The value of graduate destination survey data in understanding graduate unemployment : a focus on the universities of technology

• Published date: 01 January 2015
• Author: Andre Kraak
• Pages: 93-113
• DOI: 10.10520/EJC170746
• Date: 01 January 2015

South African Journal of Labour Relations: Vol 39 No 1 2015 93

Correspondence to: Dr A Kraak, Researching Education and Labour Centre, School of Education, University of the

Witwatersrand, P/bag 3, Wits, 2050. E-mail: andre.kraak@wits.ac.za

The value of graduate destination survey

data in understanding graduate unemployment:

A focus on the universities of technology†

by Andre Kraak*

Abstract

This article contributes graduate destination survey (GDS) evidence to the debate

about graduate unemployment in South Africa. There has been lively contestation

on the topic for several years, including several contributions from the commercial

press arguing that graduate unemployment is very high. In contrast, academic

evidence (based on national labour force data for the period 1995–2011) has

been presented suggesting that the unemployment of graduates in South Africa is

minimal, on average only 4.9% in 2011.

New evidence has emerged from two recent GDSs – one comprehensive

survey of all 2010 graduates across all qualification levels at all four universities in

the Western Cape, and a second focusing only on the 2012 Bachelor of

Technology (BTech) graduates at the Vaal University of Technology (VUT) in

Vanderbijlpark, Gauteng. These two studies, using the same methodology and

online questionnaire, provide a more accurate picture of the graduate

unemployment problem in two important economic regions of the country.

The results show that although rates of unemployment are low at the elite

University of Cape Town (UCT) and Stellenbosch University (SU) (graduate

unemployment is between 3 and 6%), rates are much higher at the Cape

Peninsula University of Technology (CPUT) – a former historically disadvantaged

technikon – with 15.8% unemployment among CPUT students. African

unemployment at CPUT reached 20.2% among all first-time entrants (as

compared with 4% for whites), suggesting the continuation of a racially stratified

labour market for highly skilled labour long after apartheid’s demise.

Similarly, unemployment rates among the BTech VUT graduates of 2012

reached 18%. This is an extremely high rate for fourth-year graduates of a

polytechnic-type institution whose primary mandate is to place qualified graduates

in jobs in the mainstream economy.

Key words: graduate unemployment, higher education, Graduate Destination

Surveys (GDSs), intermediate skills

†Acknowledgement

The author hereby acknowledges the important role of the Cape Higher Education Consortium (CHEC)

and the Office of the Executive Director of R esearch at the Vaal University of Technology (VU T) in

commissioning and funding the two Graduate Destination Surveys reported in this article. Permission to

use the data has also been granted by CHEC, the Planning Office at the Cape Peninsula University of

Technology (CPUT) and VUT. These studies would not have been possible without their active support.

* Dr Andre Kraak is a Visiting Associate Professor at the Researching Education and Labour (REAL)

Centre, School of Education, University of the Witwatersrand.

94 South African Journal of Labour Relations: Vol 39 No 1 2015

1 Introduction

The aim of this article is to present graduate destination survey (GDS) evidence in the

debate about graduate unemployment in South Africa. Data currently available suggest

highly divergent trends – for example, extremely high levels of graduate unemployment

(32% in 2005) suggested by a research council (Letseka, Cosser, Brier & Visser 2010),

and extremely low levels argued by a university-based econometrician – 4.8% in 2011

(Van der Berg & Van Broekhuizen 2012). Adding to the debate in 2013, the Cape

Higher Education Consortium (CHEC) undertook a GDS of the entire 2010 cohort who

graduated at the four public universities in the Western Cape. They concluded that

graduate unemployment was 10.1% on average. And more recently, the VUT has

determined through a GDS that 18% of its BTech graduates (third-year diploma

students who do a fourth year to acquire a bachelor’s degree) are unemployed.

Trying to make sense of the divergent trends indicated above is a huge challenge. A

major part of the problem is that analyses based on labour force surveys and other

Statistics South Africa data (see DPRU 2006; Van der Berg & Van Broekhuizen 2012)

cannot be compared and contrasted with data generated by GDS surveys – which are

done at the level of individual institutions or regional groupings of institutions. The

former data sets are national in coverage; they are repeated regularly and are

managed and quality assured by the national statistical service of the state. They do

not reflect specific rates of unemployment for specific cohorts of graduates, but rather,

they reflect the prevalence of unemployment among graduates at particular points in

time irrespective of when they graduated. In contrast, the GDS studies are cohort

studies of graduates in a particular year and in particular institutions or groupings of

institutions. They reflect more accurately the extent to which specific cohorts are

exposed to the risk of unemployment.

This article does not focus on Stats SA data. It argues in favour of using GDS results

on a wider scale to determine the extent of graduate unemployment in the current

period. The choice of GDS methodology is supported by the fact that it is widely utilised

as part of the planning activities of higher education institutions globally that are

seeking a better understanding of graduate pathways into employment.

The structure of this article is as follows. The analysis first examines the benefits of

GDSs for university managers and planners. It then examines the extent of graduate

unemployment across the globe. Thereafter, it summarises the results of two GDSs

that have been undertaken in South Africa – the only GDSs to have been executed and

publicised in the recent past. The author of this article was the principal researcher in

both the Western Cape and the VUT studies. This article publishes the results of these

surveys in an academic journal for the first time.

The concluding section provides a comparison of BTech graduate destinations

across two universities of technology located in very different regions of the South

African economy. It provides fresh insight into the differential treatment of graduates

seeking employment in differing regions of the national economy. It also highlights the

importance of two niche areas in the higher education system which do not get

sufficient analytical attention in studies on South African higher education: (1) mature-

age learners (those students who were employed while or prior to studying); and (2)

continuing students who, after graduating in 2010 and 2012 respectively, have

proceeded to study further towards higher degrees. Finally, the article examines the

important role of social networks and “social capital” in assisting young graduates to

obtain employment.