@article{MAKHILLJEAS2019142318688,
    title = {Multi-Factored Linear Models for Predicting Learning Outcomes of
Computer Science Students in Private and Public Universities in Nigeria},
    journal = {Journal of Engineering and Applied Sciences},
    volume = {14},
    number = {23},
    pages = {8840-8850},
    year = {2019},
    issn = {1816-949x},
    doi = {jeasci.2019.8840.8850},
    url = {https://makhillpublications.co/view-article.php?issn=1816-949x&doi=jeasci.2019.8840.8850},
    author = {Temitayo and},
    keywords = {Private-university,public-university,student-learning-outcome,multifactored-predictive-model,computer-science,MSPEINU},
    abstract = {The apparent disparity in the academic learning outcomes of Computer Science (CS) students in
private and public universities in Nigeria is currently a big concern. In this study, two multi-factored evaluation
models are developed to investigate and predict learning outcomes of CS students in a privately-owned Caleb
University (CU), Imota and a government-owned University of Lagos (UNILAG), Akoka-Yaba, both situated
in Lagos State Nigeria. The two universities were chosen for this study using convenience sampling. The data
used in this study was collected from 267 CS student volunteers (200-500 level) in UNILAG and 139 CS student
volunteers (200-400 level) in CU who were enrolled between 2012/2013-2017/2018 academic sessions via. a
developed closed-ended questionnaire tagged "Multifactor Student Performance Evaluation Instrument for
Nigerian Universities (MSPEINU)" with a r eliability coefficient of 0.86. 18 factors were investigated with their
associated 65 independent variables that largely affect performance of students. Regression and correlation
are the descriptive statistics used to analyze and examine the cause-and-effect impact of the factors as well as
the degree of that impact on the student learning outcomes. The findings from this study show that the actual
factors affecting performance of computer science students in UNILAG are student’s attitud e to learning,
student attendance, student background knowledge, lecturer attitude, lecturer teaching style, class population,
family income and parent education while student attendance, student opinion, proper guidance, parent
education, family income, lecture time, student background knowledge, electricity, student health, lecturer
teaching style and lecturer attitude are the factors affecting the computer science students in CU. The predictive
models developed in this study present potential cost and performance improvement benefits as it guides the
university administrators, the lecturers, the students and other relevant university stakeholders towards sound
decision making. They are also robust and exhaustive enough to be generalized to other similar universities in
Nigeria.}
    }