本研究應用吸收性馬可夫鏈，分別建立專科和大學教育背景之台灣護理人員流動率預測模式，模擬不同學歷護理人員停留於不同專業階段之人數、機率和停留時間，並以某大型區域醫院為實證模擬對象。研究結果顯示模式精準率高達94.35%至99.93%之間，且發現護理人員在晉升至N3級前，於N、N1、N2級停留時間過長，且由N2級晉升至N3級人數過少，導致N2級以下護理人員離職率過高。此外，發現大學學歷護理人員晉升時間較專科護理人員短2倍以上，但離職率卻是專科學歷護理人員之2倍。最後，本研究發現可提供相關行政管理單位與醫療機構，培育人才、制定不同職階之留任及晉級輔導措施之規劃參考。

This study employed the Markov chain to establish a model to predict the turnover of nursing staff of different education backgrounds in Taiwan, which could simulate the number, probability and retention time of nursing staff of various professional levels. The simulation data of this study came from the nursing staff of a large hospital in Taiwan. According to the simulation results, the accuracy of the nursing staff turnover prediction model was up to 94.35~99.93%. The results suggested that, before being promoted to the N3 level, the average retention time of nursing staff was too long and the number of promotions was too small, resulting in an excessively high turnover rate of nursing staff below the N2 level. In addition the turnover rate of nursing staff with a university education background was higher than that of nursing staff with a junior college education. The research findings could provide medical institutions in Taiwan with a reference for the planning of retention and promotion counseling for nursing staff of various professional levels.