I believe that we want our students at all levels to be well educated, so that everyone can be employed.  At the masters, and PhD levels it is challenging to ensure that students complete what they set out to achieve.  Have a look at this paper.  I hope it is useful to you.

"The ability of the United States (U.S.) to compete in a global market is threatened

by the loss of skilled workers. By 2012, the number of jobs requiring advanced skills will grow at twice the rate of those requiring basic skills (Hecker, 2004; U.S. Department of Labor, 2000). Each person who drops out of college adds to an already overcrowded unskilled and semi-skilled work force..."

https://libres.uncg.edu/ir/asu/f/Lindheimer,%20Jacob_2011_Thesis.pdf

For the online classes, you want to monitor their participation rates in activities as once that starts waning, shortly after they drop out. International students do language barriers and cultural sensitivity issues that if not addressed can also lead to their dropping out of online classes. The attached describes some of these challenges that I hope can help in addressing student retention.

Many thanks,

Debra

Data Generation I: International and invisible in a workforce edu...

Article Cultural Sensitivity Needed in Online Discussion Rubric Language

Student retention is worldwide problem. There are numerous academic and non-academic factors that contribute to it. Some colleges and universities use what is called a risk prediction instrument. Students who are at risk of attrition can be identified early enough to get the various ways of support that may help in retaining them.

The following may be helpful:

- Barefoot, B. O. (2004). Higher education's revolving door: confronting the problem of      student drop out in us colleges and universities. Open Learning, 19(1), 9-18.

- Campbell, M. A. (2013). The impacts of intrusive advising on the persistence of first-year   science, technology, and mathematics students identified using a risk prediction instrument. (Doctoral dissertation). Temple University, Philadelphia.

- Hausmann, L. R. M., Schofield, J. W., & Woods, R. L. (2007). Sense of belonging as a     predictor of intentions to persist among African American and White first-year           college students. Research in Higher Education, 48(7), 803-839.

- Parmar, D., & Trotter, E. (2004). Keeping our students: identifying factors that influence student withdrawal and strategies to enhance the experience and retention of first-   year students. Learning & Teaching In The Social Sciences, 1(3), 149-168.

- Tovar, E., Simon, M. A., & Lee, H. B. (2009). Development and validation of the college mattering inventory with diverse urban college students. Measurement and Evaluation in Counseling and Development, 42(3), 154-178

Retention has two different meanings in the context of education.  One is retention of knowledge, meaning can they still recall and use the information days or weeks after being exposed to it.  

The other meaning of retention is students who return next semester or next academic year and do not drop our or transfer to another school.

Fully agree with the opinion given by Michael W. Marek about retention.

http://www.teach-nology.com/highered/

http://www.truth-out.org/opinion/item/33767-higher-education-is-a-key-to-decarceration-let-s-pass-the-real-act

http://www.thehindu.com/news/national/andhra-pradesh/real-issues-of-students-in-higher-education-ignored/article8147990.ece

https://www.linkedin.com/pulse/real-disruptive-force-higher-education-collaboration-michael-crow

The notion of retention is grounded in student success. Retention related activities focus on providing a campus environment where students successfully complete their goals and complete their academic program from an institution. The possible ways of  measure retention are: Retention consulting,Student Satisfaction Inventory Institutional Priorities Survey, Adult Learner Inventory, Priorities Survey for Online Learners, Retention Management System/College Student Inventory, Advising Webinar Series.

Since you mention persistence as well as retention in your question, I will assume you are interested in measuring dropout rates among students who are initially enrolled in a formal program of study. If so, some of the earlier answers to your question provide some good resources to look at. My point in writing is to add a criticism of the usual ways of measuring retention of students. Perhaps this will be relevant to your study.

So often, retention is measured by looking solely at the percentage of students who successfully complete a program of study within a certain time-frame at the same institution in which they were situated at the beginning of the measurement period. This way of thinking about retention, focused as it is on just one educational institution and its students, and with a set time-frame, has value for showing something about the characteristics of that institution. I do not mean to imply that such a measure of retention is necessarily wrong. But before buying into that single-institution-based measurement, be aware of ways in which it can be very misleading in certain social conditions.

In a social situation in which there is considerable instability, due to economics, politics, or other factors, the standard approaches to retention measurement may not provide much useful information. For example, in the US, five-year graduation rates are typically used for institutions of higher education. But if a particular institution enrolls a large number of individuals who are self-supporting, with children to support as well, these individuals have no chance to enroll in enough courses within five years to graduate. Or, if an institution serves families of itinerant workers, it is very unlikely that those students will be able to complete their studies at just one institution, even though they may persist all the way to graduation by attending two or three institutions.

For social situations in which stability and steady progress at just one institution is not likely, other, much more difficult measures of retention may be appropriate. One needs a way to track student progress over a fairly long time frame, presumably by keeping in touch with those students directly in order to gather information from them. This is quite difficult from a practical point of view, but it is as far as I know the only way to gather the needed data, unless you have access to some sort of national system of keeping track of individual progress.

To sum up, the first thing is to decide what information you need in order to learn what you want to know. Do you really only care about retention at a given institution? Have you taken into account what that approach will leave out?