We had agreed that we would meet with our University of Birmingham friend at a kind of outdoor sports centre in which people engage in canooing, kayaking and similar pursuits but in which there is a cafe overlooking what I suspect is an artifical lake. This is called the Aztec Adventure Centre and spectators are welcome – in practice, most come along just to have a coffee overlooking a local stretch of water. When we arrived, the café was not yet open and our friend had forgotten our arrangement, until prompted by a quick phone call. I took Meg along in her wheelchair, the better to negotiate the walkways to access the cafe which was not actually open at the time at which we arrived. There was quite a cold wind blwoing across the water, so after our coffee we went inside to escape some of the colder blasts. Our friend and I then engaged in a long and some might say, arcane, discussion about the role of statistics in researches presented in the typical PhD. In particular, I knew that one of my erstwhile colleagues at De Montfort University had been failed his PhD because it was argued that he had used the wrong statistical test and in this long and protracted argument he came off the loser and was failed his PhD which, I suspect rankled with him for the rest of his life particularly as he subsequently wrote a statistics primer for students. There are two very broad classes of statistics – one being called ‘parametric’ statistics where the population assumes the shape of what is known as the ‘bell shaped curve’ and the other broad class being known as ‘non-parametric’ statistics. The whole argument revolved around the extent to which could depart from the assumes of ‘normality’ for the parametric statistics still to be legitimate and to be valid or, at what point, the departure frm ‘normality’ is such that a non-parametric category of statsitics should be deployed. One can see that there an area for disagreement and debate about how far, and to what extent, a departure from ‘normality’ is to be tolerated to invalidate the use of parametric statistics. Meg very widely dozed off throughout this particular discussion but our University of Birmingham friend and myself had very similar thoughts and approaches to this problem. We both concluded that the degree of understanding of this issue could be quite small, even amongst supervisors of PhD students, let alone their supervisors and examiners, and from this discussion, we went on to share our experiences of the PhD examination process.
In the rugby World Cup, the match of the day was Australia vs. Fiji (Fiji had recently in the warmup matches beaten England and Wales only beat them narrowly in the World Cup). Fiji beat Australia for just the third time in their history as they blew Pool C wide open with a thrilling win at the Rugby World Cup. It really was an entertaining match and I think it fair to say that the Fijians seemed to outplay Australia in every department of the game. It was the first time they had ever beaten Australia in the World Cup as well and the celebrations around the stadium, and I imagine in Fiji itself, was something to behold. So after the thrilling match this afternoon, this evening it is going to be England vs. Japan. England ought to win this game but nothing is certain in this World Cup as the smaller nations are often playing well above what their world rankings might suggest. I speculate to myself whether the Japanese being such a rule-governed society will infringe the laws of rugby less than the English – we shall see.