Assessing the progress and the underlying nature of the flows of doctoral and master degree candidates using absorbing Markov chains

Abstract

In this paper, absorbing markov chains are used to analyse the flows of higher degree by research candidates (doctoral and master) within an Australian faculty of business. The candidates are analysed according to whether they are full time or part time. The need for such analysis stemmed from what appeared to be a rather poor completion rate (as reported by the University concerned). However, this reported completion rate was a 'macro' figure that aggregated PhD and Master (by full time and part time) completions together. If there really was a problem, then the Faculty needed to know where in the system it was and what potentially might then be done to remedy the problem. The assembling of data into an appropriate database was difficult and required considerable care, but an adequate database was developed. The application of the absorbing markov chains to the problem resulted in an increase in the level of understanding associated with the long term underlying probabilities of completion and also expected durations of candidature. This information highlighted one area of concern. With the information gained from this study, appropriate strategies can subsequently be put in place for the rectification of some of the areas of concern discovered in the analysis. Areas for further research are also highlighted in this paper.

A control rule for planning promotion in a Nigerian university setting

“…The Markov chain model provides a means of unifying the states of the manpower system. More so, the Markov models have been used as a theoretical framework to assess the underlying nature of the flows of students through educational systems [8,19,20]. This study complements the existing literature on the mathematics of manpower planning by focusing on the aspect of promotion control.…”

“…(19) Finally, the question as to whether implementation of the optimal policy on the manpower system leads to retrenchment is addressed. This is done by introducing a vector…”

Projection Model of Postgraduate Student Flow

“…The Markov chain method can also measure detailed information on the students' progress such as the average time students spend in an education system whereby other techniques like regression and ratio are unable to measure this [5], [1], [6], [7]. Moreover, the Markov chain model is effective for predicting enrolment for a long range planning horizon which is over 5 years [8]. The capability of certain approaches to make a long-range planning is important for any institutions to give them directions and establish control over its own future.…”

“…Transient state refers to the state where subject moves from state i during one period to state j in the next period, while the term absorbing state refers to the state where subject enters one state which is impossible to leave [9]. The transient states used in this study are the year of course enrolment [10], [11], [8], [12], [13] and the age of course enrolment [14], [12], [13], while the absorbing states for this study are dropouts and graduates from courses [14], [11], [8], [3], [12], [13]. A one-year period is used as the period of transition states [3], [12], [13], [8].…”

“…The transient states used in this study are the year of course enrolment [10], [11], [8], [12], [13] and the age of course enrolment [14], [12], [13], while the absorbing states for this study are dropouts and graduates from courses [14], [11], [8], [3], [12], [13]. A one-year period is used as the period of transition states [3], [12], [13], [8]. In this study, the specified states and year of duration are used to develop the Markov chain model for postgraduate students flow.…”