Meeting on the Guide on Statistical Storytelling



Meeting on the Guide on Statistical Storytelling

(London, 16 February 2005)

Participation:

Chairman: David Marder (ONS, UK)

Participants: Natalia Pashintseva and Andrey Maslyanenko (Rosstat), Dag Elingsen and Hege K. Fossen Pectersen (Statistics Norway), Steve Matheson (ABS), Eileen Caponi (OECD), Helena Kolackova (Czech Statistical Office), Tim Allen (Eurostat), Colleen Blessing (US Dpt. Of Energy), Vicki Crompton (Statistics Canada), Yael Nathan (CBS Israel), John Kavaliunas, and Kenenth Meyer (US Census Bureau), Marianne Zawitz (US Bureau of Justice Statistics), Frances Comerford (CSO Ireland), Don Weijers and Sebastian van den Elshout (Statistics Netherlands), Armando de la Torre and Donald Peña (INE. Spain)

UNECE: Tiina Luige and Juraj Riecan (UNECE)

1. Review of examples of statistical stories

David will digest the raw material for the good examples of statistical storytelling. This will be circulated by e-mail to the Group. The e-mail consultation will aim at selecting the most suitable ones, and on complementing them (as the examples circulated may inspire to add or replace some examples).

Examples should cover a wider variety. There should be few (or one) examples from more countries rather than many from one. They should also cover stories made for the Internet, for printed releases, for publication, journals, act.

Different target groups and purposes should be covered. Some of the stories are intended to be used as they are, while other are to enable journalists to create their own stories.

Should be there also a “bad example”? That is how not to make the story. Vicki and Steve have supported this idea. Tim offered to create one.

Three kinds of stories by their origin: (i) authored by statisticians; (ii) authored by the journalists within the organisation; (iii) combination of the two and with involvement of the outside experts.

2. Contents of the Guidelines on Statistical Storytelling

A draft guideline was prepared by Dag. He emphasised that this was a sketch, and

As a fundamental principle, suggested by Dag and Vicki and agreed by the Group, is that the storytelling will differ from country to country, based on various cultural and institutional conditions. Therefore the guide’s style should be presented as some suggested good examples, which the countries can adapt to their conditions. In summary, it should be more “principles” than “prescriptions”.

For whom the guide is? It is for the people working in dissemination, that is for the staff of NSOs (statisticians, press officers, journalists, …) who are represented at the typical meeting on dissemination.

John proposed that monitoring of the impact of the story, like media coverage, visitors to the website, interpretation in the media, etc. might be added to the guide.

Similar, but not identical, idea came from Steve: What is the impact of story on the person who reads it? Kenneth confirmed that this is important. Vicki formulated a basic question to be asked before creating the story: What this information tells to the people, what do they learn from this social or economic information?

Sebastian suggested out that the structure/sequence of the text is somehow missing in the first outline. This has to cover the issues like short sentences, short paragraphs, no difficult numbers, catchy titles, etc.

Sebastian and Don suggested treating the question of authors’ signature. This, as a recognition, might be a motivating factor. Kenneth suggested that it may be more practical to give the contact details of the person responsible for answering. Hege said that Statistics Norway puts the name of the author with communication details, and in case of absence the call is routed to the press office. Steve tried to brainstorm on whether the storyteller or subject matter specialists are more appropriate to answer the calls, provided that the storyteller should be a person understanding the data. Vicky emphasised that the story has to reflect very conservative views of the agency and not of the individual, so that the contact details should reflect on this. A Code on Storytelling could be useful thing, supported by several people in the Group. The stories originating from NSOs refer to the official statistics, and the agency standards have to be respected. These differ from country to country, as some offices give more liberties than others, but some synthetic guidelines can be put together, which would allow countries to develop their own Code.

Marianne and Yael suggested to treat the sources of misinterpretation and address them in the recommendations.

David was applauded for having volunteered to update the draft outline on the basis of the discussion. This will be circulated to the Group and agreed through the e-mail consultations.

3. Brainstorming on recommendations for statistical storytelling

3.1. Definition of a statistical story

Dag and Steve suggested that it is important to explain in short statements the journalists the background behind the information at the appropriate (not too deep) detail. Vicki warned that the guide should not slip come too much to the media relations, but should focus on statistical storytelling. Marianne draw on these short statements, and try to get statisticians to write their stories in the same/similar style. Donald, and some others, suggested to add “writing for non-specialists” to the definition.

According to Steve the statistical story is not easy to define, it is more-less intuitive understanding of a communication satisfying the tow sides (agency, recipient) in giving the information and feeding the appetite of recipients (journalists,…). Vicki and Yael suggested that it is important to define this term that it gets the appropriate weight. Perhaps storytelling should be replaced by another term, but as David and Tiina emphasised this can lead to diluted titles and suggested “story behind the numbers”, Marianne and Steve would be fore “story in the numbers”. Kenneth would prefet to “tell” rather than “write” the story. Colleen suggested to use the typical journalistic approach to the title: a short title, extended in the subtitle (e.g. “storytelling; ho to tell ….”).

3.2. Reasons for telling statistical stories

There were several suggestions including: getting credibility, communicating statistics to people who are not used to numbers, addressing policy makers and civil society in a simple language. Vicki thought that the story may also give a pointer to more detailed information to potential users, who would come back to the agency database. These users may, include policy and decision makers, but also education.

Sebastian’s suggestion was to include the aspect of orienting the information infrastructure users so that they can better understand various phenomena of the society.

Steve considered useful not only to explain what the storytelling is, but also what storytelling is not.

Colleen would make this part as a list, rather as a paragraph, in order to have the guide simple, and she gained a support from the Group.

3.3 How to…?

Marianne recommended to include something on the headline writing, wiring for the web, use of graphics and tables. Sometimes the issues the story wants to cover require multiple sources, so there should be encouragement to look beyond one single survey. Inclusion of bullets summarising the story may be also one of the recommended techniques, what may attract more interest.

Hege spoke about addressing various socio-economic phenomena at the time when they are topics (e.g. women’s day, non-smoking day, various events, etc.). Story has to fit the purpose and data have to fit the story.

Don’s suggestion was to speak about what vehicle to use for communicating the story.

Hege and Dag are trying to use the interview form for a series of articles, and can share their experience in the Guide.

Steve spoke about the “propositional complexity”, when the propositions grow, get to a very complex construction, and need to be simplified.

Vicky thought that it might be useful to include/attach a bibliography.

There was a discussion whether training, recruiting and rewarding of statisticians is within the scope of the guide. “Rewarding” depends on the importance and credits given by the management to writing. Dag thinks that “rewarding” is an important keyword. There was a discussion how this can stimulate and help in planning.

3.4 Evaluation

Yael’s office has a practice to evaluate the best, most successful, etc. releases. Media coverage and other parameters are taken into account. This may be one of the techniques to build enthusiasm.

3. Timetable and tasks for finalising the Guide

Definition – David

Reasons for telling statistical stories – Vicki & Steve

Considerations – Sebastian

How to write or tell a statistical story – Don, Hege & Colleeen

Evaluation the impact – John & Kenneth

John Flanders (Statistics Canada) will be asked to act as a final editor together with David.

It was agree to have the Guide on the Internet first (July), and than to consider a print version..

Those who volunteered for individual parts are asked to send to Juraj

draft outline by 10 March 2005

first drafts by end March 2005

Juraj will circulate to the whole group for comments.

e-mail distribution list:

david.marder@.uk

info@gks.ru (Andrey Maslyanenko)

pashinceva@gks.ru

dag.ellingsen@ssb.no

hege.pedersen@ssb.no

steve.matheson@.au

eileen.capponi@

kolackova@gw.czso.cz

timothy.allen @cec.eu.int

colleen.blessing@eia.

vicki.crompton@statcan.ca

john.flanders@statcan.ca

yael@.il (Yael Nathan)

john.c.kavaliunas@

kenneth.c.meyer@

marianne.zawitz@

erford@cso.ie

dwys@cbs.nl (Don Weijers)

belt@cbs.nl (Sebastian van den Elshout)

atrio@ine.es (Armando de la Torre)

donaldpe@ine.es (Donald Peña)

tiina.luige@

juraj.riecan@

diane.serikoff@

................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download