488 



-7- 



eacb lato 4 tub-tuiplct aauhcd 4t ftr •• pottiblt by eluiccr. Th« 

 eo^ofidofl of chest 8 tub-tuiplcs It tbow la Appeodix IX. In ch« 

 pltceaeoc dttlfft (latllx tdopccd (tct App«ad{jt til) tach product vas 

 btltnced ovtr^ch Cttt (caking che 4 tub-Msplet togechtr), tad by 

 Che end of che 3t^£eic etch retpondenc htd taoked til 4 produces. Tbc 

 ouaber of peeks of etal^ prodoec chac were |iven co t respondenc depended 

 on his norvtl cigtrecce cwuumpcloa: respondencs snsking up Co 4 cigarettes 

 per day were given one pack of each product ac each placeaenc; all ocbert 

 were given two packs of each. Foe any given paired cooparison, for 

 liucance Sub-saaple I comparing A and ^.ac Cb« first placemenc, half 

 Che'rt^pondents sacked A first and half npked • first, the appropriate 

 IntCruction appearing on the pack. ' v^ 



Thlt f ic^c consuocr test took place over che period June- July 1974. 



Soae 4 snachs haf^lapsed since complecioo of che sur^^ and taevicably 

 ere was so«* fall<$h£ of respondents during this period^'in addition 



th 



to those in Cluster 2 wKa were deliberately excluded. Out of 126} 



saokers theoretically tvtila^e for the coosuacr test 240 did not ttke 



part; of these, 33t had given up" smoking, ISZ had aoved house, 14Z were 

 either on holiday or out after rcpeactd calls and 17Z were unwilling 

 CO parcicipate. Froa the reaaining 10Z3 cespondents 32 dropped ouc 

 after che first paired coaparison and t further 33 after the second, 

 leaving 958 who coapleted all three stages of tVe test. Only 3 ssokers 

 discontinued the test because they disliked che cigarettes. 



At the placement interview, i.e. when cigarettes for. the first 



c 



paired comparison were left with respondents, certain ^uestlof^ were o 

 asked relating to their usual brand of cigarette. On couplet ioA iBf o» 



