This study includes 6,141 pupils belonging to eight different high schools and distributed throughout two states. The cumulative, official, school records for these pupils formed the basis of the data used. The schools were selected primarily for their possession of adequate records. More dependable school records than those employed are not likely to be found, yet they tend to understate the facts of failure. It is quite possible that a superior school, and one with a high grade teaching staff, is actually selected by the requirements of the study. REFERENCES: 1. _Annual Report of United States Commissioner of Education for 1917._ 2. Josslyn, H.W. Chapter IV, in Johnson's _Modern High School_. 3. _The Money Value of Education._ Bulletin No. 22, 1917, United States Bureau of Education. 4. New York and New Jersey _State School Reports for 1917_. CHAPTER II HOW EXTENSIVE ARE THE FAILURES OF THE HIGH SCHOOL PUPILS? 1. A DISTRIBUTION OF ALL ENTRANTS IN REFERENCE TO FAILURE With no purpose of making this a comparative study of schools, the separate units or schools indicated in Chapter I will from this point be combined into a composite and treated as a single group. It becomes possible, with the complete and tabulated facts pertaining to a group of pupils, after their high school period has ended, to get a comprehensive survey of their school records and to answer such questions as: (1) What part of the total number of boys or of girls have school failures? (2) To what extent are the non-failing pupils the ones who succeed in graduating? (3) To what extent do the failing pupils withdraw early? The following tabulation will show how two of these questions are answered for the 6,141 pupils here reported on. ALL ALL ENTRANTS FAILING GRADUATES FAILING Totals 6,141 3,573 (58.2%) 1,936 1,125 (58.1%) Boys 2,646 1,645 (62.1%) 796 489 (61.4%) Girls 3,495 1,928 (55.1%) 1,140 639 (55.8%) From this distribution we readily compute that the percentage of pupils who fail is 58.2 per cent (boys--62.1, girls--55.1). But this statement is itself inadequate. It does not take into account the 808 pupils who received no grades and had no chance to be classed as failing, but who were in most cases in school long enough to receive marks, and a portion of whom were either eliminated earlier or deterred from examinations by the expectation of failing. It seems entirely safe to estimate that no less than 60 per cent of this non-credited number should[5] be treated as of the failing group[6] of pupils. Then the percentage of pupils to be classed as failing in school subjects becomes 66 per cent (boys--69.6, girls--63.4).