Academic performance of engineering students: A predictive validity study of first‐year GPA and final‐year CGPA

Students' academic achievement is regarded as the scholastic standing of students at the end of a given study period that is expressed in terms of grades. This study focused on students' first‐year GPA as a predictor of final‐year CGPA and the relationship between demographic attributes of the students to academic achievement. An ex‐post factor research approach was adopted, and Pearson's correlation and Regression Analysis were fitted with the data using Minitab software. The results of the study highlighted that first‐year GPA had a strong positive relationship with final‐year CGPA which is an indication that first‐year GPA greatly influences final‐year CGPA meanwhile, demographic characteristics have no significant correlation with students' final‐year CGPA. Finally, the correlations drawn from this study indicated that first‐year GPA had a strong positive relationship with final‐year CGPA. Therefore, the students should be properly guided, University administrators, Faculty members, policy‐makers and other stakeholders should effectively develop a monitoring blueprint for students' academic progress as poor performance has severe consequences on students, teachers should inculcate effective learning culture among students, while government and its supervisory agencies should provide adequate teaching and learning resources, create enabling learning environment for effective service delivery that focuses on improving students' performance in Engineering.

Student's success in progression and retention are the major indicators of academic performance and attainment gaps, students with low academic performance can easily be identified and effective measures can be put in place to provide interventions to bridge the attainment gap.The attainment gap in this context is the qualitative differences between degree classification that is, First class or second-class upper division awarded to students Al-sudani & Palaniappan. 2onsidering Nigeria as a case study, a Country located in the Western part of the African continent, the most populous country in Africa with an estimated population of over 200 million people, as a multi-ethnic country, we can base our attainment gap on its six geopolitical zones that housed the three predominant tribes (Hausa, Yoruba and Igbo) who were awarded either First Class or Second Class Upper as a Degree Classification to get the proportions of each geopolitical zone based on the Class of Degree Classification.
Engineering principles are apt to improve the quality of life and Tertiary Institutions comprising Universities, Polytechnics and others are tasked with the mandate of producing competent engineers, technologists, technicians and engineering craftsmen that can compete and solve real-life problems Al-sheeb et al. 3 In Nigeria, the Council empowered by law to ensure the quality and professionalism of Engineering Education and professionals are the Council for the Regulation of Engineering in Nigeria (COREN) and other bodies for Technologists, Technicians, and Craftsmen to regulate practices and ensure the best standards.The benchmark for minimum academic standards (BMAS) for Engineering qualification is set by the Council and other Tertiary Bodies such as the National Universities Commission (NUC), and National Board for Technical Education (NBTE) to ensure engineering undergraduates produced are competent and able to solve life problems Akeel et al. 4 Students' admissions into Nigeria's Tertiary Institutions are based on among other things performance at the Unified Tertiary Matriculation Examinations (UTME) conducted by Joint Admission and Matriculation Board (JAMB) located in the Nation's Capital, Abuja.It is expected that students admitted into tertiary Institutions should be successful in the field of their choice, but this is not always the case as some students may end up with carryover of some courses, and others may even withdraw from the programme based on poor performance. 5,6he mandatory schooling years for all citizens of Nigeria is nine years which is part of the Nigerian Policy on Education; students will spend six years of elementary school, three years of middle school, then proceed to three years of senior secondary school education and if interested, the student can proceed to any Tertiary Institution of learning having met the basic requirements for admission in such Institutions Raji. 7hese objectives are similar to those set by the National Academy of Engineering with its project vision of the Engineer of 2020.The project was tailored to provide a framework that will be effective in positioning education in the United States of America, to tackle future challenges across all engineering fields in the country. 8lso, Science, Technology, Engineering and Mathematics (STEM) inclined Institutions are under extreme pressure to produce competent engineers due to the shift in STEM-related jobs.Therefore, studying the impact of the Engineering Curriculum on students' Cognitive, Affective and Psychomotor domains is necessary for achieving global conformance to best practices as enshrined in some National Visions such as Qatar's Vision 2030 Taylan et al. 9

Design of the study
In this study, an ex-post factor design approach was used because of its simplicity, and the inability of researchers to change the data because the study is based on factors that have occurred Gipson. 10This offers slight advantages over the machine learning approach that is widely used to analyze complex classification problems and when there is a change in data could lead to huge variance Rajendran et al. 11 Pearson's Product Moment Correlation was used to analyze the strength of the linear relationship between two variables of interest, the magnitude of the relationship is given by Pearson's Correlation Coefficient Zhou et al. 12

Aim and objectives of the study
The study was aimed at investigating the relationship between first-year GPA and Final-year CGPA as well as the relationship between gender, age, geopolitical zone, and course of study to final-year CGPA.The Objectives include: 1. Investigate the degree to which first-year GPA predict Final-year CGPA among Engineering students.
2. Investigate the relationship between demographic characteristics, and course of study on student's academic performance in Engineering, 3. Identify which predictor variables best relate to final-year CGPA?

Research questions
The following research questions guided the study; 1. What is the relationship between first-year GPA and final-year CGPA among Engineering students?2. How does the relationship between demographic characteristics (age, gender, geopolitical zone), course of study and type of Engineering study affect students' final-year CGPA? 3. Which of the predictor variables best predicts engineering students' academic performance in bridging the attainment gap?
It is, therefore imperative to analyze the relationship between first-year GPA and final-year CGPA and the effect of Demographic Characteristics on students' academic performance from four Departments of the Faculty of Engineering, University of Abuja Nigeria.

BACKGROUND
The authors perceived academic performance as the scholastic standing of a student at the end of a given period of study which is categorized by grades obtained during the period and are cumulatively graded to a desired scale.While other researchers Mallory, 13 and Jansen, 14 think that academic performance is the ability of students to learn knowledge, skills and attitudes to develop capacities and potential to be successful in their society Apata. 15Students' academic achievement is regarded highly by all stakeholders and used in selecting schools for student scholarships as well as self-funded studies.Academic achievement is an important variable used for selecting students for schools and programmes as well as attracting lucrative job opportunities after graduation Şahin & Çoban. 16

Literature review
Abdelfattah et al., 17 conducted a predictive validity study on entrance scores and short-term performance for long-term success in Engineering Education.The study examined the effect of high school coursework, general ability tests, and achievement tests on CGPA.The study observed that correlations were significant between entrance scores and the preparatory year GPA.While first-year to third-year GPAs recorded significant correlations in predicting cumulative GPAs of Engineering students during graduation, the study also observed that girls outperform boys in their entry scores and GPAs and hence obtained higher CGPAs than their counterparts.
An interaction between mathematical disposition and independent learning in distance learning success using an ex-post factor study was conducted by Kusmaryono et al., 18 the study showcased the interaction between mathematical disposition and learning independence on learning outcomes of elementary school students in distance learning, the study concluded that the level of mathematical disposition was high and independent learning.
An investigation was conducted to predict first-year academic performance using Entry requirements for Faculty of Science students at Kaduna State University Nigeria.The study analyzes the relationship between Ordinary Level (OL) and UTME to predict the first-year students' academic performance using correlation and regression analyses which were evaluated using Statistical Package for Social Sciences (SPSS) software.The results showed that OL and UTME have a weak correlation with students' first-year academic performance Abdulkadir and Ogwueleka. 19awal et al., 20 conducted a similar study to investigate the predictive validity of the first-year GPA and final-year degree classification among management and social science students in a Nigerian University using ex-post facto research design and Pearson correlation to examine the strength of the relationship.The study observed that there was a significant but negative correlation between first-year GPA and final-year CGPA among management science graduates.Al-sheeb et al., 3 investigated student academic achievement in engineering education using cognitive and non-cognitive factors; the study adopted exploratory factor analysis and regression model selection to check the efficiency of ten critical-to-success factors on attitude and skill-related behaviors (KAB model) of 320 first-year students to investigate the causes of students' premature withdrawals and proffer solutions to ensure the success of engineering students.The findings indicated that the KAB model is effective in predicting student performance.It also highlighted the importance of conducting such measurements at an early stage for engineering students to help them in improving their performance.Kennedy and Ebuwa 6 investigated University entry score (UTME) by Joint Admission and Matriculation Board (JAMB) and Post Unified Tertiary Matriculation Examination (PUTME) as predictors of undergraduate final-year CGPA; the study had a population of 436 undergraduate students from four departments and the data were analyzed using Pearson Product Mean Correlation Coefficient and Linear Regression.At the end of the study, it was realized that the UTME and PUTME scores combined do not significantly predict undergraduate final-year CGPA in Nigeria University.
Unfried et al., 21 studied the development and validation of a measure of student attitudes towards Science, Technology, Engineering and Math (S-STEM) using iterative design and methodological approaches to find reliability, validity and fairness in measuring student attitudes towards (S-STEM).The results showed that students have separate but consolidated attitudinal responses to the academic areas of STEM.Also, the study observed that attributes towards 21st-century skills be considered as a separate factor, while upper elementary S-STEM and Middle/High S-STEM surveys showed evidence of configural, metric and scalar invariance across grade levels, race/ethnicities and genders have a different intuition towards STEM.
Meanwhile, Jaafar et al., 22 conducted a correlation study of student achievement at the Pre-University level and their Corresponding Achievement in the year-one undergraduate course of Circuit Theory I and II for students of Electrical Engineering at The University Kebangsaan Malaysia (UKM).The study analyzed the students' performance in a pre-test in the course of Circuit Theory I and II to find the relationship in the final examination.The results showed that there was a significant correlation between final examination marks for Circuit Theory I and pre-test scores with R 2 = 0.47.
Oguntunde et al., 23 studied the inter-relationship between students' first-year results and their final graduating grades.In the study, the correlation between first-year results and graduating grades in a Nigerian University was performed, regression analysis showed good fitness with the model and the regression equation can predict the final year CGPA of the students.The results indicated a strong linear relationship between GPAs and students' progression in their academic journey.
Another study also examined the predictive validity of pre-admission assessment on medical students' performance.The study had a population of 737 drawn from students in preclinical and clinical years.The study observed that only the National Achievement Test (NAT) and TOEFL scores significantly predicted the performance of students in preclinical years Debaliz et al. 24 In a similar endeavor, Almarabheh et al., 25 conducted a predictive validity study to predict the academic performance of medical students, the study observed that there was a statistically significant positive correlation between admission criteria and student performance among year one and year four medical students.The overall results indicated that the predictor variables correlate significantly with the outcome variables.
The observed relationship between Grade Point Average and Cumulative Grade Point Average on the academic achievement of students was studied by Abdelfattah et al., 17 Abdulkadir and Ogwueleka, 19 Lawal et al., 20 Debaliz et al., 24 Almarabheh et al., 25 Oguntunde et al., 23 the researchers concluded that GPAs have a strong linear relationship with final year CGPA of students.We anticipate to have similar findings in this study.Meanwhile, the observed relationship between demographic characteristics and the academic achievement of students was studied by Şahin & Çoban, 16 Shaaban & Reda, 26 Spinath et al., 27 concluded that demographic characteristics have an influence on students' academic achievement.
Researchers Stienstra & Karlson, 28 Qu et al., 29 Nunes et al., 30 Holmlund et al., 31 Nagahi, et al., 32 and Spinath et al., 27 were of the opinion that gender and age differences are not the only factors in students' academic achievements but other factors such as genetic influences, type of family, cultural orientation, type of school and environmental influences greatly affect students' performances.Due to these disparities in literature, we do not expect to see a strong relationship between age and gender in Engineering students' academic achievement or that it will be a very weak relationship.

Methods
Data collection: the participants' population were (N = 160) students that were admitted to the Faculty of Engineering to study Civil, Chemical, Electrical and Mechanical Engineering at the University of Abuja between 2017 and 2021.The data were obtained from the office of the Academic Secretary and the Management Information System Unit of the University.The raw data were de-identified to contain only GPA, CGPA, and Demographic characteristics (Age and Gender).

Research design
This study adopted an ex-post facto design, a research design exploring the impact an independent variable has on the dependent variable.The research study is based on the actions that have occurred being used to predict certain behavior.This feature of the research design means that the researchers cannot manipulate the characteristics of the dataset Abdulkadir & Ogwueleka. 33

Research variables
The independent variables in the study were GPA, Demographic characteristics, Age and Gender.The dependent variable used was Cumulative Grade Point Average (CGPA).In Nigeria, Engineering students studied for a period of five ( 5) years; the first year is regarded as the General Sciences year, second to fifth years are the core Engineering years.CGPA is classified on a scale of 0.00-5.00with 4.50-5.00regarded as First-Class Honors; 3.50-4.49as Second-Class Upper Honors; 2.40-3.49as Second-Class Lower Honors, 1.50-2.39as Third-Class Honors while 1.49 below will be on probation for a year and if the student fails to obtain a 1.50 CGPA at the end of the probationary period, will lead to the student's withdrawal from the Course of study.

Statistical analyses
Descriptive analysis was performed on the dataset, the essence was to extract meaningful conclusions.This numerical approach described the information on central tendency, distribution's shape and width of the distribution of data Kusmaryono et al. 18 The measure of variation is an indication of the degree to which the scores are either clustered or spread out in the distribution.Standard Deviation shoed the measure of variation from the mean or central points of the distribution.The standard deviation is given by the Equation.
where x = individual score, X = sample mean, N = number of scores in the distribution, S = Sample Standard Deviation.Kurtosis: is a measure that highlights the degree of dispersion among the scores or whether the distribution is short or fat Jackson. 34earson's Product Moment Correlational Analysis investigates the relationship between two variables as they relate to each other and enables the prediction of magnitudes from one variable to the other Selvamuthu & Das 35 (Table 1).

TA B L E 1
Provides the guidelines for accessing the magnitude of the correlation of a relationship.

Correlation coefficient
Strength of the relationship A correlation Coefficient of either −1.00 or +1.00 indicates a perfect correlation highlighting the strongest possible relationship, while a correlation coefficient of 0.00 indicates no relationship between the variables Jackson. 34In addition, R 2 correlation tests were conducted between the parameters of interest Abdulkadir & Ogwueleka. 19catterplots are figures depicting the graphical relationship between two variables; these plots showed points that cluster towards an imaginary line through their center in a linear manner.The stronger the correlation, the more compact the points cluster around the imaginary line through the center.Scatterplots can be positive, negative or no relationship and curve linear relationships; a positive relationship is an indication of a direct relationship between two variables that isfrom, an increase in one variable is also related to an increase in the other variable.A negative relationship is an indication that when one variable increases the other variable decreases.No relationship indicated that no meaningful relationship exists between the two variables, also the points are scattered in a random fashion Jackson, 34 Gipson. 10A curve linear relationship is experienced when two variables increase together to a point and then the other variable decreases while the initial variable increases, the data points are clustered in a curve linear pattern Jackson. 34egression Analysis is a procedure used in the prediction of an individual's score on one variable based on knowing one or more variables, in other words, it is the building of mathematical expression that describes the relationship between one or more predictors and a single output variable using multiple linear regression approach Gipson. 10he data of this study were grouped according to departments and clustered to produce an overall relationship between GPA and CGPA and the influence of Demographic characteristics on students' academic performance in Engineering.
All analyses were performed using IBM Statistical Package for the Social Sciences Version 23 and Minitab Statistical Software Version 21.2. a p-value of (<0.05) was considered in the test for significant differences between the variables.

RESULTS AND DISCUSSIONS
The study included students from four departments of the Faculty of Engineering, the data used were first-year GPA, final-year CGPA and Demographic characteristics (Age, Gender and Geo-political zone) to enable the investigation of students' first-year GPA as a predictor for final-year degree classification and check the effect of Demographic characteristics on students' academic performance in Engineering.The results of the study are presented in Tables and graphs according to the research questions that were adopted to guide the study.

Population of the study
The study's sample size comprises (n = 132) students who studied Civil, Chemical, Electrical, and Mechanical Engineering between 2017 and 2021.About 28 students were not listed due to their inability to complete the program or were withdrawn because of poor performance.

Research hypotheses
1. H1 = First-year GPA, the type of Engineering study and Demographic Characteristics influence the Final-year CGPA of students' academic achievement in Engineering.
Ho = There is no significant relationship between First-year GPA, the type of Engineering study and demographic Characteristics on Final-year CGPA of students' academic achievement in Engineering.

Analyses of data
The dataset investigated in this study contained the first-year GPA, CGPA, Gender, Geo-political zones and Age of students at the end of the program, (n = 132) students were able to graduate from the four Departments in the Faculty of Engineering; these departments are Civil Engineering, Chemical Engineering, Electrical Engineering and Mechanical Engineering.The dataset was analyzed using Minitab Statistical Software 21.Preliminary analyses were conducted to evaluate the reliability and validity of the dataset.The data were examined to ensure assumptions for conducting correlation and regression analysis were feasible, normality plots indicated that the data were normally distributed.Correlation and regression analyses were conducted using students' Cumulative Grade Point Average (CGPA) as response and other variables such as Grade Point Average (GPA), Age, Gender, Geopolitical zone and type of Engineering were used as predictors.The scatter plots and scatter plot matrices were used to showcase the summary of the regression relationship between the response and the predictors, the analysis of variance showed the means of comparing the fitness of the mean functions from the same dataset Weisberg. 36he statistical attributes of the dataset were determined and presented in Table 2A showed the descriptive statistics for Civil Engineering students, Table 2B showed the descriptive statistics based on demographic characteristics for Civil Engineering, Table 2C highlighted the descriptive statistics for chemical engineering, Table 3A showed the demographic characteristics of chemical engineering students, Table 3B showed the descriptive statistics of electrical engineering students, Table 4A showed the descriptive statistics based on demographic characteristics for electrical engineering students, Table 4B showed the descriptive statistics for mechanical engineering students and Table 4C showed the descriptive statistics based on demographic characteristics for mechanical engineering.
Research question one: Is there any significant relationship between First-year GPA and Final-year Degree classification among engineering graduates?From Figure 1A the scatter plots are the indication that the values of correlation analysis results have a very strong relationship between first-year GPA and Final-year degree classification as a positive relationship with r and p-values (0.756, 0.000, N = 38).the p-value of 0.000 which is less than 0.01 a strong statistically significant that the hypothesis is accepted and that GPA influences the student's academic performance in their final-year CGPA.
Figure 1B highlighted the matrix plot of the correlation analysis conducted on Chemical Engineering students to investigate the relationship between First-year GPA and Final-year CGPA the results (r = 0.994, p = 0.000) were recorded which showed a very strong positive correlation between GPA and CGPA of Chemical Engineering students and as such we accept the hypothesis and reject the null hypothesis.
An observation from Figure 1C showed the correlational analysis of Electrical Engineering students' GPA and CGPA, the matrix plot showed (r = 0.838, p = 0.000) an indication of a strong positive relationship between GPA and final-year CGPA and therefore we reject the null hypothesis and accept the alternate hypothesis that says there is a significant relationship between GPA and final-year CGPA of Electrical Engineering students.
From Figure 1D below, the results of the correlation analysis for Mechanical Engineering students' performance in the first-year GPA and how it relates to the final-year CGPA are shown, with r and p values given as 0.886 and 0.000 respectively.This is a perfect positive correlation and statistically significant for us to reject the null hypothesis and accept the alternate hypothesis that conforms with the statement "is there a significant relationship between First-year GPA and Final-year CGPA for students studying Mechanical Engineering" as such we accept the alternate hypothesis.
Research Question Two: H2 = Is there any significant relationship between students' demographic characteristics and type of Engineering on students' academic performance?
Figure 2A above addresses the second research question "Is there any significant relationship between students' age, gender, geopolitical zone and type of engineering on student academic performance"?From the graph it could be seen that the recorded correlation values of the relationship between Age, Gender and Geopolitical on GPA and CGPA were (r = −0.160,p = 0.388), (r = −0.069,p = 0.679), (r = −0.110,p = 0.509), (r = 0.072, p = 0.669) respectively.A careful observation of the relationship between age and gender resulted in a very weak negative correlation and p-values greater than 0.05 and an indication that Age, Gender and Geopolitical zone does not affect the student's academic performance in Civil Engineering.Figure 2B above showed the analyses of the relation between Age, Gender, and Geopolitical zone on first-year GPA and Final-year CGPA of students studying Chemical Engineering.From the graph, the relationship between Age, Gender and Geopolitical zone on GPA was recorded as (r = −0.341,p = 0.045), (r = 0.074, p = 0.672), (r = 0.110, p = 0.531) respectively, the interpretation showed that age has a weak negative correlation but statistically significant influence on GPA, then the correlation between gender had a weak positive relationship but is statistically insufficient to affect GPA, also, the correlation between Geopolitical zone and GPA was a weak positive relationship but statistically insufficient to affect students' performance in their first year of study.Meanwhile, when these characteristics were fitted against students' final year CGPA, similar results ensued for Age, Gender and Geopolitical zone.
A careful observation of Figure 2C addressed the second research question that seeks to investigate the relationship between Age, Gender, and Geopolitical Zone on students' GPA and final-year CGPA.The graph showed that Age and Gender (r = −0.159,p = 0.314), Age and Geopolitical Zone (r = −0.038,p = 0.813), Age and GPA (r = −0.313,p = 0.044), Age and CGPA (r = −0.259,p = 0.098) from all these results, it indicated that there is no significant relationship between Figure 4A shows the graph of the Quadratic Model of the Regression Analysis; from the graph, it was observed that GPA influences final-year CGPA with an R-Squared Value of (75.13%) as the percentage variation for the prediction of the regression equation (CGPA = 6.478 − 3.749 GPA + 1.151 GPA 2 -0.09535GPA 3 ) over the total of 132 students from four Departments of the Faculty of Engineering, University of Abuja.While Figure 4B showcased the correlational plots of GPA and CGPA with r and p values of 0.851 and 0.000 respectively.This was a clear indication of a perfect positive correlation and a statistically significant relationship between first-year GPA and Final-year CGPA.
The interpretation of the Regression model equation (CGPA = 6.478-3.749GPA + 1.151 GPA 2 -0.09535GPA 3 ) for all Engineering students of the Faculty of Engineering University of Abuja means that for every unit increment in first-year  GPA, there will be a decrease in of 3.749 in the final-year CGPA, while for every square increment in GPA, there will be an increase of 1.151 in the final-year CGPA and for every cubic increment in GPA, there will be a decrease of 0.09535 in the final-year CGPA provided all other factors are kept constant Oguntunde et al. 23 A careful observation of Figure 4C showed that the dataset of all Engineering Students in the Faculty of Engineering is normally distributed as the diagonal line is straight and not skewed.
The Analysis of Variance (ANOVA) shown in Table 5B above highlighted the test for the significance of the regression model of all Engineering students, the p-value of 0.000 means that the regression model fits well with the data and as such we reject the null hypothesis.
Table 5C above shows the demographic distribution of all engineering students in the Faculty of Engineering, from the data shown, Male students constitute 87.1% of the population while Female students have 12.9% of the total number of Engineering students.

DISCUSSION
This study was conducted to investigate the predictive validity of first-year GPA on final-year CGPA, the relationship between Age, Gender, Geopolitical zone and type of Engineering studied on students' academic performance as measured by the final-year CGPA in Engineering.The data comprised 87.1% males and 12.9% females totaling 132 students drawn from four departments of the Faculty of Engineering, University of Abuja.The study discovered that there was a statistically significant positive correlation between first-year GPA (predictor) and final-year CGPA (response).However, there was no statistically significant relationship between Age, Gender, Geopolitical zone and type of Engineering study on the academic performance of Engineering students.The highest R 2 values were 98.82%, 78.57%, 70.23%, and 57.19% for Chemical, Mechanical, Electrical and Civil Engineering respectively, these values served as an indication of the variation in the accuracy of the regression equations in predicting the response variable (CGPA) of Engineering students.the results were similar to the findings 20,23,33,37,38 that highlighted that first-year GPA has a strong relation to student academic success to bridge the attainment gap.Some researchers believed that poor academic performance hugely affects students' success after graduation as it leads to struggles for employment Shaaban & Reda. 26lthough there was no significant relationship between age and gender on student academic performance in this study which also conforms with Rajendran et al. 11 Some researchers found that the female gender in Engineering tend to perform lower than their male counterparts due to among other things emotional inclination 39 discovered that females perform less than males because they feel that their contributions were undervalued by the males, lack of female role models and mentors in Engineering.However, some researchers confirmed that female students outperform their male counterparts in non-engineering courses. 40,41While at the same time, Sideridis & Alamri 42 observed higher female academic achievements than males when other demographic characteristics were adopted in their study such as the level of education of parents, student-to-teacher ratios and students' attendance in schools.Similar findings were also observed between male and female academic achievements of medical students during undergraduate studies but a slight shift in pediatrics and surgery as the females performed better than their male counterparts Albalawi, 43 Eya & Ezeh. 44Meanwhile, Ossai et al., 45 believed age and gender significantly influence the academic performance of students especially females, when Academic Integrity is considered, females outperformed males while as the age of the student increases, the poorer the academic performance.To ensure diversity in Engineering studies, the study suggested the need to increase the participation of females in Engineering programmes as females constitute 12.9% of the population of the study.Gender balance in Engineering studies increases gender diversity, increased talent pool expansion, and fosters collaboration and teamwork among Engineering students with different backgrounds and perspectives, eliminating gender-based discrimination Kovaleva et al., 46 Annette. 47

Limitations of this study
This study had some limitations which could be used to make suggestions for future research.First, the data of the study was limited in terms of quantity and geography which limits the generalization of the findings, the data covers only Engineering students admitted between 2017 and 2022, other disciplines could express different patterns of conclusion with regard to the effect of First-year GPA and Demographic characteristics on Final-year CGPA.Other variables such as cognition, psychomotor, and attitudes of students also have an effect on the academic performance of students, a study towards these variables could also guide in drawing a holistic conclusion on students' academic performance in Engineering.The second limitation of this study was that it was purely an ex-post factor research design which will lead to a restricted conclusion.

Directions for future study
The authors suggest the following as future directions; the study be replicated on other Engineering Faculties within other regions of the country to have a generalized output, this will provide more evidence from the inferences observed in this study.Adopt a modern methodology such as machine learning in predicting the academic achievement of students Kanetaki et al., 48 or other research approaches that entail mixed-method research design to cater for student perspectives, content and structure of the programmes to generate a more robust conclusion on the factors influencing students' academic performance in Engineering study.

CONCLUSIONS
Despite the limitations discussed above, this study provided useful observations that contribute to our understanding of student academic performance in Engineering studies.The correlations drawn from this study indicated that first-year GPA had a strong positive relationship with final-year degree classification.Therefore, the students should be properly guided, University administrators, Faculty members, policy-makers and other stakeholders should effectively develop a monitoring blueprint for students' academic progress as poor performance has severe consequences on students after graduation and teachers should inculcate effective learning culture among students, provide adequate and appropriate teaching and learning resources, while government and its supervisory agencies provide an enabling learning environment and resourceful teachers etc.The study also observed a lesser number of females in Engineering studies from the four departments only 12.9% of students were of the female gender.Gender diversity is a global trend, there is a need to encourage female students to enroll in Engineering studies to ensure inclusiveness and attainment of Vision 2030 and sustainable development goals that seek to increase the participation of females to 25% in Science, Technology, Engineering and Mathematics (STEM), having diverse Engineering programs will improve the talent pool of engineering students, enhance problem-solving, reduced gender bias, mentorship and role model opportunities and socioeconomic benefits to the society as well as the Engineering graduands which will further boost creativity, skills, improved collaboration and promoting a more equitable distribution of opportunities and resources for females and should therefore be mentored and encouraged to rise above stereotypes and biases of our culturally oriented society.
Graph of the relationship between GPA, CGPA and demographic characteristics of civil engineering students.(B) Matrix plot of the relationship between demographic characteristics, GPA and CGPA.(C) Correlation plot between GPA, CGPA and demographic characteristics for electrical engineering students.(D) Correlation Plots for GPA, CGPA and demographic characteristics for mechanical engineering students.The r and p-values indicate how weak the correlation is between the variables.

F
I G U R E 3 (A) Fitted line plot of the regression analysis of civil engineering students.(B) Fitted line plot of regression analysis of chemical engineering students.(C) Fitted line plot of the regression analysis of electrical engineering students.(D) Fitted line plot of the regression analysis of mechanical engineering students.The plots showed the fitness of the regression equation and how it can be used to predict student's academic achievements.R-sq values showed the accuracy of model.
U R E 4 (A) Generalized regression plot for all engineering students.(B) Matrix plot of correlation analysis between GPA and CGPA for all engineering students.(C) Normality plot for final-year CGPA of all engineering students.The graphs are an indication of how well the equation can predict academic achievement, Generalized correlation plots also highlighted the positive relationship between GPA and CGPA while the Normality plot showed the data is drawn from a normally distributed population.

Descriptive statistics based on demographic characteristics of civil engineering students.
Descriptive statistics and demographic characteristics for each of the departments.Descriptive statistics and demographic characteristics for each department.
TA B L E 2Note: The tables presented a summary of the student's data such as age, gender, GPA, CGPA and geo-political zone.TA B L E 3 (A)

Descriptive statistics based on demographic characteristics of chemical engineering
Note:The tables presented a summary of the student's data such as age, gender, GPA, CGPA and geo-political zone.

Descriptive statistics based on demographic characteristics for electrical engineering students
Descriptive statistics and demographic characteristics for each department.
TA B L E 4

Descriptive statistics based on demographic characteristics of mechanical engineering students
Note:The tables presented a summary of the student's data such as age, gender, GPA, CGPA and geo-political zone.

Analysis of variance for the fitted model of all engineering students
Showed the summary of regression equations, ANOVA summary, and demographic descriptive data for all engineering students.
TA B L E 5