the preliminary framework to achieve the proposed model of pro ject success evaluation. The objectives and expected outcomes of this research is a practical list of indicators to evaluate project success which can be applied in construction industry. It requires that all indicators in the proposed model have to satisfy three conditions that are possible to collect information to evaluate, importance and applicability. In order to achieve these criteria, a preliminary survey, which includes three issues, was performed. They are feasibility study, importance study, and applicability study.

A list of o rig inal indicators, which expected representation of project success, has been established. This list is gathered from literature review and interviews with

five experts in the construction field. They have more than ten years’ experience working in the construction industry and have participated in more than five completed projects. The list of ind icators is described in Table 2 below along with the survey results.

The survey is conducted by interviews and documentary searching. Each visited construction company is asked to provide all documents from one typical project and one or two representative engineers who are the most familiar with the project. All documents were examined carefully, and engineers were interviewed to find out whether or not it is possible to collect the necessary information. Furthermore, researching project documents helped to discover more informat ion that could be used to evaluate project success and had not been mentioned before. There are three options which provide information capacity for each criteria which are: “Pro ject has this information and it is possible to


provide information or evaluation op inion”, “Project has this information, but it is difficult to provide information or evaluation opinion”, and “Project does not have this informat ion or cannot provide evaluation opinion”. For the second option, reasons for the d ifficu lty need to be described.

The second survey was developed to explore the importance and the applicability of each criteria. The questionnaire contained three main sections including general informat ion of respondent, importance survey, and application capacity survey. Respondents express their opinion in five point Likert scale:

Not important at all : rate “1” Little important : rate “2” Moderately important : rate “3” Very important : rate “4” Extremely important : rate “5”

3.2. Data Collection for Preliminary Survey

The sampling units are owner, contractor, and pro ject manager who work in the construction industry. It is difficu lt to obtain a complete list of the target population. Moreover, informat ion after a project is completed is sensitive, so most companies refuse cooperation. So, interviewing these companies is very complicated unless one has a personal relationship with them. In addition, th is research is performed on limited time and budget. From these reasons, purposive sampling is selected as a suitable tool for this research.

During January and February 2012, the first survey was carried out in Vietnam, at Hochiminh city construction companies. An open and supporting letter from Hochiminh city University of Technology was prepared and sent to thirty construction companies. As an encouraging sign, twenty-three companies allowed v isits and provided informat ion. Finally, twenty-eight interview questionnaires and document checklists were completed. The companies that were visited have been coded from one to twenty-eight.

Second survey was conducted in February and March 2012 in Vietnam. For this survey, 125 questionnaires were distributed to ten construction companies and twenty-five individual meetings outside company. Finally, forty-two questionnaires from companies and twenty-three questionnaires from individual meet ings were collected. Another sixty questionnaires were not completed because engineers in some companies are so busy with their jobs that, they do not have time to complete the questionnaires. The total of completed questionnaires was sixty-five, response ratio being 52.0 percent.

3.3. Preliminary Framework Results

There are three criteria for making decisions about which indicators should be used for evaluating project success. First, indicators having a high probability to collect informat ion, namely the probability of successful collection of information should be higher than 60%. Second, indicators are important from a respondent’s perception,

mean ing the importance level is higher than three significantly. Third, an applicab le indicator with mean value should also be higher than three. The result of the proposed framework is shown in Figure 1 and Table 2.

Table 2. Results of Indicators and Criteria in Preliminary Survey (N=28 for Probability issue, N=65 for Mean of Important and Applicability Level)

Variable Code

Probability of successful

collecting information

Mean of Important

Level

Mean of Applicability

Level Decision

COST1 100% 4.40 4.28 Accept COST2 91% 4.66 4.35 Accept COST3 95% 4.32 4.28 Accept COST4 98% 3.72 3.91 Accept COST5 57% 3.65 3.82 Reject TIME1 100% 4.54 4.42 Accept TIME2 93% 4.20 4.12 Accept TIME3 77% 4.08 4.03 Accept TIME4 73% 4.02 3.94 Accept TIME5 82% 3.91 3.94 Accept QUA1 82% 4.37 4.22 Accept QUA2 88% 4.32 4.18 Accept QUA3 86% 3.91 3.88 Accept QUA4 73% 3.80 3.83 Accept QUA5 57% 3.85 3.83 Reject QUA6 71% 3.68 3.69 Accept SAF1 84% 4.75 4.37 Accept SAF2 82% 4.40 4.22 Accept SAF3 86% 3.80 4.08 Accept SAF4 48% 3.88 4.05 Reject SAF5 46% 3.77 3.85 Reject SAF6 59% 3.55 3.75 Reject SAF7 98% 3.92 3.88 Accept SAF8 100% 4.00 4.12 Accept SAF9 96% 4.08 4.03 Accept

SAF10 95% 4.00 4.05 Accept SAF11 95% 4.00 3.95 Accept TEC1 93% 4.34 4.20 Accept TEC2 95% 4.35 4.17 Accept TEC3 88% 4.00 3.86 Accept TEC4 91% 4.25 4.09 Accept FUN1 91% 4.29 4.17 Accept FUN2 59% 4.22 4.14 Reject PRO1 68% 3.82 3.80 Accept PRO2 86% 3.83 3.83 Accept PRO3 82% 3.78 3.89 Accept WAS1 43% 3.95 4.02 Reject SAT1 84% 4.45 4.34 Accept SAT2 89% 4.08 4.12 Accept SAT3 79% 3.97 4.02 Accept ENV1 79% 3.92 3.94 Accept ENV2 75% 3.85 3.92 Accept ENV3 79% 4.31 4.17 Accept ENV4 88% 3.88 3.77 Accept ENV5 71% 3.80 3.82 Accept ENV6 75% 3.71 3.77 Accept COM1 79% 4.00 3.97 Accept COM2 70% 4.25 3.98 Accept COM3 84% 4.09 3.98 Accept LIT1 77% 4.17 4.09 Accept LIT2 88% 4.14 4.08 Accept LIT3 86% 4.17 4.15 Accept LIT4 80% 4.12 4.22 Accept


In this framework, eleven ind icators are suggested to evaluate project success. These indicators focus on evaluating four main classic targets of construction projects such as project cost, project time, pro ject quality and safety. In addition, projects also can be evaluated by assessing technical performance, functionality, construction productivity, project stakeholder’s satisfaction, assessing from surrounding environment, communicat ion, and lit igation and disputes occurring during construction time. In order to evaluate these indicators, forty six criteria are used.

The results indicated seven criteria that were not used in evaluating project success. They were Budget for

contingencies, Budget to rework unsatisfied quality requirement works, Total expenditures for safety management in project, Total expenditures to handle and compensate of accidents occur during construction, Total time lost due to accident occur, Evaluation of conformance to expectation, and Waste material in construction site. These sub-indexes belong to Cost, Quality, Safety, Functionality, and Waste material index. They were rejected because they were considered difficu lt to collect informat ion, and their probability of successful collecting informat ion are 57%, 57%, 48%, 46%, 59%, 59%, and 43% in turn, lower than 60% of crit ical value.

Project Success

Evaluation Framework

Cost

Time

Quality

Safety

Technical Performance

Functionality

Productivity

Satisfaction

Environmental Sustainability

Information

Conflict, Litigation, Dispute

Cost overrun Unit cost Rework cost Expenses incurred

Time overrun Speed of construction

Material availability

Equipment availability

Labor availability

Compatible expectation

Conformance standard

Implement certificate Defects Rework

Time

Death injures/ accidents

Heavy accidents

Slight accidents Safety sign

Protection tools/equipment

Safety level of equipment

Safety training Safety staffs

Contractor’s response

Indentifying, solving problems

Worker qualification

Technical staff capacity

Suitability between project initial objective and final product

Unit labor cost/ square meter

Unit equipment cost/ square meter

Unit labor/ square meter

Owner satisfaction

Contractor satisfaction

Consultant satisfaction

Communities complaint

Authorities reminds Suspended time

Contractor recovery as warned

Environmental expenses

Problems solving expenses

Information between members

Missing/ Delaying information Information system

Conflict level

Relationship after completed

Penalties breach contract

Outstanding claim about payment

Figure 1. Proposed project success evaluation framework


4. Research Methodology A list of 11 indicators and 46 criteria, shown in Figure 1

above, was used to develop a questionnaire. This questionnaire was designed to achieve the following objectives: — Confirm factor analysis — Quantitative weighting for evaluation index system by

structural equation modeling

4.1. Questionnaire Design

The questionnaire contained two main sections including general informat ion and evaluation of the importance of proposed indicators and criteria. First, respondents were asked their opinion about the importance of an evaluation system. Then, respondents expressed their opinion of the importance level of each indicator and criterion in five point Likert scale as mentioned in the preliminary survey. Finally, open questions were given to collect respondents’ opinions about indicators that could be important but were not mentioned in the proposed list above.

4.2. Data Collection

The survey was carried out from July to September 2012 in Vietnam. From the survey, 600 questionnaires were prepared and distributed to twenty-five construction companies. The interviews took approximately thirty to forty five minutes. Finally, only 381 questionnaires were collected, representing an average response rate of 63.50%. In the 381 questionnaires that were collected, 115 questionnaires were eliminated because they missed too much informat ion, so the total of final valuable questionnaires was 266; the adjusted response ratio was 44.33%.

4.3. Data Analysis

Prior to analysing the usable sample, it was important to check for mistake in itially. Data were screened using the complete sample (N = 266) prior to the main analysis to examine accuracy of data entry, missing values, and fit between distributions and the assumptions of necessary analysis tools. The Frequencies and Descriptive statistic command in SPSS Version 16 was used to detect any out of range values. None were found and the descriptive results are described in Table 3 below.

Table 3. List of Indicators and Criteria to Evaluate Construction Project Success and Their Descriptive Analysis Results (N=266)

Variable Code Variable Meaning Mean Std. Deviation

1. COST The degree to which the general contexts promote the completion of a project within the estimated budget 4.36 .794

COST1 Cost variation: ratio of net variations to final contract sum expressed in percentage term 4.17 .821

COST2 Unit cost is a measure of relative cost and is defined by the final contract sum divided by the gross floor area. 3.82 .836

COST3 Rework costs 3.59 .961 COST4 Expenses incurred 3.29 .960

2.TIME The degree to which the general contexts promote the completion of a project within the allocated duration 4.27 .750

TIME1 Time variation is measured by the percentage of increase or decrease in the estimated project days, discounting the effect of extension of time granted by the client. 4.18 .835

TIME2 Speed of construction is the relative time, which is defined by gross floor area divided by the construction time (number of days from start on site to practical completion of the project) 3.89 .739

TIME3 Material availability: number of days construction site delay because of supplying materials 4.08 .754 TIME4 Equipment availability: number of days construction site delay because of lack of equipment 3.98 .810 TIME5 Labor availability: number of days construction site delay because of lack of labor 4.00 .817

3.QUALITY The degree to which the general contexts promote meeting of project’s established requirements of materials and workmanship 4.59 .769

QUA1 Conformity with expectations: The different level between quality expectation of owner and real project quality after completed. 4.32 .746

QUA2 Conformity with predetermined standard: The different level between predetermined standard and real project quality. 4.28 .757

QUA3 Implement the “Evaluate the suitability project quality certificate” in the project 3.89 .855 QUA4 Number of defects need to rework when take over the project 3.77 .880 QUA5 Time to rework under-quality works 3.65 .888

4.SAFETY The degrees to which the general contexts promote the completion of a project without major accidents or injuries 4.27 .901

SAFE1 Number of death injures or accident 4.55 .967 SAFE2 Number of heavy accidents 4.25 .995 SAFE3 Number of slightly accidents 3.55 .959 SAFE4 Evaluation of safety signs 4.12 .901 SAFE5 Evaluation of providing safety tools and protection equipment 4.46 .722 SAFE6 Evaluation safety level of equipment used in construction 4.32 .778 SAFE7 Evaluation of safety training 4.24 .869 SAFE8 Evaluation of safety responsibility staffs 4.09 .892 5.TECH The degree to which the general contexts promote meeting of project’s established specifications 4.07 .764 TECH1 Evaluation of the contractor’s response to the technical requirements of project 4.40 .678


TECH2 Evaluation of technical problem identification and solution 4.21 .731 TECH3 Overall assessment qualifications of workers in the project 3.92 .712 TECH4 Evaluation of the possibility of problem solving of technical staff 4.35 .724 6.FUNC The degree to which the general contexts promote achieving the ‘‘fitness for purpose’’ objective 3.81 .984 FUN1 Degree of conformance to all technical performance specifications. 4.11 .780

7.PRO The degree to which the general contexts promote achieving effectiveness of allocated resources in order to meet the cost and time targets 3.55 .869

PRO1 Unit labor per square meter 3.65 .803 PRO2 Unit labor cost per square meter 3.81 .778 PRO3 Unit equipment cost per square meter 3.73 .796

8.SATIS Satisfaction describes the level of ‘‘happiness’’ of people affected by a project 3.72 .902 SATIS1 Owner satisfaction 4.39 .735 SATIS2 Contractor satisfaction 3.79 .778 SATIS3 Consultant satisfaction 3.73 .838

9.ENVI The degree to which the general contexts promote avoiding the effects of project on the environment 3.65 1.018

ENVI1 Frequency of complaints from the environment and communities around the construction site 3.76 .918 ENVI2 Frequency of time reminded about sanitation from the authorities 3.63 .940 ENVI3 The number of t ime and duration suspended from the authorities 3.99 1.039 ENVI4 Assessing the recovery of the contractor when warned 3.80 .827 ENVI5 Expenses for ensure environmental sustainability 3.60 .842 ENVI6 Expenses of overcoming the problems of environmental sanitation 3.59 .834

10.COMMU The degree to which the general contexts promote achieving effectiveness of communication in order to avoid misunderstanding 3.61 .977

COMMU1 Evaluation the communication in project 3.86 .825 COMMU2 The frequency of misinformation or delays affecting the project 3.94 .878 COMMU3 Information systems used in project 3.84 .903

11.LITIGA This index can be measured by number of outstanding claims, relationship among parties after project is completed, and information about penalties for breach of contract. 3.37 1.046

LITIGA1 Outstanding claim among parties about payment 3.84 .890 LITIGA2 Evaluation of conflict level among parties in check and take over the project 3.79 .865 LITIGA3 Evaluation of relationship between contractor and owner after project completed 3.83 .903 LITIGA4 Information about penalties for breach of contract 3.74 .974

Factor analysis was applied to optimize the components and faster described the index system to evaluate project success, while retaining their original characteristics. The 46 items were subjected to principal axis factoring using SPSS. Factor analysis, a multivariate statistical technique, is commonly used to identify a smaller number of relevant factors than the original number of individual variables. The application of this technique can reduce the data to a representative subset of variables or even create new variables as replacements for the original variables, while still retaining their original characteristics[31].

The main object ive of the study was the weight assignment of each indicator and criterion. It reflects the intention and preference of respondents, and affects the consequences of system. Weighting methods included subjective weighting method and objective weighting method. These were the Delphi method, the Analytic of Hierarchy Process method, the Analytic Network Process method, the Factor Analysis method, and the Entropy method. The selection of a method depended on objective and purpose. In the area of construction success measurement as mentioned in Table 1, descriptions such as ranking, summing up and Analytic of Hierarchy Process method were applied. They belong to the subjective weighting method which implied several disadvantages. Consequently, an objective weighting assignment was necessary; it was a more logical and accurate system.

Structural equation modeling (SEM) was performed to weight the importance of each indicator and criterion. SEM was an alternat ive technique fo r exp loring the interrelationship among factors in mult iple layers of linkages between variables. SEM proved an effective statistical technique in developing the causal model for exp lain ing a dependent variable with high quality in formation[32, 33]. There are two main parts to SEM: causal model between independent variables and dependent variables, and the measurement model showing the relationship among variables. This technique was applied by using AMOS 16.0 software.

5. Results 5.1. Res pondent Profiles

Of the respondents, the average age was 30.34 years and ranged from 23 to 57 years old. All of them had experience from 1 to 29 years’ experience, average 6.46 years. The level of a respondent’s academic background was one factor that influenced their opinion about construction project success. In this study, respondent’s backgrounds were classified into three groups. The data showed that 6.69% of the respondents had high school background, 78.74% had undergraduate qualification, and 14.57% had postgraduate education. Almost all respondents had acceptable education background, so they could serve as representative of the


population. Because of the purpose of this research, the number of

completed pro jects is more important than the number of years a respondent had worked. Figure 3 below separated respondents’ experience in completed project in three groups. The first group is respondents who have taken part in less than three completed projects, which make up more than 30%. The second group, which included 27.38%, was the respondents who had finished from three to five pro jects. The last group of respondents, who had more than five projects completed, appropriated high percentage of 42.46%.

Figure 3. Respondents’ experience

Before conducting further analysis, respondents were asked how important a framework was in evaluating construction project success. Figure 4 below summarises their opinions. Among 260 valid responses, 125 people believed that the proposed system is extremely important; 94 people thought that it is very important, and they comprised more than 84% of the responses. The remaining 16% of the respondents did not highly appreciate the importance of pro ject success evaluation framework. This result implies that the proposed framework is significant and necessary to study. Further analysis should be conducted.

Figure 4. Respondents’ opinion about the importance of framework to evaluate project success

5.2. Factor Analysis Result

Three assumptions were tested before conducting factor analysis to check the validation of data collected.

5.2.1. Data Validation for Factor Analysis

The first assumption was to assess the sample size. According to Hair et al.[32], the minimum sample size should have at least five times the number of variables, and should be larger than 100. In this survey, the sample size was 266, equivalent 5.78 cases to 1 variable. Therefore, the first assumption was satisfied.

The second assumption related to the correlat ion among variables in the survey. This was assessed via a correlation matrix. For factor analysis, there should be more than 20 percent of correlations larger than 0.30 at 0.01 level of significance. Analysis of the results showed that the number of correlations larger than 0.30 kept more than 50 percent.

The third assumption was the anti-image correlat ion matrix; Measure of Sampling Adequacy (MSA) was above 0.50[32]. Kaiser-Meyer-Olkin (KMO) of 46 criteria was 0.880, higher than the 0.600 suggested value. In addition, the Bartlett test of sphericity, reached statistical significance with chi-square 7333, degree of freedom 1035 and significance level of 0.000. Therefore, this factor analysis was appropriate for conducting further analysis.

5.2.2. Factor Analysis Process

Table 4. Factor Loading of Evaluation Index System of Construction Project Success (N=266)

Indicators/ Criteria

Factor Loading

Indicators/ Criteria

Factor Loading

Indicator F1 Indicator F5 SAFE7 .901 TIME2 .961 SAFE8 .859 TIME4 .951 SAFE6 .750 TIME3 .914 SAFE5 .722 TIME5 .850 SAFE4 .497 Indicator F6

Indicator F2 LITIGA1 .829 SAFE2 .848 LITIGA2 .777 SAFE1 .848 LITIGA4 .513 ENVI3 .578 LITIGA3 .439 ENVI1 .551 Indicator F7 SAFE3 .506 SATIS2 .843 ENVI2 .488 SATIS3 .740

Indicator F3 SATIS1 .439 TECH1 .751 Indicator F8 QUA1 .646 QUA4 .714 QUA2 .554 QUA5 .703 TIME1 .540 COST3 .598 TECH2 .507 COST4 .493 COST1 .450 QUA3 .311 TECH4 .383 Indicator F9 TECH3 .352 COMMU1 .766 FUN1 .332 COMMU3 .643

Indicator F4 COMMU2 .617 PRO2 .937 Indicator F10 PRO3 .868 ENVI6 .689 PRO1 .625 ENVI5 .630

COST2 .424 ENVI4 .535


An initial capture of factors was made using principal axis factoring through SPSS. Not using the rotation factor analysis pointed out ten distinct factors. Then, a promax rotation was applied to minimize the number of variables and optimize solutions. The final solution of promax resulted ten factors accounting for 58 percent of the total variance. The rotated pattern matrix and the results of factor loading are presented in Table 4 below. Finally, ten indicators with forty-six criteria comprise the evaluation

index system of construction project success. This evaluation system will be applied in the structural equation model to calculate the importance weight of each criterion and indicator.

6. Quantitative Weighting Important Indicators and Criteria based on Structural Equation Modeling (SEM)

Figure 4. Important weight of the evaluation index system of Construction Project Success

0

F14.24

Safe70, .23

e4

1.001

4.32

Safe60, .18

e3.901

4.46

Safe50, .17

e2 .821

4.12

Safe40, .40

e1.88

1

0

F23.76

Env10, .37

e9

3.55

Safe30, .55

e8

4.25

Safe20, .31

e7

4.55

Safe10, .42

e6

1.001

.8811.21

1 1.05

1

3.63

Env20, .35

e10

1.071

0

F3

4.28

Qua20, .36

e15

4.32

Qua10, .44

e14

4.18

Time10, .56

e13

4.17

Cost10, .61

e12

4.11

Fun0, .45

e16

1.001 .74

1 .79

1.54

1

.871

0

F4

3.65

Pro10, .31

e22

3.82

Cost20, .52

e21

3.81

Pro20, .17

e23

1.001.72

1

1.1413.73

Pro30, .16

e24

1.191

0

F54.00

Time50, .17

e28

3.98

Time40, .12

e27

4.08

Time30, .15

e26

3.89

Time20, .47

e25

1.001

1.031

.921.38

1

0

F63.83

Lit30, .54

e31

3.79

Lit20, .27

e30

3.84

Lit10, .29

e29

1.001

1.3211.36

1

0

F74.39

Sat10, .38

e35

3.79

Sat20, .23

e34

3.73

Sat30, .14

e33

1.001

1.561

1.901

0

F8

3.65

Qua50, .33

e40

3.77

Qua40, .34

e39

3.89

Qua30, .59

e381.00

1

.981

.561

0

F93.84

Com30, .31

e43

3.94

Com20, .40

e42

3.86

Com10, .27

e41

1.001

.861

.901

0

F10

3.59

Env60, .13

e46

3.60

Env50, .13

e45

3.80

Env40, .38

e44

1.00

1

1.011

.731

3.29

Cost40, .60

e37.841

3.59

Cost30, .59

e36 .851

Chi-square=3599.403 ; df=1529 ; P=.000;Chi-square/df=2.354;

RMSEA=.071

4.09

Safe80, .27

e5

1.001

4.40

Tech10, .24

e17

1.031

4.21

Tech20, .24

e18

1.191

3.92

Tech30, .30

e19

.99

14.35

Tech40, .26

e20

1.11

1

3.99

Env30, .43

e11

1.18

1

3.74

Lit40, .40

e32

1.421

0, .31

z1

1

0, .24

z2

1

0, .09

z3

1

0, .22

z4

1

0, .33

z5

1

0, .16

z6

0, .11

z7

0, .23

z8

0, .25

z9

1

0, .30

z10

1

1

1

1

0, .06

PROJECT SUCCESS

4.36

COST

0, .57

k1

1.00

1

4.27

TIME

0, .48

k2

1.21

1

4.59

QUALITY

0, .52

k3

1.10

1

1.46

1.47

1.77

1.43

.94

2.02

2.13

2.15

2.04

1.95

4.27

HEALTH &SAFETY

0, .59

k4

2.00

1

4.07

TECHNICALPERFORMANCE

0, .44

k51.61

1

3.81

FUNCTIONALITY

0, .76

k61.93 1

3.55

PRODUCTIVITY

0, .61

k7

1.61

1

3.72

SATISFACTION

0, .76

k8

.94

1

3.65

ENVIRONMENT

0, .66

k9

2.59

1

3.61

COMMUNICATION

0, .62

k10

2.45

1

3.37

DISPUTE &LITIGATION

0, .82

k11

2.20

1


Structural modeling was undertaken using the SEM technique to reveal the significant interrelat ionships between the factors retained from EFA. The conceptual model was described in Figure 4 below. Ten constructs related to the indicators which were exp lored from EFA, forty six criteria represented for criteria to evaluate project success and one construct represented for project success were in this model. In order to achieve a higher Goodness-of-Fit model, some links between errors were sequentially added based on the result from Modification Indices (MI). The final model is described in Figure 4. It was the optimum model that achieved the most criteria for several fit indexes without too complex relat ionships.

This model has the fo llowing fit coefficients: CMIN/DF = 2.354; RMSEA = 0.071. The final model satisfied more than 50% of the critical standards and was above the threshold of the most important standards. So, we can thus safely conclude that the model is valid and can continue analysing the outcome of the causal effects. Figure 4 provides the results of testing the structural links of the proposed research model using AMOS program. The estimated path coefficients are given. All path coefficients can be considered significant at the 90% significance level providing support for all relationships. It implies that ten proposed indicators from F1 to F10 can be used to evaluate the construction project success. In addition, these indicators can be assessed by the forty six criteria as discussed above. This system accounted for over 50% of the variance in construction project success; it is an indication of the good explanatory power of the model. A summary of the developed structural equations, path coefficients and significance levels is provided in Table 5. The following section discusses the practical implicat ions of each structural equation and its’ associated construction project success.

The weight of the indicator can be assigned based on the loading coefficients on the loading chart. The Figure 4 showed that the significant coefficient of F1, F2, F3, F4, F5, F6, F7, F8, F9, F10 were 1.953, 2.037, 1.470, 1.462, 1.772, 1.425, 0.940, 2.021, 2.130, 2.151 respectively, and their sum was 17.361. So the weight of each coefficient was calculated by div ide each coefficient by their sum as shown in Table 5. The result of criteria weight which described in Table 6 was calcu lated in the same way.

An interesting result from SEM modeling showed that indicator F10 was the most important indicator impacting a construction project’s success, contributing 12.4%. Indicator F10 contained three criteria related to environmental sustainability, which are Envi4, Envi5, Envi6. Other criteria related to environment retained an important position when associated with safety criteria built up indicator F2, which is the th ird important indicator and contributes 11.7% to project success. In addition, indicator F9, which contained criteria Commu1, Commu2, and Commu3, showed their importance as contributing 12.4% of project success. The criteria Envi5, Envi6, Commu3, Commu1, Commu2 ranked as the first five criteria in project evaluation system.

The SEM results were surprising and interesting for authors. In tradit ional opin ion, as summaries in the literature reviewed, original pro ject objectives such as cost, time, and quality are considered as the most important criteria in the evaluation of project success. The results pointed out the opposite. Environmental sustainability and safety held the central position in the evaluation framework. However, the results are reasonable and explainable. As a global tendency, environmental sustainability and safety are more and more important in an evaluation of p roject success[23][25][20][13][28][30]. Tab le 5 and Table 6 below summarize the important weight of all indicators and criteria and their rankings.

Table 5. Path coefficients between Evaluate Indicators and Construction Project Success (N=266)

Indicators/Criteria Estimate S.E. C.R. P Weight Ranking

F1 <--- Project Success 1.953 0.447 4.366 *** 0.112 5 F2 <--- Project Success 2.037 0.465 4.384 *** 0.117 3

F3 <--- Project Success 1.470 0.345 4.257 *** 0.085 7








Sum 17.361 1.000


Table 6. Path coefficients between Criteria and Indicators to Evaluate Construction Project Success (N=266)

Indicators/Criteria Estimate S.E. C.R. P Weight in

each Indicator

Weight Ranking in

each indicator

Ranking in all criteria

system

SAFE4 <--- F1 0.881 0.069 12.673 *** 0.191 0.022 4 24 SAFE5 <--- F1 0.817 0.053 15.516 *** 0.177 0.020 5 27 SAFE6 <--- F1 0.903 0.056 16.095 *** 0.196 0.022 3 21 SAFE7 <--- F1 1.000 0.217 0.024 2 14 SAFE8 <--- F1 1.004 0.065 15.401 *** 0.218 0.025 1 13

Sum 4.605 1.000 ENVI1 <--- F2 1.000 0.156 0.018 5 31 ENVI2 <--- F2 1.067 0.085 12.51 *** 0.167 0.020 3 28 ENVI3 <--- F2 1.182 0.094 12.54 *** 0.185 0.022 2 23 SAFE1 <--- F2 1.054 0.088 11.98 *** 0.165 0.019 4 29 SAFE2 <--- F2 1.208 0.090 13.43 *** 0.189 0.022 1 20 SAFE3 <--- F2 0.882 0.088 10.006 *** 0.138 0.016 6 32

Sum 6.393 1.000 COST1 <--- F3 0.540 0.123 4.401 *** 0.065 0.006 9 46 FUN1 <--- F3 0.875 0.124 7.059 *** 0.106 0.009 6 43 QUA1 <--- F3 0.736 0.116 6.338 *** 0.089 0.008 8 45 QUA2 <--- F3 1.000 0.121 0.010 4 41

TECH1 <--- F3 1.033 0.116 8.911 *** 0.125 0.011 3 40 TECH2 <--- F3 1.190 0.128 9.32 *** 0.144 0.012 1 36 TECH3 <--- F3 0.991 0.119 8.358 *** 0.120 0.010 5 42 TECH4 <--- F3 1.114 0.124 8.981 *** 0.135 0.011 2 39 TIME1 <--- F3 0.791 0.129 6.121 *** 0.096 0.008 7 44

Sum 8.270 1.000 COST2 <--- F4 0.722 0.094 7.693 *** 0.178 0.015 4 35 PRO1 <--- F4 1.000 0.247 0.021 3 26 PRO2 <--- F4 1.139 0.089 12.817 *** 0.282 0.024 2 15 PRO3 <--- F4 1.185 0.092 12.94 *** 0.293 0.025 1 12

Sum 4.046 1.000 TIME2 <--- F5 0.377 0.064 5.877 *** 0.113 0.012 4 38 TIME3 <--- F5 0.917 0.051 18.015 *** 0.275 0.028 3 9 TIME4 <--- F5 1.035 0.054 19.23 *** 0.311 0.032 1 7 TIME5 <--- F5 1.000 0.300 0.031 2 8

Sum 3.329 1.000 LITIGA1 <--- F6 1.360 0.148 9.156 *** 0.267 0.022 2 22 LITIGA2 <--- F6 1.324 0.144 9.166 *** 0.260 0.021 3 25 LITIGA3 <--- F6 1.000 0.196 0.016 4 33 LITIGA4 <--- F6 1.416 0.159 8.923 *** 0.278 0.023 1 19

Sum 5.100 1.000 SATIS1 <--- F7 1.000 0.224 0.012 3 37 SATIS2 <--- F7 1.556 0.184 8.435 *** 0.349 0.019 2 30 SATIS3 <--- F7 1.899 0.225 8.423 *** 0.426 0.023 1 18

Sum 4.455 1.000 COST3 <--- F8 0.850 0.096 8.838 *** 0.201 0.023 3 16 COST4 <--- F8 0.841 0.096 8.747 *** 0.199 0.023 4 17 QUA3 <--- F8 0.555 0.086 6.485 *** 0.131 0.015 5 34 QUA4 <--- F8 0.981 0.09 10.912 *** 0.232 0.027 2 11 QUA5 <--- F8 1.000 0.237 0.028 1 10

Sum 4.227 1.000 COM1 <--- F9 0.900 0.079 11.45 *** 0.326 0.040 2 4 COM2 <--- F9 0.860 0.082 10.515 *** 0.312 0.038 3 5 COM3 <--- F9 1.000 0.362 0.044 1 3

Sum 2.760 1.000 ENVI4 <--- F10 0.730 0.06 12.178 *** 0.266 0.033 3 6 ENVI5 <--- F10 1.014 0.055 18.326 *** 0.370 0.046 1 1 ENVI6 <--- F10 1.000 0.364 0.045 2 2

Sum 2.744 1.000 1.000

7. Conclusions In this paper, authors exp lored the SEM technique to

calculate the importance weight of evaluation indexes

system for construction project success. Project success is a foundation to manage and control current projects, and to plan and orient future pro jects. The SEM model, along with factor analysis, represents a new approach to weighting the


importance role of indexes system that has not been discussed before. Weight assignment plays a very important role in the evaluation index system. Some methods and approaches were applied in weight assignment. Each of them has its own advantages and disadvantages. By doing fitting index, this paper provided a fit of model and assigned quantitative weight indexes system. The results of this paper suggested a system of ten indicators and forty-six criteria to evaluate construction project success. In this system, the importance of criteria related to environment, safety issue, and communication among project was stressed. They rank in the top of the project evaluation model. Th is implies a tendency in respondents’ opinions about the importance level of these issues. Up to now, a complete system included both list of indexes and importance weight of them were achieved. So, p roject success can be quantitative evaluation. However, this system should be applied in a real construction project and the feedback obtained to optimize; this is a part of our future mission. Ten indicators and 46 criteria is a complex system to evaluate pro ject success. It helps to describe the project comprehensively. However, it make the evaluator encounter a litt le difficulties. So, future researches may concentrate on this issue to optimize the evaluation system.

REFERENCES [1] Hatush Z, Skitmore M. Contractor selection using

multicriteria utility theory: An additive model. Building and Environment. 1998;33:105-15.

[2] Parfitt MK, Sanvido VE. Checklist of Critical Success Factors for Building Projects. Journal of Management in Engineering. 1993;9:243-9.

[3] Hatush Z, Skitmore M. Evaluating contractor prequalification data: selection criteria and project success factors. Construction Management and Economics. 1997;15:129-47.

[4] Chan APC, Ho DCK, Tam CM. Design and Build Project Success Factors: Multivariate Analysis Journal of Construction Engineering and Management. 2001;127:93-100.

[5] Terry C-D. The “real” success factors on projects. International Journal of Project Management. 2002;20:185-90.

[6] Chan APC, Scott D, Chan APL. Factors Affecting the Success of a Construction Project. Journal of Construction Engineering & Management. 2004;130:153-5.

[7] Chu P-Y, Hsiao N, Lee F-W, Chen C-W. Exploring success factors for Taiwan's government electronic tendering system: behavioral perspectives from end users. Government Information Quarterly. 2004;21:219-34.

[8] Nguyen LD, Ogunlana SO, Lan DTX. A Study on Project Success Factors in Large Construction Projects in Vietnam. Engineering, Construction and Architectural Management. 2004;11:404-13.

[9] Salminen JM. Measuring performance and determining

success factors of construction sites[C821040]. Finland: Teknillinen Korkeakoulu (Helsinki) (Finland); 2005.

[10] Chan APC, Lam PTI, Chan DWM, Cheung E, Ke Y. Critical Success Factors for PPPs in Infrastructure Developments: Chinese Perspective. Journal of Construction Engineering and Management. 2010;136:484-94.

[11] Tabish SZS, Jha KN. Identification and evaluation of success factors for public construction projects. Construction Management and Economics. 2011;29:809-23.

[12] Sanvido V, Grobler F, Parfitt K, Guvenis M, Coyle M. Critical Success Factors for Construction Projects. Journal of Construction Engineering and Management. 1992;118:94-111.

[13] Chan APC, Scott D, Lam EWM. Framework of Success Criteria for Design/Build Projects. Journal of Management in Engineering. 2002;18:120-8.

[14] Menches CL, Hanna AS. Quantitative Measurement of Successful Performance from the Project Manager's Perspective. Journal of Construction Engineering and Management. 2006;132:1284-93.

[15] Shenhar AJ, Levy O. Mapping the dimensions of project success. Project Management Journal. 1997;28:5-13.

[16] de Wit A. Measurement of project success. International Journal of Project Management. 1988;6:164-70.

[17] Songer AD, Molenaar KR, Robinson GD. Selection Factors and Success Criteria for Design-Build in the U.S. and U.K.: University of Colorado; 1997.

[18] Griffith AF, Gibson GE, Hamilton MR, Tortora AL, Wilson CT. Project Success Index for Capital Facility Construction Projects. Journal of Performance of Constructed Facilities. 1999;13:39-45.

[19] Chua DKH, Kog YC, Loh PK. Critical Success Factors for Different Project Objectives. Journal of Construction Engineering and Management. 1999;125:142-50.

[20] Shawn WH, Donald DT, Warren KT. Measuring Project Success in the Construction Industry. Engineering Management Journal. 2004;16:31-7.

[21] Shahrzad Khosravi HA. A Success Measurement Model for Construction Projects. International Conference on Financial Management and Economics. Singapore: IPEDR; 2011. p. 186-90.

[22] Al-Tmeemy SMHM, Abdul-Rahman H, Harun Z. Future criteria for success of building projects in Malaysia. International Journal of Project Management. 2011;29: 337-48.

[23] Ahadzie DK, Proverbs DG, Olomolaiye PO. Critical success criteria for mass house building projects in developing countries. International Journal of Project Management. 2008;26:675-87.

[24] Bryde DJ, Robinson L. Client versus contractor perspectives on project success criteria. International Journal of Project Management. 2005;23:622-9.

[25] Chan APC, Chan APL. Key Performance Indicators for Measuring Construction Success. Benchmarking: An International Journal. 2004;11:203-21.

[26] White D, Fortune J. Current practice in project management


— an empirical study. International Journal of Project Management. 2002;20:1-11.

[27] Tukel OI, Rom WO. An Empirical Investigation of Project Evaluation Criteria. International Journal of Operations & Production Management. 2001;21:400-16.

[28] Lim CS, Mohamed MZ. Criteria of project success: an exploratory re-examination. International Journal of Project Management. 1999;17:243-8.

[29] Crane TG, Felder JP, Thompson PJ, Thompson MG, Sanders SR. Partnering Measures. Journal of Management in

Engineering. 1999;15:37-42.

[30] Liu AMM, Walker A. Evaluation of project outcomes. Construction Management & Economics. 1998;16:209-19.

[31] Pallant J. SPSS Survival Manual. 2nd ed: Open University Press; 2004.

[32] Hair JF, Black WC, Babin BJ, Anderson RE. Multivariate Data Analysis. 7th ed: Pearson Prentice Hall; 2010.

[33] Tabachnick BG, Fidell LS. Using Multivariate Statistics. 5th ed: Allyn & Bacon; 2006.

project evaluation, weight assignment,...

Documents