Further questions about the 'Prove It' poll.

Mike Post made this Freedom of Information request to Science Museum Group

This request has been closed to new correspondence. Contact us if you think it should be reopened.

The request was partially successful.

Dear Sir or Madam,

On 23rd November 2009 I submitted a request to the Science Museum summarised as: “Questions about the 'Prove It' poll”. For your convenience I include my original enquiry below.

The poll has now closed. Could you please tell me:

The precise time and date that the poll closed?
The ‘in’ and ‘out’ counts at that moment on the website?
The ‘in’ and ‘out’ counts at that moment from the exhibition?
The precise adjustments that were made to the website scores after the closure of the poll?
The reason for the adjustments?
The precise adjustments that were made to the exhibition scores after the closure of the poll?
The reason for the adjustments?
To whom, if anyone, the Museum submitted the final scores and what was the response or were the responses?

My original enquiry was:

My request is in connection with the Science Museum's 'Prove It'
poll. Can you please tell me:

When the Museum decided to run the poll?
Who was responsible for providing the ‘evidence’?
When it was that it was realised that the pollsters had failed to
implement the email confirmation security measure?
Why on 12 November 2009, between 10.00 and 12.00 UTC, approximately
1500 votes were deducted from the ‘out’ score?
Why on 12 November 2009 between 14.00 and 16.00 UTC, approximately
1500 votes were added to the ‘in’ vote and approximately 2000 votes
were added to the ‘out’ vote?
What instructions were given to Museum staff regarding the poll?
What information the Museum circulated to schools and other
educational institutions about the poll?
When the Museum intends to complete the poll?
What use the Museum intends to make of the poll?
How much the poll has cost?

Thank you in anticipation.

Yours faithfully,

Mike Post

Freedom of Information \(SCM\),

Thank you for your enquiry.

We will reply as soon as possible. There may a delay in your receiving a
response, depending on the volume of enquiries we have received. However,
a response will definitely be forwarded to you within 20 working days.

Kind regards,

National Museum of Science and Industry

For updates on all Science Museum news and events sign up to our free
e-newsletter at www.sciencemuseum.org.uk

This e-mail and attachments are intended for the named addressee only and
are confidential. If you have received this e-mail in error please notify
the sender immediately, delete the message from your computer system and
destroy any copies. Any views expressed in this message are those of the
individual sender and may not reflect the views of the National Museum of
Science & Industry. This email has been scanned for all viruses by the
MessageLabs Email Security System.

Dear Mike Post,

Thank you for your emails dated 23rd November 2009 and 3rd December 2009
with regards to the Science Museum's Prove It Poll. I have forwarded
your enquiries onto the relevant Museum Section; someone from that team
should be in contact with you in due course to further discuss this
matter.

Please note as the Museum received two separate requests for information
on differing dates the following Freedom of Information Act 2000
deadlines apply:

Enquiry dated 23rd November 2009 (received after 5pm and therefore
counts as being received by the Museum on the next working day, the 24th
November 2009) - Deadline: 21st December 2009

Enquiry dated 3rd December 2009 (Please note public holidays over the
Christmas period do not count as a working day) - Deadline 4th January
2010

I trust this information is of some use to you and I wish you all the
best with your research.

Rory Cook,
Documentation Centre,
Science Museum,
London, SW7 2DD

show quoted sections

Carroll Victoria,

Dear Mr Post

Thank you for your enquiries. Below are answers to the questions raised in
your emails of 23.11.09 and 03.12.09.

In response to your email of 23.11.09:

· When the Museum decided to run the poll?

The Science Museum decided to develop an exhibit and website about the UN
Climate Change Conference in Copenhagen in July 2009. The aims of the
project were to raise awareness of the Copenhagen conference, to enable
visitors to explore the evidence for climate change and the significance
of Copenhagen; and to provide a platform for visitors to express their
views.

During the development of the project it was agreed that one component of
the experience would be an opportunity for visitors to decide what they
felt Copenhagen should deliver and then express their views to the UK
government ahead of the Conference. The final format of this component -
which invited visitors to be `counted in' or `counted out' to the
statement: "I've seen the evidence. And I want the government to prove
they're serious about climate change by negotiating a strong, effective,
fair deal at Copenhagen," was agreed 6th October 2009.

· Who was responsible for providing the `evidence'?

All of the content for PROVE IT! was researched and written by the Science
Museum's Exhibitions team, in consultation with experts from Oxford
University, the Met Office's Hadley Centre, London School of Economics,
Overseas Development Institute, National Oceanography Centre, Open
University, Imperial College and University of York.

· When it was that it was realised that the pollsters had failed
to implement the email confirmation security measure?

The email validation measure was implemented, but when moved into the live
environment suffered a conflict with our wider email security systems. We
got the first indications of this on Thursday 22nd October but its full
implications only became apparent on Monday 26th. As a result of this we
removed a number of duplicate votes from both the "count me in" and "count
me out" totals from our database retrospectively.

· Why on 12 November 2009, between 10.00 and 12.00 UTC,
approximately 1500 votes were deducted from the `out' score?

· Why on 12 November 2009 between 14.00 and 16.00 UTC,
approximately 1500 votes were added to the `in' vote and approximately
2000 votes were added to the `out' vote?

A database synchronisation issue occurred on Thursday 12 November. This
was spotted by the Science Museum's web team and the problem was rectified
on the same day.

· What instructions were given to Museum staff regarding the
poll?

No instructions specifically regarding the poll were given to Museum
staff. As for all new exhibitions, a briefing sheet was issued to
customer-facing staff. The sections relating to the poll read as follows:

"The action stations on gallery invite visitors to decide whether or not
they would like to support the statement: "I want the government to prove
they're serious about climate change by negotiating a strong, effective,
fair deal at Copenhagen". We will pass on the results of the poll to the
government before the conference. The aim of the poll is to encourage
visitors to reflect on the content and decide where they stand, and to
enable them to make their view count.

The web site offers users the opportunity to participate in this poll, to
explore the content in depth, and also to send messages to friends and
family via email, Twitter and Facebook about climate change."

· What information the Museum circulated to schools and other
educational institutions about the poll?

PROVE IT! was included twice in the e-newsletter which the Museum sends to
schools on a monthly basis.

10 October text

"Explore the evidence that human activity is behind climate change and
share your views. This small new exhibition is suitable for secondary and
post-16 students."

II November text

"Visit PROVE IT! our exciting new exhibition and website, to find out more
about climate change. Why it's happening. Why time is running out. What
the world is planning to do about it. "

· When the Museum intends to complete the poll?

The poll closed on the morning of 1 December 2009

· What use the Museum intends to make of the poll?

The results were announced to the Department of Energy and Climate Change
on 1 December 2009.

· How much the poll has cost?

It is not possible to say exactly how much the poll component of PROVE IT!
cost, since it was developed as an integral part of a larger website and
exhibition. The following is therefore an estimate:

Total spend on website: £10,000 (including staff time)

Components of website included:

· Poll

· Deep content, including evidence for climate change and
information about Copenhagen

· Make your own message to send by email, Twitter, Facebook

· About prove it, including send a comment

Estimated spend on poll component of website: £3000 (including staff
time)

Total spend on new media development on gallery: £15,000

This included:

· New interactive interface for deep content station projections

· Voting stations, with additional `find out more' and `comment'
sections

Estimated spend on development of voting station software: £5000

This does not include the hardware (computers and screens) since they were
bought for use in a future exhibition.

Total estimated spend on poll: £8000.

In answer to your further questions submitted 03.12.09:

· The precise time and date that the poll closed?

Tuesday 1 December at approximately 10.45.

· The `in' and `out' counts at that moment on the website?

The in and out counts which were displayed on the public site were noted
at approximate time of closure (10.45am) as 7532 in, 8989 out. These
figures included both web votes and gallery votes. Once the poll was
closed the database was scrutinised and some duplicate votes were
identified. These were removed from the totals.

· The `in' and `out' counts at that moment from the exhibition?

The in and out counts from the exhibition were included in the above
totals.

· The precise adjustments that were made to the website scores
after the closure of the poll?

· The reason for the adjustments?

· The precise adjustments that were made to the exhibition scores
after the closure of the poll?

· The reason for the adjustments?

Once the poll was closed the database was scrutinised and some duplicate
votes were identified. These were removed from the totals. The final
results once further duplicates had been removed from the database were:

Gallery

Counted in = 3408

Counted out = 626

Web

Counted in = 2650

Counted out = 7612

Total

Counted in = 6058

Counted out = 8238

· To whom, if anyone, the Museum submitted the final scores and
what was the response or were the responses?

The results were forwarded to the Department of Energy and Climate Change
(DECC). No comment on the results was received. For a response to the
results I suggest you contact DECC directly.

Kind regards

Dr Vicky Carroll

Copenhagen Project Leader

Science Museum

Exhibition Road

London SW7 2DD

show quoted sections

Dear Carroll Victoria,

Thank you for your reply.

Can you please explain the nature of the database synchronisation issue which occurred on Thursday 12 November when approximately 1500 votes were added to the 'in' vote.

Is there any connection between the approximately 1500 votes added to the 'in' vote on 12 November and the 1474 subtracted from the 'in' vote after poll closure on 1 December?

Does the Museum accept that the running 'in' vote score was falsely inflated from 12 November until poll closure?

Can you please tell me what steps the Museum is taking to re-establish its reputation for competence in these matters?

Thank you in anticipation.

Yours sincerely,

Mike Post

Carroll Victoria,

Dear Mike Post

Thank you for your message. In response to your further enquiries:

Can you please explain the nature of the database synchronisation
issue which occurred on Thursday 12 November when approximately
1500 votes were added to the 'in' vote.

The count for Prove It collected data from several different databases,
reflecting web counts and counts from several different terminals in the
Prove It gallery. Due to a technical error, this count was briefly
configured to harvest data from a test database rather than the correct
databases. In fixing this error we had to resynchronise the master count
to all of the correct databases.

Is there any connection between the approximately 1500 votes added
to the 'in' vote on 12 November and the 1474 subtracted from the
'in' vote after poll closure on 1 December?

It is not possible to trace each individual "count me in" or "count me
out" through the resynchronisation process.

Does the Museum accept that the running 'in' vote score was falsely
inflated from 12 November until poll closure?

Once the poll was closed the database was scrutinised and some duplicate
votes were identified in both the "in" and "out" counts. These were
removed from the final totals.

Can you please tell me what steps the Museum is taking to
re-establish its reputation for competence in these matters?

The Science Museum is committed to continuing to engage the public with
the science of climate change. For this reason we are planning a major new
climate change gallery to launch later this year. There will also be an
associated website, outreach activities and events delivered over a three
year programme.

Kind regards

Vicky Carroll

Science Museum

show quoted sections

Dear Carroll Victoria,

You have not answered my questions. Please answer my very serious questions. Public money is involved.

I am requesting an internal review.

Yours sincerely,

Mike Post

Yates Emma,

Dear Mr Post

Thank you for your email dated 14 January 2010, which was addressed to my
colleague, Victoria Carroll, requesting an internal review of the response
(dated 14 January 2010) to your request (dated 15 December 2009) which was
made under the Freedom of Information Act 2000 ("Act").

I am the Corporate Information and Enquiries Manager for NMSI and I have
carried out a thorough review of this matter ("Review") in accordance with
the guidance set out in part IV Secretary of State for Constitutional
Affairs' Code of Practice on the discharge of public authorities'
functions under Part I of the Act ("Code of Practice").

I am sorry that you are dissatisfied with the response which was provided
to you and I trust that this email will provide you with the appropriate
information to bring this matter to a close. Wherever possible we
endeavour to provide full and frank replies to requests for information
under the Act and recognise the importance of doing so in order to comply
with our obligations and to provide full access to our collection and
collection information to its owners, the public.

Upon reviewing all of the correspondence relating to your enquiry I feel
that, whilst we have not deliberately attempted to withhold any
information from you, we have not fully responded to some of the questions
contained in your email of 15 December 2009. Accordingly, please find
below our revised responses:

1. Can you please explain the nature of the database synchronisation
issue which occurred on Thursday 12 November when approximately
1500 votes were added to the 'in' vote?

The count for Prove It collected data from several different
databases, reflecting web counts and counts from several different
terminals in the Prove It gallery. Due to a technical error, caused
by a database configuration setting mistakenly being moved onto the
live webserver as part of wider content publishing, this count was
briefly configured to harvest data from a test database rather than
the correct databases. In fixing this error we had to resynchronise
the master count to all of the correct databases to reflect the
total overall number of votes cast.

2. Is there any connection between the approximately 1500 votes
added to the 'in' vote on 12 November and the 1474 subtracted from the
'in' vote after poll closure on 1 December?

As it is not possible to trace each individual "count me in" or "count
me out" vote through the resynchronisation process we are unable to
provide you with a conclusive answer to this question without
speculating. We do feel, however, that in making these adjustments we
endeavoured to accurately reflect the votes cast in the face of
technical difficulties in order to ensure that the public were not
mislead as to the outcome.

3. Does the Museum accept that the running 'in' vote score was
falsely inflated from 12 November until poll closure?

We accept that there were periods of time when the running totals of
both the `in' and `out' votes were inflated by the inclusion of
duplicate votes. However, this was not through any attempt on the
part of the Science Museum to misrepresent the votes cast. Once the
poll was closed the database was thoroughly scrutinised and all
duplicate votes were identified. These were removed from the final
totals in order to more fairly represent the voting. You will note
that both before and after the duplicates were removed there were a
greater number of votes cast in the `out' category.

4. Can you please tell me what steps the Museum is taking to
re-establish its reputation for competence in these matters?

We accept that errors occurred in the technical management of the
vote collection in respect of the Prove It poll, however, we do
feel that, in this matter, we acted competently and with the desire
to ensure that the opinions that were expressed were represented
fairly, and to ensure that the public were not mislead by quickly
identifying technical issues and working to correct them at the
earliest available opportunity. The Prove It poll, like all web
polls, could not represent a statistically representative sample of
public opinion, as all web audiences are to a degree
self-selecting. As with all of our exhibitions, we have taken on
board the difficulties which we faced with this exhibition in the
gathering of votes as part of the Prove It poll and will learn from
these experiences in future exhibitions in order to ensure a high
standard for our visitors. In addition, the Science Museum is
committed to continuing to engage the public with the science of
climate change. For this reason we are planning a major new climate
change gallery to launch later this year. There will also be an
associated website, outreach activities and events delivered over a
three year programme.

I trust that my Review has been acceptable to you and that this
information will be useful to you in completing your research.

Kind regards

Emma Yates

Corporate Information and Enquiries Manager

NMSI

The members of the NMSI family include:

Science Museum, National Media Museum, National Railway Museum, Locomotion

National Media Museum

Bradford

West Yorkshire

BD1 1NQ

e. [1][email address]

If you feel that your enquiry has not been handled effectively, please
contact the Science Museum's Freedom of Information Team at the address
below to request a full internal audit.

Freedom of Information Complaint

Records Manager

Science Museum

Exhibition Road

London SW7 2DD

Email: [2][email address]

Alternatively, you may also wish to contact the Information Commissions
Office on the following:

The Information Commissioner

Wycliffe House

Water Lane

Wilmslow

Cheshire SK9 5AF

Tel: 01625 545700

Fax: 01625 545510

<< OLE Object: Picture (Metafile) >>

From: Mike Post [[3]mailto:[FOI #24207 email]]
Sent: Thu 14/01/2010 10:55 AM
To: Carroll Victoria
Subject: Internal review of Freedom of Information request - Further
questions about the 'Prove It' poll.

Dear Carroll Victoria,

You have not answered my questions. Please answer my very serious
questions. Public money is involved.

I am requesting an internal review.

Yours sincerely,

Mike Post

show quoted sections