Sources of Data

Main topics

Data categorisations:
 Primary vs secondary
 Internal vs external
 Numerical vs categorical
 Discrete vs continuous
 Nominal vs ordinal

Big data analysis

Sampling
Primary vs secondary

Primary
 Data obtained from first hand sources; origin of data
 Surveys, observation, experimentation, interviews etc.
 Original research on the specific topic
Primary vs secondary

Secondary
 Data already obtained by third parties
 Libraries, newspapers, journals, magazines, governments, banks, internet etc.
 Research conducted on the specific topic or another topic
 Considerations: accuracy, reliable, suitability etc.
Internal vs external data
Internal sources:
 Accounting system(records): invoices, timesheets, journal entries, budgets etc.
 Department wise data: Payroll, marketing, production etc.
 Strategic planning system
Internal vs external data

External sources:
 Libraries, newspapers, journals, magazines, governments, banks, internet, customers etc.
Numerical vs categorical

Numerical:
 Expressed in numbers
 It can be discrete or continuous
 Examples: Marks, attendance, height

Categorical:
 Descriptive data rather than numeric
 It can be nominal or ordinal
 Examples: Eye colour, educational qualification, feedbacks
Discrete vs continuous

Discrete:
 non-continuous data
 include only countable data
 ex.: Class attendance, goals scored by a team

Continuous:
 unbroken data
 include uncountable data
 measured using appropriate measures
 ex.: Distance, height, weight
Nominal or ordinal

Nominal:
 Cannot be measured
 No set order or scale
 Examples: Name, eye colour

Ordinal:
 Has a set order or scale
 Examples: Feedbacks measured through ratings
Impact of general economic environment:

 General economic environment affects costs and revenues of firms at national

and international level.
 The forecast state of the economy will influence the planning process for
organisations which operate within it.
Impact of general economic environment:

Discussion points:

 GDP
 Economic trends
 Inflation
 Interest rates
 Exchange rates
 Tax levels
 Government policy
 Economic cycles: boom, bust or recession
Big data analysis

Data analytics:
The process of collecting and examining data in order to extract meaningful business
insights, which can be used to inform decision making and improve performance.

5 Vs of big data:
 Volume: large quantity of data
 Velocity: speed of generating data
 Variety: different types of data
 Veracity: reliability of data
 Value: cost-benefit analysis
Variety
Structured:
 Well organized
 Example: Government statistics, mark list of students in an excel

Unstructured:
 Unorganized data
 Example: Social media

Semi-structured:
 Mix of structured & unstructured; not totally unorganized
 Example: Sorted mailbox
Uses of big data
 Direct access to customers
 Cheaper marketing
 Quicker decision making
 Better costing and pricing
 New product design & features
Big data analysis

Pros:
 Faster
 Cheaper
 Direct access to customers

Cons:
 Unreliable - Veracity
 Lack of technical know-how
 Difficulty in understanding
Sample vs population

 What is sampling?
 When do we use sampling?

Key terms:
 Population
 Sample
 Census
 Sampling frame
Sample vs population
 If all members of a population are examined, the survey is called a census.
 If it is not possible to survey the entire population, a sample is selected.
 The results from the sample are used to estimate the results of the whole.
 A sampling frame is a numbered list of all items in a population.
Sample vs population
A sampling frame should have the following characteristics:
 Completeness
 Accuracy
 Adequacy
 Current
 Non-duplication
 Convenience
Probability sampling vs non-probability sampling

Probability sampling: sampling method in which there is a known chance of each member of
the population being selected as sample
 Random
 Stratified random
 Systematic
 Multistage
 Cluster
Probability sampling vs non-probability sampling

Non-probability sampling: sampling method in which the chance of each member of the
population appearing in the sample is not known
 Quota sampling
Random sampling

 Samples selected such that every item in the population has an equal chance of
being selected
 Sampling frame is required

Benefits and drawbacks:

+ Free from bias
+ Easy to select
‒ Unrepresentative

Quasi random sampling methods: Stratified, systematic & multistage

Stratified random sampling
 Method of sampling in which population is divided into strata or categories.
Random samples are then taken from each strata or category
 Sampling frame is required

Benefits and drawbacks:

+ Free from bias
+ Reflect structure of population; representative
+ Increased precision
‒ Laborious: prior knowledge of strata required
Systematic sampling

 Sampling method which selects every Nth item after a random start
 Sampling frame is required

Benefits and drawbacks:

+ Easy to use
+ Cheap
‒ Unrepresentative
‒ Not entirely random
Multistage sampling
 Sampling method which involves dividing the population into several sub-populations and
then selecting a small sample of these sub-populations at random
 Each sub-population is then divided further, and then a small sample is again selected at
random. This process is repeated as many times as required

Benefits and drawbacks:

+ Fewer investigators required
+ Cheap
‒ Unrepresentative
‒ Not truly random
Cluster sampling

 Sampling method that involves selecting one definable subsection of the

population as the sample
 Non-random sampling method

Benefits and drawbacks:

+ Fewer investigators required
+ Cheap
‒ Unrepresentative
‒ Not truly random
Quota sampling

 Samples selected until a certain quota is covered

 Non-probability sampling method

Benefits and drawbacks:

+ Fewer investigators required
+ No sampling frame required
+ Cheap
‒ Unrepresentative; very biased
‒ Not truly random
Thank you