motivational differences. On the other 

 hand, relatively few hor.emakers in the 

 three cities remain to be convinced that 

 canned tuna does have these desirable 

 characteristics . 



SUGGESTIONS 



The following suggestions have been 

 derived from the analysis and sa^mary of 

 the data: 



The domestic tuna canning industry- 

 should pack and distribute tuna packed in 

 brine in all three markets . There is no 

 need for another can size for tuna. 



T!;e selection of Binr.ingham, Boston, 

 and Detroit was nade, in part, with the 

 intention of giving representation to three 

 regions of the country — the South, North- 

 east, and North Central — rather than to 

 three particular cities. V^^hen vievred from 

 this perspective it should be noted that 

 the factors vfith the greatest potential for 

 motivating consumers to becorie heavy asers 

 of tuna are quite different for the three 

 areas. Cnly one motivating characteristic 

 appears among the leaders for all three 

 cities; thus the problem of increasing tuna 

 consumption cannot be solved by the use of 

 a single national advertising program. 



Advertising csLmpairriS in all three 

 cities should stress that caxmed tuna is a 

 desirable food for all population groups 

 and that it has a pleasant taste . i'e rional 

 promotional themes should be developed for 

 Birmingham underscoring the facts that 

 canned tuna is an inexpensive, nourishing, 

 and easily digested food. In Boston the 

 emphasis should be shifted so as to accen- 

 tuate the ideas that the product is popular 

 v/ith those interested in losing weight and, 

 in order to combat the brine market, that 

 tuna packed in oil is not undesirably oily. 

 Detroit advertising should be based on the 

 concepts that tuna is an inexpensive food 

 and that it is used by experienced cooks. 



It is no longer necessarj' to emphasize 

 convenience and ease of preparation themes 

 in canned tuna advertising since av/areness 

 of these characteristics is widespread 

 among homemakers. 



It is apparent fran the results of the 

 survey that the promotional efforts made by 

 tuna canners should be directed toward 

 increasing consumption among consaters 



already using tuna and particularly among 

 those using the product cnly occasionally — 

 the light and sporadic user groups. 



TABLES 



The tables showing the percentage dis- 

 tribution of the responses to each of the 

 questions pertinent to this report are in- 

 cluded in the next section. A weighted base 

 was employed for the computation of each 

 percentage distribution. This base is 

 shown at the bottom of each table column. 

 Weighting the actual number of interviews 

 corripleted in each city was necessary since 

 a small number of the sampling units were 

 sub-sainpled to avoid an excessive number 

 of interviews in any one interviewing 

 assignment. This procedure was necessary 

 in those sample area segments which had 

 grown considerably in number of households 

 since the 1950 Census. In addition, the 

 total Detroit area was divided into zones 

 which were either predominately white or 

 non-*Jhite with the former sampled at one- 

 half the rate of the latter zone; weighting 

 vjas employed to restore the proportionality 

 of the race distribution in this city. 



No weighting was attempted for house- 

 holds selected for the sample but not inter- 

 viewed (refusals, unable to contact, etc.). 

 The actual number of completed interviews 

 and the weighted base for the total re- 

 spondent population in each city are shown 

 below. 



Cit£ 



Birmingham 



Boston 



Detroit 



Actual number 

 of interviews 



553 

 609 



'.veighted 

 base 



669 

 572 

 916 



Percentage distributions were computed 

 whenever the weighted base was 25 or more. 

 Only the number of responses in each 

 category is shown whenever this criterion 

 is not met. In percentage distributions, 

 each percentage vjas computed separately and 

 no effort was made to force the column to 

 add to exactly 100 percent. The occasional 

 discrepancies which occur because of 

 rounding should not affect use of the data. 

 In instances where the percentages add to 

 more than 100 percent because of multiple 

 answers by respondents, a footnote to this 

 effect is included in the table. 



13 



