1. 1 研究方法報告 Controlling Information Systems Development Projects: The View from the Client 楊曉微 許全佑

2. 2 研究動機 the need to better understand how client liaisons exercise control during systems development projects.

3. 3 研究目的 a need to examine the ways in which client liaisons structure their use of control on IS projects.

4. 4 研究模型 (Research Model) Behavior observability Outcome measurability Behavior control Outcome control Clan control Self control Client’s understanding of the ISD process Predicting the Choice of Control Mode 自變數 (IV) 應變數 (DV) 調整變數 (MV)

5. 5 Definition Control: the set of mechanisms designed to motivate individuals to work in such a way that desired objectives are achieved(Kirsch 1996,Jaworksi 1998) client liaison: the person whose business unit is or would be significantly impacted by the system and who was responsible for oversight of the project from a client perspective

6. 6 Definition(cont.) Behavior Observability: the extent to which the client liaison has access to information that reveals individual behaviors and actions Outcome Measurability: the client liaison’s ability to measure outcomes and results

7. 7 Definition(cont.) Behavior control: The client liaison expects the IS project leader to follow articulated procedures and rules to transform inputs during ISD process Outcome control: the client liaison expects the IS project leader to achieve specific project outcomes and targets

8. 8 Definition(cont.) Clan Control: the client liaison becomes part of a project team clan by instilling,embracing,and fostering shared values and goals among the projectteam,and common approaches to working on the project Self Control: The IS project leader autonomously defines goals and behaviors for the project

9. 9 研究方法論 (Research Methodology) Large-scale questionnaire survey 2 separate questionnaires: reduce common methods bias IS project leaders Client liaisons Provide data about the the use of 4 control modes Behavior Outcome Clan Self Indicate their ability Behavior Observability Outcomes Measurability Client’s understanding of the IS development process

10. 10 Organizations and projects the researchers contacted senior IS executive in a variety of industries in the South, East, and Midwest The data collected between July 1997 and January 1999 77 projects from 32 firms data collected from most large for-profit firms, three large public universities and four government agencies sixty-nine usable pairs(90%response rate)

11. 11 2 criteria of Project Selection The development projects should have been completed during the last 12 months have been underway for at least 12months and have completed one major client deliverable Access to both IS and client personnel on systems development projects was mandatory for data gathering

12. 12 假設 (Hypotheses) (cont.) Hypothesis1 High levels of outcome measurability will be associated with the exercise of outcome control.

13. 13 Hypothesis2 High levels of behavior obervability and client’s understanding of the IS development process will be associated with the exercise of behavior control. Hypothesis3 High levels of behavior observability and low levels of client understanding of the IS development process will be associated with the exercise of clan control. 假設 (Hypotheses) (cont.)

14. 14 Hypothesis4A Low levels of outcome measurability will be associated with the exercise of self-control. Hypothesis4B Low levels of behavior observability will be associated with the exercise of self-control. 假設 (Hypotheses)

15. 15 研究工具的使用 Control Mode = β 0 +β 1 (Project size) +β 2 (Project leader experience) +β 3 (Behavior Observability) +β 4 (Outcome Measurability) +β 5 (Client Understanding of the Development Process) +ε

16. 16 相關係數矩陣

17. 17

18. 18

19. 19

20. 20

21. 21 Sensitivity analysis Goal: check the robustness of the results 1. add an additional control variable 2. concern the effects of project size 36 additional regressions were run in which combinations of the largest and smallest projects (by cost,effort,and duration) were removed from the sample