

bags, (2) all fruit in bulk, and (3) combination of packaged (polyethylene 

 bags) and bulk fruit. Oranges were packaged and priced in units of 12 and 

 grapefruit in units of 6. The test was conducted in nine stores during three 

 time periods of 1 week each. Each type of display was offered in three stores 

 each week and was rotated weekly so that it would be offered in each store 1 

 week during the test. The experimental design is shown in the appendix, 

 table 6. 



Sales data for the displays tested in the experiments described above were 

 obtained by audits of store records. Supplemental data on produce sales and 

 customer count were obtained from the chain headquarters. The supplemental 

 data were used as a basis for adjusting sales data if the number of customers 

 patronizing the stores or other outside influences affecting store sales, such 

 as factory or office pay periods, did not follow the same pattern each week for 

 all stores. 



Experimental designs and procedures used in these tests provide a basis 

 for estimating the reliability of the data or the probability of obtaining 

 similar results in repeated tests. 



Repeated time and motion studies were made in retail stores to determine 

 the labor costs for packaging, pricing, and displaying the fruit using each 

 type of package and display included in the test. The costs of packaging mate- 

 rials for all test items were also obtained. 



Quality control tests under simulated retail conditions were conducted at 

 Beltsville, Md., to evaluate the efficiency of the packaging materials and 

 displays in maintaining the quality of the fruits used in the merchandising 

 tests. A mechanically refrigerated produce display cabinet and two nonrefrig- 

 erated produce display racks were used in the quality control tests (appendix, 

 figs. 1 and 2). 



FINDINGS 



Packages for Grapes 



Sales of grapes in home toter bags were 18 percent greater than sales in 

 overwrapped trays. It was observed, however, that some of the customers who 

 bought grapes in home toter bags supplemented their purchases with grapes from 

 other packages. This resulted in an estimated k percent shrinkage or loss to 

 the store and was due partly to incomplete filling of the bags. Thus, complete- 

 ly filling the bags when weighing and pricing them would reduce losses. But, 

 even after allowing for the loss, the sales difference in favor of the home 

 toter bag was found to be statistically significant at the 20 percent probabil- 

 ity level- -that is, in 8 out of 10 repeated tests under similar conditions, 

 sales of grapes in home toter bags could be expected to be greater than sales 

 in overwrapped trays. The following tabulation shows the average number of Im- 

 pound packages of grapes sold per store : 



- 6 - 



