Abstract:
The use of social media is on the rise globally. There is also a decline in performance of students in academics in Teachers’ Training Colleges. There is limited evidence on the effect of social media use on academic performance in Vihiga County in Kenya from past studies. This study purposed to examine the effect of social media use on academic performance of students in Teachers’ Training Colleges in the county. The four objectives used to fill the research gaps of this study were: to examine the relationship between types of social media used for communication and level of students’ academic performance, to find out the relationship between frequency of social media use and level of students’ academic performance, to determine the relationship between interpersonal relationships and level of academic performance of students, and to evaluate the relationship between online attacks and level of academic performance of students in the colleges. This was achieved using students’ performance in end of term examinations as well as five social media platforms namely: Twitter, Facebook, Instagram, WhatsApp and YouTube. The study was guided by technological determinism theory. Descriptive survey research design was used in the study with the aid of mixed method approach. The target population of 1,584 comprised of 6 principals, 1,478 students and 100 tutors in 6 Teachers’ Training Colleges. Purposive sampling technique was used to sample the 6 colleges and 6 principals. Simple random sampling technique was used to sample 306 students and 80 tutors from the table made by Krejcie and Morgan (1970) giving a total sample size of 392. Data was gathered using interview guide and questionnaires. Validity and reliability were tested by piloting the tools in 2 Teachers’ Training Colleges in Kakamega County. Experts at the School of Education and Human Resource Development in Kisii University scrutinized the tools to determine its validity. Cronbach’s alpha was used to determine the reliability. Acceptable values of 0.78 and 0.80 were obtained from questionnaires issued to tutors and students respectively. Statistical Package for Social Sciences Software Version 20 was used to analyze quantitative data descriptively and inferentially and presented in tables and graphs. Pearson Product Moment Correlation, Regression and ANOVA were used to test if a correlation existed between the variables. Qualitative data was analysed thematically in line with the study objectives and reported in textual form based on the emerging themes. The study observed that a positive correlation which was statistically significant existed between the variables in the first null hypothesis since the p-value got (0.000) was less than 0.05. Hence, the rejection of this null hypothesis concluded that types of social media affected students' academic performance. It was noted from null hypothesis two that a positive significant association existed since p = 0.000 < 0.05. The rejection of this null hypothesis led to a conclusion that frequency of social media use influenced academic performance. The p = 0.000 < 0.05 obtained in the third null hypothesis showed a statistically significant positive relationship meaning that interpersonal relationships influenced academic performance. The fourth null hypothesis was finally rejected since a statistically positive correlation existed (p = 0.000 < 0.05). It was concluded that magnitude of online attacks correlated with academic performance significantly. The study recommended that students should use the various types of social media frequently as well as their positive interpersonal relationships with other people on social media and knowledge of online attacks to enhance their academic performance. The findings would sensitize principals, students and tutors on effect of social media on students’ academic performance in Teachers’ Training Colleges. It may also be significant to administrators by assisting them to adopt and form suitable policies on the use of social media among students in such colleges.