Royal Wedding, William and Catherine
Wedding of Prince William, Duke of Cambridge, and Catherine Middleton
From Wikipedia, the free encyclopedia
On Friday, 29 April 2011, Prince William, Duke of Cambridge married Catherine Middleton at Westminster Abbey, London at 11:00 am BST (UTC+1). William, who is second in the line of succession to Elizabeth II, first met Middleton in 2001, while both were students at the University of St Andrews. Their engagement, which began on 20 October 2010, was announced on 16 November 2010.
The build up to the wedding and the occasion itself attracted much media attention, with the service broadcast live around the world, and with it being compared and contrasted in many ways to the last high profile British royal wedding in the United Kingdom, the 1981 marriage of William's parents, Prince Charles and Lady Diana Spencer. It was watched by a global television audience of over 2 billion. Much of the attention focussed on Middleton's status as a commoner (ie. not of royal blood or a part of the aristocracy) and such marrying into royalty. Hours before the service, William was granted the titles of the Duke of Cambridge, Earl of Strathearn, and Baron Carrickfergus by the Queen.
As William was not the immediate heir to the throne, the wedding was not a full state occasion, with many details left to the couple themselves to decide, such as much of the the guest list of about 1,900. It was, however, a public holiday in Britain and featured many ceremonial aspects, including use of the state carriages and roles for the Foot Guards and Household Cavalry. It was attended by most of the Royal Family, as well as many foreign royals, diplomats, and the couple's chosen personal guests.
Middleton wore a white dress with a 270 centimetres (110 in) train, by English designer Sarah Burton, as well as a tiara lent to her by the Queen. William, commissioned as a Royal Navy Lieutenant, Army Captain and Air Force Flight Lieutenant, chose instead to wear the uniform of his honorary rank of Colonel of the Regiment for the Irish Guards. William's best man was his brother, Prince Harry, while the bride's sister, Pippa, acted as her maid of honour. The Dean of Westminster officiated most of the service, with the Archbishop of Canterbury conducting the marriage ceremony itself and the Bishop of London giving the sermon. A reading was also given by the bride's brother, James. After the ceremony, the newly married couple travelled in procession to Buckingham Palace for the traditional appearance on the balcony and a fly past before crowds assembled in The Mall.
The GCP event was set to begin at 11:00 and continue for 6 hours. The wedding vows were taken about half an hour later, and the ceremony was about an hour. The wedding party went to Buckingham palace, where the newly wed couple appeared at about 1:20 to 1:30 on the balcony to "seal" the marriage with a kiss. The result is Chisquare 21578.136 on 21600 df, for p = 0.541 and Z = -0.102.
I had a useful exchange with Michael Duggan on how to more completely codify the specification of "global events", in which he suggested that categories of events should have set durations for all cases. He gave some examples, like +6 hours for earthquakes and other natural disasters, +4 for terrorist attacks, +2 for speeches and important addresses, etc. I challenged him to provide a time frame for the Royal Wedding, which he answered with the surprising suggestion of 30 hours. I had already done the formal analysis, but present here the result for Michael's prediction as kind of contextual background -- and an object lesson in the challenges of setting fixed time periods to accommodate often unique global events. It starts 5 hours earlier and continues for 19 hours after the formal period (5 to 11 on the X-axis here). It is, as noted in the caveat at the bottom of the page, not appropriate to interpret single event outcomes, but the only notable aspect here is the lengthy period of positive deviation in the latter half of the graph, which starts some 12 hours after the wedding, long after the peak of global attention.
It is important to keep in mind that we have only a tiny statistical effect, so that it is always hard to distinguish signal from noise. This means that every "success" might be largely driven by chance, and every "null" might include a real signal overwhelmed by noise. In the long run, a real effect can be identified only by patiently accumulating replications of similar analyses.