Sign In

 MOTHS Project

This project offers a contribution to the our understanding of the global phenomenon of the mobility of the highly skilled, with a focus of doctoral and post-doctoral graduates.

Such mobility has characterized shared learning over the ages.  What makes the present phenomenon different however, is the scale, scope and speed of mobility. Mobility comprises three main dimensions, often categorized as brain drain, brain gain and brain circulation.  

These three categories may be further disaggregated as suggested in the table below. 

Table 1: Categories of mobility

Category Internal External
Brain drain Job mobility Emigration
Brain gain Job mobility
Staff development Immigration
Brain circulation Job mobility Exchange
Staff development

Such disaggregation is understood from the perspective of a given organization. So, for example, seen from the perspective of a local university:
Brain drain might entail a graduate who remains in country, but who moves to a non-research position. 
Brain gain refers to the recruitment of a staff member from abroad, and possibly even from another local institution.   
Brain circulation refers to both inward and outgoing movement of graduates who exit and return either to the local host or a foreign host.

The MOTHS Project focuses on a large subgroup of the mobile, namely students from African countries who come to study in South African universities. 
As Table 2 shows, South Africa plays a major role as a site for inward student mobility from across Africa.  At postgraduate level she serves as the hub of Sub-Saharan Africa. 

Table 2: Origins and destinations, 2015


Table 2 includes a column showing the proportion of students studying abroad for each country.  One notes the exceptionally high value for Zimbabwe.

South Africa is a signatory to the SADC Protocol on Education and Training, that conjoins Member States to allocate and accommodate 7% of SADC students on ‘home’ fees in their respective universities. 

It so happens that South Africa is the only Member State that meets this target, implicitly supporting high level skills development for the SADC and across the sub-continent.

Some might argue that these transfers amount to a repayment of the historic debt that South Africa owes to the sub-continent through its more than a century of migrant labour exploitation.   The grandchildren and children of those migrant miners who worked below the ground are now the migrant students who fly way above it.  At postgraduate level the complement of foreign students is much higher than the 7% average.

As Table 3 shows, in academic year 2015, foreign students comprised 17% of Masters enrolment, 37% of doctoral and 63% of post-doctoral students. 

Table 3: Postgraduate enrolment, 2011/12-2015/16
Source: National R&D Survey, 2015/16.

These foreign students are a very important resource to their countries of origin, and given South Africa’s skills shortages, an important resource for this country as well.

The National Development Plan is quite clear on the issue, suggesting that upon graduation, doctoral students should be furnished with seven-year work permits.  

No such policy has as yet been enacted.

Over 2012-16, the universities graduated a total of 8036 doctoral candidates, of whom 4170 are South Africans, including 1904 classified as ‘white.’  

The adjective ‘white’ is shown in quotation marks as the application of the old apartheid era Population Act classification is quite arbitrary when applied to foreign students.  

Table 4 lists doctoral awards for countries with 40 or more graduates.

Table 4: Doctoral degrees by selected countries, 2012-16

Country Number
Botswana 74
Cameroon 122
Canada 45
Congo 42
Dem Rep Congo 55
Ethiopia 131
Germany 54
Ghana 121
India 68
Kenya 178
Lesotho 108
Malawi 118
Mozambique 50
Nigeria 529
Rwanda 71
Sudan 52
Swaziland 59
Uganda 141
United Kingdom 56
Tanzania 99
USA 122
Zambia 113
Zimbabwe 728

This implies in that in the order of 2900 African migrant students, slightly more than a third of all doctoral graduates students, acquired a doctorate in a South African university during 2012-16.   

This number is greater than the 2200 awards to South Africans other than ‘white.’  No further racial breakdown will be applied in what follows.  

It is noteworthy that some 450 degrees were awarded to students from the European Union and the USA.

Accordingly, one seeks to understand the career path of the African doctoral graduate, posing the question ‘where are you now?’  Have you gone home; are you in Seattle or Zurich; are you around the corner doing a post doc?

Various aspects of the MOTHS have received prior attention in the hands of local scholars, notably Kaplan, Meyer, Kahn, Pogue, Meyer, and Hopli, but the above questions elude a definitive response.

A full literature study on the MOTHS phenomenon has been completed as a separate study. 

The MOTHS Survey

The Department of Home Affairs no longer collects emigration data, so that studies on brain drain that wish to use official data must rely on foreign sources, that are often incomplete. 

This data problem was experienced by all of the authors cited above.

In addition, HEMIS data on the nationality of students is not available at the individual level.   Table 4 is more or less the best available data on nationality.

A consequence of the above information gap is that data on graduate production does not provide an accurate picture of domestic production, a matter that makes planning that much more difficult.

To address some of the above shortcomings, the MOTHS Project sought to capture primary data on doctoral graduates, the intention being to conduct a statistically valid survey.

The instrument chosen for this purpose was a modification of the questionnaire that was developed for the OECD/Eurostat Careers of Doctorate Holders project of the early 2000s. 

Ideally a university would provide the Principal Investigator with a list of doctoral graduates, with appropriate biodata and contact details. 

Subject to the usual ethical clearance procedure the Principal Investigator could then contact the graduates and bring them into the intended survey frame.

In practice, this proved to be nigh impossible. 

Universities restrict access to such data for a number of reasons:  foremost among these are the risk of abuse of the information for commercial purposes, and secondly the complexities associated with the Protection of Personal Information Act. 

Accordingly, a compromise solution was invoked, whereby the Alumni Office of a university elected to identify the target group of graduates and send them an email request to participate in the online survey.  

The agreed survey protocol thus rendered the PI blind as to the identity of a respondent.

All online surveys experience problems with return rates, the more so where direct interaction with the target group is disallowed. 

Accordingly, one may have to accept a return rate in the range 7% to 10% as the best that might be hoped for.  

A total of eleven ‘traditional’ universities were selected for the survey. Their total doctoral production accounted for some 65% of the country total. 

The alumni offices were requested to identify foreign African doctoral graduates for each year over 2012-2016, and to send each graduate an email solicitation that hosted a link to the online questionnaire.

In the event eight of the alumni offices sent out questionnaires, with four sending a follow up reminder. 

The return rate among these eight universities varied considerably, from as low as 1,5% to a ‘high’ of 10,5%.  One leading research university failed to participate.

This indirect approach resulted in an overall response rate of 6,5%.

The return rate was further hindered by the likelihood that contact information becomes compromised with time, and that no robust means of follow up was possible.

Alternative Approach

Faced with such a low response rate, an alternative way of assembling the survey frame had to be sought.

One possibility might have been through collecting university graduation ceremony lists that in some cases include a mini-biography of the doctoral candidates.  Unfortunately, most universities do not publish such lists and the information is not readily available on web pages. 

The microdata collated by the Department of Higher Education and Training to the HEMIS system is also unavailable for general research purposes.

This left but one possibility, namely to search the National Electronic Theses and Dissertation Portal that is maintained and held by the National Research Foundation, with each university also hosting its own parent repository.  

The elegance of such an approach is that one is using information that is already in the public domain.

Unfortunately, the database is unstructured, with no specified fields, and no metadata, so that search is a very cumbersome labour intensive process.  Moreover, there is no field specification of the nationality of an awardee.

By comparison, the equivalent database in Germany was established in 1969, hosts metadata, and is structured for search.

How then to determine who the foreign African doctoral graduates might be?

In a few cases, an extensive dissertation acknowledgement page might furnish direct information or clues as to the home base and other circumstance of the graduate, as for example where the graduate thanked an identified employer for granting study leave.

Such ‘hits’ are the exception rather than the rule.

It is at this point that experience and serendipity offered a way forward. It so happens that the PI spent thirteen years across the Limpopo teaching at the University of Botswana where he met with staff and students from across the continent, and thereby gained insight into naming patterns of various ethnic groups.

Accordingly, it was possible to sift through the dissertation records of a university for a given year; parse out the doctoral dissertations, and then to identify the foreign African awardees.  This required visual search of anything up to a thousand records per university, per year. The total number of foreign graduates so identified showed a 95% agreement with the HEMIS country totals for reach of the selected universities. 

The next step was to triangulate those so identified by enlisting the services of a foreign African scholar who would produce their own records independently, thus allowing a cross-check.

The concordance between PI and African scholar was better than 95%.

So, the first step toward compiling a database of foreign awardees became a possibility.

The method was applied to the five research universities for the years 2012 and 2013, that led to the identification of 420 foreign African graduates.

Question was, how to contact the graduates?  The solution – social media.  Extensive search of LinkedinTM and FacebookTM gave rise to 360 email addresses. 

In the first round of dispatch, 336 emails were transmitted without rejection.  

Preliminary results and comments

Some preliminary observations are available from the accumulated data.

A thumbnail sketch serves to highlight what may be gleaned at a high level.

Quite naturally all fields in all records have not been completed. However, this much may be stated:
Age on graduation was ± 43 years
24% of graduates are female
44% of students enjoyed South African financial support
28% had foreign financial support from employers, government or donors
19% made use of loans or own funds 

Moving into employment after graduation was seamless, since in many cases the students were on study leave from their home employer. 
70% of graduates are found to be employed in Africa, in their country of origin, of whom 75% are in universities, and 15% in research organizations.
19% of graduates are employed in South African universities or science councils.
7% of graduates are employed outside Africa.

Provisional estimates suggest that the above hold true at the 95% confidence level.

It is clear that the majority of graduates are employed outside South Africa.

This phenomenon arises through an in-built selection bias in that the majority of inward bound students are mid-career professionals, often on study leave from their employer at home, where they also have young families. 

It is thus to be expected that they will return home.

Among those employed in South Africa, most are in temporary positions, while a few were employed as post docs.

The overall picture is that South African universities succeed in educating competent, high level skilled people, who are equipped to contribute at home base.

They complete their doctorates in the minimum 3-year period, publish, and in many cases maintain academic contact with universities in South Africa.

The MOTHS pilot project identifies the doctoral programs as an example of brain circulation. 

Table 5 summarizes these flows from the perspective of a source country. 

Table 5: Migrant student flows

Category Internal External
Brain drain Unknown 19% to South Africa
7% out of Africa
Brain gain Unknown Unknown
Brain circulation Unknown 70% return

Together with the post-doctoral programmes these contribute to maintaining an open scientific culture in our universities that interacts across the world. 

Graduating students from 97 countries is testimony to this.

Mining of dissertation databases has been shown to be a viable alternative to tracer studies, and may be applied to other groups of interest.

The data mining approach supplements bibliometric techniques that have become a standard way of tracking the  mobility of researchers who publish in the academic literature.


UWC Supports Sustainability
© 2013 UWC | Disclaimer | Sign-in

Contact Centre

University of the Western Cape,
Robert Sobukwe Road,
Bellville, 7535,
Republic of South Africa | +27 21 959 2911