tiểu luận kinh tế lượng FACTORS AFFECTING GPA OF FOREIGN TRADE UNIVERSITY STUDENTS IN 2019

Statistical significance of variables using T-test...25 CONCLUSION...27 REFERENCES...28 Exhibit 1: Summary statistics of variables in the GPA model....13 Exhibit 2: The min - max - media

FOREIGN TRADE UNIVERSITY FACULTY OF INTERNATIONAL ECONOMICS

=====o0o=====

ECONOMETRIC REPORT

Group: 4 Class: English 7 - VJCC - K57 Members: Lê Thị Thu Phương - 1815520216

Nguyễn Hải Linh - 1815520191 Nguyễn Minh Trang - 1815520232

Instructor: Ph.D Từ Thúy Anh

Hanoi, October 2019

Table of Contents

ABSTRACT 4

INTRODUCTION 5

LITERATURE REVIEW 6

I Questions of interest 6

II Procedure and software used 7

THEORETICAL BACKGROUND 8

I Research history in the world 8

II Research history in Vietnam 8

III Econometrics Methodology 9

1 Specifying the object for modeling 9

2 Defining the target for modeling by the choice of the variables to analyze 9

3 Econometric Model 10

4 Model specification 11

5 Estimation technique 11

DATA DESCRIPTION 13

I DATA OVERVIEW 13

II ESTIMATION OF ECONOMETRIC MODEL 14

1 Checking the correlation among variables 14

2 Regression model 16

III CHECK MULTICOLLINEARITY, NORMALITY OF RESIDUAL AND HETEROSCEDASTICITY 20

1 Multicollinearity 20

2 Normality of residual 22

3 Heteroscedasticity 23

2

4 Statistical significance of model (using F – test) 25

5 Statistical significance of variables (using T-test) 25

CONCLUSION 27

REFERENCES 28

Exhibit 1: Summary statistics of variables in the GPA model. 13

Exhibit 2: The min - max - median value of each variable 14

Exhibit 3: Correlation matrix 15

Exhibit 4: Scatterplot of variables in GPA model 16

Exhibit 5: Ordinary Least Square Model 17

Exhibit 6: Ordinary Least Square Model Explanation 18

Exhibit 7: Multicollinearity test 21

Exhibit 8: Result of White’s test 24

With the increasing diversity of students attending university, there is a growinginterest in the factors predicting academic performance This study analyzes the effect ofgender, working hours (part-time jobs), times absence, self-study hours, marital status(having boyfriend/girlfriend), extra activities participation hours on the total GPA ofForeign Trade University (FTU) students in 2018 We have conducted a survey withonline questionnaire to collect data of 50 FTU students, their overall grade point averageswere collected at semester completion - which were used to estimate multiple linearregression models The results showed some significant evidence for the hypothesis,concluding that the number of times absence is more likely to affect GPA with highsignificance The gender variable is shown to affect GPA less importantly Identifying thefactors that influence academic performance (GPA scores in particular) can improve thetargeting of interventions and support services for students at risk of academic problems

Econometrics is the study of the social science in which the tools of economic theory,mathematics and statistical speculation are applied to analyze economic problems.Econometrics uses the mathematical statistics methods to find out the essence ofstatistics, make conclusions about the collected statistics that can make predictions abouteconomic phenomenon

As much as Economy is a meaningful science that determines the social development ingeneral and national growth in particular, Econometrics is the use of statistical techniques tounderstand those issues and test theories Without evidence, economic theories are abstractand might have no bearing on reality (even if they are completely rigorous) Econometrics is

a set of tools we can use to confront theory with real-world data

Since its inception, econometrics has provided economists with a sharp instrumentfor measuring economic relations As economics students, we recognize the need to studyand learn about Econometrics in logical and problem analysis To better understand how

to put the Econometrics into reality and to apply the Econometrics effectively andcorrectly, our team would like to develop the econometrics report to analyze the topic

"Factors that affects Foreign Trade University students’ GPA in 2019"

Our groups follows the methodology of econometric comprising eight steps toanalyze the data Note that because of the lack of information on the data set, allinferences of abbreviations and others are based on assumptions and self-research As aresult, we hope to have shown clearly our logic and reasoning of analysis

To the extent of purpose and resources, there are still deficiencies in this report, but welook forward to providing readers with a decent view of the overall of the data set given andthe knowledge that we have gained through Ph.D Tu Thuy Anh’s Econometrics course

LITERATURE REVIEW

Grade Point Average is a number representing the average value of the accumulatedfinal grades earned in courses over time More commonly called a GPA, a student’s gradepoint average is calculated by adding up all accumulated final grades and dividing thatfigure by the number of grades awarded This calculation results in a mathematical mean

— or average—of all final grades The most common form of GPA is based on a 0 to 4.0scale (A = 4.0, B = 3.0, C = 2.0, D = 1.0, and F = 0), with a 4.0 representing a “perfect”GPA— or a student having earned straight As in every course

GPA is the main criteria used to assess and rank students applied in most universities

in Vietnam Today, in the context of fierce competition in the job market, the requirementfor job seekers is increasing, so as soon as they enter colleges, students need to setthemselves a certain goals to strive for In most universities including Foreign TradeUniversity in particular, the average score is used as a main criterion to consider the type

of diploma Some companies and employers also set a minimum average score as a way

to decide the number of candidates who apply for a position In addition, many studentsdesire to continue studying at higher levels or win scholarships to study abroad aftergraduation Therefore GPA is highly valued at the time because it is considered a factorthat plays a important role in the achievement of these goals

I Questions of interest

Because this score is calculated by dividing the total score of the subjects, to achieve ahigh average, it is necessary to strive continuously throughout the university In fact, thereare always students with high scores, and a lot of students do not Most students think thatfocusing on studying will get good results, but there are some who find themselves studyinghard but still not achieving the desiring grades From that fact, a question arose: “What is thestudent's average score determined by?” - which is the basic question to which

this report targets to find the answer Understanding these factors can help students have

a clearer direction when pursuing their target GPA

II Procedure and software used

● Procedure

Step 1: Questions of interest

Step 2: Online questionnaire

Step 3: Econometric model

Step 4: Data collection

Step 5: Estimation of econometric model

Step 6: Check multicollinearity and heteroscedasticity

Step 7: Hypothesis postulated

Step 8: Result analysis

● Gretl software is primarily used to analyze the data and run the regression

THEORETICAL BACKGROUND

I Research history in the world

According to Coleman’s (1961) zero-sum time-allocation model, working student’sGPA’s may be lower than GPA’s of students not working

A Astin, 1977; Lunneborg et al., 1974 has indicated that women maintain higherGPAs than men

Pantages and Creedon (1975) and Abbott-Chapman, Hughes, and Wyld (1992) foundthat students with poor study habits are more likely to withdraw from university or tohave academic adjustment problems in the transition from high school to university.Kirsten McKenZie & Robert Schweitzer (2001) identified full-time students withlimited employment responsibilities will have higher GPA’s than full-time students withmore demanding employment responsibilities

According to Bratti and Staffolani (2002), student learning outcomes are largelydetermined by the student's learning attitude because the allocation of time for studydepends on their decisions If Gi be the learning result of a student, depending on the timespent on self-study (Si), class time (who) and his ability (ei) then Gi = G (si, ai) ei

Mussie T Tessema, Kathryn J Ready & Marzie Astani (2014) indicated work haspositive effect on both satisfaction and GPA, when students did work fewer than 10 hours.Thus, part-time job may not always be detrimental to students’ satisfaction However, whenstudents work for more than 11 hours a week, students’ satisfaction and GPA were found todecline for each additional category of work, although the change is very small

II Research history in Vietnam

A number of studies in Vietnam have initiated this problem, such as Huynh Quang Minh(2002), a survey of factors affecting the academic performance of students at Ho Chi MinhCity University of Agriculture and Forestry The study results (with significance level of

about 10%) show that the average score of students in the second stage of semester isdetermined by the level of reading reference materials, class time, self-study time,average score whereas in the first stage is the number of drinks a month and the entranceexamination scores.

Another study by Nguyen Thi Mai Trang, Nguyen Dinh Tho and Mai Le Thuy Van(2008) on the main factors affecting the learning knowledge of economic students Theresearch results show that students' learning motivation has a strong impact on theiracquired knowledge, faculty capacity has a high impact on learning motivation andlearning knowledge of students Both elements: aim for studying and faculty capacityexplain 75% of the variance of acquired knowledge

Research gaps: Results of studies show that there are differences in academic resultsamong different groups of students, resulting in heterogeneity of the influence of factors

on academic performance of students

III Econometrics Methodology

1 Specifying the object for modeling

GPA score = f (x)

As such, this report finds the relationship between GPA scores, which is the objectfor modeling, and each of relating factors including sex, time spending, class attendanceand marital status

2 Defining the target for modeling by the choice of the variables to analyze

As mentioned above, there are four main categories that are expected to affect GPAscores: sex, time spending, and marital status Hence, the choices of would be suchvariables that constitute them After thorough research, factors have been narrowed down

to six significant ones: (sex) gender, (time spending) part-time job hours per day,

extra-activities hours per week, self-studying hours per day, (class attendance) times absence and (marital status) having boyfriend/girlfriend.

A brief description of each variable:

2nd-year students)

0 is the intercept of the regression model

is the slope coefficient of the independent variable u

is the disturbance of the regression model

5 Estimation technique

We test for the relationship between gender, part-time hours per day, times absenceclass per month, self-study hours per day, boyfriend/girlfriend, extra-activities hours perweek (school clubs), data description This procedure will reveal the existence, nature,strength, direction, and magnitude of possible predictive relationships It must be notedthat modeling using OLS requires certain assumptions to hold true, one of which is forthe error term to have a constant variance – that is, no heteroskedasticity – to obtain thebest linear unbiased estimator (BLUE)

11

 Test for Normality

To determine whether a sample data set is well-modeled by a normal distribution and

to compute how likely it is for a random variable underlying the data set to be normallydistributed

Meanwhile, the presence of multicollinearity among independent variables preventsthe isolation of the effects of one variable to the dependent variable which means that one

or more regression coefficients could be estimated imprecisely To ascertain that no suchrelationship exist among the independent variables, we make use of correlation checktool in Gretl

To ensure that the OLS assumption of homoscedasticity is met, the initial model usingthe homoscedasticity – only variance formula was subjected to the White test to test forheteroskedasticity

DATA DESCRIPTION

I DATA OVERVIEW

- Data source: This set of data is collected from survey questionnaires from FTU students

- This database is used to run on Gretl software and the result includes 50 observations with one dependent variable, GPA, and 6 independent variables:

* From Gretl, we get the summary statistics of variables The result is shown in the Exhibit

1 below:

Exhibit 1: Summary statistics of variables in the GPA model.

Where:

S.D is the standard deviation of the variable

Min is the minimum value of the variable

Max is the maximum value of the variable

* We also get the table of min, max and median of independent and dependent variables

in Exhibit 2:

Exhibit 2: The min - max - median value of each variable

II ESTIMATION OF ECONOMETRIC MODEL

1 Checking the correlation among variables

First of all, the correlation of GPA and Gender, Parttime, Absence, Study, Lover, Actsare checked by calculating the correlation coefficient among these variables Thecorrelation coefficient r measures the strength and direction of a linear relationshipbetween two variables

The result is shown in Exhibit 3:

Exhibit 3: Correlation matrix

From the matrix above, we have the correlation coefficient r <0.8

⇒ We have enough evidence to conclude that the correlation between GPA and each of the

independent variable is decent enough to run the regression model Specifically:

- GPA and Gender have a weak downhill relationship

- GPA and Parttime have a weak downhill relationship

- GPA and Absence have a moderate downhill relationship

- GPA and Study have a weak uphill relationship

- GPA and Lover have a weak downhill relationship

- GPA and Acts have a weak uphill relationship

The correlation between each pair of variables can be visualized in Exhibit 4 below:

Exhibit 4: Scatterplot of variables in GPA model

2 Regression model

2.1 Ordinary Least Square Model

Having checked the required condition of correlation among variables, the regression model is ready to run The result is shown in Exhibit 5

Exhibit 5: Ordinary Least Square Model

The regression function:

where:

0 is the intercept of the regression model

is the slope coefficient of the independent variable

e is the disturbance of the regression model

gender is the gender of student Male=1, female=0

parttime is the number of hours per day for doing part-time

jobs absence is times absence from class study

study is the number of hours per day for self study

17