A structural equation model for success measurement of development projects : The case of ISTKA projects

In recent years, research studies on determination of success criteria and critical success factors have gained a large share in project management research. In this area, it is not possible to define a standard project success. It is also difficult to determine the success criteria and critical success factors to cover each project type. In this study, the success criteria of the projects financed by Istanbul Development Agency, one of 26 development agencies in Turkey, were analyzed and efforts were made to determine the project success perception in the development agency projects. In addition, by examining the possible relationship between success criteria and critical success factors, it is aimed to find the most important factors affecting success. In order to achieve this goal, a Partial Least Squares Structural Equation Model has been established that includes variables of training, project team, project design, risk, sponsor, monitoring and project success. It is aimed to test whether the links established with this model, i.e. hypotheses, have been verified in terms of the target group. In addition, it has been examined whether there is a meaningful relationship between the project type and perceived success of projects, the type of project owner institution and the perceived success of the projects. It was also desired to measure the moderator effect of the project team-project coordinator, project team-monitoring specialist and project team-target group relationship variables. Finally, it was tested whether the model predictions support various relationships for different groups. In the research, firstly literature studies were carried out to establish the general framework of success criteria and success factors. Then, the hypotheses to be tested and the research model were established. After that, the documents and data of the projects supported by ISTKA were examined. After this phase, sample selection was made and data were collected by means of survey study from the determined sample. The collected data and obtained findings were analyzed. Finally, the results obtained were evaluated and suggestions were made to improve project success and project management. The data used within the scope of the research consist of 270 completed projects. These projects were financed under the 15 Financial Support Programs announced by ISTKA in 2014 and 2015 with the call for proposal method. In the study, a survey was conducted in order to determine the success criteria and critical success factors of the projects. The respondents consist of project coordinators, coordinator assistants and project team members. A project member may have been involved in more than one project, supported by İSTKA at different times. For this reason, they were asked to answer the questionnaire considering their completed projects. In the questionnaire preparation process, the literature survey was conducted and the questionnaires used to determine success criteria and critical success factors were specified. In order to understand the conformity of the statements obtained by the literature search, opinions were received from monitoring specialists and the project coordinators who worked on the projects supported by the ISTKA. Then it has been decided to conduct a pilot survey in order to test whether the expressions used in the questionnaire formed are understandable and to prevent any meaningfulness due to language differences. A comprehensive questionnaire was developed after the pilot application. The questionnaire consists of three main parts. The first part contains basic questions such as the demographic characteristics of the respondent, project code, and project team size. The second part consists of the questions about project success criteria and the third part consists of the questions about determining the success factors. In the questions in the second and third sections, it was determined how much participants agreement with statements by using the 7-point Likert scale. The questionnaire was prepared electronically using the Survey Monkey application and sent via e-mail to the participant. E-mail was successfully delivered to 460 people who were involved in 270 projects and 156 people were responded the survey. As a result, analyzes were made using valid responses of 148 participants. According to the analytical method used, this number was found to be sufficient. The data obtained on the basis of the questionnaire were analyzed in three steps. Firstly, the demographic data of respondents were examined. After that, all data were analyzed in terms of outliers, missing data, normality and reliability by using various tests on SPSS program. At the next section, the measurement model and the structural model were analyzed on SmartPLS program. The reliability and validity of the expressions used for each variable in the measurement model were tested. For this purpose; composite reliability, indicator/expression reliability, average variance extracted and discriminant validity were examined. Afterwards, the meaning and degree of the relationship between the structural model and the variables in the model are examined. Hypotheses have been tested on this count. As a result, eleven of the twenty hypotheses established were significant and supported for the selected sample. According to the results, it was understood that the project team perceived the projects as successful. However, when evaluating these projects according to ten success criteria (schedule, budget, project worked, target group, project team satisfaction, target user benefit, best solution, minimum problem, multiplier effect, target group effect), perceived success was found to be much higher. The findings of the research showed that the projects supported by ISTKA were perceived more successfully by the project team than they were in reality. When the answers to the time, budget and quality in the iron triangle were examined, it was understood that more than half of the projects supported by ISTKA could be considered as unsuccessful. This situation is in parallel with the studies examined in the literature study. According to project team perception, it was determined that ISTKA projects were not perceived as successful according to criteria such as project team satisfaction, multiplier effect. However, when the criteria related to the target group were examined, it was seen that the perception of success was high. When we look at the relationship between six success factors (education, monitoring, project team, project design, risk and sponsor) and project success in the Partial Least Squares Structural Equation Model, it was determined that the most important critical success factors affecting project success are project design, risk and project team. According to the results, project design was found to be the most important success factor affecting project performance. This factor includes design of activities, identification of roles of key personnel in the project, determination of the target group to take advantage of the project outputs, and planning of the resource plan such as time, budget and labor required for the implementation of the project. Another important critical success factor to increase project performance is risk. The risk factor concerns the planning of any kind of glitch that may occur in the project. According to the analyzes, the project team factor has been identified as the final critical success factor affecting project success. According to this factor, the selection of the project team in accordance with the project and the members of the team understand the roles of the project and their adoption increases the project performance. This study also examined whether there is a meaningful relationship between the type of project and the type of institution that owns the project and the perceived success of the projects. It was seen that the success of the projects was not affected by the diversity of different project types and different project-holding institutions. Another goal of the research is to measure the moderator effect of the project team-project coordinator, project team-monitoring specialist and project team-target group relationship variables. When the analysis results were examined, it was seen that a negative moderator effect of the project team-project coordinator relation variable was found on the relation of project design and project success. Therefore, if the relationship between the project team and the project coordinator is strong, then the impact of the project design on the project performance can decrease. But, in the same analysis, it was seen that the project team-project coordinator relation directly and positively affected the success of project. The moderator effect of the project team-monitoring specialist variable on the relation of training and monitoring was investigated and this hypothesis was supported. It is understood that the strong relationship between the project team and the monitoring specialist, assigned by the agency, affects the relationship between the training and the monitoring negatively. But, in the same analysis, it was seen that the project team-monitoring specialist relation directly and positively affected the monitoring factor. Another aim of the study is to test whether model predictions support different relationships for different groups. For this purpose, it was researched whether the structural model relations differed between groups in terms of project implementation year, total project budget size and amount of support. Finally, the actual/targeted performance indicators of the projects are compared with the project performance that is shaped by the perception of the project team. According to project team's perceptions, there was no significant relationship between the project performance and the actual performance of the projects. This finding is the result of the project team perceiving their projects more successfully than they are. In the literature, no empirical study was conducted to examine the relationship between the performance and critical success factors of projects supported by regional development agencies in Turkey. Therefore, this study is thought to provide important contributions by filling this gap in the literature.