(TIỂU LUẬN) assignment 2 individual case study inferential statistics

The individuals using internet indicator refers to the percentage of internet users in the global population.. Three income groups expressed an upward trend in the percentage of individu

ECON1193 BUSINESS STATISTICS 1 –

Assignment 2 Individual Case Study - Inferential Statistics –

Word count

2

Part 1 Introduction

Sustainability is the development that fulfills present needs without undermining future

generations' potential, maintaining a balance between economic growth, concern for the

environment and social well-being (Acciona n.d.) In today world, the term 'sustainable

development' is gradually becoming popular not only in businesses but also in our society

Initially, this term was originated from the Brundtland report in 1987, defined sustainable

development as ‘the human ability to make development sustainable-to ensure that it meets the needs of the present without compromising the ability of future generations to meet their own needs’ (Robert, Parris & Leiserowitz 2005, p.10) In 2015, the United Nations adopted the 2030 Agenda, which includes the Sustainable Development Goals (SDG) a plan developed to protect – the planet and ensure the well-being of people around the world SDGs consist of 17 goals which are planned to achieve in 15 years, from 2015 to 2030 (Appendix A) One of the goals is SDG Goal 8 specifies Decent work and Economic growth To make this goal more coherent, there is a factor we can consider using: Individuals using Internet

As defined, Internet users are individuals who have spent the last three months using the Internet from any area worldwide The Internet can be accessed through a mobile phone, a computer, personal digital assistant, games machine, digital TV and so on (Roser, Ritchie & Ortiz-Ospina 2015) The individuals using internet indicator refers to the percentage of internet users in the global population According to the International Telecommunication Union n.d (ITU), the number of internet users increased sharply after 10 years due to the rapid growth of technology and the popularity of social media, from only 1.5 billion people in 2008 to 4.1 billion people by the end of 2019 which accounts for 53.6% of global population (Appendix B) In some regions such as Europe, Americas, there has been higher number of individuals using the Internet per

100 inhabitants than other areas (Appendix C), in which he Internet will produce significant ‘t cost savings in many sectors of the economy, resulting in faster productivity growth It will also produce lower prices for consumers, resulting in faster growth in living standards (Rivlin & ’ Litan 2001) Hence, the use of the internet and technology for administrative, social affairs and economic development should be highly encouraged to achieve the SDG Goal 8

In addition, the relationship between Individuals using internet and gross nation income (GNI) has been noticeable The income classification is determined using the Atlas process, based on a calculation of national income per person, or GNI per capita Prydz & Wadhwa (2019) defined the low-income economies with the GNI per capita of $1,025 or less in 2018; middle-income countries with GNI per capita between $1,026 and $12,375; lastly, high-income nations achieved GNI per capita of $12,376 and above Three income groups expressed an upward trend in the percentage of individuals using internet, recorded by World Bank Low-income countries have the lowest percentage of internet users, at 15.8% of population; followed by is middle-income with 45.8% of population who went online On the contrary, countries with high GNI per capita showed a large gap with the two first groups, account for 84.6% of population, which is also higher than the world’s individuals using internet in 2017 (Appendix D) Nevertheless, GNI is

not the only factor that affect the proportion of individuals using the Internet, it can be driven by other factors

Part 2 Descriptive Statistics and Probability

1 Probability

1.1 Consider whether income and Individuals using internet are statistically independent

events

38 chosen countries in the data set #9 are divided into 3 groups of income:

• Low-Income countries (LI): countries with a GNI less than $1,000 per capita

• Middle- Income countries (MI): countries with a GNI between $1,000 and $12,500 per

capita

• High-Income countries (HI): countries with a GNI greater than $12,500 per capita

These countries are also sorted based on the Internet usage:

• Low usage of internet (A): individual using the Internet less than 40% of population

• High usage of internet (A’): individual using the Internet more than 40% of population

We create a contingency table (unit: Number of countries):

Low usage of internet

(A)

High usage of Internet (A’)

Total

Low-income

Middle-income

High-income

Next, we will compare two probabilities to check if individuals using internet and income are statistically independent or not The two probabilities are: the marginal probability that a random chosen country has low internet usage, which is P(A); and the conditional probability that a random chosen country has low internet usage, regarding low income, denoted P(A|LI)

4

P(A) = 14

38 =

7

19 = 0.37 (1)

P(A|LI) = P(A ∩ )LI

P(LI) =

5

5= 1(2)

The outcome presents that the marginal probability that a random chosen country has low internet usage is different from the conditional probability that a random chosen country has low internet usage, given that low income As a result, the number of individuals using internet and income are statistically dependent In other words, individuals using internet and income are related to each other

1.2 Which country categories are more likely to have high usage of internet?

Based on the Contingency table above, to identify which country categories are likely to have high usage of Internet (above 40% of population), we calculate the conditional probability of countries have high usage of Internet regarding each category: Low-income, Middle-income and High-income

P (A’|HI) = P (A ∩ ) ′ HI

P(HI) =

13

13 = 1 = 100%

P (A’|MI) = P (A ∩ ) ′ MI

P(MI) =

11

20 = 0.55 = 55%

P (A’|LI) = P (A ∩ ) ′ LI

P(LI) =

0

5 = 0 = 0%

In conclusion, high income countries are the most likely to have high usage of internet with 100% of population using the internet – the highest usage among three categories Follow by

is middle income countries still recorded high internet usage with 55% Except for low income countries, it is recorded with 0% individual using the Internet

2 Descriptive Statistics

Measures of Central Tendency

From (1) and (2)

→ P(A) ≠ P(A|LI)

P (A’|HI) > P (A’|MI) > P (A’|LI)

Table 1 The measures of Central Tendency of Individuals using the Internet (% of population)

We have created histograms of three country categories to decide whether any outliers are apparent or not Based on the observation from the chart, there

is no outlier showing in the three country categories Therefore, since no outlier is found, Mean

is the best measure of central tendency

According to Table 1, the average of individuals using Internet in selected HI countries the is

highest with 82.92%, come after is MI countries and LI countries with 42.15% and 18.41%

respectively In conclusion, HI countries tend to have a higher average of individuals using

Internet in population than MI and LI countries

Part 3 Confidence Intervals

a We suppose a 95% confidence interval for the world average individuals using Internet in

2017

3

0

0.5

1

1.5

2

2.5

3

3.5

Individuals using Internet (% of population)

Histogram of LI countries' individuals using

Internet

Figure 2 Histogram of MI countries

1

3

5

6

5

0 1 2 3 4 5 6 7

11.92 26.53 41.13 55.73 More

Individuals using Internet (% of population)

Histogram of MI countries' individuals

using Internet

Figure 1 Histogram of LI countries

4

7

0

1

2

3

4

5

6

7

8

Indivuduals using the Internet (% of population)

Histogram of HI countries' individuals using

Internet

Figure 3 Histogram of HI countries

6

We have 1- = 0.95 α → significance level α = 0.05

As population standard deviation is unknown, we have to use T-table instead of Z-table σ Significance level α = 0.05; 𝛼 2⁄ = 0.025

Degree of Freedom d.f = n 1 = 38 - 1 = 37 –

Confidence interval estimates: X ± t × √nS =

→ 44.13 61.83 ≤ 𝜇 ≤

→ We are 95% confident that the true world average of individuals using the internet is

between 44.13% and 61.83% of population in 2017

b Although the population standard deviation is unknown, the sample size of the σ

assigned dataset is 38, greater than 30, which is large enough to apply for Central Limit Theorem (CLT) CLT is applicable so the distribution of the sample mean will become approximate normal distribution, regardless of the shape of the population Therefore, no assumption is needed to calculate these confidence intervals

c In case the population standard deviation is known, the confidence interval will show a σ decrease (the gap becomes smaller) Since the sample standard deviation varies from S sample to sample, it induces some confusion and the results may not convincing enough Instead, the population standard deviation can increase a more accurate and precise 𝜎 result McLeod (2019) stated that ‘The narrower the confidence interval (upper and lower values), the more precise is our estimate Besides, the width of confidence interval ’ values will become narrow as the sample size increases, which means the bigger the sample size n, the more accurate the outcomes show

Part 4 Hypothesis Testing

a According to a report published by the World Health Organization (WHO), in 2016, the world average individual using the Internet (% of population) is 44.7% In part 3a, we are 95% confident that the true world average of individuals using the internet ranged from 44.13% 61.83% in 2017 However, it is uncertain to assert the global average of to individuals using the internet will remain constant, increase, or decrease in coming years Still, the sample mean (the point estimate of confidence interval) 52.98% in 2017, is

According to T-table

➔ t(n-1, 𝛼

2) = t(50,0.025) = 2.0262

52.98 + 2.0262 ×26.92√38 = 61.83 52.98 - 2.0262 × 26.92√38 = 44.13

which is seen to be higher than 44.7% in 2016 Hence, the world’s individuals using the Internet is predicted to rise in the future A hypothesis test we conduct below will reinforce our statement

Confidence level (1 – α) *100% 95%

Step 1: Check the CLT

As the sample size n = 38, is greater than 30, CLT can be applied And the sampling distribution

of all possible mean becomes normally distributed since sample size n increases

Step 2: State the Null Hypothesis, H and the Alternative Hypothesis, H 0 1

H0; μ 44.7 ≤

H1; μ > 44.7

Step 3: Choose level of significance α = 0.05; sample size n = 38

Since H sho1 ws the sign “>”, we use an upper-tail test

Step 4: Determine which table to use

Since the population standard deviation is unknown, and the sampling distribution of all sample mean is normally distributed, we will use the T-table

Step 5: Determine Critical value (CV)

Level of significance α = 0.05

Degree of freedom d.f = n -1 = 38 1 = 37 –

Step 6: Calculate test statistic t

ttest = X− μ

S

√n

= 52.98 − 7 44

26 92

√38

= 1.896

Step 7: Make statistical decision

The ttest > tCV (1.896 > 1.6871), the test statistic belongs rejection region so to

→ Reject the Null Hypothesis H 0

Because it is upper tail test, the t critical value (t ) 1.6871 CV is

8

Step 8: Make a managerial decision in the context of the problem

As we reject H , hence with 95% level of confidence we can conclude that the world average 0

individual using the Internet will not increase

Step 9: Discuss the possible errors

As we reject H , we might have committed Type I error (0 α) We are concluding that the world average individual using the Internet will not increase in the future, but there are still

opportunities that it may increase The probability of Type I error can be reduced by minimizing the significance level α Because level of significance picked by researchers, it can be changed is easily The significance level, for instance, can be reduced to 0.01 (1%), which means there is a chance of 1% that the null hypothesis is mistakenly rejected (Corporate Finance Institute n.d.)

b Suppose the number of countries in the assigned data set triples, the result will be more precise To be more specific, we will inspect the two formulas which are the sample standard deviation (S) and the test statistic t

Firstly, considering the sample standard deviation formula, S and degree of freedom (n 1) have –

an inverse relationship, when sample size n increases, S will decrease

Then, considering t test with the sample standard deviation formula, S and t test are inversely related; plus, sample size n and t test show a direct relationship As a result, when n increases and decreases, t test will increase, shift to the right but remains in the rejection region area From the Figure 4 above, we can observe that t test and t value will be quite far from each other Hence, the statistical decision could be said to remain unchanged

The triple in the sample size n will increase the accuracy of the results Determining Standard error’s formula SE = SD/√n, the standard error falls when we increase the sample size n as they

S = √∑ (Xi−X)ni=1

n−1 ttest = X− μ

S

√n 1.6871 1.896

are inversely related, and contrast, the standard deviation will likely not to change due to the in increase of the sample size (Altman & Bland 2005) Thus, in our case, the three times increase in the sample size will certainly decrease the standard error of the test by 1.7 times The smaller the error we get, the more reliable the results turn out, the more appropriate of the sample value of the whole population (Kenton 2020)

Part 5 Conclusion

In brief, the sustainable development goals established by United Nations has been positively responded by many countries around the world They analyze the global problems we are facing today related to hunger, poverty, climate change, environmental degradation, peace and justice (United Nations n.d.) The 17 goals are set to eliminate the challenges above and are committed

to accomplish within a 15-year period One of the goals related to this analysis is Goal 8 - economic growth and decent work Through this goal, the number of individuals using the internet - the main topic of this article, partly reflects the development of the economy Some significant findings can be derived from the calculation and analysis of individuals using the Internet in 38 selected countries

Firstly, we concluded that income and individuals using the Internet are statistically dependent

in the first part From the calculation and conclusion, it is noticeable that high income countries (GNI above $12,500) recorded the highest usage of internet with 100% individuals using the Internet, follow by is middle income countries with 55%, and low income at 0% Next, with the descriptive statistics, after calculating the Mean of three categories, one more time we conclude that HI countries observed the highest average of individuals using the Internet

Besides, we are 95% confident that the true world average of individuals using the internet is between 44.13% and 61.83% of population in 2017 Besides, according to WHO report, the ’s world average individual using the Internet (% of population) is 44.7% in 2016, and in 2017 our data set recorded 52.98%, which is predicted to rise in the future However, after conducting the hypothesis test, there are still chances of this number to decrease which is type II error

Moreover, an increase in the sample size by three times is proven to have no impact on the statistical decision of the true world average number of individuals using the Internet Also, it can improve the accuracy of the test

Finally, from all findings above, it is recommended that government should innovate and

highlight the use of the Internet and technology not only for our daily life, but also for economic and social purposes By applying the Internet and modern technology, census and administrative tasks are quicker and easier for both citizens and government to achieve Therefore, improving the quality of life as well as the income

10

REFERENCES

Acciona n.d, 'What is sustainable development and how to achieve it?', Acciona, viewed 15 August 2020,

<https://www.acciona.com/sustainable-development/>

Altman GD & Bland JM 2005, ‘Standard deviations and standard errors’, PMC, viewed 23

August 2020,

<https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1255808/

https://corporatefinanceinstitute.com/resources/knowledge/other/type-i-error/

International Telecommunication Union n.d, 'Statistics', International Telecommunication Union, viewed 16 August 2020,

<

https://www.itu.int/en/ITU-D/Statistics/Pages/stat/default.aspx#:~:text=%E2%80%8B%E2%80%8B%E2%80%8B%E2%80

%8B%E2%80%8B,people%2C%20are%20using%20the%20Internet.>

Kenton, W 2020, ‘Standard Error’, Investopedia, viewed 23 August 2020,

<https://www.investopedia.com/terms/s/standard-error.asp>

Leod, S 2019, 'What are Confidence Intervals in Statistics?', Simply Psychology, viewed 19 August 2020,

<https://www.simplypsychology.org/confidence-interval.html>

Prydz, BE & Wadhwa, D 2019, 'Classifying countries by income', World Bank, viewed 16 August 2020,

< https://datatopics.worldbank.org/world-development-indicators/stories/the-classification-of-

countries-by-income.html#:~:text=As%20of%201%20July%202019,between%20%243%2C996%20and%20

%2412%2C375%3B%20high%2D>

Rivlin, MA & Litan, ER 2001, 'The Economy and the Internet: What Lies Ahead?', Brookings, viewed 16 August 2020,

<

https://www.brookings.edu/research/the-economy-and-the-internet-what-lies-ahead/#:~:text=The%20Internet%20will%20produce%20significant,faster%20growth%20in%20 living%20standards.>

Robert, WK, Parris, MT & Leiserowitz, AA 2005, 'What is Sustainable Development? Goals,

Indicators, Values, and Practice', Environment: Science and Policy for Sustainable Development,

vol 47, no 3, pp 8-21,

<https://www.tandfonline.com/doi/pdf/10.1080/00139157.2005.10524444?casa_token=Fm39b9

bC2NwAAAAA:RHM4O0LmRzN5b8yFr-8yWPbaACI5HF48xQ-_Og57HaST7-Zr6z3Om3l5vUYgr2dmEnLIm3aSQ_GX_A>