Research of relationship between risk and performance in it projects

In order to create a model that examines the relationship of risk and performance, risk, the performance and features of the project were examined in the literature. The most commonly risk groups have been identified. Risks have been analyzed in 6 groups: requirement risk, users risk, organizational environmental risks, planning and control risk, and project complexity, team risk. Requirements risk includes factors such that "continually changing system requirements", "system requirements not adequately identified", "unclear system requirements" and "incorrect system requirements". Organizational environmental risks includes factors such that "change in organizational management during the project", "unstable organizational environment", "corporate politics with negative effect on project", "organization undergoing restructuring during the project". Team risk includes factors such that "team members lack specialized skills required by the project" and "inadequately trained development team members", "lack of commitment to the project among development team members", "frequent conflicts between development team members", "frequent turnover within the project team" and "inexperienced team members". "users resistant to change" and "lack of user participation". User risk contains "users resistant to change", "lack of user participation", "lack of cooperation from users", "users with negative attitudes toward the project" and "conflict between users". Planning and control risk includes factors such that "inexperienced project manager", "project milestones not clearly defined", "ineffective communication", "lack of an effective project management methodology", "inadequate estimation of required resources", "project progress not monitored closely enough" and "poor project planning". Project complexity risk includes factors such that "project involves use of technology that has not been used in prior projects", "project involved the use of new technology" and "immature technology". The studies examined the performance in the literature and decided to examine in two separate groups: process performance and product performance. Process performance is to complete the desired time and budget and product performance is to completed with the desired properties such as ease of use, product qualiy and etc. 3 important feature of the strategic importance of the project, project duration and project sourcing were added to the model. Performance, risk and project characteristics question combined and survey has been prepared. The main hypothesises are the negative effect of the risk dimensions on performance. The modaretor effect of project characteristics such as project duration, strategic importance and project sourcing relationship between risk and performance was observed. After the model was created, two method was selected in order to test the hypothesises. First method was an explanatory factor analysis method. The reason of factor analysis was that the classification of the performance and risk classification in literatüre and the classification of performans and risk classification which collecting data determine whether it overlaps with each other. Second method was Structural Equation Modeling (SEM). The purpose of the SEM was that all relationships could examinate by one model and the margin of error is minimum. Survey method was used in order to verify hypothesizes. Survey included demographic information, project characteristics, risk and performance questions. Risk, performance and project characteristics questions were asked to in mind recently completed project. Project characteristics contain strategic importance, project duration and project sourcing. Strategic importance options are strategic, informational, transactional. Project duration is project completion time. Project sourcing is that project is done internal or outsource. Risk dimensions were examined with 30 questions in total. The performance was examined by 9 questions. Answers questions of risk and performance is measured by Likert scale 7s . Survey has been applied to people working in the Information Technology projects in Turkey. Data were collected by two methods. First method was that survey was formed by "surveey" internet site and the survey link was sent to people. Second method was that survey hard copies were given to people and survey hard copies then picked up. A total of 306 surveys were collected. Model analysis was carried out in six stages. SPSS 21.0 and AMOS 21.0 programs were used in order to analyze model. First, data analysis was performed. The purpose of outlier data analysis is to determine whether a data to mislead the investigation of the relationships between variables. After analysis of data outliers, 274 remained. Kurtosis and skewness value of the data has been checked for compliance with the normal distribution. Normal distribution of the data was found to be suitable. Factor analysis was conducted with 274 data . Before reviewed factor analysis , KMO and Bartlett test results were checked and were found to be suitable for factor analysis. As a result of exploratory factor analysis, if more than one factor under significant question is loaded or located at the bottom of the inverted factor analysis with literature, this factor has been removed. As a result of factor analysis, risk remained 16 questions. After the factor analysis for performance, 7 questions remained. Risk and performance dimensions reliability was checked after factor analysis. Requirement risk has been excluded because response was not reliable. As a result of factor analysis of the remaining variables analyzed SEM work was started. SEM analysis was performed two stages. The first phase of SEM was that forming measurement model and the second phase of SEM was that forming of the structural model. The purpose of the measurement model is to show how much can be measured by the implicit variable observed variables. The analysis of the measurement model was carried out by confirmatory factor analysis. After measurement model created, goodness of fit values has been checked. Some observed variables have been removed because the absence of the expected value of the goodness of fit values of the observed variables and latent variables in the model measure due to the low rate of releases. Achieving the desired value as a good fit in the measurement model and the desired level of measurement models come with variable explained by the observed variable modification was completed. Structural models created and the goodness of fit values have been checked. After providing values of goodness of fit model it was introduced to the model interpretation stage. It has proven planning and control risk has strong negative impact and process performance. It has proven to team negatively affect product performance. The relationship between requirement risk and product/process performance were not observed because requirement risk reliability was too low for analysis. Relationships between project complexity and product performance, between organizational environment risk and product performance, between user risk and product performance, between planning and control risk and product performance were not supported. Relationships between project complexity and process performance, between organizational environment risk and process performance, between user risk and process performance were not supported. In contrary to literature, team risk affected positively process performance. "When product performance increase, process performance increase" has proven. When project is strategic, the effect of planning and control risk on performance is greater. Strategic importance effect of relationship between team risk and product performance has not investigated because team risk didn't affect negatively product performance in operational projects. Planning and control risk effect level was close short duration projects and long projects for this reason, project duration didn't affect relationship between planning and control risk and process performance. Project duration effect of relationship between team risk and product performance has not investigated because team risk didn't affect negatively product performance in long duration projects. Outsource project the effect of team risk on performance is greater than insource project. The level of effect team risk on product performance in outsource project was higher the level of effect team risk on product performance in insource projects. The level of effect planning and control risk on process performance was same in outsource and insource projects consequently project sourcing hasn't affected relationship between planning and control risk and process performance. Information technologies projects in Turkey, the desired time and cost; paying attention team risk and planning and control risk for completing the desired properties should concluded. When completion of project in desired time and budget is important, planning and control processes are placed emphasis. When completion of project in desired features, team education and team communication are placed emphasis. In the project beginning and during the project, if project characteristics (strategic importance and project sourcing) can be pay attention and take actions, the impact of risk on performance can be decrease.