tiểu luận kinh tế lượng FACTORS AFFECTING THE INTELLIGENCE QUOTIENT (IQ) INDEX OF FTUS STUDENTS

FOREIGN TRADE UNIVERSITY FACULTY OF ECONOMICS & INTERNATIONAL BUSINESS ---○○○---ECONOMETRICS REPORT FACTORS AFFECTING THE INTELLIGENCE QUOTIENT IQ INDEX OF FTU'S STUDENTS Group:14 Class:

FOREIGN TRADE UNIVERSITY FACULTY OF ECONOMICS & INTERNATIONAL BUSINESS

-○○○ -ECONOMETRICS REPORT FACTORS AFFECTING THE INTELLIGENCE QUOTIENT (IQ)

INDEX OF FTU'S STUDENTS

Group:14 Class: KTEE218.1 Lecturers: MSc.Quynh Thuy Nguyen

MSc Phuong Mai Vu Thi

GROUP MEMBERS

1 Nguyen Thi Thuy Tram -1814450067

2 Nguyen Thi Hue - 1816450042

3 Tran Thi Thuy Dung - 1814450022

4 Hoang Thu Thuy -1814450066

5 Karan Singh - 1890190113

Ha Noi, September 2019.

Table of content

Table of content 3

ABSTRACT 5

INTRODUCTION 6

SECTION 1: OVERVIEW OF THE TOPIC 8

1.1 Intelligence Quotient Index and related terms 8

1.2 Theoretical Background 9

SECTION 2: MODEL SPECIFICATION 11

2.1 Methodology in the study 11

2.1.1 Method to derive the model 11

2.1.2 Method to collect and analyze the data 11

2.1.2.1 Method to collect the data 11

2.1.2.2 Method to analyze the data 11

2.2 Theoretical model specification 11

2.2.1 Specification of the model 11

2.2.1.1 Population Regression Function: 11

2.2.1.2 Sample Regression Function: 13

2.2.2 Explanation of the variables: 13

2.2.3 Description of the data 14

2.2.3.1 Source of data 14

2.2.3.2 Statistical descriptions of the variables 14

2.2.3.3 Correlation matrix between variables 15

SECTION3:ESTIMATED MODEL AND STATISTICAL INFERENCE 16

3.1 Estimated model 16

3.1.1 Estimation result 16

3.1.2 The sample regression model 16

3.1.3 The coefficient of determination 16

3.1.4 Meanings of estimated coefficients 16

3.1.5 Other results analysis 17

3,2 Hypothesis Testing 17

3.2.1 Verifying the suitability of individual regression coefficient 17

3.2.2 Verifying the suitability of regression model 19

3.2.3 Explanation for coefficient that is not statistically significant 19

3.3 Some recommendations 20

3.3.1 Some recommendations for improving Intelligence Quotient Index 20

3.3.1.1 Improve GPA 20

3.3.1.2 Improve height 20

3.3.1.3 Others workable solutions 21

a) Learn an instrument 21

b) Start meditating 21

3.3.2 Some recommendations for future research 21

CONCLUSION 23

REFERENCES 24

APPENDIX 25

INDIVIDUAL ASSESSMENT 32

For this, we decided to identify what are the key factors that influence theIntelligence Quotient Index of Students from the Foreign Trade University inHanoi.

This Report shows the findings from a questionnaire survey, whereempirical studies (n = 90) on factors which may influence the IntelligenceQuotient Index among students are reviewed

The review details how the Intelligence Quotient Index among studentsrelates to the gender, the average GPA from the previous year, religion, height

In this report, multiple ordinary least-squares (OLS) regression was used toverify the suitability of the model The model can sufficiently explain therelationship between the observed and implied variables The result revealed thatstudents who have impressive height, achieve high academic achievements andare not religious often have a higher IQ

Econometrics is "the quantitative analysis of actual

economic phenomena based on the concurrent development of theory and

observation, related by appropriate methods of inference"( P A Samuelson,

T.C Koopmans, and J R N Stone (1954)- "Report of the Evaluative

Committee for Econometrica," Econometrica 22(2), p 142.) Econometricians

try to find estimators that have desirable statistical properties including

unbiasedness, efficiency, and consistency

There are 2 classifications of Econometrics: Theoretical econometrics

(concerning methods and is closely related to mathematical statistics) and

models, analysing economic history, and forecasting)

Numerous research studies conducted over a period of almost a centuryhave shown that intelligence is not only essential to an individual’s life but also

an important correlate and determinant of a wide range of economic and socialphenomena including educational attainment, socioeconomic status, earningsand life achievement Low intelligence is a significant determinant ofunemployment, poverty, welfare dependency, mortality and crime (Brand andSmith, 1987)

That’s the reasons why our group opted for the topic “Factors affecting Intelligence Quotient Index of FTU’s students” to aquire additional

knowledge about some determinants of Intelligence Quotient Index as well assuggest feasible solutions to improve Intelligence Quotient Index for students.This essay incorporates the following content:

 ABSTRACT

 INTRODUCTION

 SECTION 1: OVERVIEW OF THE TOPIC

 SECTION 2: MODEL SPECIFICATION

 SECTION 3: ESTIMATED MODEL AND STATISTICAL

SECTION 1: OVERVIEW OF THE TOPIC

1.1 Intelligence Quotient Index and related terms

In one sense, human intelligence is something all humans share: it is what

traditionally marks us out from other animals and has made Homo sapiens one

of the more successful species on the planet It involves language and thecapacity to develop and transmit a culture, to think, reason, test hypotheses, andunderstand rules, and so on

For the greater part of the twentieth century, however, the psychological

study of human intelligence attempted to understand how and why people differ

in intelligence Until the development of cognitive psychology in the 1960s,indeed, this focus on individual differences quite overshadowed any attempt tostudy the general nature of human intelligence, what people share rather thanwhat sets them apart (Mackintosh, 2011)

Intelligence is measured by intelligence tests These testes typically consist

of several different kinds of tests of verbal reasoning, non-verbal reasoning,mental arithmetic, vocabulary, verbal comprehension and perceptual, spatial andmemory abilities (Mackintosh, 2011)

The total score derived from these several standardized tests designed toassess human intelligence is called Intelligence Quotient (IQ) Index

The scores obtained on intelligence tests are expressed in a metric in whichthe mean IQ of a representative sample of a national population is set at 100 andthe standard deviation is set at 15 Thus, approximately 96% of the populationhave IQs in the range of 70 to 130 The highest IQs that have been recorded arearound 200 (de Jong and Das Smaal, 1995)

1.2 Theoretical Background

Intelligence quotient is determined by several factors which include bothgenetic as well as non-genetic factors Even though genetic factors play themajor role in determining IQ, various other modifiable environmental influencescan influence the IQ of an individual IQ tests generally are reliable enough thatmost people ages ten and older have similar IQ scores throughout life But someindividuals score very differently when taking the same test at different times orwhen taking more than one kind of IQ test at the same age It has been noted that25% of assessed individuals will obtain a 10-point IQ score difference withanother IQ battery Even though not all studies indicate significant discrepanciesbetween intelligence batteries at the group level the absence of differences at theindividual level cannot be automatically assumed Variations in IQ scores arebased on an individual’s specific knowledge, vocabulary, expressive languageand memory skills, visual special abilities, fine motor coordination andperceptual skills Moreover, one’s emotional anxiety, tension and unfamiliaritywith the testing process can also influence the IQ score (Mackintosh, 2011).The genetic and environmental factors that determine IQ have beendifficult to pin down scientifically, but several aspects of the environmentincluding socioeconomic status and education are correlated with IQ, and it hasbeen shown that malnutrition can reduce IQ The variability in cognitive abilitiesamong different individuals is due to the interaction of genetic andenvironmental factors Genetics account for around 50% as per many studiesand increasing with age Shared and non-shared environment account for 25%and 20%, respectively, the latter 5% being represented by errors in theevaluation of the cognitive abilities Environment can modify geneticallydetermined cognitive abilities, and an enriched environment can improve the

performance (Aiello and Dean, 1990) Therefore, we want to focus on factors like gender, religion, previous GPA and height which are all environmental and genetic factors.

We may be genetically predisposed to a certain brain volume, structure andpathways – a certain level of intelligence set by our biology – but how much weachieve isn’t based in biology alone The type of life we lead also affectsintelligence (Gonzales, Cauce, Friedman, Mason, 1996).

SECTION 2: MODEL SPECIFICATION

2.1 Methodology in the study

2.1.1 Method to derive the model

In our research, we use Multiple Linear Regression, which is a linearapproach to modeling the statistical relationship between a dependent variableand one or more than one independent variables In particular, IntelligenceQuotient Index is statistically dependent on Gender, Height, Religion, andHeight

2.1.2 Method to collect and analyze the data

2.1.2.1 Method to collect the data

We want to measure factors affecting Intelligence Quotient index of FTU’sstudents in the previous school year (2018- 2019) 90 people were selectedrandomly from that population We use a cross section of that particularpopulation range The data source is taken from our survey

2.1.2.2 Method to analyze the data

In order to analyze the dataset and interpret the correlation matrix betweenvariables, we use STATA

2.2 Theoretical model specification

2.2.1 Specification of the model

Having considering previous research as well as theoretical background,our group has built a function to analyze the influences of some factor onIntelligence Quotient Index:

Intelligence Quotient Index= f(Gender, Height, GPA, Religion)

With a view to determining the effects of these above- mentioned factors

on Intelligence Quotient Index, our group decided to choose the regressionanalysis models

2.2.1.1 Population Regression Function:

Intelligence Quotient Indexi = β0 + β1 Genderi+ β2 Heighti + β3 GPAi + β4

Religioni +µi

In which:

β0 : the intercept term of the model

β1: the regression coefficient of Genderi β2: the regression coefficient

of Heighti

β3: the regression coefficient of GPAi

β4: the regression coefficient of Religioni

µi: the disturbance term of the model, which represents

Intelligence Quotient Index, but these factors are not cited in the model

2.2.1.2 Sample Regression Function:

Intelligence Quotient Index = + Gender+ Height+ ̂ GPA +

: the estimator of µi- residual

2.2.2 Explanation of the variables:

- Dependent variable is Intelligence Quotinent (IQ), which is ranged

from 1 to 10 in the questionnaire

 1: Normal

 5: Intelligent

 10: Genius

- There are 4 independent variables:

 Variable Height (m) from 1 to 10

 Yes, i have a religion: 0

 No, i do not have any religion: 1

 Variable GPA: The score is measured in number, it’s positive and the

After running sum IQIndex Gender GPA Religion Height, we obtained a

resulting table, which tells us the number of observations (Obs), the averagevalue (Mean), the Standard Deviation (Std Dev.), the Minimum (Min), theMaximum (Max) values of each variables

No Variable Obs Mean Std Dev Min Max

2.2.3.3 Correlation matrix between variables

In order to analyze the correlation between the variables, we run

corr IQIndex Gender GPA Religion Height

Gender has a positive but pretty low correlation coefficient of 0.1016

Therefore, it has a positive effect on the dependent variable IQIndex

GPA has a positive and extremely high correlation coefficient of 0.8621

Threfore, it has a positive effect on the dependent variable IQIndex,

Religion has a positive but pretty low correlation coefficient of 0.1016

Therefore, it has a positive effect on the dependent variable IQIndex

Height has a medium correlation coefficient of 0.3082 Therefore, it has

a positive effect on the dependent variable IQIndex

SECTION3:ESTIMATED MODEL AND STATISTICAL INFERENCE

3.1Estimated model

3.1.1 Estimation result

Variables Coefficient T P-value Confidence interval (95%)Constant -5.770058 -11.16 0.000 [-0.3532235; 0.3758733]Gender 0.113249 0.06 0.951 [2.751548;3.338474]

Height 0.267508 5.82 0.000 [0.014334;0.6437907]

GPA 3.045011 20.63 0.000 [0.1761593;0.3588567]Religion 0.3290625 2.08 0.041 [-6.797711;-4.742406]

3.1.2 The sample regression model

Intelligence Quotient Index i = + Gender i + Height i + GPA i + Religion i +

According to the estimated result from STATA and the Ordinary least squares (OLS) method, we obtained the sample regression model:

Intelligence Quotient Index i = -5.770058 +0.113249 Gender i +

0.267508 Height i + 3.045011 GPA i + 0.3290625 Religion i +

3.1.3 The coefficient of determination

R2 is 0.8547 means that the estimated model explains 85.47% of the

total variation in the value of Intelligence Quotient Index in this sample

3.1.4 Meanings of estimated coefficients

̂

= -5.770058 means that if all 4 independent variables equal zero, the expected value of Intelligence Quotient Index will be -5.770058 In fact, it does not happen as the minimum of the variable Height is 1 and GPA of a student never equals 0.

̂

= 0.0113249 > 0 means that holding other factors fixed, when you are male , the expected Intelligence Quotient Index increases by 0.0113249 unit of

3.1.5 Other results analysis

Sum squared residual - RSS: 34.4189844: measures the sample

variation in the

The standard error of Gender variable is 0.1833497 which means the

standard distance between the observations and the regression line is about 18%.Similar with GPA, Religion and Height variable, the distance are

about 15%, 4% and 51% respectively

3.2 Hypothesis Testing

3.2.1 Verifying the suitability of individual regression coefficient

β 4 Religion i + µ i

Hypothesis: (βj = β1, β2, β3, β4)

Testing formula:

Ts=

To verify the suitability of individual regression coefficient we use STATAand achieve the result as below:

+ P-value of Gender is 0.951 so PGender > α (α = 5%)

 At 5% level of significance, we accept the hypothesis

+ P-value of Height variable is 0.000 so PHeight < α (α = 5%)

 At 5% level of significance, we reject the hypothesis

+ P-value of GPA is 0.000 so PGPA < α (α = 5%)

 At 5% level of significance, we reject the hypothesis

+ P-value of religion is 0.041, so Preligion <α (α = 5%)

We can conclude that at 5% level of joint significance, we have

enough evidence to reject the hypothesis H0 and accept H1

In conclusion, the model is consistent at the significance level of 5%

3.2.3 Explanation for coefficient that is not statistically significant.

According to “Introduction to Psychology” by Plotnik R,

Kouyoumdjian H (2013), numerous researchers support the notion that there are

no significant sex differences in general intelligence, while others have argued for slightly greater intelligence for males, and others for a female advantage These results depend on the methodology, tests which researchers used for theirclaims, and the personal performances of the participants Our outcome is that ifyou are male, the expected Intelligence Quotient Index increases by 0.0113249 unit of measurement, which is an extremely minimal rise So there are no

significant differences between females and males in terms of intelligence This