Lecture 1

Task 1

“Statistics” is said to have four semantic interpretations (or meanings). These

typically are:

1. Statistics as numerical facts (data)

2. Statistics as a method or technique
3. Statistics as a science (discipline)
4. Statistics as a government or organizational activity (institutional sense)

Now let’s define each in our own words and give specific Uzbekistan-related
examples (using sources like [Link], the State Committee on Statistics, or [Link],
the Central Bank of Uzbekistan).

1. Statistics as Numerical Facts (Data)

Definition (in own words):

Statistics in this sense refers to actual numerical information that describes some
aspect of reality — for example, figures showing economic growth, population, or
trade. It’s the raw, quantitative facts collected about different phenomena.

Example (Uzbekistan):
According to [Link], Uzbekistan’s GDP growth rate in 2024 was 6.2%. This
number itself — 6.2% — is a statistic in the first sense, representing a measurable
economic fact about the country.

2. Statistics as a Method or Technique

Definition (in own words):

Here, statistics refers to the set of methods used to collect, organize, analyze, and
interpret data. It includes sampling, regression, hypothesis testing, and other
analytical tools.

Example (Uzbekistan):
Economists at the Central Bank of Uzbekistan ([Link]) might use time series
analysis to forecast future inflation rates based on past consumer price index
(CPI) data. The use of this analytical method represents statistics in the
methodological sense.
3. Statistics as a Science (Discipline)

Definition (in own words):

In this sense, statistics is the scientific field that develops principles and theories
for working with data — it’s an academic and research discipline that guides how
we understand and model uncertainty.

Example (Uzbekistan):
Researchers at Tashkent State University of Economics might conduct studies on
improving the accuracy of agricultural output forecasts using statistical models.
Their work contributes to the science of statistics — developing and refining
statistical knowledge.

4. Statistics as a Government or Organizational Activity

Definition (in own words):

Statistics here refers to the organized system or agency responsible for collecting
and disseminating official data — usually by the state or large institutions.

Example (Uzbekistan):
The State Committee on Statistics of the Republic of Uzbekistan ([Link])
conducts national population censuses, collects industrial output data, and
publishes monthly economic reports. This institutional activity represents
statistics as an organizational function.

✅ Summary Table:

Interpretation Meaning (in own words) Example (Uzbekistan context)

Numerical facts about
1. Data GDP growth = 6.2% ([Link])
phenomena
CBU uses regression to predict
2. Method Techniques to analyze data
inflation
Academic field developing University researchers improving
3. Science
theory forecast models
4. Government body collecting and State Committee on Statistics
Organization publishing data ([Link])

Task 2
1. The regions of Uzbekistan (Tashkent, Samarkand, etc.)

Type: Qualitative – Nominal

Justification:
These are categories or names of regions that represent geographic locations, not
measurable quantities. The regions don’t have a natural order or ranking —
Tashkent isn’t “greater” or “less” than Samarkand in any numeric sense. Thus, it’s
qualitative nominal data.

2. Monthly family income in Uzbek Soum

Type: Quantitative – Continuous

Justification:
Income is a numerical measure that can take on any value within a range (e.g.,
3,250,000 soums; 3,250,001 soums, etc.). Since income can be measured in
fractions of a soum (in theory) and represents a continuous scale, this is
quantitative continuous data.

3. Customer satisfaction rating for a bank's service (1–Very Dissatisfied to 5–

Very Satisfied)

Type: Qualitative – Ordinal

Justification:
Although the variable uses numbers (1–5), these numbers represent ranked
categories — the difference between 1 and 2 is not necessarily the same as
between 4 and 5. The key property is order (from dissatisfaction to satisfaction),
not numeric distance. Therefore, it is qualitative ordinal data.

4. The number of employees in an SME

Type: Quantitative – Discrete

Justification:
The number of employees is a count — you can have 12 or 13 employees, but not
12.5. Since it takes only whole number values, it’s quantitative discrete data.

5. The exact weight of cotton harvested from a field

Type: Quantitative – Continuous
Justification:
Weight is a measurable quantity that can take on any value within a range — for
example, 1,235.75 kg. Even small fractions are meaningful. Therefore, it’s
quantitative continuous data.

Task 3

1. Qualitative Side – Nature and Essence of the Phenomenon

Internal migration in Uzbekistan refers to the movement of people from rural

regions to urban centers such as Tashkent, Samarkand, and Fergana.
This phenomenon is driven by economic, social, and educational factors —
people seek better employment opportunities, higher living standards, and
improved access to education and healthcare in cities.

However, it also brings challenges such as urban overcrowding, housing

shortages, and pressure on public services. Culturally, internal migration affects
family structures (e.g., younger people moving away from parents) and regional
labor patterns, influencing the overall socio-economic development of the
country.

In essence, this is both an economic necessity and a social transformation process

reshaping Uzbekistan’s demographic and urban landscape.

2. Quantitative Side – Measurable Variables

To study this phenomenon empirically, we can identify measurable indicators

that capture its intensity and effects:

No. Variable Description

Number of internal migrants per Total people officially changing residence
1
year from rural to urban areas
Annual percentage increase in population
2 Urban population growth rate (%)
in major cities
Average monthly wage in urban Measures income disparities motivating
3
vs. rural regions (UZS) migration
4 Housing cost index in urban areas Reflects demand pressure in city housing
No. Variable Description
markets
Employment rate among new Indicates how successfully migrants
5
urban residents integrate economically

Task 4

1. Subject of the Study

Subject:
To describe the typical student at my university in quantitative and qualitative
terms.

Example (if you’re at a university in Uzbekistan):

“The socio-economic and academic characteristics of students at Tashkent State
University of Economics.”

This subject focuses on summarizing key features — not predicting or generalizing

beyond the current student population.

2. Population

Population:
All currently enrolled undergraduate students at the university.

Explanation:
The population includes every individual who is registered as a student in the
chosen academic year. If the study is limited (for practical reasons), you could
specify one faculty (e.g., Faculty of Economics).

3. Variables to Be Collected

Type Variable Measurement/Scale Description

Quantitative Numerical age of the
1. Age Continuous (years)
(3) student
2. Monthly Continuous Average amount spent
 Type Variable Measurement/Scale Description
expenditure (in
per month
soums)
3. GPA (Grade Academic performance
Continuous
Point Average) indicator
Qualitative
4. Gender Nominal Male / Female / Other
(2)
Field of study
5. Faculty or Major Nominal (Economics,
Management, IT, etc.)

These variables together give a comprehensive view of the “typical” student’s

demographic, academic, and financial profile.

4. Method: Descriptive or Inferential Statistics?

Chosen Method: Descriptive Statistics

Justification:
The aim is to describe the characteristics of the current student population — to
find averages (mean age, average GPA), frequency distributions (gender or faculty
proportions), and possibly measures of variation (standard deviation of monthly
expenditure).
We are not trying to make predictions or infer patterns beyond the current
university or population. Therefore, descriptive statistics — not inferential — are
appropriate.

✅ Summary Overview:

Component Description
Subject Description of a typical student at the university
Population All currently enrolled undergraduate students
3 quantitative (age, expenditure, GPA) + 2 qualitative (gender,
Variables
faculty)
Component Description
Method Descriptive statistics — to summarize and describe existing data

Task 5

. Using Inferential Statistics

To test the claim that families in Tashkent spend 4 million soum monthly on
groceries, I would:

 Define the population as all families in my district.

 Select a sample (e.g., 50–100 households) and record their actual grocery
spending.
 Use hypothesis testing (a one-sample t-test) to compare the sample mean
with the claimed 4 million soum.
If the difference is statistically significant, the claim is likely inaccurate for
my district.

2. Hypotheses

 Null hypothesis (H₀): μ = 4,000,000 soum

→ Average spending equals 4 million soum.
 Alternative hypothesis (H₁): μ ≠ 4,000,000 soum
→ Average spending differs from 4 million soum.

Task 6

1. Using Statistical Thinking in Program Design

Before launching a program to support young entrepreneurs, policymakers would

use statistical thinking to make evidence-based decisions.
They would collect and analyze data on:

 Number and age distribution of current entrepreneurs (from [Link])

 Start-up success/failure rates
 Access to credit and average loan sizes (from [Link])
 Regional differences in youth employment and business activity

This helps identify who needs support most, how large the target group is, and
what resources are required.
2. Using Statistical Methods to Evaluate Success

After implementation, policymakers could apply key statistical methods for

economists such as:

 Descriptive statistics: Compare before-and-after averages (e.g., number of

new businesses started, youth employment rate).
 Time series analysis: Track trends in entrepreneurship over time.
 Regression analysis: Assess whether participation in the program
significantly increases business survival or income, controlling for other
factors.
 Surveys and hypothesis testing: Test if satisfaction or access to finance
improved among young entrepreneurs.

These methods allow the government to measure the program’s real impact and
decide whether to expand, adjust, or discontinue it.

Task 7

1. Effect of “Every Family is an Entrepreneur” Program on Household Income

Method: ANOVA (Analysis of Variance) or t-test

Why:
We’re comparing average household income between two or more groups
(participating vs. non-participating regions).
ANOVA or a two-sample t-test shows whether the differences in mean income are
statistically significant.

2. Relationship Between Central Bank’s Key Rate and Loan Volume

Method: Correlation or Regression Analysis

Why:
We want to see if changes in the key interest rate are related to loan volumes.

 Correlation measures the strength and direction of the relationship.

 Regression can go further — showing how much loan volume changes
when the key rate changes.

3. Forecasting Electricity Demand in Tashkent

Method: Time Series Analysis
Why:
The data involves electricity consumption over time, and the goal is to predict
future demand.
Time series models (like trend or ARIMA) analyze patterns and seasonal effects to
make reliable forecasts for the next 5 years.

Task 8

. Article Selected

The article titled “Uzbekistan’s economy grows 7.2 % in H1 2025, but household
income growth slows to 9.5 %” from [Link]. [Link]+1

 It reports that in the first half of 2025 GDP growth was 7.2% (vs 6.6% a year
earlier) and that real growth in household income slowed to 9.5%.
[Link]+1
 It uses data from the National Statistics Committee of Uzbekistan (the
national statistical agency) and presents various sector-growth figures and
inflation rate. Кун.uz+1

2. Critical Evaluation

a) Transparency of Data Presentation

 Positives: The article states the source (“released by the National Statistics
Committee”). It gives specific growth figures (GDP 7.2 %, real income
growth 9.5 %) and also notes the previous period’s number (11.9% for
income). Кун.uz+1
 Gaps / questions:
o It is not clear how “real growth in household income” is defined
(which deflator, representative sample, whether median or mean).
o It does not show raw values (e.g., average income per household) or
how many households are included in the survey.
o It mixes many sectoral growth figures (industry, transport, exports)
without clarifying the methodology or data quality for each.

b) Potential for Misleading Presentation

 The headline emphasises that “income growth slows to 9.5%” which may
suggest that households are doing much worse, but 9.5% is still a
 substantial growth rate. The framing could lead a reader to believe a crisis
when it might simply be a moderate deceleration.
 Because only growth rates are shown (percentages) rather than actual
income levels or distribution, the reader cannot assess whether the growth
was from a low base, whether the benefits are widely shared, or whether
inflation eroded the gains (though inflation is mentioned).
 The article gives many sectors growing at double-digits (e.g., services,
exports) which might suggest strong economy overall—but without linking
those sectors to actual household welfare. A reader might misinterpret
“economy grows strongly” as “households are better off”, though the
slower income growth suggests nuance.

c) Questions I Would Ask to Assess Objectivity

1. What is the definition of “household income” used? Does it include

transfers, informal earnings, remittances, self-employment income? The
article mentions that part of income comes from self-employment but
without full breakdown. [Link]+1
2. What is the sample size and methodology of the data collection? Are the
same households tracked over time (panel) or is it a cross-section? How are
rural vs urban households weighted?
3. What deflator or inflation adjustment is used to derive “real” income
growth? The article mentions inflation (CPI) fell from 5.2% to 4.2%. [Link]
Does the “real income” growth take that into account, and how precisely?
4. What is the distribution of income growth across regions, income levels,
and demographic groups? The article gives some regional numbers in a
related piece. [Link] But without that one can’t tell if the
“average” hides large disparities.
5. Are there revisions or preliminary estimates? The article says data for H1
are preliminary (in some cases). For example “according to preliminary
estimates from …”. Кун.уз+1 So how reliable are they?
6. Could any of the growth be due to methodological changes rather than
real changes? For example changes in data collection, coverage, or
classification can affect growth rates.

✅ Summary

The article does provide useful statistics and identifies a slowdown in income‐
growth relative to past years, which is valuable. It uses an official data source and
gives some relevant indicators. However, from an ethical statistics perspective
there are concerns:
  The lack of transparency about methodology, distribution and sample
details means the reader may not fully understand how robust the claim is.
 There is a risk of misinterpretation because focusing on growth rates
without context (base levels, inequality, inflation, regional differences) can
lead to misleading impressions.
 To be objective and ethical, the article should provide more metadata,
clarify assumptions, show possible limitations, and ideally give more than
one measure (e.g., median income, income by decile) rather than only
average growth.

Task 9

1. Hypothetical Dataset (10 Companies)

Company_ Number_of_Emplo Annual_Revenue_Mln Growth_Prosp

Sector
ID yees Soum ect
C001 IT 45 12.5 High
C002 Agriculture 120 25.8 Medium
C003 Retail 80 18.3 Medium
Manufactur
C004 200 40.6 High
ing
C005 Services 60 15.1 Low
Constructio
C006 150 30.9 Medium
n
C007 IT 30 10.2 High
C008 Agriculture 95 22.7 Medium
C009 Retail 50 13.4 Low
C010 Energy 300 75.3 High

2. Excel Calculations

Once you enter this data in Excel:

 Mean Number of Employees:

Formula → =AVERAGE(C2:C11)
 Median Number of Employees:
Formula → =MEDIAN(C2:C11)
 Mean Annual Revenue (Mln Soum):
Formula → =AVERAGE(D2:D11)
 Median Annual Revenue (Mln Soum):
Formula → =MEDIAN(D2:D11)
3. Example Results (based on the data above)

Variable Mean Median

Number_of_Employees 113 87.5
Annual_Revenue_MlnSoum 26.48 21.75

Task 10

1. Descriptive Statistics:
The typical company in the dataset has around 50 employees and generates an
average annual revenue of $5 million, suggesting most companies are small to
medium-sized. The median years in operation is 8 years, indicating that half of the
companies have been established for less than a decade.

2. Inferential Thinking:
A possible question could be: “Do companies with more employees tend to
generate higher revenue?” To investigate, I would use correlation analysis to see
if there is a linear relationship between the number of employees and revenue,
and then potentially run a simple linear regression with revenue as the
dependent variable and number of employees as the independent variable to test
the strength and significance of the relationship.