CHAPTER 



15 



Statistical Treatments 



A formal mathematical analysis of the results of experimental research 

 is almost a necessity if the data are to be interpreted in terms of some 

 hypothesis. This analysis is needed because of the natural variation in 

 the numbers obtained from all measurements. Most hypotheses cannot 

 be answered with an unquestionable "yes" or "no"; the variability of 

 the results makes us say "probably." The formal system known as statis- 

 tics, or more specifically as biometry, allows us to say "probably" with 

 some estimate of how closely this "probably" approaches a definite "yes" 



« » 

 or no. 



Biometry is now a method for the analysis of the quantitative data 

 resulting from biological research. Although it is a branch of the division 

 of mathematics known as statistics, biometry includes and uses almost all 

 of statistics. "Statistics" originally were numbers, particularly numbers 

 applied to populations of people. Thus "vital statistics" include such fig- 

 ures as the number of births, deaths, marriages, the incidence of diseases, 

 and the number of persons employed by the automobile industry. Mod- 

 ern statistics includes a great deal more than measurements or records 

 concerning people. Even the grammatical usage of the word has changed. 

 When "statistics" is construed as a plural noun, each statistic is a number 

 or bit of information. "Statistics" used as a singular noun, however, refers 

 to a field of mathematics. Usually we can judge the meaning from the 

 context. 



In addition to their usefulness in analyzing data, biometric principles 

 are useful as a guide in planning experiments. If we know in advance 

 what form of statistical analysis will be required, we can be sure to take 



199 



