39 



and weight of the ditferent meniber.s of the family; the number of 

 meals taken by each; the kinds, amounts, and cost of food consumed 

 during- a given period, and, so far as possible, the financial and hygienic 

 conditions at the time of the study. Methods had to be devised for 

 the collection of such data. In some cases it was possible for the 

 investigator to enter the homes and gather the statistics personally, 

 while in others dependence had to be placed on the statistics furnished 

 by the families themselves. To facilitate the work during the prosecu- 

 tion of the studies in Philadelphia, questions of a general nature were 

 formulated w^hich it was hoped would elicit the desired information. 

 These were printed in the form of a small account book, the first part 

 of the book being devoted to the questions concerning the family 

 itself, while the remainder was arranged so that the quantity and cost 

 of each food material purchased each day could be entered on the line 

 on which the name of the material was printed. These books proved 

 useful and were also employed in the Chicago studies, and served to 

 lessen the work when the investigator made the entries and to simplify 

 the matter so that there would be as little chance for error as possible 

 when the statistics were recorded by the housekeeper. 



The information gathered in both series of studies was, on the whole, 

 fairly satisfactory, though in several cases where records were kept 

 entirely by the family, some of the statistics recorded were manifestly 

 incorrect. To discover errors the accounts wei"e carefully examined 

 as soon as a study was finished, and questions were asked concerning 

 doubtful entries. In this wa}^ explanations and corrections were 

 obtained while the matter was still fresh in mind, and greater accuracy 

 was secured. 



The calculations of the results of the studies as given in the present 

 bulletin were made b}^ the same methods as have been noted in pre- 

 vious bulletins reporting studies carried on under the auspices of the 

 Office of Experiment Stations." None of the foods used was analyzed. 

 The composition of nearly all of them was assumed to be that given 

 for similar materials in a former publication of this Office.'' The com- 

 position of a few cooked foods was computed from the composition of 

 the materials used in preparing them and the proportions of each 

 material taken according to a recipe believed to be representative. 

 The percentages of nutrients assumed for any food material used in 

 these studies may be found in Table 29 of the Appendix. The refer- 

 ence numbers in the first column of that ta))le correspond to those 

 given in parentheses following the weight and cost of the food materials 

 in the detail tables of the studies. 



The studies as given below have been grouped according to the 

 nationalities of the families, as it was believed this would present the 



« See list on cover. 



^ U. S. Dept. Agr., Office of Experiment Stations Bui. 28, revised. 



