Compare parametric and non-parametric statistics:																				
																				
NON-PARAMETRIC:																				
			Q-test:	H0 = There is no outlier in the data set.							Confidence interval:									
Data set	sorted	Q		H1 = There is one outlier in the data set.								Data set sorted								
30.1	30.1	0.351										30.1		N=	8					
32.56	32.56			alpha=	0.05							32.56			"<30,1;35,14>"					
33.33	33.33			range=	7							33.33								
34.1	34.1			N=	8							34.1								
34.45	34.45			Qcrit=	0.526							34.45								
35.12	35.12											35.12								
35.14	35.14			"The calculated value 0,351 is smaller than the critical value 0,526."								35.14								
37.1	37.1	0.280		"Therefore, the null hypothesis is accepted, no  outlier present."								37.1								
																				
																				
																				
																				
																				
																				
																				
																				
																				
																				
																				
																				
																				
PARAMETRIC:																				
		Grubb´s test:	H0 = There is no outlier in the data set.																	
Data set			H1 = There is one outlier in the data set.																	
30.1												Confidence interval -  Student:								
32.56			Max value=	37.1		T (max)=	1.500787589													
33.33			Min value=	30.1		T (min)=	1.874477671						Data set							
34.1			Mean=	33.99		critical=	2.1266						30.1							
34.45			St. dev.=	2.073911074									32.56			N=	8		s.e.m.=	0.73
35.12						"Both calculated values are lower than the critical value 2,1266."							33.33			Mean=	33.99		L1=	32.25
35.14						"Therefore, the null hypothesis is accepted, there is no outlier."							34.1			St. dev.=	2.07		L2=	35.72
37.1													34.45							
													35.12			t=	2.36		"<32,25;35,72>"	
													35.14			alpha=	0.05			
													37.1							
																				
CONCLUSION:																				
Non-parametric tests are more suitable as there is a small sample of data and it is not normally distributed.