Prefatory remarks

The medical statistician is someone who looks up the number of patients you need in a table and comes up with the answer too many . Then, when the trial is over, he presses a button on his computer and produces a report that nobody can understand. 

Statistical Issues in Drug Development, 2nd Edition. Stephen Senn 2007 John Wiley Sons, Ltd. ISBN 978-0-470-01877-4 

What makes a good experiment Is it one which has been precisely controlled and for which the protocol is full of phrases like the volunteer will eat a standard breakfast consisting of two croissants, one cup of coffee, a pat of butter and 30 g of jam Is an even better experiment one which specifies the type of fruit to be used in making the jam, whether the croissants are to have been made with butter or not, and how many spoons of sugar (if any) are allowed in the coffee  

One may put it more strongly than this. Just as with a chemical analysis, so with a clinical trial you have to have established the way in which the results will be analysed in order to perform it properly. If you have designed a clinical trial effectively, then you must know what the analysis will be. If you know what the analysis will be, there can be no reason for you to consult the statistician for advice as to how to analyse it once the trial is run. Hence, the only sort of person who needs the statistician s help once the trial is run is the sort of person who needs his or her help when the trial is designed. Thus, the time to first consult the statistician is either before or when the trial is planned or not at all. 

To return to our discussion of experiments, a good experiment is one which may reasonably be expected to be salient. Obviously, other things being equal, a large experiment is better than a small one, and one which is too small is unlikely to be salient. Hence the belief has arisen that the major function of the statistician in designing an experiment is to use his theories to determine the sample size. In fact, sample size is only one aspect of the problem and its importance has been overrated. 

---

Of course, this idealized Southern lady never really existed in fact. More accurate was Dabney s careful prefatory remarks reminding his reader that all this elevated character was to be displayed wholly within the domestic sphere. xhe woman s place was in the... 

So we come to the last prefatory remark, which returns us to the opening. What follows is not authoritative the intention is to lay some ideas before the reader that he or she may cogitate about independently. 

It is apparent that up to a critical volume fraction Vj of free polymer, the particles displayed long-term stability (1/1F=0). Any further increase in the free polymer concentration resulted in the onset of instability, which was manifested by an increase in l/W. In all cases, however, l/W reached a maximum value and then declined. Flocculation was thus not evident at high concentrations of free polymer. It usually occurred only over a finite range of free polymer concentrations. This lack of flocculation at higher concentrations of free polymer is the phenomenon of depletion stabilization mentioned in the prefatory remarks to this chapter. It will be considered in detail in the next chapter. 

The purpose of these prefatory remarks has been to explain the underlying theoretical basis for the data that appear in the tables that follow and the experimental difficulties that are involved. Such tabulations must always be taken cautiously and critically. However, with proper care, and understanding, the data should be very useful. 

As for the second set of prefatory remarks, remarks regarding what won t be treated in what follows, this book doesn t spend much time... 

In order to obtain some evidence of the actual strength.of concrete, and to compare several varieties of compositions, the experiments contained in the following table were made at Fort Adams some prefatory remarks are necessary in relation to them. 

---