Telling a Story: Turning Numbers Into Information and Knowledge Jon Hall, PARIS21.

Telling a Story: Turning Numbers Into Information and Knowledge

Jon Hall, PARIS21

NSOs: Data Disseminators or Knowledge Builders?

People’s Knowledge of Official Measures of Economic Performance What U.S. Consumers Know About Economic Condition

Richard Curtin Research Professor and Director Survey of Consumers - University of Michigan

Statistics are worthless unless they are used!

Boring

Not relevant to people’s lives

Statistics are often perceived as

The last 2 cm is the most important

PUBLIC DATA DISSEMINATION SCENARIO

DATADATANational Statistical SystemNational Statistical System

MEDIAMEDIAInternet, CD, Books, TV, Newspapers, Presentations

INTERPRETERSINTERPRETERS

AUDIENCEAUDIENCEPublic, Policymakers, Students

Journalists, Activists, Researchers, Teachers

Source: Gapminder

DATADATA

National Statistical SystemNational Statistical System

MEDIAMEDIA

AUDIENCEAUDIENCE

INTERPRETERSINTERPRETERS

BUT THIS IS NOT THE CASE...BECAUSE...

BORINGBORING

DIFFICULTDIFFICULT

EXPENSIVEEXPENSIVE

Source: Gapminder

10

5 7445 7 9 112112 8 347 779 6,3

8 5656 2006=102006=10

135,4 9

45 + 7 12%

5 7445 7 9 112112

7687 8 2 3477 779 6,3

8 5656 2006=102006=10

135,4 9 45

+ 7 12%

Methods/calculations

Data entry & editing

Data collection/surveys/adm. registers

(Master plan/strategy/MDGs/PRSPs, etc.)

Statistics

Data

Tabulation

NSOs = Numerical Volcanoes?

11

Methods/calculations

Data entry & editing

Data collection/surveys/adm. registers

(Master plan/strategy/MDGs/PRSPs/etc.)

User-friendly presentation

Tabulation

12

• Find the story

• Tell the story

• Spread the story

13

Find the interesting stories

• Don’t be afraid to make statistics interesting – they are!

• But how to find the interesting story?

14

Requires analysis…

• In a complex society, readers need to be

guided through the numerical jungle: “What do

the numbers really mean”?

• Informed users and the general public need

explanations, interpretation and comment

15

Analysis also

• Forces us to look closer at the data: concepts,

definitions, measurements, sampling, etc.

• … and provides feedback into the statistical

production process, increasing the quality of

statistics, by uncovering errors and mistakes

16

Analysis means …

• Selecting among all the possible numbers

What is… – Important?– Interesting?– New…?

• Making the figures comparable• And comparing: pointing out differences, trends and

tendencies – Over time– Between groups

17

Finding the Story• Pick an indicator and ask ….

• What is the level and the rate of change for the nation?

• How do different subgroups/regions etc differ?

• What is driving the change?

• How much change affect other aspects of life?

• How does this compare with other countries?

18

Telling the Story

• Short Simple Communication – Use Plain Language

19

General motto: KISS!

• Keep It Short and Simple!

• This applies to:– Tables

– Graphs

– Titles

– Text

20

General motto: KISS!

User-friendly presentationof statisticsGuide to creating a dissemination strategy and disseminationguidelines for developing and transition countries

On www.paris21.org

21

User-friendly?

From: 2004 Census, Qatar

22

Literacy rate. Adult men and women

  Men Women

1935 30.8 8.0

1945 44.3 13.5

1950 47.7 16.7

1955 56.3 21.3

1960 54.8 21.1

1965 64.7 27.6

etc. : :

From: Statistical Indicators 1923-2004, Turkstat

User-friendly?

23

General motto: KISS!

Men & Women not Males & Females

People not Persons

24

General motto: KISS!

A female born in 2010 has a life

expectancy at birth of 77.3 years

Or

In 2010 a baby girl can expect to live to

be more than 77.

25

Telling The Story

• Use new ICT tools and other techniques

Telling The Story

Ask 4 questions:

• What do we have/use traditionally to present our statistical data?

• Are all these options effectively helping us to reach our different audiences?

• What are the new possibilities in this area?• Are they better suited for our targets and will they

help improve communication on statistics?

Gapmider

27Factbook on Gapminder

Fusion Charts

You Tube

30

Spreading the Story

Web 2.0 – Information Explosion• Search engines• Government and NGO websites• Communication tools

– youtube for video, • Interactive documents • Discussion forums on data

– Many eyes, swivel• Wikis • Mash-ups

– E.g. google maps plus data on [war, water, mobility…] from 3G phones

• Information and misinformation explosion

32

Traditional dissemination model

Media

The public

Libraries

Education

Government/org.

Business

NSI

The role of the media

33

“New” dissemination model

The public

Libraries

Education

Government/org.

Business

NSI

Media

34

The media are of vital importance because:

• They represent an important channel of dissemination

• They help increase the visibility and public recognition of the NSO– And can improve the publics trust in statistics

35

Statistician and journalist: Like cat and dog?

36

We must cooperate

Journalists are our best friends …

37

Two basic rules when communicating with the media:

• Equal treatment– No pre-release access

• Establish a release calendar

38

In many NSOs dissemination is still limited, because:

• Focus on data collection and processing

• Lack of experience (and coordination)

• ”Fear of dissemination”?

• Lack of dissemination strategy and guidelines

“Statistics are people with the tears washed away”

[email protected]

—Victor Sidel