Unit 1: Introduction to statistics & sampling
Business Statistics notes · PTU syllabus (BCOMGE 201-18)
On this page
Unit summary
Business decisions rely on numbers — sales trends, customer surveys, quality checks. Statistics provides the tools to collect, organise and interpret such data. This unit covers the meaning, features, importance, functions, scope and limitations of statistics, sources of primary and secondary data, classification and frequency distributions with their graphs, population and sample, descriptive and inferential statistics, and the probability and non-probability methods of sampling.
After this unit you can
- Define statistics and explain its features, functions, importance, scope and limitations
- Distinguish primary and secondary data and name the methods and sources of each
- Build a frequency distribution and draw a histogram, frequency polygon and ogives
- Distinguish population and sample, parameter and statistic, descriptive and inferential statistics
- Explain probability and non-probability sampling methods and sampling and non-sampling errors
PTU syllabus topics
- Meaning
- features
- importance
- functions
- scope and limitations of statistics
- sources of primary and secondary data
- presentation of data
- frequency distribution
- population and sample
- descriptive and inferential statistics
- probability and non-probability sampling methods
Topic 1
Meaning, features, importance, functions and scope
Statistics (singular) is the science of collecting, classifying, presenting, analysing and interpreting numerical data. Statistics (plural) are numerical facts collected for a purpose — "the statistics of sales".
Meaning
Numerical facts or data
Science and method of handling data
Example
Sales figures of five years
Methods such as averages and correlation
Definition by
Horace Secrist
Croxton and Cowden
Aggregates of facts
A single figure is not statistics
Numerically expressed
Quantitative, not qualitative
Affected by many causes
Sales depend on price, income and taste
Reasonably accurate
Estimates allowed, not guesses
Collected systematically
For a predetermined purpose
Comparable
Same units and time, so comparison is possible
Related to one another
Facts about the same field
Simplification
Large data reduced to a few figures
Presentation
Tables, diagrams, graphs
Comparison
Between periods and groups
Formulation of policy
Evidence for decisions
Forecasting
Trend and regression
Testing hypotheses
Inference from samples
Relationships
Correlation and regression
Marketing
Market research, demand forecasting, advertising effect
Production
Quality control, inspection by sampling
Finance
Risk, returns, ratio analysis, stock trends
Human resources
Attrition, performance, pay scales
Economics and planning
National income, price indices, census
Banking and insurance
Credit scoring, premium calculation (probability)
Government
Policy, planning, budgets
Limitations of statistics
| Explanation | |
|---|---|
| Studies groups, not individuals | Results are true on average |
| Only quantitative data | Qualitative facts such as honesty cannot be measured directly |
| Can be misused | Selected figures can mislead — "lies, damned lies and statistics" |
| Needs expert handling | Wrong method gives wrong answer |
| Results are not exact | Based on samples and estimates |
| Depends on homogeneous data | Mixed data gives meaningless averages |
Topic 2
Sources of data
Meaning
Collected first-hand for the present purpose
Already collected and published by others
Sources
Own survey, interview, observation, experiment
Government publications, reports, journals, websites
Cost and time
High
Low
Reliability
High, if collected carefully
Must be checked for suitability
Example
Questionnaire to 200 customers
RBI Handbook, Census, Economic Survey
Direct personal investigation
Interviewer meets the respondent
Indirect oral investigation
Ask people who know
Questionnaire (mailed)
Written questions sent to respondents
Schedule (enumerators)
Form filled by trained enumerators
Observation
Watching behaviour
Telephone and online surveys
Quick and wide reach
Experiments
Controlled trials such as test marketing
Published
Government — Census, NSSO (MoSPI), RBI Bulletin, Economic Survey, Handbook of Statistics
Published
Non-government — trade associations, research bodies, newspapers, journals
International
UN, World Bank, IMF
Unpublished
Company records, research theses
- Suitability
- Does it match the objective?
- Adequacy
- Enough coverage
- Reliability
- Reputation of the source and the method
- Definitions
- Same units and definitions as needed
- Date
- Up to date
- Cross-check
- Compare with another source
- 1
Fix the objective
- 2
Decide what information is needed
- 3
Write short, clear, unbiased questions
- 4
Arrange in logical order
- 5
Add instructions and an introduction
- 6
Pre-test on a small group
- 7
Revise and administer
Topic 3
Classification, tabulation and frequency distribution
Classification arranges data into groups with common characteristics — by time, place, quality or quantity.
Chronological
By time — yearly sales
Geographical
By place — state-wise population
Qualitative
By attribute — gender, religion
Quantitative
By measurable variable — marks, income
A frequency distribution groups observed values into classes and shows the number of observations in each.
| Marks (class) | Frequency f | Mid-value x | Less-than cumulative frequency | More-than cumulative frequency |
|---|---|---|---|---|
| 0–10 | 5 | 5 | 5 | 50 |
| 10–20 | 8 | 15 | 13 | 45 |
| 20–30 | 15 | 25 | 28 | 37 |
| 30–40 | 12 | 35 | 40 | 22 |
| 40–50 | 10 | 45 | 50 | 10 |
- Class limits
- Smallest and largest values of a class — 10 and 19 in inclusive classes
- Class interval (width)
- Upper limit − lower limit
- Mid-value
- (Lower limit + upper limit) ÷ 2
- Frequency
- Number of observations in a class
- Cumulative frequency
- Running total of frequencies
- Class boundaries
- Limits adjusted to remove gaps (9.5–19.5)
- Relative frequency
- Frequency ÷ total
Types of frequency distributions
| Meaning | |
|---|---|
| Discrete | Variable takes separate values — number of children |
| Continuous | Variable takes any value in a range — height, income |
| Inclusive and exclusive | Class 10–19 includes both ends; 10–20 excludes the upper limit |
| Open-ended | "Below 10" or "50 and above" |
Exam tip
To convert inclusive classes (10–19, 20–29) into exclusive ones for the median and mode formulas, subtract half the gap (0.5) from each lower limit and add it to each upper limit: 9.5–19.5, 19.5–29.5.
Number of classes
k = 1 + 3.322 log₁₀ N
Class width
h = (Largest value − Smallest value) ÷ k
Tally marks
Count observations in groups of five
Example
How many classes? For N = 50 marks ranging from 3 to 48: k = 1 + 3.322 × log₁₀ 50 = 1 + 3.322 × 1.699 = 6.64, so use 7 classes. Class width h = (48 − 3) ÷ 7 = 6.4, rounded up to a convenient 7 (or, for neat classes, 5 classes of width 10 as in the table above).
Tabulation
Table number and title
Caption
Column headings
Stub
Row headings
Body
The data
Head-note and foot-note
Source note
Diagrams and graphs of a frequency distribution
The five bars show classes 0–10, 10–20, 20–30, 30–40 and 40–50 with frequencies 5, 8, 15, 12 and 10; the tallest bar locates the mode
The ends are brought down to the axis at the mid-points of the empty classes before and after
The two ogives cross at the median — cumulative frequency N/2 = 25 at marks about 28
Choosing a diagram
| Best for | |
|---|---|
| Histogram | Continuous frequency distribution |
| Frequency polygon | Comparing two or more distributions |
| Ogive | Medians, quartiles, number below or above a value |
| Bar diagram | Comparing categories |
| Pie chart | Parts of a whole |
| Line graph | Time series |
Topic 4
Population, sample, descriptive and inferential statistics
Meaning
Entire group under study
Part of the population selected for study
Measure called
Parameter (μ, σ)
Statistic (x̄, s)
Size
N
n
Study
Census — complete enumeration
Sample survey
Example
All customers of a bank
200 customers chosen
Cost and time
High
Low
Accuracy
Complete but may suffer from non-sampling errors
Sampling error but fewer non-sampling errors
Suited to
Small populations
Large populations, destructive testing (bulbs, crash tests)
Purpose
Summarise and describe data
Draw conclusions about a population from a sample
Tools
Averages, dispersion, charts
Estimation, hypothesis testing, regression
Example
Average marks of a class
Whether average marks of all students exceed 60
Topic 5
Sampling methods
Sampling is selecting a part of the population to represent the whole.
Probability (random)
Every unit has a known chance — simple random, systematic, stratified, cluster, multistage
Non-probability
Selection by judgment or convenience — convenience, judgment, quota, snowball
Simple random
Every unit equally likely; lottery or random number table
Draw 50 roll numbers from 500
Systematic
Every k-th unit after a random start, k = N ÷ n
Every 10th customer in the register
Stratified
Divide into homogeneous strata, sample each in proportion
Students by programme
Cluster
Divide into clusters, randomly select whole clusters
Select 5 colleges out of 40 and survey all students
Multistage
Sampling in stages
State → district → village → household
Example
Systematic and stratified samples: a college has 1,000 students — 600 in B.Com, 300 in BBA and 100 in BCA — and needs a sample of 50. Systematic: k = 1,000 ÷ 50 = 20; pick a random start between 1 and 20, say 7, then take roll numbers 7, 27, 47 and so on. Stratified (proportional): B.Com 600/1,000 × 50 = 30, BBA 15, BCA 5, each chosen at random within its course — so every course is fairly represented.
Exam tip
In sampling answers, always state whether the method is probability or non-probability, give the selection rule, one example and one limitation.
Convenience
Whoever is easily available
Biased
Judgment (purposive)
Researcher chooses typical units
Subjective
Quota
Fixed numbers from each group, chosen non-randomly
Selection bias
Snowball
Respondents refer others
Not representative
Selection
Random, known chance
Judgment or convenience
Sampling error
Can be measured
Cannot be measured
Generalisation
Valid
Doubtful
Cost
Higher
Lower
Cause
Only a part is studied
Faulty data collection, non-response, wrong recording
Reduction
Larger samples
Careful design, training and checking
Occurs in
Sample surveys only
Census and sample surveys
- Larger sample
- Smaller sampling error
- Variability
- More variable population needs a bigger sample
- Confidence and precision
- Higher confidence needs larger n
- Cost and time
- Limit the size
- Formula idea
- n = (Z × σ ÷ E)²
Key terms
- Statistics
- Science of collecting, presenting, analysing and interpreting data
- Primary data
- Data collected first-hand for the purpose
- Secondary data
- Data already collected by others
- Frequency distribution
- Table showing the number of observations in each class
- Histogram
- Bar diagram for a continuous frequency distribution
- Ogive
- Cumulative frequency curve
- Population
- The whole group under study
- Sample
- A part of the population selected for study
- Parameter and statistic
- Measures of a population and of a sample
- Stratified sampling
- Random sampling within homogeneous groups
Quick revision
- Statistics: aggregates, numerical, many causes, systematic, comparable.
- Primary data (first-hand) vs secondary data (existing sources).
- Frequency distribution: class interval, mid-value, cumulative frequency; histogram, polygon, ogive.
- Parameter (population) vs statistic (sample).
- Probability sampling: simple random, systematic, stratified, cluster, multistage; non-probability: convenience, judgment, quota, snowball.
- Sampling errors can be measured; non-sampling errors arise from poor procedure.
Important exam questions
Practice questions written to the PTU exam pattern for this unit's syllabus: short answers (Section A style) and long answers (Sections B and C style).
Short-answer questions
- Q1.Define statistics and state two features.
- Q2.Name two limitations of statistics.
- Q3.Differentiate between primary and secondary data.
- Q4.Name four methods of collecting primary data.
- Q5.What is a class interval and a mid-value?
- Q6.What is an ogive?
- Q7.Differentiate between population and sample.
- Q8.Differentiate between a parameter and a statistic.
- Q9.What is stratified sampling?
- Q10.Differentiate between probability and non-probability sampling.
- Q11.What is a sampling error?
Long-answer questions
- Q1.Explain the meaning, importance, scope and limitations of statistics.
- Q2.Explain the methods of collecting primary data and the sources of secondary data.
- Q3.Construct a frequency distribution and draw a histogram, frequency polygon and ogives for the given data.
- Q4.Distinguish between descriptive and inferential statistics with examples.
- Q5.Explain the probability and non-probability methods of sampling.
Stuck on this unit?
Message SBS on WhatsApp for help with Business Statistics, or to ask about studying B.Com at Synetic.
