Policy makers believe that there are ghost workers, persons who do not work yet receive salaries, on Ghana's public sector payroll. However, little is known about the factors that create opportunities for ghost workers. The purpose of this quasi-experimental study was to extend Cressy's conceptualization of the fraud triangle theory to test the applicability of Reinikka and Svennson's graft estimation model by using nonprobability quota sampling to select 85 heads of public agencies for participation in a cross-sectional survey. The research questions focused on the relationship between size of government agency, the estimated number of opportunities for ghost workers, and the dependent variable of occurrence of ghost workers in the public sector in Ghana. Correlational and multiple regression analysis was used to discern the relationship between the independent variables of agency size and opportunities for ghost workers and the dependent variable of number of ghost workers. Results revealed a statistically significant, positive relationship between the number of opportunities for ghost workers and the number of ghost workers. However, there was a negative relationship between the size of government agencies and both the number of opportunities for ghost workers and the number of ghost workers. The implications for social change include recommendations to revise the Financial Administration Act of Ghana by introducing new controls in the payroll administration at the Controller and Accountant General's Department to eliminate ghost workers from the payroll so that public funds can be saved to provide more public services for Ghanaians.