Data Analysis Flashcards
Data definition
Numbers, letters, symbols, raw facts, events and transactions
Recorded but not yet processed into a form suitable for management use
Information
Data which has been processed
So it is meaningful to the person receiving it
Information ‘formula’
Data + Meaning
Uses of information
- Planning
- Decision making
- Controlling
When preparing for a budgeting exercise, management accountants must identify what?
Appropriate sources of information
Types of data
- Quantitative
- Qualitative
- Discrete
- Continuous
Quantitative data
Numerical data
Measurements or quantities
Can be analysed using statistical methods (risks management)
Qualitative data
Cannot be expressed as numbers/values
Harder to analyse
E.g. nationality, hair colour
Discrete data
Non-continuous data
Can only take certain values e.g. integers
Discrete data is counted
Continuous data
No gaps
Can take on any value
(within a range)
E.g. time/distance
Continuous data is measured
Types of sources of data
Internal
External
Internal data sources
E.g.
Accounting records
HR records
Payroll records
Machine logs
Computer systems
Procurement data system
Timesheets
Communication with staff
Two types of external information
Formally gathered
Informally gathered
Formally gathered data examples
Marketing research
E.g. new trends, customer tastes, competitor products
R&D
Tax and accounting specialists
E.g. new legislation/standards
Legal specialists info
E.g. changes in health and safety at work
Informally gathered data
Data gathered on an ongoing basis
E.g. newspapers, internet, meetings with external colleagues
Qualities of good information
- Accurate
No typos. roundings, categorised, assumptions - Complete
All information provided for the purpose - Cost beneficial
Benefit > cost of producing info - User-targeted
Understandable and useful to recipient - Relevant
For purpose intended - Authoritative
Genuine, highest quality for purpose, source should be knows and reliable - Timely
Produced in advance when needed - Easy to use
Clear, concise, constructive, communicated appropriately
Data analysis steps
- Identify information needs
- Collect the data
- Analyse the data
- Present the information
- Use the information
Three types of data analysis
- Inferential statistics
- Exploratory data analysis
- Confirmatory data analysis
Inferential statistics
Uses random sample of data from pop
To describe and make inferences about it
Exploratory data analysis
Pattern identified in a set of data
May be:
Regression analysis
Correlation analysis
Confirmatory data analysis
Confirms(/disconfirms) hypothesis
Using statistical methods
E.g. price increase of 3% will reduce demand by 5%
When is sampling appropriate
When possible to select units from the population
Three reasons sampling is necessary
- Whole population may not be known
- Testing whole population costly (time+money)
- Items may be destroyed in testing
2 rules of sampling
- Must be a certain size
- Must be representative
Compatibility bias
Comparing data from different sources
Data bias
Sample not representative
Selection bias
Not selected randomly
Observer bias
Observer assumptions inadvertently influence observations
Cognitive bias
(Subconscious) Perception of data by user that leads to misinterpretation of results.
Hypothesis testing
Confirming whether a hypothesis is true
Statistical significance
Results occurred due to hypothesis not chance
Type 1 error
False positive
Null hypothesis falsely rejected
Type 2 error
False negative
Null hypothesis falsely accepted
Big data
Datasets whose size is beyond the ability of the typical database software to capture, store, manage, analyse
Four Vs of big data
Volume
Variety
Velocity
Veracity
Big data Volume
Amount of data der into organisation
Do they have resources to store and manage this data?
Or have the money to upgrade IT?
Big data variety
Various formats of data received
Are their systems compatible for and capable of accessing the various forms of data?
Legally, is the data owned by the organisation or the 3P?
Big data velocity
Speed data fed into organisation
Are the systems able to capture and process real time data?
Do they have skills to analyse the data in a timely manner?
Big data veracity
The reliability of the data received
Can they challenge data received data from 3P?
Is the data received representative?
Importance of big data
Potential to achieve competitive advantage
More data sources
E.g. social media, internet of things
Exponential growth in computing power and storage capacity
New avenues of knowledge creation
E.g. crowd sourcing, open source software
Data science
Collecting, preparing, managing, analysing, interpreting and visualising large and complex datasets
Scientific approach applying mathematics, statistics and computing.
Increased demand for employees with data science skills
Data analytics
Value extracted from big data
Converting data into useful information
Benefits of big data, data science and data analytics
Significant opportunities
Abundance of data, potential to capture and harness data
- Decision making
Speed of analysis - Customer analysis
Market segmentation and customisation - Innovation
- Risk management
Risks of big data, data science and data analysis
- Storage
- Skills
- Data dependency
- Overload
- Data privacy
- Data security
Data storage challenge
Systems Must be reviewed and upgraded to cope with data
Big data skills challenge
Data scientists and analysts rare
Hard to recruit and retain
Data dependency risk
If decisions made on weak, erroneous, corrupted data.
Data privacy risk
May break legislation
