The Quality of Data in Britain Jil Matheson, National Statistician - - PowerPoint PPT Presentation

the quality of data in britain
SMART_READER_LITE
LIVE PREVIEW

The Quality of Data in Britain Jil Matheson, National Statistician - - PowerPoint PPT Presentation

The Quality of Data in Britain Jil Matheson, National Statistician 15 January 2013 A tradition of data Florence Central Statistical Domesday Book 1801 census Nightingale Office East India Statistical societies Computing age Vital


slide-1
SLIDE 1

The Quality of Data in Britain

Jil Matheson, National Statistician

15 January 2013

slide-2
SLIDE 2

A tradition of data

East India Company statistics 1801 census Domesday Book Statistical societies Central Statistical Office Computing age Florence Nightingale Vital statistics

slide-3
SLIDE 3

Why is quality important?

  • Around a thousand official statistical outputs,

covering the economy and society

  • Used in decision making and public debate
  • Influence on financial markets
slide-4
SLIDE 4

Quality in official statistics

The Board is …to promote and safeguard— (a) the quality of official statistics, (b) good practice in relation to official statistics, and (c) the comprehensiveness of official statistics. …references to the quality of any official statistics includes— (a) their impartiality, accuracy and relevance, and (b) their coherence with other official statistics. Statistics and Registration Service Act 2007

slide-5
SLIDE 5

Quality in official statistics

Principle 4: Sound methods and assured quality Quality should be monitored and assured, taking account of internationally agreed practices.

  • …a level of quality that meets users’ needs, and that

users are informed about the quality…

  • Adopt quality assurance procedures…
  • Publish quality guidelines...

Code of Practice for Official Statistics

slide-6
SLIDE 6

What is quality?

Accuracy Trust? Fit for purpose

slide-7
SLIDE 7

European Dimensions of Quality

Accuracy Trust? Fit for purpose

  • Relevance
  • Accuracy
  • Timeliness & punctuality
  • Accessibility & clarity
  • Comparability
  • Coherence
slide-8
SLIDE 8

Public Trust in Statistics

Trust? Fit for purpose

slide-9
SLIDE 9

Census: Quality since 1801

  • Quality of census figures have always been challenged
  • Improved quality and confidence in the census

through various methods…

Post- enumeration surveys Comparing with other data Technology: improve process, reduce errors Question testing and field testing Adjusting for an under count Community involvement

slide-10
SLIDE 10

The census, internationally

  • Standardised concepts
  • Harmonised methods

Growing importance of quality internationally

slide-11
SLIDE 11

Crime: Admin and Survey Data

slide-12
SLIDE 12

GDP: Quality Trade-Offs

Accuracy Timeliness

slide-13
SLIDE 13

Migration: Improving Quality

  • Questions about

quality

  • Meet user needs by

keeping pace with real world changes

  • Improvements in

timeliness, accuracy and coherence

slide-14
SLIDE 14

Importance of migration

slide-15
SLIDE 15
  • Enormous growth in

information (5 exabytes = all the words ever spoken by mankind)

  • Should statisticians make

better use of this type of data? How?

Brave new world of data

“Data is the 21st century’s new raw material”

slide-16
SLIDE 16

Big Data, Big Questions

Is big data competing with traditional statistics? How much of this information is “data”? Can statisticians use this new type of data? How? Are we the “wise

  • ld men/women” of

statistics? Or can we challenge these new kids on the block? What is the quality of this information? Can it be trusted?

slide-17
SLIDE 17
  • Quality

assurance?

  • Nowcasting?
  • An alternative

data source?

Using big data

House price inflation and estate agent searches

slide-18
SLIDE 18

Intro Details Conclusion CRUCIAL information HELPFUL information Nice to know

Communicating statistics

Changing ways of accessing data… …leads to changes in how we communicate

slide-19
SLIDE 19

Quality and Trust

  • Do people

understand what they see?

  • Do people trust what

they see?

  • How can we convey

the quality?

Access to data and statistics is more and more removed from the original source.

slide-20
SLIDE 20

What’s our brand?

Trusted brands with an expected level of quality Are statistics trusted? Do people know what quality to expect?

slide-21
SLIDE 21

Challenges for official statisticians

Budgets Outputs People

slide-22
SLIDE 22

Right Statistics Right Methods Right Impact

Rebalancing our activities…

Responding to the challenge

slide-23
SLIDE 23

What is a statistician?

Adding value Assessing quality Telling the stories Finding the messages Collaborating

slide-24
SLIDE 24

The Quality of Data in Britain Our understanding of quality keeps changing Adapt Uphold our reputation