Anecdotal evidence indicates that there is growing interest and concern about what actually happens in the classrooms since the Government of Kenya successfully implemented the Free Primary Education (FPE) policy in 2003. Teachers may be well-trained and yet effective learning still fails to take place. This study provides the first opportunity in Kenya to understand whether classroom interactions, including aspects such as ‘Opportunity to Learn’ and teacher subject knowledge explain why some schools are consistently ranked at the top while others dominate the bottom performance ranks in Kenya’s KCPE1. Research evidence has shown that an important aspect of quality education is the teaching and learning processes that go on in the classrooms. For APHRC, this is ground breaking study that complements its other education research studies which have focused on assessing the impact of FPE on access for the poor residents of the slums of Nairobi.

The study uses cross-sectional data and involved several steps to obtain the sample. Firstly, using KCPE examination results for 2002 to 2005, districts which have consistently been ranked within the top 10%, middle 20% and bottom 10% were indentified. Secondly, from this pool of identified districts, two were randomly selected from each of the three categories. Thirdly, a pool of schools that have consistently been ranked in the top and bottom 20% categories respectively in the KCPE results between 2002 and 2005 in the selected districts were identified. These schools were selected taking into consideration stratification by type so as to include both public and private schools. Finally, 12 schools in each district were randomly selected for the study based on the following criteria: six schools that have consistently been ranked in the top 20% in the district and another six that have been ranked at the bottom 20% in the KCPE results. To ensure proportional representation of both public and private schools, weighting was done according to the number of public and private schools in each of the selected districts. Using this step-by-step process, the following six districts were sampled for the study: Baringo, Embu, Garissa, Gucha, Murang’a and Nairobi. The selected districts represent Western, Central, Eastern, Rift-Valley and Nairobi regions of Kenya. It can therefore be said that the schools included in this study constitute a nearly nationally representative sample.

The instruments for data collection were developed in collaboration with researchers and practitioners with expertise in classroom observations, curriculum development, primary school teaching and assessment. The instruments were piloted in six primary schools, which enabled improvements to be made on various aspects of the questionnaires.

Data collection included the use of video recording and a classroom checklist to capture classroom interactions during lesson teaching; survey instruments for the teachers, head teachers and students; teacher numeracy assessment tool; and a learner numeracy assessment tool. Sixteen field interviewers were trained on how to administer the instruments and in the use of the video cameras.

The first round of field work was carried out between the months of May and July 2009 in 72 primary schools. 211 teachers, 72 head teachers, and 2,437 primary grade 6 pupils participated in the study. The second round of the field work was carried out during the months of February and March 2010. The video clips were analyzed by two experts using a video analysis rubric and the questionnaires were used to develop the analytical dataset on teaching practice and quality of learning in the visited schools. As is required of us by the Kenya Government, the National Council for Science and Technology and the Ministry of Education issued the requisite research authorization for the study, and the draft report has been shared with the Ministry of Education, Kenya.

Summary of key findings and policy implications

• The mean pupil score in Mathematics is 46.89% which is below 50% usually considered to be the pass mark.
• With the exception of Baringo district, girls in the other five districts have on average scored slightly higher (between 0.3 and 3.6 percentage points) than boys.
• The difference in maths mean scores is large (23.83 percentage points) between the top and bottom performing schools in Nairobi and smaller (4.9 percentage points) between top and bottom schools in Garissa.
• The maths mean score for teachers is 60.5% which is only slightly above the average pass mark of 50%.
• Male teachers from top-ranked schools scored higher than male teachers from bottom-ranked schools by an average of 7 percentage points.
• There is a linear but weaker relationship between pupil mean score and teacher score in the bottom ranked schools – which is to say, teachers in the bottom-ranked schools made some difference in the performance of their pupils in Mathematics wheras in the top-ranked schools, this relationship is non-existent. Individual seatwork was a dominant teaching activity in Mathematics lessons; recitation is the dominant activity in English lessons; and whole-class responses was the dominant activity in the Science lessons. However, in the bottom schools, whole-class responses was dominant in both English and Science lessons.
• Use of relevant teaching aids such as a manila paper illustrating a concept and placed on a classroom was found to be important in teaching and learning.
• Learners exposed to more interactive classrooms, for instance recitation with question and answer activities, scored higher marks.
• High performing schools (and by implication students) had higher gain scores than low performing schools (students).
• There is need to increase the availability of non-basic teaching materials through working in subject teams so that the developed materials can be shared across streams and that their effectiveness is evaluated by the subject team.
• There is need for a clear school policy on the development and utilization of teaching aids with a view to ensuring the availability of effective aids in the classrooms.
• There is need to encourage the head teachers, deputy head teachers and senior teachers to institutionalize lesson observation, feedback and professional guidance at school level.
• There is need to periodically assess teachers’ level of competency in the subjects they teach; and to review the policy on teachers teaching all subjects so that teachers can teach subjects they are competent in.
• On social relations, more parental and teacher/head teacher interactions should be encouraged, including parental involvement in what happens in the classroom.
• In low-performing schools, teachers may require more pedagogical skills-upgrading with a view to enabling them shift their lessons to more learner-centered approaches.

* Persons wishing to know more about the study can contact Dr Moses Ngware, African Population and Health Research Center, Nairobi, at: