data standards and governance

Question 1

data standards

Answer 1

Data consistency is essential for everyday situations:
* Choosing the right battery (AA or AAA)
* Putting the right fuel in a car (diesel or unleaded)
* Ensuring that dates are in the correct format (i.e. DD/MM/YY or MM/DD/YY)
* If different groups have different standards, data exchange is difficult
* Therefore it is important organizations document how they organize
data

Question 2

data stewardship

Answer 2

Stewardship is the careful, responsible management of something entrusted to
one’s care on behalf of others (DAMA Dictionary of Data Management, 2011)
* The stewardship of data is the responsibility of groups/organisations who hold
onto the data
* There should be clearly designated people responsible for the formulation
and enforcement on how data is stored an accessed
* There are also legal responsibilities (i.e. General Data Protection Regulation
[GDPR]) more on this a bit later!
* With all of these factors in mind, a data governances policy is usually a good
idea

Question 3

GDPR

Answer 3

The Data Protection Act 2018 is the UK’s implementation of the GDPR (General
Data Protection Regulation)
* In a nutshell, the GDPR states:
* You should aim to collect only the minimal amount of data needed
* You only collect data that is relevant
* You take steps to protect the data and report breaches
* You retain the data for the shortest time possible
* Relevant people can request access to their data and can request their data is deleted
Anonymous data does not fall under GDPR regulation as it’s not linked to a
person
* So if we asked you if you have enjoyed this module and only collected your views on the
module, this is not covered by GDPR
* However pseudonymous data does!
* This includes taking some details like just the postcode or student number
* Could be reverse-engineered to obtain the original person’s identity, e.g. by combining with
other data sources
* Example: If we asked if you enjoyed this module and also noted your student numbers, we
could combine these with data from NESS to find out who each student was

Question 3

data governance

Answer 3

In a nutshell, data governance defines how an organization collects,
stores, processes and uses data
* It should comply with relevant laws
* It should also cover the entire lifecycle of data
* The four “pillars” of data governance are: [1]
* Data stewardship
* Data quality
* Data management
* Use cases

Question 4

data value and cost

Answer 4

Every piece of data produced has a value and associated cost
* The value of data is dependent on the relevance of the data and/or
it’s potential to make a company money
* The cost is associated with storing and retrieving the data and
ensuring it is appropriately safeguarded
* Small amounts of data is usually straightforward to manage, but for
companies, retaining every piece of data collected is not feasible
* Therefore, companies should develop data governance mechanisms

Question 5

data quality

Answer 5

In it’s raw form, data may be of low quality due to lack of context, clarity or
spurious information
* Therefore data is frequently processed or “cleaned up” based on how it
will be used. Examples of processing data include:
* Classification and coding data based on themes (thematic content analysis)
* Merging of different data sources
* Recording relevant metadata (such as when the data was collected and by whom)
* Ideally, any changes to data should be recorded and original copies
retained (data versioning)
* The overall aim should be to ensure that data remains accurate, complete
and consistent

Question 6

information life cycle

Answer 6

Tier 1 represents data at its highest value and use. It should be processed and interpreted to
maximize its value
* Tier 2 represents “brand new” unprocessed data or older data that may not be as relevant. Both
factors lower the value of the data
* Tier 3 represents old data that is to be archived, it is unlikely to be useful though this cannot be
guaranteed
Time

Question 7

data management

Answer 7

Depending on the data lifecycle, data should be available to for use at the right time and in the
right way
* I.e. recent and/or useful data should be easy to access
* Older or less useful data should be stored for possible future use
* Metadata should be used where appropriate to support the data

Question 8

costs and use cases

Answer 8

cost is an important factor due to:
* The need to provide some types of data quickly
* The cost of storing data
* Hardware and software costs
* Staff costs
* Legal costs
* Hoarding data is not ideal, therefore when creating a data
governance policy, the authors should consider how and why data is
used.

Question 9

data governance practices

Answer 9

An organisation’s data governance policy can be defined in multiple
ways:
* A single “master” document
* Spread across multiple groups depending on how they use data
* Incorporated into staff training
* Formed as part of standard working practices (i.e. the use of data
management software)
* The main thing is that people are aware of their obligations