| dc.description.abstract | There exists long standing challenge of low test scores among girls in mathematics at secondary school level in Kenya. A number of studies both quantitative and qualitative have attempted to find out causes of this challenge. Factors which have been suggested in relation to low test scores in mathematics among girls have majorly focused on school environment, biological factors and attitude. There was therefore need to conduct a study to establish the influence of classroom based predictors on test scores in mathematics among girls in public secondary schools in Kisumu County. Classroom based predictors studied were pedagogical techniques applied by teachers of mathematics, teaching load, textbook ratio and class size as independent variables and Kenya Certificate of Secondary Education (KCSE) examination results of girls who sat for the examinations between 2010 and 2014 as the dependent variable. The objectives of the study were to determine the influence of pedagogical techniques on test scores in mathematics among girls at form four level, to establish the influence of teaching load on test scores in mathematics, to find out the influence of text book: student ratio on test scores in mathematics and to determine the influence of class size on test scores among girls in mathematics at form four level. Convergent parallel design under mixed method research approach was applied where both quantitative and qualitative data was collected and analyzed to examine the degree of influence that exists between the variables. The study was conducted in Kisumu County, Kenya. The target population of the study consisted of 142 public secondary schools which presented female candidates for KCSE between 2010 and 2014, 142 Principals, 142 Heads of mathematics department and 390 mathematics teachers who taught the girls in Kisumu County. Saturated sampling technique was applied to select all the 18 girls’ secondary schools in Kisumu County while systematic random sampling was used to select 38 out of 124 mixed secondary schools in Kisumu County. The sample constituted 39% of the study population. Data was collected using questionnaires, interview schedule and document analysis. Instruments of data collection were validated by the researcher’s supervisors and reliability of the instruments was established through test retest method by carrying out a pilot study in 5 schools which were not part of the study sample. Quantitative data was analyzed using descriptive statistics in form of frequency counts, percentages, means, Pearson’s product moment correlation and regression analysis. Qualitative data from interview schedule were analysed by using thematic analysis approach. Statistical Package for Social Sciences (SPSS) version 22 was applied to assist in analyzing data. All tests of significance were computed at α = 0.05. The findings of the study concluded that there was a positive significant correlation [t (99.92) = 2.23, p = 0.28] between pedagogical techniques applied in teaching mathematics and test scores among girls at form four level. It was established that teaching load was negatively related (n = 180, r = -.265; p < .05) to test scores in mathematics among girls. The findings indicated that there was a positive significant correlation (n = 180, r = .677; p <.05) between text book: student ratio and test scores in mathematics. The study also revealed a significant negative correlation (n = 180, r = -.525; p <.05) between class size and test scores in mathematics among girls. The study recommended that teachers of mathematics should apply varied pedagogical techniques depending on the ability and understanding of students in order to attain higher scores in mathematics, more mathematics teachers should be employed to reduce the teaching load for effective performance, textbook: student ratio of 1:1 should be strictly adhered to and class size should be 45 or below to enable teachers cater for individual interest of students. | en_US |