英文摘要 |
Hospital managers usually encounter such problems as how to make cost comparisons between departments, and how to compare the past cost with the present one or the standard one at the same department. The purpose of the study is to explore an effective way to propose unusual cost reports for further study. We picked up telephone fees as the indicator in a regional hospital, scientifically sifting out the departments with unexpectedly-high telephone fees. We used Rasch analysis (Rasch, 1960) to transform raw data with natural logarithm (Ln) function into a Likert-type scoring scales to examine whether all items meet model’s expectation. WINSTEPS computer software was performed and the rating scale model was applied to investigate the unidimensionality of measured characteristics (consumption of telephone fees) and to generate easily-read reports, on which items (months in this study) with Outfit and Infit MNSQ beyond 0.05 and 1.5 are deemed as misfits to the Rasch model and considered for deletion from the assessed sequence. Successive Rasch analyses were performed until a final set of items satisfied the model fit requirements. The results show that Rasch analysis could (1) pick up abnormally suspected departments for further examination in telephone cost consumption per month, (2) disclose the most unexpected units for responding to monthly telephone fees for further improvement, and (3) illustrate a visual KIDMAP diagram to explain the most unexpected units in telephone fees across months. KIDMAP can help managers detect the suspicious units for consuming telephone fees in a month effectively. |