Biometrics training for MSc students in agriculture: Approaches being undertaken in the Faculty of Agriculture, University of Nairobi

L. Akundabweni

Department of Crop Science, Faculty of Agriculture, University of Nairobi
P.O. Box 29053, Nairobi, Kenya
E-mail: campusblooms@swiftkenya.com

Introduction

This presentation uses a Rockefeller Foundation supported project in the Faculty of Agriculture at the University of Nairobi (UON) to show how it is possible to leverage outside support to strengthen areas of weakness and build partnerships that can enhance biometrics training in universities.

Existing applied biometrics in MSc training

MSc courses are available within the Faculty of Agriculture, UON for a number of disciplines, e.g. crop science, soil science, animal production, crop protection, agricultural economics, range management, applied nutrition and food science and technology. A lecturer in biometrics in the Department of Crop Science teaches biometrics to all the students together, except for those in agricultural economics. The needs of the students in agricultural economics are not catered for in this curriculum. In addition, owing to poor previous training in statistics, students in applied nutrition have currently been separated to follow a less rigorous course. The MSc training is for two years. The first year is entirely course work when most of the biometrics teaching is given; students carry out their research projects in their second year. Funding was provided in 1998 by The Rockefeller Foundation to strengthen the biometric input into MSc student training. The funding was primarily used to set up a modest computer laboratory in the Department of Crop Science. Funding was insufficient to hire a computer manager. Instead we hired an assistant with some computing knowledge to help with hardware and software maintenance, and organised a crash course for other staff in order to develop a general level of basic computing competence among those involved in teaching in the laboratory. This worked reasonably well and we were able to put in place adequate practical computing and statistical skills to support the research phase in year two of the MSc course (Figure 1). So we thought!

Figure 1. The progression in biometric and computing skill training needed for successful completion of the research project in year two of an MSc in Agriculture.

Improvements required

Figure 1 demonstrates the progression that students are expected to follow during the execution of their research projects. During the preparatory phase at the end of the first year students are expected to have defined research objectives, have undertaken bibliographic searches, defined null hypotheses, planned the design of their experiments, the measurements to be recorded and the methods of data collection. They are also expected to have acquired sufficient skills in spreadsheet and statistical software to plan the statistical models they will fit to their data. However, students received a major shock at the end of year one when suddenly they found themselves thrust into the situation of applying what they had learnt for real. Indeed, a questionnaire to students at this time demonstrated the need for extra training prior to the writing of research proposals. At this stage the assistance of the International Livestock Research Institute (ILRI), where one of the junior biometricians on the staff of the Department of Crop Science was currently on attachment, and also the Department of Mathematics at the UON, was sought. A week-long workshop was organised in which the students participated in sessions on development of research objectives and hypotheses, data handling and management, statistical analysis by Genstat and interpretation of results.

Achievements of workshop

The week-long course resulted in some major achievements.

• Fuzzy experimental objectives were clarified and the hypotheses to be tested better understood.
• Experimental designs were in many cases revised and often simplified, e.g. a split-split-split-plot reduced to a split-split plot. There appeared to be the mistaken idea in many students' minds that the more complicated the experimental design the more impressive would be the thesis.
• All four members of staff teaching biometrics in the department attended and interacted with the students. Hopefully this will have increased the likelihood of increased consultations during the research and analysis phase. Previously students had worked entirely under the direction of the supervisors with little interaction with the biometrics members of staff.
• An improved understanding was achieved for the measurements that were necessary to collect during the students' experiments and those that were not. The importance of writing down statistical models and doing dummy statistical analysis runs prior to starting their project was better appreciated.
• The possibility of a stronger link between biometricians in the Faculty of Agriculture and those in the Department of Mathematics, UON and ILRI was initiated.
• So successful did this approach appear to be that it was felt that a similar course for supervisors might be equally beneficial!

Future

A questionnaire completed by students at the end of the workshop highlighted the areas in which they felt they were strong and those in which they were weak. The weak areas will be considered for emphasis during the normal classwork sessions.

A major lesson that we have learnt over the past year is that, given the will, great things can be achieved from a small beginning, and, by working together, a little funding can go a long way to ensuring the relevance and usefulness of the existing postgraduate syllabus to the quality of future agricultural research in the region.

We should now be able to build on what we have achieved so far by adopting our research methods to meet the needs of the changing scene in agricultural research. Through the linkages being established with other biometricians in Nairobi there should be improved opportunities to work together on the tools required to support the new areas of on-farm agricultural research being developed. Not only will this require close working relationship between biometricians themselves but also between biometricians and scientists, both within and outside the faculty.