Profile of Research Methodology and Statistics Training of Undergraduate Medical Students at South African Universities

  • J Dommisse
  • G Joubert


Background: Medical practitioners need to have knowledge of statistics and research principles, especially with the increasing emphasis on
evidence-based medicine. The aim of this study was to determine the profile of research methodology and statistics training of undergraduate
medical students at South African universities in terms of which topics are taught, by whom teaching is done, when these topics are taught and how
they are taught.
Method: Respondents for this descriptive study were persons responsible for the teaching of statistics and research methodology at the eight
medical schools in South Africa. They were identified by the head of each school who also gave permission for the school to participate. The
respondents completed a questionnaire and checklist after giving informed consent. No response was obtained from one university. Responses were
compared to international guidelines.
Results: At five universities the material is taught in the first year, at one in the second year and one in the third or fourth year, depending on when
it is selected as an elective. The material is reinforced in other modules in the medical programme at three universities. The persons responsible for
teaching are mainly statisticians (six universities). Class sizes vary from 40 to 320 students with four universities having 200 or more students per
class. At two universities the current course has been in place since 2003, at two since 2000, and at two since the 1970/80s. The following topics
are taught at the majority of universities: study designs in medical research, exploring and presenting data, summarising data, probability, sampling, statistical inference, analysis of cross tabulation and critical reading. At four universities there are practical classes, three of these mainly for computer work. At three universities tutors are used, at two of these the tutors are postgraduate students in statistics whereas at one university registrars, doctors and researchers are used as tutors. Students at three of the universities complete a research project, at two of these the students complete the full research process from planning up to reporting, whereas the project at the other university focuses mainly on the analysis of data.
Conclusion: Recommendations have been made regarding topics which should be covered and teaching methods which should be used at all
universities. Doctors should be involved in the training to ensure clinically appropriate material and examples.

Journal Identifiers

eISSN: 2078-6204
print ISSN: 2078-6190