43 oe « praca ae? 7 - oe Sy a ui ot ag si an ai peo De fal eee =oe tH gy He Paes a Nt oe aie ate te Ge 2 Shi 8 a bat a oe ry ee pe iat 3 BF Hy bs i. ‘1 +. .2 4 ts * ees ie i bist H f ‘ i ae se fae ae ee poet rent cay ae Ford aes Pe bs + i a a a ae ie eg | 4 A ix oes | a, vee ae a pe i ‘ io ne H id ee : ae hy TREN he rH anese yi SE Sa He boat ache 4 i ati | § tee er ee a ny % i at perevanriladtereaienhitss JUVENILE EMPLOYMENT PONte LABOR MOBILITY IN T 3 BUSINESS. CYCLE Mohs a Toa: University of Virginia Library HD6250.U4 M54 1927 ALD Juvenile employment and labor TU MX O00 cOl SeeSee epee sR ae Bearer peo ne ae ee Rpcpareret tapers ET nat ee Pap sae fort bed Sf a ae erent eee anes i Sp ae ae rerere Se Se Se or aera ae op ee ran aero fn 3S oe $e ao aed ot A nt ht Sere ee et Tee ag et oem * a 5 <> abner = Sea ae a NEEN Se ererce eee x oe ee eae eee pores ed ede Sale rae ieee’ pee Ee Paranda ono ak od oh ne ate nl Ak ok see ax ptetass ~ os ets SP PRP a oe eS SPE? = Srenas ae ae reside ares arrears tne ba end a, Soest aeae epee oad pene Peetabal toda pe ete SASL ae Poder 5 Eee re aie eto Sere Ss Ree Saad ta tenia oi Sot Perea eres eee Piet ores a ee ea ras et Pareto ee SF Spe ed Pa eee ae aoa rae One tad en ad s SP arse naa pene aia Poiana ee mb adactamear ae pe reas Soo oe ss sribars aes See aeedn eee ye ar sete ; = ‘i “ Ses : = epee rere reer papete rag Sis aoe SF nearer Maine ; eases ee < < = Pa ae Poe Meee yateweatusstgret 2 ee "ai nese ae ear Berens SSeS testes Seat Pe Se ee ae a ab iat ah ? 3 Fae ee fe Se Sa ere af Sart ee ae wat res mica Seo oe Sop ae near esr va oak epee : cere Se Beene Peer ra sbrad oh ae Ae 3 eoen Ea s a3 bak spree ere <. Se 7 Oe as oe: Sonne Ea Me oe eee siepepe Bag ne ee Peepers kare ae peepee “me ee) * uv pe roe se oe oe sheen i = ee 3 ee - 7 cree ‘ = rare reese aban a al lone ee oa Pe REA SP Pear ne BELO er eee = ‘ Pee aspen ze Meta lr aE peerer z near seee Bots te Bene nein d Sian at ape Pear eee aise Se See “so Se ea ad we haha td eb J. peaese ae. ren? = crane Seb ce mtn arn SESE aes eae, a> > : se esis = Seaiae spate Sepa Perenee arorae Sa IS aS Sere eZ eae ards Sse = ae MEO : Serre Sent mane we is aaa es ae oe Se Sout na fre eee oer cree SPS ae eeietats aparece earlobe seni ee ea serine: ee > a = ~~ ae Nap ad are 5S Soe ire eae — Soe ae 2 SNe ae Sad gab ae Se es Sat preretet ew ee PP ss aaa ae So aioe aN | Pata Saenroe, oa Senne peer th = » “ S Es on eee SS ee, ae ees act ee en SES Stes tie Pte peers cpio oe ze - or ; a pease Pe Serer ores peer Pee nred PP a Bie See ariae Sirah ess ¥ a tee AEM He * a eset tae nee Sed eee Siar Sere 2 Z a pape ee een Seppe ea oe a ee Bars: apeoeet 23 eee Spare SS feaate “po i OSS oz pee) aoe : 2 we te sal . ps ea ard Set eae pare sees ~~ aa ad ne nae oe ne er Pl a haeed teat i ob abe bea ae seer a see Sera Soa a abate 2S 573 sar ee ae Pa near ae Oe Fab end ae oe oe “a ah NEO gt a aese Pe rrae SS a 2 sf Rett : = ad os Sea sbueeeeieree eae ta seers a 2 = Eee naa eater eee eae == a. aoe year SS ae PIAS alr ne er ete sre et Foe Toms re = Sia ee or 6. ree peepee seeigniisvenstrieeen Eos eres ~< es 5 Sef emcee orn ate Ei ee Aobaae”LIBRARY OF THE UNIVERSITY OF VIRGINIA PRESENTED BY FRANK WILLIAM HOFFERJUVENILE EMPLOYMENT AND LaABorR MOBILITY IN THE BusINEss CYCLEHE Massachusetts Child Labor Committee is publishing this book as the second of a series on phases of the child labor problem, the first book being ‘‘Child Labor in Massachusetts.”’ The Committee was organized in 1908 and incor- porated under the laws of the Commonwealth in 1920. Its purposes are stated in the charter as follows: “To raise the standards of public opinion concerning the welfare of children; to aid working children to secure further education, improved health and good recreation; to protect children against premature or otherwise injurious employ- ment; to improve the conditions under which children are employed.” The directorate of the Committee is composed of Grafton D. Cushing, Chairman; Frank Leveroni, Vice-Chairman; Charles F. Bradley, Treasurer; Miss Esther G. Barrows, Dr. Merrill E. Champion, Richard K. Conant, Roy M. Cushman, Mrs. Ed- ward C. Mason, Rev. George P. O’Conor, Mrs. George W. Perkins, Rev. Malcolm Taylor, and Mrs. Philip E. Tripp. The offices are at 41 Mt. Vernon Street, Boston.Juvenile Employment and Labor Mobility in the Business Cycle By MAURICE BECK HEXTER, Ph.D. Executive Director, Federated Jewish Charities of Boston Instructor in Social Ethics, Harvard University BosTON MASSACHUSETTS CHILD LABOR COMMITTEE PUBLISHERSPURCHASE COPYRIGHT, 1927, BY MASSACHUSETTS CHILD LABOR COMMITTEE Printed in the United States of America Price, in paper, 50 cents per copy; im cloth, $1.00 per copy.To JUDGE A. K. COHEN, ALBERT W. KAFFENBURGH, and FRANCES STERN Civic servants and friends. as a slight token of appreciationFOREWORD Students of child labor sometimes lose sight of the fact that their field of inquiry is closely bound up with other similar fields. This volume proves the close alliance of child labor with at least one other such field, the economics of the labor market. We have read and heard much of senior employment indexes. As far as we know, however, this is the first time that such an index has been made for the junior employees. And at the same time a new angle of the child labor problem has been presented. Dr. Hexter has done real pioneer service in showing the possibilities of such an index. Not only has he made an index for one city and a special group of children, but he has proved the feasibility of using the same type of data for other cities and has shown the possi- bility of constructing junior indexes parallel to the senior ones. Special mention should be made of the charts, which are one of the most important parts of the book. Clearer than words, these charts present the story of the fluctuations in the juvenile labor market and how these fluctuations compare with those of the indexes of the senior market. They enable us to view the juvenile labor market in the same way that the business man follows the ebb and flow of the work of his concern by the use of charts and tables. The author has had entire freedom in the writing of this study. We are publishers only, although we are proud that one of the first contributions in this ixx FOREWORD Feld of endeavor is being published under the auspices of the Massachusetts Child Labor Committee. Thus in presenting the second volume in its series of publications, the Massachusetts Child Labor Com- mittee feels that it is rendering both to the students of child labor and of business variations a most unique and valuable contribution. Like the preceding volume, Child Labor in Massachusetts, the conclu- sions and results although based on local statistics are applicable to the country as a whole. GRAFTON D. CUSHINGPREFACE The Federated Jewish Charities of Boston has attempted from time to time to make modest con- tributions to knowledge of certain aspects of social problems which, though not directly associated with the sectarian nature of the range covered by that organization in its daily tasks nevertheless, intimately conditioned the welfare of its own specific clientele as residents of the wider community. Recently, these eftorts have concerned analyses of possible connec- tions, causal or coincidental, between the recurrent ebb and flow of business and cyclical variations of sociological phenomena. The present monograph attempts such a study in the province of juvenile employment and_ labor mobility. It attempts to add to our body of knowl- edge of the processes through which children go who are leaving or who have recently left school for work. As happens so frequently with similar studies, this one raises aS many or more questions than it solves. When more studies of such a nature become available it may be possible to do as much for children as now is done for adults in the employment field. The writer gratefully acknowledges here the aid and stimulus of Dr. Julius Goldman. For a careful revision and proof-reading of the manuscript he is under a deep obligation to Miss Mabel A. Strong, Secretary of the Massachusetts Child Labor Com- mittee. He is grateful, too, to Richard K. Conant, Commissioner of Public Welfare of the State of Massachusetts, who first suggested the interest the Massachusetts Child Labor Committee has 1n issuing such monographs. MAURICE B. HEXTER Boston, Massachusetts, August 26, 1927.TABLE OF CONTENTS Foreword Preface. List of Tables List of Charts Introduction The Trend Seasonal Indexes lhe Business Cycle. $2 New Employment Certificates. vi, New Educational Certificates, further School Attendance Compulsory , , | New Educational Certificates, further School Attendance Not Compulsory it mployment Certificates Reissued Educational Certificates Reissued, further School Attend- ance Mandatory. Educational Certificates Reissued, further School Attend- ance Not Required Comparison With Certain Other Economic Series... . New Employment Certificates. New Educational Certificates, further School Attendance Compulsory. New Educational Certificates, further School Attendance Not Compulsory Employment Certificates Reissued... Educational Certificates Reissued, further School Attend- ance Compulsory.. Educational Certificates Reissued, further School Attend- ance Not Compulsory Total Educational Certificates Reissued All Reissues Combined. Conclusions Appendix lables Giving Standard Units For All Series Studied At —_ ae A bo v- 70 79 58 99 104I]. II] lV. Gey XIb. LIST OF TABLES PAGE Occupational Distribution of Employed Children in the United States and Massachusetts (1920, United States Census). Oc upational Distribution of Juvenile Employees in Massachusetts, 10-15 Years of Age (1920, United Jct upational Distribution of Massachusetts ( on- tinuation School Children, 1920-1921. Principal Massachusetts Industries Employing Con- tinuation School Children, 1920-1921. Occupations of Children Who Went to Work Before Their 16th Birthday Between September 1, 1914, and August 31, 1918 Number of Certificates Issued by the Superintendent of Schools, Boston, 1915-1925 Seasonal Indexes of the Six Certificate Series Cyclical Fluctuations of the Several Certificate DETICS.... Coefficients of Correlation, New Employment ’ r 7 Certificates and Day s Index of Manufacture : Coefficients of Correlation, New Employment Cer- tificates and Federal Reserve Board Employment Index Coefficients of Correlation, New Educational Cer- tificates (attendance voluntary) and Day’s Index of Manufacture Coefficients of Correlation, New Educational Cer- tificates (attendance voluntary) and Federal Ke- serve Board Employment Index. . Coefficients of Correlation, Employment Certificates Reissued and Day’s Index of Manufacture Coefficients of Correlation, Employment Certificates Reissued and Federal Reserve Board Employment Index... Ll 30 42 60XV1 XIlIa. XIIb. XIIJa XITIb. MTT; X1Va. XIVb. a IIl. VI. VII. VIII. List or TABLES PAGE Coefficients of Correlation, Educational Certificates Reissued (attendance voluntary) and Day’s Index OfoMiantitacture: ake ne ict h ee loe e Coefficients of Correlation, Educational Certificates Reissued (attendance voluntary) and Federal Reserve Board Employment Index...... 74 Coefficients of Correlation, Total Educational Cer- tificates Reissuedand Day’s Index of Manufacture 82 Coefficients of Correlation, Total Educational Cer- tificates Reissued and Federal Reserve Board Pimploymentslndex: . skis.) ¢ dois ot cede Oe Coefficients of Correlation, Total Educational Cer- tificates Reissued and Berridge’s Mobility Index 84 Coefficients of Correlation, All Reissues Combined and Day’s Index of Manufacture........:....:. 90 Coefficients of Correlation, All Reissues Combined and Federal Reserve Board Employment Index.. 92 APPENDIX TABLES PAGE _ Total Educational Certificates Reissued, Standard Writes eee ee Ee ee ee ee Educational Certificates Reissued, School Attend- ance Voluntary, Standard Units............... 105 Educational Certificates Reissued, School Attend- ance Compulsory, Standard Units.............. 106 ’. Employment Certificates Reissued, Standard Units 107 _ New Educational Certificates, School Attendance Voluntary; otandatd Units: . 3 .- <2 «0. ieee New Educational Certificates, School Attendance Compulsory, Standard Units.................. 109 New Employment Certificates, Standard Units.... 110 Total Employment and Educational Certificates Reissued: Standard Units, :.. ¢ .cuics 65s eee eeI]. III. Vik Vid. Vile. Vila VIIb. Villa. VIIIb. Ville. VIlId. LIS OF CHARTS Annual Totals for Each Series Trends it Vert Les Kd lal \ ertificates I< ¢ issued (atte Pace } MLIISOT \ ) Kd Lit Lh, Soe tes \ (attendance lun- tarv): Ed il Cert ites, New (attendan: ; : ( } bisOUl } I CaTIO! i \_ ert Les Reissurs | . : (atte C \ LI »* Ota rd U tS * wf ' Vi > ( eT rec | d STAT Bureau of Labor Statistics, | I ent Index be « t Reser rATe rl i ent | it a ha ~~ | ’ I I ‘ i , 4 \ ; 7 4 vew | ment Cert tes d Dav’'s Index of Vianut re Ne VW } iT) i | ‘ | ' ? ’ t qd bt | Vj iT é ' 7 i hres | eT \ eS Ne ent ert es and Labor Ouit Rate in eC | ra Edt tional Cert ates Nev (attendance CO puls } ted ot tes Bur ii f | es LIST] Employment Index: Federal Reserve Board EK mpl ent Index: Harvard Employment Index New Educational Certificates (atte ndance not com- pulsory ); United States Bureau of Labor Statistics, ment Index: i ede! | Reser\ ; Board, Employment Index; Harvard Employment Index New KE ducati nal ( ertif ‘ ate >» (attendan: ee not COTIIi- pulsory) and Day’s Index of Manufacture New Educational Certificates ( ittendance not com- pulsory) and Harvard “B’ Curve New Educational ¢ ertificates (attendance not com- pulsory) and Help Wanted at Three Employment (Jtmmces 2 on 7; ‘ )XVIll Ville. I Xa. List oF CHARTS PAGE New Educational Certificates (attendance not com- puleory)i2 and Labor eu Rate in Selected Fac- tories. Employ ae Getibentes, Reisueds Wiited States Bureau of Labor Statistics, Employ ment Index; Federal Reserve Board, Employment Index; Harvard Employment Index. . Employment Certificates Reissued and Day’ S aden of Manufacture. . . Employment Gertifeites essa and arvana bo GUnve’. . Employment C serine ates R peenedand Help Wanted at Three Employment Offices. . Employment Certificates Reissued and icihee Quit Rate in Selected Factories. . Educational Certificates Reissued (attendance com- pulsory); United States Bureau of Labor Statistics, Employment Index; Federal Reserve Board, Employment Index; Harvard Employment Index _ Educational Certificates Reissued (attendance com- pulsory) and Day’s Index of Manufacture...... _ Educational Certificates Reissued (attendance com- pulsory) and Harvard “B” Curve.. Educational Certificates Reissued (Cetendancete com- pulsory) and Help Wanted at Three Employment Offices... . Educational C Spincates Resaned (attendance com- pulsory) and Labor Quit Rate in Selected Fac- tories. : Batiioral @rtheaten Remened \Gttendaneee, not compulsory); United States Bureau of Labor Statistics, Employment Index; Federal Reserve Board, Employment Index; Harvard pupae: ment Index. . Educational Certificates URewaued Gttendaneel not compulsory) and Day’s Index of Manufacture. _ Educational Certificates Reissued (attendance not compulsory) and Harvard “B”’ Curve. _ Educational Certificates Reissued (treadancen not compulsory) and Help Wanted at Three Ee ment Offices. . . Educational Gert pentcs| Reread (attendance not compulsory) and Labor Quit Rate in Selected Factories. . 3 56 58 59 61 63 65 66 68 69List OF CHARTS X1X PAGE XIla. Total Educational (Certificates Reissued: United states Bureau of Labor Statistics, Employment Index; Federal Reserve Board, Employment Index; Harvard Employment Index 81 XIlb. Total Educational Certificates Reissued and Day’s Index of Manufacture ; 83 Alle. Total Edu ational Certifi ates Reissued and Harvard “B”’ Curve 85 XIId. Total Educational Certificates Reissued and Help Wanted at Three Kmployment (Jtnces R6 XIle. Total Educational Certificates Keissued and labor uit Rate in Selected Factories 8 / Alila. Total Employment and Educational Certificates Ie ¢ issued: | nited States Bureau of abor statis- tics, Employment Index; Federal Reserve Board, Employment Index; Harvard Employment Index 8&9 XIIIb. Total Employment and Educational Certificates : Reissued and Day’s Index of Manufacture , 9] AllIc. Total Employment and Educational Certificates Reissued and Harvard “B”’ Curve 93 XliId. Total Employment and Educational Certificates Keissued and Help Wanted at Three Employ- ment Ofhces... 94 Allle. Total Employment and Educational Certificates Reissued and Labor Quit Rate in Selected Fac- tories 95 X1Va. All Series Combined (compulsory attendance ex- cluded); Educational Plus Employment Reissues: New Educational Plus Educational Reissues; and Snyder's Clearings Index... 97 XIVb. All Series Combined (compulsory attendance ex- cluded); New Employment Certificates Plus New Educational; Employment Certificates, New Plus Keissues; and Person’s Index of Trade ee XV. Worcester—Educational Certificates (New and Re- issued); Boston Educational Certificates (Re- issued); Day’s Index of Manufacture; Standard UnitsJUVENILE EMPLOYMENT AND LABOR MOBILITY IN ITHE BUSINESS CYCLE In the much tilled field of child labor we seek in vain for studies covering the cyclical variations of juvenile employment. ‘Time has been spent, and well spent, upon trends but little effort has been applied to studying the possible effects of the variations in com- merce and manufacture upon the juvenile labor market. And yet, the need of knowing more about these cyclical fluctuations is often stressed by the students of child labor. Within the past five years our knowledge of the connection between the gradual “ebb and flow of business’’ and various aspects of that group of ques- tions termed ‘“‘the labor problem’’ has greatly in- creased. Immigration,! strikes,2 attempts at in- dustrial democracy,*® labor turnover,* employment,” tH r Mier nand ti Bustr Cycle, Put t ‘\ i National ! ] I mic Research (New York, 192¢ \ en S I ‘ ] Sir - I Le! Ba | il DD imertican | non Review, VI, Number 4 (Decem! 1916 39769 by tl riter “‘“Jndt if Gener Busin ( ‘ n Cambridge, 119 . ' The bli : id ' P 1 Ky in 1919, Alvin H. H er Cycle t Strik lmer nl non Revte De- iber, 1921), XI, Nun r 4, 616-621 - Paul H. D is, ‘*Per | Prot it ( idminis- lrattion (Jul 19: W. F. Ogburn, r) Kf Condition i > lal Pro 1 1 rl Mi v1 ‘ r 19 ‘ Lan! ‘Busin lor the Ameri Lal I ement Columbia udies in Histor E-conon and Public 1 CX. Number 2 (New York, 19274 15-16 ‘ Paul F. Brissenden and Emil Frankel, Labor Turnover in Industry (New York. 1922). pp. 34-35; also by same writcrs The Mobility f Industrial | abhor.”’ Political Science Quarterly, 35, Number 4 December, 1920), 566-600; S H. Slichter. The Turnover of Factory Labor (New York, 1919), Chapter V; ee also rel nce to Professor Lanfear’s contribution ! ted in footnote (3) just above: Anne Bezanson and others, “A Study in Labor Mobility, Anmmnats the American Academy of Political and Soctal Scsence, CIII, Number 192, Business Cycle: Some Industrial 161~—234: William A. Berridge, ‘‘Labor and the busi! Aspects, Revteu Economic Statistics, VIIJ, Number 3 (July, 1926), 154 143: by the same writer, ‘‘Your Labor Turnover Good or Bad?” Factory (septem-2 JUVENILE EMPLOYMENT CYCLEs and trade union affiliation,’ are some of the aspects which have been under scrutiny during this time. As more has been learned of these suspected con- nections the search for further knowledge has become intensified and individualized. Weare no longer con- tent with a general or total index of unemployment. We now want to establish isolated indexes for at least the primary divisions of industry,’ as well as specific indexes for individual businesses. We also desire to create separate indexes for each occupation. And for each state and city we aim at establishing special data of many varieties. It is also man- 5 William A. Berridge, ‘‘Employment and the Business Cycle,’’ Review of Economic Statistics, 1V, Number 1 (January, 1922), 12-51; by the same writer, ‘Cycles of Employment and Unemployment in the United States, 1903-—1914,’’ Journal of the American Statistical Association, XVIII (March, 1922), 42-55 and (June, 1922), 227-240; by the same writer, Cycles of Unemployment in the United States, 1903-1922, Pollak Publication, Number 4 (Boston, 1923). 6 George E. Barnett, ‘‘Growth of Labor Organization in the United States, 1897-1914," Quarterly Journal of Economics, X X X (August, 1916), 780-795; by the same writer, ‘“‘The Present Position of American Trade Unionism,”’ American Economic Review, Supplement, XII, Number 1 (March, 1922), 44-55; Leo Wolman, The Growth of American Trade Unions, 1880-1923, Publication Number 6 of the National Bureau of Economic Research (New York, 1924). 7 Note Diagram 3 entitled ‘‘Indexes of Employment in Eight Primary Divisions of Industry in Canada, 1920 to 1925,’’ Mintstry of Labor Gazette, X X XV, Number 1 (London, January, 1927), 13. 8 The experiment of the Walworth Company is extremely suggestive. See American Labor Legtslation Review (March, 1923), p. 23. 9‘*The Federal Bureau’s figures are now published in a special section of the Monthly Labor Review, which is issued as an advance bulletin, with news releases sent out earlier, as promptly as the data are available. The data are given for the main industries and their subdivisions, and are then recapitulated for the nine geographical divisions used by the United States Census Bureau. It has not been proved feasible for the federal bureau to publish the data for each state, or for federal reserve districts, which, it has been suggested would make the data directly useful to the 12 federal reserve banks. All the collect- ing state bureaus, however, now publish their own data currently for local use.’’ Ralph G. Hurlin and W. A. Berridge, Employment Statistics for the United States, a plan for their national collection and a handbook of methods recom- mended by the committee on Governmental Labor statistics of the American Statistical Association (New York, 1926), p. 15. Another indication of this trend is recognized in the following: ‘‘In recent months a number of manufacturers and distributors have applied to the Minis- try of Labor to enquire whether they could be regularly supplied with tables showing the course of unemployment from month to month in each town or area served by an Employment Exchange. The information in each case was required to enable the firm to adjust its sales activities to the changing pros- perity of the various local areas or to make due allowance for such changes in judzing the effectiveness of its advertising or salesmanship.JUVENILE EMPLOYMENT CYCLES ifestly advisable, especially from a social viewpoint, to create separate indexes for the various age groups. Skilled and unskilled labor each should have a re- liable index established for them. At the present time the Federal Department of Agriculture is carry- ing on a promising experiment in the collection of national employment statistics for farms."° As has been already noted, one of the important refinements should be a special index for juvenile employment, a branch of the labor market which has not yet re- ceived the measure of attention from the economists of the business cycle which it merits. At the same time as this tendency toward in- dividualization of the goal of statistical research ap- plied to the economics of the labor market has de- veloped, there occurred a most interesting evolution in the application of the same statistical methods to sociological data.'' The strong academic effort to break down the walls which hitherto have artificially divided one group of data from another has borne I I Labor, but 31 of the kind ti I Mintstr [ r ( levotes the | f its unempl! I to pr t la lustry Da I pul entitled ‘‘Lo I Index will sive, for 63 r ul in Great Brit bers of in- sured 1 h area, t ther ¥v thi ti iployment ror me! ind j yn the M ir t 14 ‘ ionth ( npa ' n { ind t m I , ; f order that ret i ns may readily | n. ‘‘Arrangements have been made for tabulating the returns recel\ 1 from Employment Exchanges, for preparing the hgurs in percentage form, and for printing and despatching in the shortest pra ticable time, ind the in! rmation will be in the hands of subscribers within two to tnres weeks of the day to which the figures relate..... The Ministry of Labor Gazette, op. cit. p. 11. 9 Ibid, pp. 13-14. \l See by the present writer, Social Consequences of Bussness Cycles, Pub- lication Number 7 of the Pollak Foundation (Boston, 1925); Dorothy Swain Thompson, Soctal Aspect of the Business ¢ London, 1925); R. H. Hooker, ** Journal of ‘On the Correlation of the Marriage Rate with Foreign Trade, the Royal Statistical Soctety, LXIV (1901), 485; George R. Davies, ie Aspects of Busine Cycles, Quarleriy Journal of the Universtiy ' = cial if North4 JUVENILE EMPLOYMENT CYCLES fruit. Already, as a consequence, there is appearing an increasing number of studies from the fertile borderlands between different fields of inquiry. Apparently the method of statistical analysis so greatly developed by the economists’? within the past decade, on being applied to sociological time series leads to similarly valuable findings and judgments. As yet few sociological problems have been subjected to this type of analysis. For the most part they have been analyzed only from the viewpoint of social statics. There are, however, very few sociological problems which would not gain immeasurably in clarity if they were so analyzed. Dakota, XII, Number 2 (January, 1922), 108-21; Katherine E. Howland, “‘A Statistical Study of Poor Relief,’’ Journal of the American Statistical Asso- ciation, XVIII (December, 1922), 480-89; W. F. Ogeburn and D. S. Thomas, ‘“‘The Influence of Business Cycles on Certain Social Conditions,’”’ Journal of the American Statistical Association, XVIII (September, 1922), 340; G. U. Yule, ‘‘On the Changes in the Marriage and Birth-Rates in England and Wales during the past Half-Century,’’ Journal of the Royal Statistical Society, LXIX (1906), 88-132; Ethel M. Elderton, England North of the Humber, Report on the English Birth Rate, Part I’ (London, 1914) diagram opposite page 12; August Meitzen, Hztstory, Theory, and Technique of Statistics (translated by R. P. Falkner) (Philadelphia, 1891), p. 138; E. Levasseur, La Population Francaise, II] (Paris, 1889) diagram opp. p. 12; Heinrich Rauchberg, Die Bevilkerung Oesterretchs (Vienna, 1895) diagram opposite p. 26, giving the fluctuations in births, deaths and marriages, together with the price of various grains, and average wages for the years 1820-90; L. M. Moreau-Christophe, Du Probleme de la Misere et de la solution ches les peuples anciens et modernes, III (Paris, 1851), 222; G. V. Mayr, Stattsttk der gerichtlichen Polizei in Kéniz2- reiche Bayern (Munich, 1867), p. 136, et passim; A. von Oettingen, Die Moral- Stattsttk 1m thre Bedeutung fiir etne Socialethtck (Erlangen, 1862), p. 239; L. |tuld, Der Einfluss der Lebensmittelpreise auf die Bewegung der strafbaren Hand- lungen (Mainz, 1881), p. 193; W. Starke, Verbrechen und Verbrecher in Preus- sen, 1854-78 (Berlin, 1884), p. 77, et seg.; A. Mever, Die Verbrechen in threm Zusammenhang mit dem Wirtschaftlichen und Soztalen Verhdlinissem in Kanton Zurich (Jena, 1895), p. 60, et passtm; M. Tugan-Baranowsky, ‘‘Die Socialen Wirkungen der Handelskrisen in England,’’ Archiv fiir Sociale Gesetzgebung, XV, 36, et seq.; ibid., Studien zur Theorie und Geschichte der Handelskrisen in England (Jena, 1901); G. Schnaepper-Arndt, Sozialstatisttk (Leipsic, 1908), p. 624, et sez.; Paul Lafargue, ‘“‘Die Kriminalitit in Frankreich,’’ in Dte neue Zeit (1890), p. 120; Gustav Aschaffenburg, Crime and its Repression (Boston, 1913), p. 112; F. S. Crum, “The Marriage Rate in Massachusetts,’’ Journal of the American Statistical Assoctation, IV (December, 1895), 322-39; F. Stuart Chapin, ‘‘Dependency Index for Minneapolis,’’ Social Forces, V, Number 2 (December, 1926), 215-224; by the present writer, ‘‘The Business Cycle, Relief Work, and Desertion,’’ The Jewish Social Service Ouarterly, I, Number 1 (February, 1924), 3-33, Number 2 (May, 1924), 1-30, Number 4 (February, 1925), 1-24. 12 See the extremely valuable contribution upon this point by Professor Warren M. Persons, ‘‘Statistics and Economic Theory,’’ The Review of Eco- nomic Statistics, VII, Number 3 (July, 1925), 179-197.JUVENILE EMPLOYMENT CYCLES The question of what response, if any, the juvenile labor market makes to business cycles is important for further careful study. In this connection, Dr. W.I. King” says that ‘‘manifestly there has been no tendency worth mentioning for sons or daughters to change occupations because of the fluctuations of the business cycle. The number of factory workers in this group declined by less than a dozen persons. The only shift of movement was the movement of those not gainfully occupied into the ranks of pro- fessional service, commerce, and trade, and this change is explained mainly by the fact that a con- siderable number of boys and girls finished school during the period and began clerking in stores or teaching school.” After presenting data indicating the number of 14-year-old and 15-year-old children employed in twenty cities in Massachusetts from 1919 through 1925. Fuller'4 and Strong conclude that ‘“‘it is dif- hcult to draw general conclusions as to the tendency of child employment in Massachusetts. The year 1919 was undoubtedly a year of abnormally high em- ployment of children, due largely to the continuing effects of the war period. The years 1924-1925 were undoubtedly years of low employment, at least in certain industries and certain cities, but what is a normal amount of child employment, the available facts and figures do not disclose, nor do they indicate what may be the amount of child employment 1n 1926. VW |. King. Emodi meni Hours and karnin in P7 ertl Depres- sion, United States, 1920-1922, Publicati Jul the Nat LB 1U of Economic Research, Second Edition (New York, 1923), p. 25 if Raymond G. Fuller and M ibe] A. Strong, Citid Labor in Ma aQchtuUseé! (Bost« n. 1926 pp. 23-24: see also “Child Labor i Wisconsin, 1917-1922, 1923," Wisconsin 7 Industrial Commtsstor i ird Bur Our B New York, 1923).6 JUVENILE EMPLOYMENT CYCLES “The fact of rapid fluctuations of child employ- ment, locally and in the state as a whole, over a very wide range of variation, is, however, entirely clear; and this is the real significance of the tables. The instability of child employment, with its ups and downs, its cyclical movements, is by no means an un- important aspect of the child-labor problem—dquite as important, indeed, as the instability of child workers.” Fortunately there is a vast amount of data con- cerning the juvenile labor market which has emerged as a result of protective labor legislation for employed minors. This study which is both an attempt to in- dicate the possibilities of erecting a juvenile employ- ment index as well as an attempt to analyze the mobility of juvenile labor, is based upon such data. Massachusetts has an interesting history con- cerning such protective legislation.!® By 1900 the statutes prohibited employment of children under 14 in stores, factories, and workshops, and included special provisions prohibiting the work of older chil- dren in certain occupations regarded as either physi- cally or morally dangerous. By the enactments of 1913, children under 16 years could no longer work more than eight hours daily and forty-eight hours weekly; greater restrictions were also cast about night work. A street trades law, revised and enlarged in 1921, was passed. Hazardous occupations were prohibited to children under 16 and the list of interdicted dangerous occupations for chil- dren between 16 and 18 was greatly enlarged. Author- ity to enlarge the list still further was granted to the 15 Fuller and Strong, op. ctl., pp. 13-14, portray this history very briefly.JUVENILE EMPLOYMENT CYCLEs State Board of Labor and Industries. School at- tendance was made compulsory up to 16 years of age for children not lawfully employed; continuation schools might be established and four hours weekly attendance at such schools was required of employed minors. By 1919 the establishment of continuation schools became compulsory in such communities where during the preceding year two hundred or more minors between 14 and 16 years of age had been em- ployed. Likewise the hours of labor for children below 18 were limited to nine hours per day and forty- eight hours per week; the educational qualification for employment certificates was increased from the fourth to what amounted to the sixth grade. In 1921 the sixth grade requirement for the educational qualification was actually reached. To present a picture of the individuals behind the hgures which are analyzed in this study and so reach later a clearer interpretation of the findings of our analysis it is worth while to describe bri fly the con- ditions and the mechanism which surround the is- suance of the pt rmits re quired by the Massachusetts law for employed minors." The regular employment certificate is required for every kind of full-time resident employment for minors 14 to 16 years of age except home work, farm labor, and private domestic service. Each minor in this age group who, at any time, leaves school to work per- manently while schools are in session—if that em- ployment occupies at least six hours a day—must procure this certificate and file it with the employer For a description in detail of the mechanism see Dr. Helen Sumner Woodbury, The Working Children of Boston, | S. Department of Labor, 7 , : ’ . . . ' 1 >>. ne “5 Children’s Bureau, Bureau Publication Number 89 (Washington, 1922), pp. /-12.8 JUVENILE EMPLOYMENT CYCLES before he may be legally “employed in,” “about,” or ‘‘in connection with,” any factory, workshop, man- facturing, mechanical or mercantile establishment, or enter any employment. Employment is defined in Section I, Chapter 149 of the General Laws, as “any trade, occupation or branch of industry, any particular method or process used therein and the service of any particular employer, but it shall not include private domestic service or service as a farm laborer.” — For minors 16 to 21 years of age educational cer- tificates are required for employment in a factory, workshop, manufacturing, mechanical or mercantile establishment, in a bootblack stand or establishment, barber shop, or in the construction or repair of build- ings, or by an express or transportation company. Minors over 16 are not required to secure an educa- tional certificate unless they are employed ‘‘zn’’ the above enumerated establishments. Two forms of educational certificates are issued; the regular cer- tificate for minors who are not required to attend evening schools; and the special certificate for those who are required to attend evening schools when estab- lished. These certificates are issued by the superin- tendent of schools or his representatives as in the case of the issuance of employment certificates. Edu- cational certificates carry the same requirements as the employment certificates with respect to return by the employer to the issuing officer within two days after the employment of the minor terminates. The regular educational certificate is issued to min- ors who can meet the requirements for completion of the sixth grade of the public schools in the city orJUVENILE EMPLOYMENT CYCLEs 9 town where they live. Evidence of age and of ed- ucation must be furnished by the minor, who is re- quired to appear in person before the issuing officer, and sign the certificate. The issuing official in signing the certificate testifies that the minor in question can meet the specified educational requirements, and authorizes his employment subject to the provisions of the law which forbid employment in certain class- ifications of labor. Before granting the certificate, therefore, he must satisfy himself that the work in which the minor is to engage does not fall in the prohibited list, which is as follows: Work in or about blast furnaces; in the operation or manage- ment of hoisting machines; in oiling or cleaning hazardous ma- chinery in motion; in the operation or use of any polishing or buffing wheel; at switch tending; at gate tending; at track re- pairing; as a brakeman, fireman, engineer, motorman, OF conduc- tor upon a railroad or railway; as a fireman or engineer upon any boat or vessel; in operating motor vehicles of any description; in or about establishments wherein gunpowder, nitroglycerine, dynamite, or other high or dangerous explosive is manutfas tured or compounded; in the manufacture of white or yellow phos- phorous or phosphorous matches; in any distillery, brewery, or any other establishment where malt or alcoholic liquors are manufactured, packed, wrapped or bottled; in that part of any hotel. theatre, concert hall, place of amusement or other estab- lishment where intoxicating liquors are sold. This section does not prohibit the employment of minors in drug stores, but the Department of Labor and Industries has prohibited the employment at sandpapering of lead painted surfaces indoors, and the employment of girls under 18 years of age in carrying messages for a company, corporation, or association. The special educational certificate is issued to minors who are over 16 and under 21 years of age, who cannot meet the requirements for the completion of the sixth grade. Such minors are required to attend10 JUVENILE EMPLOYMENT CYCLEs evening school (if established) until they have re- moved the deficiency. They must present to their em- ployer each week a record of such attendance, other- wise their employment is unlawful. These descriptions provide the explanation of the first three of the six series used here. It is most im- portant for later interpretations to note again that before these various documents issue from the office of the superintendent of schools there is a definite promise of employment. The latter three series evolve from another provision of the law. This pro- vision reads that within two days after the cessation of employment the certificate must be returned to the office of the superintendent of schools. When re- issued these three main forms comprise the last three series in our analysis. Consequently the ‘‘reissue’”’ series connote juvenile labor mobility, comprising voluntary quits, lay offs, or shutdowns, and subse- quent re-employment in either the same occupation and with the same employer or in some other place or in a different occupation. These series are most im- portant ones, for labor turnover and the mobility of labor are critical questions, especially in understand- ing the social and economic effects of the business cycle. One other description must be presented before proceeding with the analysis. This concerns a statis- tical picture of the occupations in which children in Massachusetts are found. |} In 1910, 18.4 per cent of all the children in the United States between the ages of 10 and 15 were gainfully employed. This pro- portion dropped to 8.5 per cent at the time of the 1920 census. Since corresponding per cents for Mass-JUVENILE EMPLOYMENT CYCLEs 11 achusetts are respectively 9.6 and 8.6, it is manifest that the state presented a poorer picture than the country as a whole. The occupational distribution of these children which is given in the following table is of interest for Our present purposes. TABLE |! OCCUPATIONAL DISTRIBUTION OF EMPLOYED CHIL- DREN IN THE UNITED STATES AND MASSACHUSETTS (1920. United States Census UNITED \IASSA- STATES CHUSETTS Agricultural pursuits 61.0 Bi Manufacturing and mechanical ie 70.4 All other ox cupations 71 5 27 l Obviously, we are concerned with a group which is engaged in industry and consequently has much to add to our knowledge of cyclical fluctuations. It is important, however, to know more in detail of the specific tasks and occupations performed by these juveniles. The following table!’ also shows the heavy preponderance of the industrial group of oc- cupations which forms so important a part of the juvenile labor market:12 JUVENILE EMPLOYMENT CYCLEs TABLE II OCCUPATIONAL DISTRIBUTION OF JUVENILE EM- PLOYEES IN MASSACHUSETTS, 10-15 YEARS oF AGE (1920, United States Census) Number of Occupations Children Employed Spb sa eee Coe Rie ISAS cre srs bo Slsdeie ie ous Sees 33,723 AST ACUI Cra MUPSUItS) Geli ceicie fe 3 les omelet 831 Manufacturing and mechanical .............. 23,753 Domestic and personal service ............... 1,226 Wlerical OCCU NALIONS fiche oo ee eee eae eke 4,317 Trade (including stores, newsboys, etc.) ...... 2,749 Transportation (including messengers) ....... 655 ESxtrachion O1mminerals® of 0.6% Ss oosicee 0 Se cdot 7 OtherZoccupavionst hee chokes ie 185 The next two tables indicate some information concerning the 14- and 15-year-old group of children attending the continuation schools of Massachusetts in 1919, 1920, and 1921.8 TABLE III OCCUPATIONAL DISTRIBUTION OF MASSACHUSETTS CONTINUATION SCHOOL CHILDREN, 1920-1921 Occupations in Which _Number of Percentage of Continuation School C hildren Em- Children Children were Employed] Ployed in Speci- Employed fied Occupations ‘Lotal eee reece erase 19,044 100.0 Manufacture <<. .-... 13,110 69.0 Textile) .::3 2h Se ce 7,467 39.0 Others ’y. 28 eee. Ge 5 643 30.0 Non-Manufacturing ....| 5,934 31.0 Stores!o'. sk eee cen 2,298 12)0 Miscellaneous) 3-04 4. 3,636 19.0 18 From a study made by the State Department of Education and quoted by Fuller and Strong, op. ctt., p. 157.JUVENILE EMPLOYMENT CYCLES ~ eA ee Ley PRINCIPAL MASSACHUSETTS INDUSTRIES EMPLOYING CONTINUATION SCHOOL CHILDREN, 1920-1921 ‘ ' " q ‘y ; r> N er of Children rercentage EK mployed in of Children ; : : @ . e Industry Specified Industrie Employed In : Specified Potal Rov ir] | i OT} I ; \ vITis Industries l otal ) O44 85 059 100_0 Textiles 7/467 5 261 +, ZI 39.0 Stores. SU . : Z | 115 12 U . , Bake , ( ont { C lot] Depart Drv il yf (,roce I “atner YOOdsS, icil, leatherandrubbers 2,130 1 408 662 Le Cand " / U ; 5. Printing and lithog ing . ' S700 es f,2 2 Age ) 67 52 Others U0 38 2,61 32.0 aD a aS He au: : . at er Concerning the actual work performed Dy tnese children the Massachusetts Child Labor Committee, after studying the continuation school chi 1922 and 1923, wrote that ‘“‘two-fifths of the children are working in textile mills as doffers, sweepers, oilers, wire boys and girls, weavers, and bobbin boys and girls. In the shoe factories they are employed, as a rule, as helpers on nailing machines and stitchers of very simple seams. They carry boxes and are floor girls in candy factories. In paper box factories, they fold and paste and pack. Hundreds are em- ployed in errand work either inside or outside the stores and plants. In stores they are cash girls,14 JUVENILE EMPLOYMENT CYCLES errand boys, helpers in the stockroom, and occasion- ally sales girls.// Wherever these children are, it is mainly their arms and legs that are wanted., Of the two hundred forty pupils in one continuation school ‘only six were seriously engaged in tasks that lead directly to the trades.’ " ” This same condition was true in 1906 when the Massachusetts Commission on Industrial Education found that ‘33 per cent of the children of this state who began work between the ages of 14 and 16 were employed in unskilled industries, while 65 per cent were in low-grade industries; thus a little less than two per cent were in skilled or high-grade indus- tress = After a most careful study of the occupational histories of children under 16 in Boston, Cambridge, Somerville and Chelsea, for the Children’s Bureau of the United States Department of Labor, Dr. Helen S. Woodbury” found that practically two-thirds of these children went into clerical occupations; wrap- ping, selling, or delivery of goods. The balance went into factories or mechanical pursuits. One-third of the clerical group was represented by messenger and errand service and delivery work. Important, too, was cash and messenger service in retail stores. The following table22 shows in detail the occupational dis- tribution of two groups of working children as studied by Dr. Woodbury. 19 Massachusetts Child Labor Committee: The Continuation Schools of Massachusetts, p. 7. 20 Report of the Commission on Industrial and Technical Education, Massachuselts Senate Report Number 349 (April, 1906). 21 Dr. Helen Sumner Woodbury, The Working Children of Boston, Urs: Department of Labor, Children’s Bureau, Bureau Publication Number 89 (Washington, 1922), p. 32. a Compiled from Table 115 and Table 1, Appendix, of Dr. Woodbury’s study.JUVENILE EMPLOYMENT CYCLEs TABLE V OCCUPATIONS OF CHILDREN WHO WENT TO WorK BEFORE [THEIR SIXTEENTH BIRTHDAY BETWEEN SEPTEMBER 1, 1914—AuGustT 31. 1918 I l by ir the ce | PAT I VS — tinu- . 123 r' Percent All Occupations: .416 1100 7.381 1100.0 Persona d domestic occupatio1 LYS AE 02 221 Pers | service (other S vants in the hi 124 | 12] 1.6 Hous d or! i] ep 5 | 9 Factor’ ind f | occ tions 154 33 el) Sed } iIcTOTYV ¢ 11 ] \) one } 14.4 IQ Shoe { 7 670 10.4 i4 iO. 1 (‘loth A ' d ot needle 1 if $39 O 533 iste Dextil | 22/ po 1S 102 Candy factor 50 . Lo / ()the fa tor HO4 7 4 S78 a Ap] rent ( nd hel er-skilled trades 164 PAO 153 Zu { ri | Or DaATIOI WraDdDbD all Oo nd del eT i oOods 1 () 4) Fh: RS 4 65 »s (Office work 176 1.4 +40) 6.U Cash and messenger work—d parti nt store SY 14.0 1.053 14 5 Packing. wra ing, labelling, an: shipping room work 310 1.8 525 i 4 Selling 54 LO. |, 213) | ns Messenger work, errand and de livery 2.095 Sd » R95 nS All other occupations 29 . 20 4 Not repe16 JUVENILE EMPLOYMENT CYCLEs There is no formal or logical reason to suppose that the occupational distribution of the children of Dr. Woodbury’s study does not comprise a random sample of the whole period which is under scrutiny in this present monograph.*4 We can, accordingly, understand why Dr. Woodbury concluded that “the occupations in which children between 14 and 16 years of age could be employed were of course decid- edly restricted by their ages, lack of physical strength, and lack of education and experience. To a certain extent they were also restricted by law, particularly by the provisions in regard to hours, continuation school attendance, and employment on machines. As a result most of the positions held by the children studied were for simple mechanical tasks or for run- ning errands or for carrying articles either inside or outside the establishment. Although none of these positions required any real skill, some of them per- mitted the development of a certain dexterity, and others made it possible for the child to acquire a little practical knowledge of the business apart from his own small task. A few of them, doubtless, offered opportunity for promotion to more skilled or respon- sible positions if the child remained, which he rarely did, until he grew older.”’ The data used here have been obtained from the records of the Boston School Committee concerning the employment certificates issued to minors under 21 from 1899 onward.25 However, the rigorous analysis is restricted to the data beginning in 1914 because the year 1943.weth its many legislative enact- 24 Fuller and Strong prestnjf‘ somie interesting data in this connection; Op. ca... 0. O24: 2> For the years 1894-1899 we were unable to find monthly data.JUVENILE EMPLOYMENT CYCLEs 1/7 ments caused a distinct break in the statistical homo- geneity of the data. The detailed provisions of these restrictions sur- rounding the juvenile labor market in Massachusetts result in a variety of statistical series,2° of which six?? were studied; namely, new employment certifi- cates, new educational certificates where school attendance is compulsory, new educational certificates where school attendance is voluntary, employment certificates reissued, educational certificates reissued where the school attendance is compulsory, and educational certificates reissued where school attend- ance is voluntary. Under analysis these six series yield information about the following reactions of the juvenile labor market; the trend, the seasonal fluctuations, the cyclical variations, and the rela- tionship between these cyclical fluctuations and certain correlatives. The Trend The following table shows the annual number of new and reissued employment and_ educational certificates granted by the Superintendent of the Boston schools. *% A cetailed recital of the different forms demanded by the certification laws may be noted in the Bulletin of the Department of Education and De- partment of Labor and Industries, Whole Number 168 (1926, Number 3), pp. 15-40. This publication has been relied upon for a description of the pro- hibited occupations. 77 We have combined the educational reissues into one series and then all the reissues into another series. These are not available in this form in the offices of the School Committee of Boston. The analysis to which they have been subjected is the one developed by Professor Warren S. Persons. For its description see Indices of General Business Conditions (Cambridge, 1919), pp. 5-36.18 JUVENILE EMPLOYMENT CYCLES TABLE VI NUMBER OF CERTIFICATES ISSUED BY THE SUPERINTENDENT OF SCHOOLS, BosTON, 1915-1925 I Educa- Educa- Educa- Educa- tional tional tional tional Employ- |Certificates|Certificates Employ- |Certificates|Certificates ment School At- | School At- ment School At- | School At- } Year of |Certificates| tendance | tendance | Certificate} tendance | tendance Issuance} | Compul- | Voluntary | Compul- | Voluntary | oaory © +I sal sory | New New | New | Reissues Reissues| Retssues “4915 | 3,686 | 957 | 10,291 | 3,370] 1,243 | 15,219 “to16 | 7,086 | 611 | 11,971 | 7,561 | 1,291 | 24,420 “1917 7,033 421 | 10,911 10,568 | 910 | 25,018 1918 | 8,689 | 217 | 11,176, 12,043 | 668 | 27,363 “oi | aor | 157 | 92s | 11005 | 285 | 20076 1920 | 6,118 613 ~ 9390 | 9,534 | AT: i 17 729 “4921 | 3,078 | 1,060 | 7,801 4875 | 654 “11,568 1922 | 4,420 | 1,024 | 9,144] 4,983 | 875 | 13,701, 1923 | 5.276 | i 912 | 10,009 | 6,049 1,070 14,979 1924 3,435 | 743. | 8,489 | 4,384 | 900 9,836 1925. 3,964 368 | 9,360 | 4.308 632 10,754 | | Chart I shows these data graphically. Note first that the two lowest (smallest) curves concern those series in which the element of further compulsory school attendance plays a part;—in other words, these curves reflect the work tendencies of children who can not meet the requirements for the completion of the sixth grade and must attend evening classes until this deficiency has been removed.ChiAraiat @NNUAL TOTALS FO R EACH SERIES | ~20 JUVENILE EMPLOYMENT CYCLES These groups in all probability consist of children mentally retarded. It is a curious feature that the dip of the chart in these series Numbers 2 and 5 occurs during the same years in which the remaining series portray the hump. We are at present unable to offer a satisfactory explanation for the dip. May it be due to the ease with which employers secured children who had a higher educational status and who came into the juvenile labor market in response to the following reasons suggested in connection with the hump in the other series? All the other curves show the hump from 1915- 1919 depicting therein the effect of the war upon the juvenile labor market. This has been noted also for other jurisdictions. McGill?’ writes that ““begin- ning with the autumn of 1915 an unprecedented rise began in the number of children entering gaintul employment, and heavy increases were practically everywhere recorded for 1916 and 1917, even before the United Statesentered the war. . . . Of twenty- three cities for which data are available for the calen- dar year 1916 and also for 1915, every one except San Francisco suffered an increase in 1916 over 1915 in the number of children receiving employment certif- icates. . . . After the entrance of the United States into the war the number of children taking out employment certificates continued to rise. The forces at work pushing children into industry included the growing cost of the necessities of life, combined in many cases with the absence on military duty of 28 Nettie P. McGill, ‘Trend of Child Labor in the United States, 1913 to 1920."" Monthly Labor Review, XII, Number 4 (April, 1921), 7-9. This study has been carried on to 1923. See Monthly Labor Review, XVII, Number 3 (September, 1923), 101-105; also Employment of Young Persons in the United S.ates. National Industrial Conference Board (New York, 1925), pp. 23-31.21 JUVENILE EMPLOYMENT CYCLEs members of the family who had previously contributed toitssupport. High wages offered by employers hard pressed for help proved a powerful magnet, drawing into business and industry many children under 16 who in normal times would have remained in school. A spirit of adventure due to the excitement of war has been mentioned also as an important cause for the increased number of children going to work, especially among boys. . . . While with the few exceptions noted 1918 was the peak year as regards the numbers of children under 16 years of age going to work, in several cities the increase in 1918 was so striking as to arrest attention even in that year of generally large increases.” Chart II,?¥ containing the trend lines for these series upon ratio ruling, better depicts the course of these diverse series of data over the whole span of years which is under review. Very striking is the parallelism of all the series with the exception of Number 3 and Number 2. This parallelism, of course indicates that the ratio of decrease in these various series has been similar throughout this span of years. The second feature to which we direct attention is that all of them incline downward with the exception of Number 2 (a reissue series concerned with com- pulsory attendance at night school) which has a slight upward slope. Even for this series it may well be that a longer span of years will indicate a negatively inclined trend since the trough years may have unduly influenced the result. We conclude, therefore, that the trend of number of juveniles leaving school for the labor market since 1914 has on the whole been 2% These lines have all been itted by the method of least squares.CHAK vil +— —4 TRENPS 1 = + 4 20 O00 15000 10000 3000 8000 7000 ¢ OOO 5 000 4 000 5000 2000 1 500 1 OOO 300 600 7100 S00 500JUVENILE EMPLOYMENT CYCLEs downward at a rather marked rate.*® And it is to be noted, of course, that simultaneously there has been a growth in population. It is impossible to state whether this decline is due to the protective legislation which now surrounds the juvenile labor market; or whether it is due to a greater reluctance on the part of employers to engage juvenile labor; or whether it is the result of the installation of labor- saving devices which kills the demand for the type of service hitherto supplied by juveniles; or whether the decline is due to a recognition on the part of the parents of the need tor more thorough intellectual and technical preparation for an industrial or com- mercial career. Whatever further research indicates to be the cause of this pronounced decline and what- ever may be the trend which the coming years will disclose, there can be no question whatsoever that this decline occasions a heavier demand upon the secondary schools and also upon the colleges and universities. The Seasonal Indexes The importance of the round of the seasons has been rather neglected in studies of the juvenile labor market. The following table presents the seasonal ° The decline is not due markedly to the war hump since the number of new employment certificates issued in 1914 was 4508 as compared with 3686 for 1915. See Table IV in Fuller and Strong, loc. cit., p. 23. We have records of employment certificates issued by the School Committee in Boston since 1894. As many legal changes occurred between that date and 1913 when the new law was pass d, the data are not homogeneous en ugh tor acute analy SIS. The linear trend, however, for that span was upward. The data are as follows: (894 21443) 190) 2208 1908 2205 1895 1784 | 1902 2528 1909 | 4702 1896 1457 1903 | 2817 | 1910 | 5046 1897 ) 1395 1904 2918 1911 | 5562 1898 | 1456 i905 | 3590 | 1912 6305 1899 ! 1976 1906 | 4350 | 1913 9973 1900 1991 1907 477124 JUVENILE EMPLOYMENT CYCLES indexes of each of our six basic series. The chart which follows presents them pictorially. TABLE VII SEASONAL INDEXES OF THE SIX CERTIFICATE SERIES \ Educa- Educa- | | Educa- Educa- Rennie Homaly Ty) thomas |) eer as tional tional Mo | &mptoy~ |Certificate: i\Certificates| “™P-°Y- |Certificates\Certificates 7 fonth C ment School At-| School At- |, ment School At-| School At- I Breck Vertincates tendance | tendance |\ertilicates! tendance | tendance ssuance| | Compul- | Voluntary Compul- | Voluntary SOTY » Ty : tg : | | Ne | ve Vew | Kerssu | -eissues| Retssues ere ee meee ee ee ee Jan 53 faa leet 8 al 208 107’ ||. 9.97 Feb 10 95 65 70 $7 76 Marcel 58 | 127 38 95 111 102 ipo tee oe a —|- eS ee April 48 | 73 i 7] 85 94 I ee |e Ee % es Oy | Ss ie tee + ao | ~ Oo? c viay 61 i : SO ' Ss / ) OO | IH —_——— |} —————_—__|___—_- ——} el -—-- ~ ee i | s June 276. | 64 | 1438 ii oi 1 109 oe a eepeiere leon on : | = E July Ties ls 4 12057184055 95 Aug L543 3|) 61 94 69 85 | 102 SEE ti eo Peto et 30 34 121 oy ——___|_______|—— | = ae Oot. eel etis.p| 156 113 52. 151 Sele } Paki Hise) SORE: Cae * oo ie a A Nov. | 16 136 100 ae 120 101 | Dec 54 | 109 92 30 SO | 83 It will be noted from even a cursory examination of both of these statistical devices that there is a profound variation which accompanies the change of the seasons. The “New Employment CertificateINDEXES CERTIFICATES CERTIFICATES, J C Z O (" C LJ /) me ~»y « ~ — . - - oe U e" ~~ EMPL‘ NEW NEW COMRA. SOR GTTENPANC AT TEN PANCE OL S5sCHOX NAL & & ‘HOO L 7 CERTIFICATES 7 ~ 7 4 e > _ ~ reis* FMMIPLOYMENT 4 COMPULS OY ATTENDANCE SCHOOL AT TENPANCE O SCHOOL NO ES ‘ REISS ErPUCA TIONAL26 JUVENILE EMPLOYMENT CYCLES Series’’ (Series Number 1) has its peak load during the summer and early fall months, and this peak has a double top, namely, in June and September—the months in which school closes and opens. When we turn, however, to the ‘““Employment Certificate Re- issues’’ (Series Number 4) we find a different seasonal variation. For thisseries the Junerise has disappeared and the peak load occurs in the fall months corres- ponding essentially with the intensification of those industries in which these juveniles are employed. The ‘‘New Educational Certificate Series, School Attendance Compulsory’’ (Series Number 2), has its peak load occurring during the fall and winter months, with a spring rise. This same seasonality is to be noted concerning the other series affecting children for whom further schooling is mandatory, namely, Series Number 5. This feature is prob- ably due to the tendency on the part of retarded children (predominantly to be found in these two series) to make their exit from school at the earliest possible moment without regard to terminal dates of the school system when they might be promoted to higher grades. After discussing retardation of child- ren who take their first position during the school year and concluding that there is a pronounced tend- ency for children of this sort not to await the begin- ning of the vacation period, Dr. Woodbury writes that ‘‘this may be ascribed partly to the fact that a child who leaves school during the school year loses the chance to complete the grade last entered, but it undoubtedly indicates also a greater tendency on the part of the retarded than on the part of other children to drop out of school at the first opportunity re-JUVENILE EMPLOYMENT CYCLEs gardless of the completion of any unit of school WOLK. When we turn to the New Educational Series,” where no further school attendance was mandatory (Series Number 3), we note essentially the same re- sults as those noted for Series Number 1, namely, the double peak, one occurring in June and one in September. The one principal difference is that these two peaks are relatively less important in the former than in the latter. This is undoubtedly the result of the same influences we noted as operative in Series Number 2 and Series Number 5, namely, the ten- dency to leave school at as early a date as possible regardless of promotion dates. Consider next the ‘‘Educational Certificate Re- issues Series’”’ where no further school attendance was imposed by law (Series Number 6) and note essen- tially the same features as in the case just discussed. In this case there is an interesting difference; the peaks are less pronounced and there is present a spring peak. Glance at the chart once more and disregard the individual series. There is to be noted an upswing in all the different series from February to March. Again, there is present a rise for all the series from August to September; then, with one exception, a decline to October and thereafter a decline to Decem- ber. On the whole the indexes rise from December 32 to January. 31 Loc. ctl., p. 138. 32One aspect of these indexes is interesting. It concerns the amplitude through which they move. The Standard Deviation for these series 1S as foll ws:28 JUVENILE EMPLOYMENT CYCLEs The Business Cycie We turn now to a presentation of the relationship between these several series and certain correlatives, representative of business fluctuations. We propose to discuss the following items in turn. 1. A description of the cyclical fluctuations of the six basic series as mentioned above and several combinations of these basic series. bo We then proceed to study graphically and math- ematically the correlation between certain of these series and a number of accepted economic series, such as a. Three well-known employment series;** (1) Employment Index of the United States Bureau of Labor Statistics, Series Number Standard Deviation ee oe ae eee eee oe 60.7 ? Sire OE ee es gee Ns ee ata: alo) ahs 24.5 Der ela 5 cle w wiw.e uw ain wie aie o ae w/a © 46.8 Re Re ee ow cia al ay whe tie ue pis 21.3 ee OO ee wie cletelmip him bunid ale sis 35 It is to be noted that the two highest standard deviations are reached when we are dealing with the Employment Certificate series, either new or reissues. In other words, the seasonal variations are greatest in children 14 to 16 years. Of the series concerned with the educational certificates, the greatest varia- bility is found for those children who are under obligations for further schooling and are going to their first job. There the standard deviation is 33.7 while for those who do not have further schooling to do the standard deviation is 24.5. The percentage of difference from one series to another is interesting. Thus, between the ‘‘Employment Certificates New” and *‘Employment Certificates Reissued”’ there is a difference of 13.9 or 22.9 per cent of the former; between Series Number Two and Five a difference of 12.6 or 37.4 per cent of the former: between Number Three and Number Six 11.0 or 44.9 per cent of the former: between Number Two and Number Three 9.2 or 27.3 per cent of the former: and between Number Five and Number Six 7.6 or 36 per cent of the former. (The standard deviation is the device which statisticians most frequently utilize to express the dispersion of statistical data. When items are expressed as fractions of the standard deviation they are called standard units. See, for examples, the tables in the Appendix.) $3 These three employment indexes are differently composed. For a de- tailed description of each see Berridge, loc. ctt., pp. 134-136. In calculating the coefficients of correlation later we utilize for reasons mentioned there the Index of the Federal Reserve Board. ‘‘Industrially it is much more compre-JUVENILE EMPLOYMENT CYCLEs (2) Employment Index of the Federal] Reserve Board. (3) Harvard Employment Index: b. Day’s Index of Manufacture: The Harvard “B” Curve: d. Help Wanted at Three Massachusetts Em- ployment Offices;*: e. Labor Quit Rate in Selected Factories:*5 5 4 “SS f. Snyder’s Clearings Index;* g. Persons’ Index of Trade. The following tabular summary depicts the cycles in the various series, individually and in combination. } . ; tr) L ror which a sultaDi ’ i 4 » + and tor any part ol compared with 1919, 1921, and made it possibl dustry, but in g the part of th or more) of the in the result the cyclical December, 192 reasons the Fed both the other t' 34 Data kis m ifacturing industry ; ; = ; ; 7 — } eries coul ce ODLaInNed Irom al bur 1, State or tederal, . ; VT yy ' . ly ' 1 ' ; ; , post-bellum period Mach specimen index was carefully monthly variations for that entire industry in the years 1923 yt if m the censu I Manufactures This te eries which did not truly represent a given in- ral the tests showed such surprising representativeness on rting samples that the great bulk (three-fourths country s tactory workers may be considered as represente It is nstructed as to reflect the secular trend as well as seas ict tions.”’ See the Federal Reserve Bulletin 1272-1279, and (May, 1925), pp. 324-330 kor these ral Keserve Board Index runs higher on the charts than do Wo 1n ie) eral Reserve Bar 33 W. A. Bern per ts, I ne Revieu 140: ibid, ““Your pp. 1-4 Thi Metropolitan Life | sv bility and kindre % Carl Snvd Journal of The (September, 192 4 : supplied by the Statistical Department of the Boston Fed- lige, ‘“‘Labor and the Busine Ls rT 1 ‘ss Cycle: Some Industrial As- of Economic Slaitstscs, VIII, Number 3 (July. 1926). 136 ibor Turnover: Good or Bad?’’, Factory (September, 1926 em contains a list of the companies co-operating with the insurance Company in producing this index of labor mo- indexes ‘A New Clearings Index of Business for Fifty Years,”’ rican Slatustscal Association, NIX, New Series Number 14730 JUVENILE EMPLOYMENT CYCLES TABLE VIII CYCLICAL FLUCTUATIONS OF THE SEVERAL CERTIFI- CATE SERIES Number of Description Series of Series 1. New Employment Certifi- cates 2 New Educational Certificates, School Attendance Compul- sory 3. New Educational Certificates, School Attendance Volun- tary 4. Employment Certificates Re- issued Educational Certificates Re- issued, School Attendance Compulsory 6. Educational Certificates Re- issued, School Attendance Voluntary Combination of Series Num- ber 2 and 3 8 Combination of Series Num- ber 1, 2, and 3 cy ~ 9 Combination of Series Num- ber 5 and 6 10. Combination of Series Num- ber 4, 5, and 6 Description of the Cycle ET The cycle is well defined and very much _ better than Numbers 2 and 3. The series is cyclical but extremely sluggish. Cyclical but very irregular The cycle is very good Cyclical but extremely sluggish Excellent cyclical move- ment Poor cyclical movement The cyclical movement is slightly better than Num- ber 7 but still poor. Excellent cyclical move- ment Excellent cyclical move- ment. For synchronous pairing the coefficient of correlation is +0.84 with Series Number 9. A comparison of the various curves as shown above suggests two very important conclusions concerning which we shall have something further to say. These two conclusions are as follows:~ JUVENILE EMPLOYMENT CYCLEs 31 a. he series composed of certificates granted to children going to their first positions, for both age groups and both schooling groups, respond with a less well defined cyclical movement than do the series composed of children who are shifting jobs or who are returning to work after a lay-off or a quit. b. Ihe two series concerning children going either to their first or to subsequent positions, and for whom further school is obligatory are extremely sluggish in their cyclical movements. It is to be recalled in this connection that these data concern children who are somewhat retarded in their schooling, especially as compared with the children in the other series. Precisely what proportion is mentally defective in this series can not be stated. The proportion of such children must, we believe, be rather large. Evidently both the ebb and flow of business, with its alternations of employment possibil- ities for juveniles, affects these retarded children less than it affects those who are more advanced in their schooling.’’ Whatever the response to these ups and downs of business this group of children makes, it surely requires much longer to register. And this lethargic response does not arise from the fact that the group bulks large in volume. Quite the contrary. Recall, 33 Helen T. Woolley, An Experimental Study of Children (New York, 1926), pp. 605-606, concludes “‘that rank in mental tests has still less to do with earn- ing capacity or regularity of employment during the first four years in industry than rank in physical tests. In this case the correlations with earning capacity are positive for both girls and boys (with one exception). The correlations with factors of regularity are so small and so inconsistent that they seem to indicate no relationship whatsoever, except that girls have consistently nega- tive correlations with the number of positions held, showing a tendency to less shifting among mentally superior girls.”’32 JUVENILE EMPLOYMENT CYCLES too, that the round of the seasons likewise af- fected this group of children very slightly. This raises a very puzzling question and one upon which further light needs to be shed ere a defin- itive answer may be proposed. It is this. If one assumes that the retarded children are the last to be affected by opportunities of new positions and consequent higher earnings which arise with an improvement in business, one would have likewise to grant the fact that these children would constitute the first ones to be discharged or layed-off with a decline in business. A glanceat Chart IV willindicate an interesting feature in this connection. It re- quires only three years from the high point on the curve representing these data to the low point (February, 1916 to April, 1919) but thence it requires four years to reach the high point again (April, 1919 to April, 1923). In other words, the descent is more rapid than the ascent. It may well be, however, that on the decline these children take positions lower in the scale of remuneration and so remain in industry. One question, however, remains. In sociological time series, one meets very frequently with data which, as in the case of the series in which further school attendance is mandatory, reflect an extremely sluggish or slow response to the cycle. Such series which are plainly cyclical but ebb and flow are ex- tremely slow to register. These series are in direct contrast with sensitive series in the economic provinceLV CHART ISSUED EMPLOYMENT CERTIFICATES R a NEW EMPLOYMENT CERTIFICATES aay ND e-—-- EDUCATIONAL CERTIFICATES REISSUED (comPuLsory) es eeepc ee ee J aa wee ee ee ion34 JUVENILE EMPLOYMENT CYCLES where sluggish series, while met with, are rarer than in the sociological province. The question which re- mains and which warrants rigorous study is to de- termine which type of series better typifies the re- percussions of the business world. We turn next to a detailed description of the course of the cycle of each series from 1915 onward. For a graphic presentation of these curves see Chart IV and Chart V. New Employment Ceritficates. Beginning with 1915 the curve is below the line, but about the middle of that year it turns upward and crosses the line about the middle of the year 1916. Except for occasional dips below, the curve hovers around the zero line, registering in a minor way in April, 1917, our declaration of war, and begins a rapid and sudden rise in February of 1918, reaching a point almost 3.5 units of the Standard Deviation, undoubtedly reflecting the demand for juvenile work- ers occasioned by the shortage of adult labor. Another and yet higher spurt is registered in July of the same year, betokening the unusually large seasonal em- ployment of children who had finished their school year at that time The curve for the years 1919, and 1920 up to July, is substantially above the line. Then begins a rather precipitous decline and for the rest of that year and the next year (1921) the curve does not get above the line until July 1922. The curve is essentially above the zero line during 1923, but early in 1924 again drops below the line to remain there until July of 1925. After a tendency to dip below again in the Spring of 1926, the curve ends the year above the line.CoE (V, EVUCATIONAL CERTIFICATES, NEW. ATTENPANCE VOLON TARY ATTENPANCE COMPULSORY EPUCATIONAL CERTIFICATES, NEW . EPUCATIONAL CERTIFICATES, REISSUEP. ATTENDANCE VOLUNTARY ONTTS STANDARC36 JUVENILE EMPLOYMENT CYCLEs New Educational Certificates, further school attend- ance compulsory. This series is one of the sluggish series to which we alluded earlier. Beginning with January 1915 there is a gradual decline which, judged by a twelve-months moving average, crosses the zero line in September of 1916. With the exception of a spurt in June and July, 1917 (these are manifestly children who are retarded mentally but who nevertheless concluded the school year), the series remains below the line until the summer of 1920 when there occurred a rise which continued—except for a sudden decline in December, 1920, and January and February, 1921—until the summer of 1924, except for several occasional and minor dips. Thenceforward the line continues down- ward and remains there through 1920. New Educational Certificates, further school attend- ance not compulsory. The year 1915 in this series was essentially one of depression, though toward its close the curve rises rapidly reaching a point almost three times the Standard Deviation. With very occasional minor dips it remains above the line until—as judged by the twelve-months moving average—February of 1919, having experienced during that epoch the immense rise in June of 1918. The smooth curve continues below the line until November, 1922, and after re- maining above the line during the year 1923, declines below the line for 1924. Since February of 1925 the curve has been above the line. In the description of the course of this curve we have relied upon the twelve-months moving average, which does not appearJUVENILE EMPLOYMENT CYCLEs upon the chart. The curve actually charted is, of course, more jagged. J = mae past "(Dy vista ratae I. -) cainp Empblowvme mt CéeTititcales IXé issued. The course of this series parallels excellently the course of Series 1, described above. The amplitudes of both curves are essentially alike. The New Em- ployment Certificates series seems to lead the present series by one or two months. ca ’ 7 mia wt ametata fst DP oa ci ssadct f Sy be ~_ =a “> + ILQAUuca UU FbUld ( Ciil pa (LUG? ALCLIOUH (i. ILUTTNel SUilOf¢ eL- tendance mandatory. This series is the sluggish series alluded to earlier in this paper. The curve depicting its course is above the line through 1915 and 1916, but in 1917 begins a gradual decline, crossing the line during the early part of the year and, practically speaking, remains there (except for a modest rise toward the close of the school year of 1918) for five years crossing the line on its upward course during the summer, 1922. From the peak reached in this swing during the early months of 1923 there has been a gradual but rather pro- nounced decline until the close of our period. Educational Certificates Reissued, further school at- tendance not required. This series shows a very excellent cyclical move- ment. From a point rather well below the line (almost twice the Standard Deviation) during the first seven months of 1915, the curve very rapidly rises, crossing the line in January, 1916, and, con- tinuing upward, reaches its maximum July, 1918. Then begins a gradual decline which becomes greatly38 JUVENILE EMPLOYMENT CYCLES accelerated in the Spring of 1919. After a pro- nounced rise in the Summer of 1919 there begins a prolonged decline, the curve crossing the line on its downward sweep in September, 1920. The lowest point (almost twice the Standard Deviation) occurs in January, 1921, whence a slow upward movement commences. On this upswing the curve crosses the line during midyears, 1922, and after remaining above the zero line throughout practically the whole of 1923, swings downward for the year 1924, and for seven months of 1925. Since that time the curve has been above the line. Comparison With Certain Other Economic Series We turn next to a presentation, an analysis of the cyclical fluctuations in conjunction with a number of accepted economic series. New Employment Certificates. Chart Vla presents the cyclical fluctuations in the New Employment Certificates and three rather well- known employment indexes: the one issued by the Har- vard Economic Service, one compiled by the Federal Reserve Bank, and thethird the product of the Bureau of LaborStatistics. The pairing obviously is very good, except since June, 1925. Greater irregularity is to be noted in the juvenile series, the result perhaps of the smaller sample with which we are concerned. Sim- ilarly, the amplitude of the juvenile curve is less than the amplitude of the senior series. Of the threex Ww G c x at — Y G4 | 2 Zz lteschae [ae 4 | > | - Bet poseanad a iearaee Soaaeminae E ' Sr \ : le 2 4 | io mite. OL Sa \ paca SE 3 ! . ! y “Ys ‘ 3 - ei x! q TT x i) Wi ul) SS en | ae y—-__—— dl 0 - § Sj Lo CS! a ala a b + i@: . Segoe ley ee a 5 Ww a Bre elie - - ¢cw> ah PO ee pe iW Ta wate: | a u a > OVy} : = Ke. W ig | = aioe bed ay ¢ — © = wo ul) a ce 4 | Suc ——_- =r le >, oe as Bah OD tan Se reese are Te = \2°ouL Ne % elm Salata: it a) iv ——— —— 4 = | eT |40 JUVENILE EMPLOYMENT CYCLES senior series, the Federal Reserve Board Index is closer to the juvenile curve and, on the whole, the latter leads the former with the exception of the year, 1920. Chart VIb shows the same juvenile data 1n com- parison with Professor Edmund E. Day’s Index of Manufacture. The pairing is excellent throughout the whole span, the amplitude of the manufacturing series being slightly larger. Note the simultaneity in the fluctuations. The following tables present the coefficients of correlation reached by comparing the juvenile series with the Federal Reserve Board Employment Index and with Day’s Index for various epochs and for dif- fering degrees of lag or lead.*”* 374. The coefficient of correlation is a statistical device for measuring the concomitance between different series of data. It will be well to explain just what is meant by “‘lag’’ and ‘“‘lead”’ in correlating time-series. If A goes east invariably at the same instant that B goes east, we shall have perfect posttive correlation. If A goes west invariably at the same time that B goes east, we shall have perfect inverse or negative correlation. This time relationship we have called a synchronous or simultaneous relationship. If A goes west invariably five minutes after B goes west, we secure perfect posstive correla- tion, with A lagging five minutes (or, what is the same thing, with B leading five minutes). If A goes east invariably five minutes after B goes west, we secure perfect negative correlation, with A lagging five minutes. This degree of covariation is expressed by the coefficient of correlation varying between +1,00 and —1.00, passing through zero. It is plain that this relationship is not necessarily causal. Since everything that happens must occur in time, in correlating time-series particularly must the coefficient be interpreted with great caution, since from a certain point of view all time-series are correlated to some degree or other. In coefficients secured from time-series, Moreover, we cannot control our results by an appeal to the statistical constant known as the ‘Probable Error.’’ See the Presidential address by Professor Warren M. Persons to the Eighty-Fifth Annual Meeting of the American Statistical Association. This is the first chapter in Number 6 of the Pollak Publications, The Problem of Business Forecasting.Vib SEAN —— NEW EMPLOYMENT CERTIFICATES SCTURE ih A MANO INDEX OF ~ * = ; ~~, —* f Se et i926 —_ i mae cl = | tx — - } eel = a > = ~ — ee ——_-- — — ns _— —— -+ —____ "=o et | ff, =. 8 ee) ‘ — : -_ ) m1 —" fe ee el ee ee —_ FF . a*h., - eda” ny42 JUVENILE EMPLOYMENT CYCLES TABLE [Xa COEFFICIENTS OF CORRELATION New Employment Certificates and Day’s Index of Manufacture New Employment New Employment Certificates Synchro- Certificates Epoch Lag | nous Lead 2 months | 1 month | 1 month |2 months | a eee emeee ye Ns SEE se EN 1920-23 + .818 + .842 + .862 + .856 +.799 24 + .687 + .722 + .752 + .721 + .656 25 + 569 + 614 + .657 + 620 + .558 1919-21 +. 889 | +.929 + .945 + .881 + .$19 22 + .775 + .797 + .794 + .740 + .676 23 + .684 + .706 + .709 + .672 + .614 24 + 597 + 628 + 644 + 591 + .526 25 + 509 + .549 + .577 +.523 | +.460 | TABLE [Xb COEFFICIENTS OF CORRELATION New Employment Certificates and Federal Reserve Board Employment Index | New Employment New Employment 3 Certificates Certificates Epoch Lao Synchro- Lead = nous 2 months | 1 month 1 month |2 months 1920-23 + .764 + . 844 + .870 + .878 + .878 24 + .661 + 743 + .776 + 791 + .790 25 + 536 + .610 + 644 + .661 + .661 1919-21 + .853 + .849 + .883 + .905 + 902 22 + .707 + .781 + .808 + .819 + .815 23 + .599 + .660 + .681 + .700 + .700 24 + .542 + .605 + .634 + .649 + .644 25 + .451 + .510 + .541 + .558 + .554Vic CHART EMPLOYMENT CERTIFICATES - NEW HARVARD B CURVE44 JUVENILE EMPLOYMENT CYCLES It will be noted that the coefficients are on the whole rather high, especially for the three year period, 1919-1921. It is plainly shown that the juvenile series leads the Federal Reserve Board data while it fluctuates synchronously with Day’s Index. Value is added to these coefficients because they rise from the period of maximum lag to the point where the coefficient is a maximum and then a decline sets in. For example the coefficients of correlation between the juvenile series and the production series were cal- culated for a time, relationship running from four months lag through simultaneity to four months lead. The coefficients are as follows: +.873, +.881, + 889, +.929, +.945, +.881, +.819, +.784, +./769. Professor Allyn A. Young has written that “when the coefficients found for different periods of lag increase steadily as the period which yields the maximum coefficient is approached, and when that coefficient is itself tolerably large, there is good reason for put- ting reliance upon it.’’*8 Note also the high value of the coefficient for the juvenile series and the produc- tion series when the former leads the latter by one month;namely, +.945, and consider this in connection with the coefficient reached by Professor Berridge*® in correlating his adult employment series with the same Index of Production and for the same period of time. In that case the coefficient was .96. We may make the general observation also that when the year 1919 is included in the epoch correlated the higher cor- 38 Introduction to Social Consequences of Business Cycles (Boston, 1925), mi UX: 39 W. A. Berridge, op. cit., p. 55. Disregarding for the present the discus- sion whether probable errors for coefficients of correlations secured in time series serve the same purpose as those reached in biometric investigations the ee coefficients are not, with respect to their probable errors, significantly different.Vid CHART CERTIFICATES IPLOYMENT . Ff Cc EPIPLOYMENT OFFI Ce >» ~~ Pile ik ANTEV ‘ Vie ‘GHAR Vie NEW EMPLOYMENT CERTIFICATES ANP IN SELECTED FACTORIES lu ee ¢ i = 5 G tv O Ce ¢ 4 i ! i ! 192+ iy92>5 i9i7 i9\ia i9ig \9 ivel I9SIG i9vItsJUVENILE EMPLOYMENT CYCLEs relations are found when the production series is the correlative. When we begin our pairings with 1920 the Federal Reserve Board data on the whole provide us with higher correlations. Chart VIc portrays the cyclical fluctuations of these juvenile employment data with the similar fluctuations in the Harvard ‘“B”’ Curve. Chart VId shows the cyclical fluctuations in the New Employment Certificates issued compared with the cyclical variations in a series reflecting Help Wanted at Three Massachusetts Employment Of- fices. Chart VIe demonstrates these same cyclical fluct- uations compared, however, in this instance, with the ebb and flow in the Labor Quit Rate in Selected Factories. New Educational Certtficates. further school attend- * ance compulsory. We present now Chart VIIa and Chart VIIb com- paring the cyclical fluctuationsin the New Educational Certificates series for children of whom further schcol- ing was required, with certain of the business correla- tives used just above. It is not necessary to repeat that the sluggishness of the juvenile employment data in this instance makes graphic and also mathematical comparisons difficult, if not entirely impossible. What does emerge from the picture is a conviction that the series is cyclical but extremely slow to register changes, except in some short periods.CHART VIla x Li) Aa zZ a(S Or 2 4 uJ ae 3 O aH ae WwW _ x YY ; 27 WA Gi Vez Ss -- = a er £5 ¢ hu EE > 89 a oe é Jud ue 2 - Scr < Yo D1CcE bn Dy Ra 4 uae qc kwu z ¢w 5: Evian q ote Os a> 45zEq au acZwt iD We a! 1 al i927 1926 1925 ~~ 1924 1923 s at = | Sea Sea ek | ale Met idee ttc ale ees eae i alee aa | ein a ee a | cae o a | eraarat af Ae = 7 On SH 8CHART VIIb COMPULSORY NPANCE ~ 1 ATT EDUCATIONAL CERTIFICATES - NEW. ------— DAY'S _[ — MANUOFACTUR INDEX OF50 JUVENILE EMPLOYMENT CYCLES New Educational Certificates, further school attend- ance not compulsory. There follows now a series of charts (VIIIa through VIlle) portraying a comparison between the series composed of New Educational Certificates issued to children who have complied with the educational re- quirements of the state code and the same five curves from the commercial world as hitherto employed. It will be instructive to present again two tables (Xa and Xb) containing the coefficients of correlation reached by pairing this juvenile series with both Day’s Index and the Index issued by the Federal! Reserve Board. ‘LABLE >Xa COEFFICIENTS OF CORRELATION New Educational Certificates (attendance voluntary) and Day’s Index of Manufacture New Educational New Educational Certificates Svnchro- Certificates Epoch Lag “nous Lead 2 months 1 month 1 month |2 months 1920-23 + 696 +710 +..748 + 726 + .672 24 + 566 + .639 + .681 + 621 + 538 25 + 577 + 559 + 652 + 576 + 504 1919-21 + 557 + 731 + .761 + 681 + .609 22 + 513 +. 538 + 605 + 599 + 536 23 + 531 + 568 + 610 + 595 + .540 24 + 467 +- 533 + 578 +- 530 + 451 25 + 476 + 480 + 568 + .504 +-.435Villa CHART @TTENPANCE NOT COMPULSORY NEW EPUC@TION@AL CERTIFICATES. UNITED STATES BOREAL OF LABOR STATISTICS. FEDERAL RESERVE BOARD. —e—- HARVARD EMPLOYMENT EMPLOYMENT INDEX INPEX EMPLOYMENT ee ee | INDEX52 JUVENILE EMPLOYMENT CYCLES TABLE Xb COEFFICIENTS OF CORRELATION New Educational Certificates (attendance voluntary) and Federal Reserve Board Employment Index New Educational New Educational Certificates Svnchro- Certificates Epoch Lag “nous Lead 2 months 1 month 1 month |2 months 1920-23 + 536 + .595 + 653 + 709 +. 849 24 + 441 + .512 + 574 + .624 + .723 25 + 303 + 381 + 444 + 492 +575 1919-21 + 510 + 671 + 663 + 649 + .795 22 + 474 + 546 + 556 + 610 +. 748 23 + 430 + 492 + 504 + 560 +657 24 +. 369 + 440 +. 463 + 505 +585 25 + .261 + 338 + .368 + .409 +..477 As in the case of the New Employment Cer- tificates the best pairing is a synchronous relation- ship when Day’s Index is used as the correlative and the timing when the Federal Reserve Index is used is best when the juvenile series leads two months. While not as large as where we paired the New Em- ployment Certificates, the coefficients are still high and regularly distributed in approaching or leaving the period for which the largest coefficient is secured. Just as in the previous set of coefficients, on the whole the larger coefficients are reached when using Day’s Index than when using the Federal Reserve Index as a correlative. Employment Certificates Reissued. We now turn to a group of charts reflecting the comparison in the cyclical variations in the Employ-NOT COMPULSORY ATTENPANCE CHART VIIIb NEW EDUCATIONAL CERTIFICATES. INDEX OF MANOFACTORE eeeGHA Viltc ATTENDANCE NOT COMPULSORY NEW EDUCATIONAL CERTIFICATES . ANS w——-=—— HARVARP “B” CORVE igiT7 i9QiS i91sSVIlId CHART OTTENPANCE NOT COMPULSORY ANP HELP WANTEP AT THREE EMPLOYMENT OFFICES NEW EDUCATIONAL CERTIFICATESGrIAR I: Value @TTENDANCE ‘NOT COMPULSORY IN SELECTED FACTORIES NEW EDPUCATIONGL CERTIFICATES. AND? —--— LABOR QUIT KATE LE Ona shi ae Ww) | a Sa N 2 | o Spaeperenenetieemats : cia = N = : Lala ‘. | papel | 0 Cow mace (ei: Male 1924 I9e2e ig25 i92I 920 ' | /Y | i =e Pic se ena : [asst ee pos, = = | ae a < : eae aL osetia | f i919JUVENILE EMPLOYMENT CYCLES 57 ment Certificates Reissued and these same economic barometers. This juvenile series is composed of children 14 to 16 years of age who are changing jobs for any reason, as distinct from the children of the same age who are leaving school for their first em- ployment. Chart [Xa shows this juvenile series in connection with the three employment indexes. Beginning with the year 1920 until 1926 the pairing is good. In amplitude the juvenile curve is closer to the Federal Reserve Board material. When we turn to Chart [Xb we note a comparison with Day’s Index. Here the pairing is good, except for the year 1921. The amplitude through which these curves range is less for the juvenile series than for the production series. Dr. Berridge’s study reached the same conclusion. Chart I[Xc compares this juvenile series with the Harvard ‘“B”’ curve and the relationship between the two is quite obvious. Chart [Ad presents a com- parison with the Help Wanted at Three Employ- ment Offices while Chart [Xe uses as a correlative Dr. Berridge’s Labor Quit Rate. In this case the pairing is good except for the periods March, 1922, to August, 1923, and June, 1925 to 1926. The ampli- tudes are essentially equal. There is no clear time relationship noticeable except a modest tendency for the Labor Quit Rate to lead. What are the coefficients of correlation for this juvenile series which were reached by using two economic correlatives, Day’s Index and The Federal Reserve Index? These we present in the following two tables numbered Xla and XIb.CHART [Xa —— EMPLOYMENT CERTIFICATES REISSOED ANP © WJ _ Oo - Ql © aie Ligsialt x | te uJ | | G } zZ | | % eet a <{ 3 - wz zh 7 ere 2 Or ys re 74 $ 4 | * ow ST Acs $235: eI Ws aS ‘ ¥ e “ALS le ad Se ve Y a } - ke -z | a co | WEL ni ‘ S = o a> | qo | A gre | ; hj & WN bod | o Al } | 4 => | oe q 4b Fee ‘ g 28| | s OES. lid | eet | | ne| | Dye ) p eee 4 | =~ | | , > > Is F 2 o O72 — ° -- 210ee REISSOED Ww 4 = Nn O og G5 5 Ee “ == ae 7 2F ye x x a 5 pte cont SS SS = - Z O a a P= Ww ee DAYSJUVENILE EMPLOYMENT CYCLEs TABLE Xla COEFFICIENTS OF CORRELATION Employment Certificates Reissued and Day’s Index of Manufacture Employment Employment Certificates Synchro- Certificates Epoch Lag nous Lead 2 months 1 month 1 month |2 months 1920-23 + .556 + 577 + .494 + .489 + 432 24 + 480 + 497 + .429 + .411 + 358 25 + 361 + 377 + 314 + .295 + .247 1919-21 + .607 + 647 + 600 + .637 + 546 22 + .522 + 545 + .516 + .578 + .483 23 + 426 + 450 + 426 + .448 + .388 24 + 382 + .403 + 384 + 390 + 333 25 + 294 + 314 +-.292 + .293 + .241 TABLE XIb COEFFICIENTS OF CORRELATION Employment Certificates Reissued and Federal Re- serve Board Employment Index Employment Employment Certificates Synchro- Certificates Epoch Lag nous Lead 2 months | 1 month 1 month |2 months 1920-23 + .645 + .659 + .961 + .649 + .592 24 + .592 + .608 + .850 + .592 + .537 25 + .526 +-.542 + .746 + .535 + .490 1919-21 + .634 + .643 + .636 + .607 + 528 22 + .620 + .641 + .768 + .642 + .590 23 + .498 + .516 + .762 + .518 + .476 24 + .479 + .497 + .701 + .492 + .448 25 + .439 + .457 + .635 + .452 + 420ANS HARVARD “B CURVE ivi? ft — > _ ——/+1 er D -_—— EMPLOYMENT CERTIFICATES REISSUOEDCHART [Xd — EMPLOYMENT CERTIFICATES REISSOEP ANP ae HELP WANTED @Tr THREE EMPLOYMENT OFFICES = ON aie o | e ay Ne =| i >| t & =< > a — i» is 3 o © ~ 0 - ©[Xe CHART REISSUED CERTIFICATES EMPLOYMENT SELECGIED. FOCIORIES IN QUIT RATE ---- LABOR64 JUVENILE EMPLOYMENT CYCLES Note first the very high coefficient ( +0.961) reached for synchronous pairing between the Federal Reserve Board and the juvenile series for the period 1920 through 1923. For the six years, 1920-1925, the coefficient is also very good—-+.746. Whereas in the previous two sets of coefficients presented the better correlative was shown to be Day’s Index, in the present case the better economic variable is the Federal Reserve Board Index. In other words, for children making their first entry into the labor market a production series is a better correlative; for the same type of child once in the labor market a better picture is secured by looking at an employment index. The juvenile series lags behind Day’s Index one month, while the best coefficients are secured for simultaneous pairing with the Federal Reserve Index. The tables also reflect a rather poor pairing between the curves for the year 1919. Educational Certificates Reissued, further school at- tendance compulsory. Now we present the same series of economic bar- ometers in connection with the Educational Certif- icates Reissued series in which the children have not complied with the educational requirements in charts numbered Xa through Xe. This juvenile series it will be recalled concerns the children who are retarded in school and who have been in the labor market for a time and are changing jobs for any of a number of reasons. As we are here concerned again with a sluggish series the pairings throughout are on the whole very poor, except for very brief epochs, such as, for example, three years subsequent to June, 1922,) . a EPUCATIONAL CERTIFICATES REISSUED (comPpuLsory — CHART EMPLOYMENT INDEX INDEX EMPLOYMENT INDEX U.S.BUREAQAU OF LABOR STATISTICS. —-—— FEDERAL RESERVE BOARD. — ae — | + a — | HARVARD EMPLOYMENTXb CHART —— EDUCATIONAL CERTIFICATES REISSUED (comPuLsory) ANP DAYS INDEX OF MANUFACTORE 2 + = N 9 otis \ eT oT ieee r v V 1922 1924 i92s 1926 1927 i923 i92i 1920 ivi? 19is IMT (9I6 ivisANP —— EDUCATIONAL CERTIFICATES REISSUED (computsory) ---— HARVARD “B CURVE enone pesto 1 — : Qo ~ nt i. coal > i- > fj} —_} 1 __ “= 5 Na —— 12 rane e 2 | neo eee ) © he << eg 7 \- 'Xd CHART REISSUED (comporsory) —— EPUCATIONAL CERTIFICATES ANP ---— HELP WANTEP @T THREE EMPLOYMENT OFFICES + © w - 0 - w " | eee : | es Pa } | Se | Ss . —, —— | eter y| Oar toe Lt ae
  • 1 ~ ene ul | 3 ; — n | een, nh nd | i | . {Fi gee fees : e——4 | , | f= + pa x 4 4 toes | e rm ———————— — mw | oe | ' > : - ep a" eo ey 2 ers | - leary al _ c 5 — Paes | were ' tT. : ‘ ; 2 4 — — — a = ——_ —— “ | by rales | } ‘ i =, - Sie — aS 10 | Al 4 : ' sel wt he ee a74 JUVENILE EMPLOYMENT CYCLES TABLE XIla COEFFICIENTS OF CORRELATION Educational Certificates Reissued (attendance vol- untary) and Day’s Index of Manufacture Educational Educational Certificates Certificates Epoch Lag Synchro- Lead nous 2 months | 1 month 1 month |2 months 1920-23 + 847 +- 893 + 934 + 943 + .891 24 + 692 + 727 + 778 +784 + .742 25 + 603 + 621 + 659 + 688 + .620 1919-21 + 839 + .920 +. 978 + 990 +928 22 +-.718 + 668 + 834 + 874 + .817 23 + 653 + 659 +-.776 + 802 + .753 24 + 557 + .560 + 672 + 693 +653 25 + .500 + 493 + 586 + 624 + .560 TABLE XIIb COEFFICIENTS OF CORRELATION Educational Certificates Reissued (attendance vol- untary) and Federal Reserve Board Employment Index. Educational Educational Certificates Certificates Epoch Lag Synchro- Lead nous ) ) i 2 mos.| 1 mo. 1 mo.|2 mos./3 mos./4 mos. 1920-231 +.764| +.843 | +.929 |+.971/+.952|/+.913/+.876 24) +.677) +.752 | +.826 |+.749|+ .862\+ .836)+.3811 25| +.542| +.619 | +.696 |+.739/+.738)|+.725|-+.710 1919-21} +.857| +.930 | +.955 |+.902)+.848/-+ .760)+ .689 22| +.716| +.788 | +.868 |+.870)/+.856|+.820)|+ .790 23| +.604| +.671 + .737 + .761|+.731|+ .702!+ .676 24, +.558} +.624 | +.688 + .704)+ .698!+ .673)+.655 25] +.450] +.529 | +.596 |+.615|+.615|+.601/+.591 | |XIc CHART STTENDANCE NOT COMPULSORY REISSOED. CERTIFICATES EDUCATIONAL HARVARD “B= GCORVE ac. ye f° <* <* . =< _- a? | ay | Soe, le ae aa 35S x7. + a , Seemed ies epee = an” ee76 JUVENILE EMPLOYMENT CYCLES The first feature to be noted is that where the series of Educational Certificates Reissued, no further schooling required, is used as a correlative with both Day’s Index and the Federal Reserve Index we reach much higher coefficients than when we use either the Employment Certificates Reissued or the New Edu- cational Certificates. Accordingly, we are warranted in concluding that children 16 to 21 years old, already in the labor market, register changes in economic circumstances of the market place more sensitively than do either younger children who have been in the labor market or children of the same age who are just entering the market. Further, on the whole, this series provides some of the highest coefficients found in the study. Note for example*?®, the value +.990 for one month lead to the juvenile employment series as compared with Day’s Index for three years, 1919-1921. Compare this coefficient with the one reached by Dr. Berridge for the same time period which was, as mentioned earlier, +0.96. We must direct attention likewise to the fact that when using the Employment Certificate Reissues series as the correlative the best timing relationship is a synchronous one for the Federal Reserve Board Index and with Day’s Index a one month lag re- lationship. In the present instance the timing relationship for the Federal Reserve Board has a tendency to shift over to one in which the juvenile series leads. This is definitely the case when we turn to Day’s Index. Here the juvenile series leads one month for all epochs. 40 For this same period the coefficient for pairing with Berridge’s Labor Quit Rate was found to be +.747. Then these coefficients successively decline through the following values as succeeding years are added to the epoch utilized in the calculation: +.656, +.607, +.603, and +.516.XId CHART ATTENDANCE NOT COMPULSORY —— EDUCATIONAL CERTIFICATES REISSUED. EMPLOYMENT OFFICES HELP WANTED AT THREE ——— + aoe - a —w G LS — Se ee —_—_-"" =. —_—_— _ —— : - - Lo? a qen” ey orn ” Gos ‘. a a te — >= ——— eae. he ‘ poh ieee ft: boro” iccetnaee tees oe N - ’ ~ ao" ee 2 ne ss 7 ~~ P —_—_—_ -~ — a - _— : - i Cun : » | < ee - 7 ‘> + gd cneeremememnemnnatprenemnenneee Zz os a Still canaeeed commpapaieed iemeetenes ee eeene eeteeeeentaeeeet - : 7 t r a. —— *~ to } ——— — + pS pe pS -——— = =~ _— SeXTe CHART —_ EDUCATIONAL CERTIFICATES REISSUED. ATTENDANCE NOT COMPULSORY AN? IN© SEEECTER FACTORIES ---— LABOR QUIT RATE Ola Ke iO ee el r : =>-— = “ ¢ > SEY mais Kt > ae e ") — a — : QQ : = \ + ra Ale ze ‘ P = = eo : ee 7 Paes —_ 1 = | | | b = 5 } I eS o | e pee | ats 2 elie | 4 © W - Oo - ~ ”JUVENILE EMPLOYMENT CYCLEs 19 Total Educational Certificates Reissued. There follows now another series of charts which concerns the Total Educational Certificates Reissued. This obviously includes children of both groups, 16 to 21, for whom further schooling is prescribed and also those who have met the educational minima prescribed by law. Again we utilize these same economic barometers to make the relationships graph- ically plain. Chart XIla portrays this series in jux- taposition with the three employment indexes. The junior-serles obviously is not so steady in its ebb and flow as are the senior data. The pairing, however, is excellent until August, 1925. The amplitude of the fluctuations is essentially alike but with a greater kinship, as in earlier series, with the Index of the Federal Reserve Board. The junior series exhibits a tendency to precede the senior data in changes in direction. The next chart, XIIb, contains these junior data in connection with Day’s Index. The pairing is very good and the amplitudes are on the whole compar- able. Note well that the pairing of the present em- ployment index is much better with Day’s Index than are the other employment indexes exhibited in this study subsequent to August, 1925. This point merits much further study by barometric economists, since the spread between the employment series and the production series could be explained only by postulating a vast increase in productivity per man employed. The Harvard “B’’ Curve appears in connection with this present series in Chart XIIc. The pairing is good, with the exception of the first nine months of80 JUVENILE EMPLOYMENT CYCLEs 1925, with the amplitude of the certificate series slightly smaller. The certificate series also appears to lead the “‘B” Curve from three to five months. In Chart XIId we utilize as an economic barom- eter the Help Wanted data. The pairing, with the exception of the fall of 1919 and the spring of 1920, is good until August, 1925. The amplitude is on the whole alike and the educational certificate series has the tendency to lead by one month rather than to lag. Berridge’s Labor Quit Rate is shown in connection with our present data on Chart XIIe. If we excepta nine months period beginning with October, 1919 the pairing is excellent. The curves part company in August, 1925, just as happened in the case of the employment indexes. The amplitudes also are with a short exception essentially equal. The Labor Quit Rate curve seems to lag. It is not unlikely that this reflects the situation under which juveniles can afford to take a chance on changing their positions much better than can their elders with their greater and more regular obligations. This may also reflect the circumstance under which the first increase or de- crease in employment volume is shown by changes in the demand for unskilled labor, in which market these juveniles are predominantly found. The coefficients of correlation in this instance are very interesting.*! We present them in Tables No. XIIla, XIIIb and XIIIc. The highest coefficient is found for the three year period, 1919-1921, be- tween Day’s Index and the juvenile data, with the latter lagging one month. In that case the value 41 The coefficient of correlation for synchronous pairing between this juvenile series and all the reissues combined is +.84.XI la CHART REISSU CERTIFICATES ED A = at 4 ' EPUCATIO TOTAL , 7 INDE NT INDEX LABOR STATISTICS. EMPLOYME U.- OF orn ie ihe Oe ae BO FEDER@GAL RES EMPLOYMENT BOARP. SERVE -——— —S$HOEVARD EMPLOYMENT INDEX. rh en a ——— —. Pr / Ps ~» * a . “<= & . | ~ é ' , i” J —82 JUVENILE EMPLOYMENT CYCLES found was +.976, and again we direct attention to the value found by Dr. Berridge for his senior em- ployment data and the same economic correlative which was, as we have mentioned, +.96 without allowing for lag or lead. For synchronous pairing our own coefficient is +.941. Of these three groups of coefficients the highest are found with Day’s Index as a correlative and the lowest with Dr. Berridge’s Mobility Index. The juvenile data lag behind the production curve, but show a distinct tendency to lead the changes in the Federal Reserve Board Index and Berridge’s data. TABLE XIlIla COEFFICIENTS OF CORRELATION Total Educational Certificates Reissued and Day's Index of Manufacture Educational Educational Certificates Certificates Epoch Lag Synchro- Lead nous 2 months | 1 month 1 month |2 months 1920-23 + .917 + .969 +- 947 + 891 + .857 24 + .767 + 811 +. .794 + .721 + .699 25 + .639 + .686 + 669 +. .612 + .603 1919-21 + 921 + .976 + 941 +. 887 + .813 22 + .815 + 871 + 814 + 734 + .696 23 +-.774 + .826 + 775 + .711 + .658 24 +672 +.718 + 676 + 599 + .559 25 + .576 + .623 + 586 +- 523 + 497XIIb CHART —— TOTAL EDUCATIONAL CERTIFICATES REISSUED AND ----— DAYS MANOFACTURE INDEX OF84 JUVENILE EMPLOYMENT CYCLEs TABLE XIIIb COEFFICIENTS OF CORRELATION Total Educational Certificates Reissued and Federal Reserve Board Employment Index Educational Educational Certificates Syn- Certificates Epoch Lag chro- Lead nous 2 months | 1 month 1 month |2 months 1920-23 + 782 + 826 +. 847 + 953 + 927 24 + 676 +729 + 746 + 846 + 835 25 + 549 + 608 + 633 + .720 + 714 1919-21 + 849 +. 896 + 829 + 881 + 816 Dy + 731 + 755 + 764 + 848 + 823 23 + 619 + 652 + .667 + 741 + .717 24 + 559 + 600 + .615 +. 687 + 675 25 + 466 + 515 + 536 + 602 + 594 TABLE XIIIc COEFFICIENTS OF CORRELATION Total Educational Certificates Reissued and Berridge’s Mobility Index Educational Educational Certificates Syn- Certificates Epoch Lag chro- Lead nous 2 months | 1 month 1 month |2 months 1920-23 + .488 + 558 + 605 + 583 + .524 24 + 466 +- 534 + 581 + .572 + .527 25 + .390 + 446 + 490 + .485 + 451 1919-21 + 362 + 591 + .674 + .667 + .684 22 + 364 + .540 + 612 + .601 + .567 23 + 405 + 517 + 579 + 561 + 516 24 + 401 + 503 + 563 + 556 + .518 25 + 344 + 432 + 488 + 484 +. 454XIIc CHART REISSUED CERTIFICATES EDUCATIONAL =— TOTAL AND? ---— HARVARD “B’ CURV = iSVisXIId CHART —— TOTAL EDUCATIONAL CERTIFICATES REISSUED AND ---~- HELP WANTED @T THREE EMPLOYMENT O S FFICE. = O - tu m N z 4 —, h 4 esis | Gr | <9 cota wa Z->| e | uae + =a | pt s., 4 | Sonera } ye =r = ~§] 6 rane is > <> | SI ‘ 2 IG 1or7 i918 in? ims—— TOTAL EDUCATION@L CERTIFICATES REISSUED ANP ---- LABOR QUIT RATE IN SELECTED FACTORIES88 JUVENILE EMPLOYMENT CYCLEs All Reissues Combined. We reach now the last group of graphic compar- isons. This one concerns a series of juvenile data in which we have combined all the material which covers the matter of reissues. In other words, these data refer to children between the ages 14 to 21 who are going to new positions for any reason, be it a quit, a lay-off, or any other type of mobility or turnover feature. Again we use the same economic barometers which have served us thus far. Accordingly Chart XIIla portrays this relationship with the three em- ployment series. The pairing in this case is excellent from the fall of 1919 to the summer of 1925; the amplitudes differ of course, but with respect to the Federal Reserve Board not very markedly. There is a tendency for the Federal Reserve Board Index to lag, especially on the recoveries. Professor Day’s Index is utilized in Chart XIIIb, and we note that the pairing is excellent throughout, and, excepting 1921, the amplitudes are alike. The junior data show a tendency tec lead the production series by one to two months, especially on the re- coveries. Chart XIIIc demonstrates the likeness of these juvenile data with the Harvard “B” Curve, and it will be noted that beginning with the middle of the year 1919 the pairing is good and the amplitude of both curves essentially the same. There is no marked tendency for either curve to lag or to lead on either the declines or the recoveries. The relationship between these juvenile employ- ment data and the Help Wanted Curve is shown on Chart XIIId. The pairing is good throughout but aXI Ila CHART CERTIFICATES REISSUED AND —— TOTAL EMPLOYMENT ANP EDUCATIONAL --—-— U.S.BUREAU OF LABOR INDEX INDEX EMPLOYMENT STATISTICS. —-— FEDERAL RESERVE BOARD. EMPLOYMENT ARVARD EMPLOYMENT oe | H a een —— ee 5 = . —_ F _—_+_-__—_90 JUVENILE EMPLOYMENT CYCLES tendency to part company is noted early in 1925. The timing seems to be synchronous and the am- plitudes are essentially alike. Dr. Berridge’s Labor Quit Rate is utilized as a barometer in Chart XIIle. The pairing is good through- out with the exception that,asin earlier comparisons, there is the same parting of the ways in the summer of 1925. Except as noted above, the amplitudes are equal. On the whole the juvenile series seems to lead the senior series. TABLE X1Va COEFFICIENTS OF CORRELATION All Reissues Combined and Day’s Index of Manufacture Epoch All Reissues Lag | Synchro-| All Reissues Lead 2 months | 1 month re 1 month |2 months 1920-23 + .694 + .694 +.138 -+-.723 + .672 24 + .568 + .569 + .617 + .600 + .556 25 +.517 = .912 + .546 + .528 + .487 1919-21 +.597 +.679 +.702 +. 7/09 + .748 22 + .504 + .516 +.599 + .624 +.551 23 +.533 +.559 + .627 +.636 | +.584 24 + .455 + .477 +.945 +.548 | +.502 25 + .427 + .441 + .494 + .494 + .451XIIIb CHART —— TOTAL EMPLOYMENT AND EDUCATIONAL CERTIFICATES REISSUED DAYS INDEX OF MANUFACTURE92 JUVENILE EMPLOYMENT CYCLES TABLE XIVb COEFFICIENTS OF CORRELATION All Reissues Combined and Federal Reserve Board Employment Index Epoch All Reissues Lag |Synchro-| ‘Al! Reissues Lead 2 months 1 month De 1 month |2 months 1920-23 + .656 + 692 + 670 + 612 + .557 24 + 566 + 592 + 569 + 518 + 467 25 + 464 + 486 + .459 + .407 + .356 1919-21 + 571 + 643 + 676 + 668 + .626 22 + .555 + 587 + 582 + 530 + 474 23 + .513 + 541 + .533 + 488 + 435 24 + 461 + 483 + 472 + 430 + 380 25 + 388 + 408 + 391 + 348 + .297 Table XIVa and Table XIVb contain the coeffi- cients of correlation reached when comparing this series as we have a number of others with the Federal Reserve Board Index and Professor Day’s Production Index. Higher values are reached when using the pro- duction data as the correlative than when using the employment material. The highest coefficient reached was +.748 when we paired the junior series with the production data allowing the latter to lag two months. As a whole the production series seems to lag behind the juvenile employment series and the latter seems to lag one month behind the senior empioyment index. This latter time relationship is just the op- posite from that disclosed in pairing only the chil- dren 16 to 21 where we found the senior data to lag behind the junior data. Reference to Table XIbTIFICATES REISSVEL} ANP —— TOTAL EMPLOYMENT ANP EDUCATIONGL CER ---— HARVARD 6 CURVECHART XITIId —— TOTAL EMPLOYMENT AND EDUCATIONAL CERTIFICATES REISSUED ANP -=-== HELP WANTED @T THREE EMPLOYMENT OFFICES - O - ©) - a ¢ —~ a $ ¢ | eee R = = 4 ¢ aes S CDE ooatn T~-> nN ex > “3p - | <3 | aan a S ce c a x o Sa o ¢ iS ‘6 2 o Ro PERE "cet — TOTAL EMPLOYMENT ANP EPUCATIONAL CERTIFICATES REISSUVED ANP ---- L@BOR QUIT RATE IN SELECTED FACTORIES ~ - =. ~ ~ ~ -— — Sire G 1 aw ; tn--4 ES - ' eis | lo ; : joes pce ees | ' oO presakot Ds 1° aa | ) Bg ee | ft ict ~———]_ | | | } Le | —t96 JUVENILE EMPLOYMENT CYCLES will show the reason for this divergence. There it was shown that the Employment Certificates Re- issued, a series concerned with younger children, namely, 14 to 16 years of age, lagged behind the Federal Reserve Board Index. In other words the younger children do not stand up well in the competi- tion of the labor market with the older children and, hence, we see reflected here the prevailing and rapidly increasing reluctance on the part of employers to en- gage these younger “‘hands.”’ The next two charts numbered XIVa and XIVb use two different economic barometers. The former uses the Clearings Index of Carl Snyder and the latter portrays Professor Persons’ Index of Trade. While the two charts contain different juvenile series, it is manifest that the better correlative is the Persons’ Index of Trade. Both charts reflect a most interest- ing feature with respect to the juvenile Jabor market since 1915. It is quite clear that the fluctuations in the war years were exceedingly violent in comparison with these well-recognized economic barometers, and that before the war and since the deflation period of 1920 the fluctuations have been more modest and more akin to the changing conditions in business. Corroboration from Worcester To confirm these Boston findings, to assure our- selves that they are not isolated, to prove that the erection of a juvenile employment index to cover a large area and many individuals is possible, we show chart XV which is only one series of many which we have analyzed for Worcester, Massachusetts. DetailederLarives) CYCLE AVERALE OF REISSOES EVUCATIONAL xilVa PLUS INDEX CHART CLEGRINGS EDUCATIONAL PLUS EMPLOYMENT REISSUES EDUCATIONAL DERS ---— ALL SERIES COMBINED. (COMPULSORY ATTENDANCE EXCLUDED) ——— New SNY 0 OQ “ wv os | : ; | - +s J 7 ha adhd vA F it wr 7 a wey eee)CHART XIVb PLUS NEW EPUCATIONAL NEW PLUS REISSOES S ——---—- ALL SERIES COMBINER (COMPULSORY ATTENDANCE EXCLUDED) EMPLOYMENT CERTIFICATES. —-— NEW EMPLOYMENT CERTIFICATE O 8 O WwW — ss CYCLE RELATIVES) ARITHMETIC AVERAGE OF INDEX OF TRAPE seers compPRiIse AND? —e—- PERSON'S (ComMPUTED TREND«(00) (EMPLOYMENT ANP EPOCATIONAL 200 150 100 (is22 Ii9IiSc iz7iSJUVENILE EMPLOYMENT CYCLEs 99 comment is obviously unnecessary. Undoubtedly a national juvenile employment index is possible. The cities which probably contain useful information for such an index are as follows: San Francisco New York City Bridgeport Rochester New Haven Yonkers Waterbury Cincinnati Wilmington Toledo District of Columbia Philadelphia Indianapolis Pittsburgh Louisville Providence New Orleans Milwaukee Baltimore Chicago Minneapolis Detroit St. Paul Reading St. Louis Springfield Manchester New Britain Jersey City Akron Newark Dayton Paterson Los Angeles Buffalo Youngstown Conclusions The necessity of applying refined statistical an- alysis to sociological data and especially such data which appear in the borderlands between two fields of inquiry is evident. The fluctuations of the juve- nile labor market contain such data. For instance, ‘1 Massachusetts there emerges as the result of the child labor law a group of interesting socio-economicXV CHART WORCESTER - EDUCATIONA@L CERTIFICATES. (NEW ANP REISSUED) BOSTON- EDUCATIONAL CERTIFICATES. (REISSUED) INDEX OF M@ANOFACTORE pays STANDARD ONITTS “<<, i918 isi7 IiziSGJUVENILE EMPLOYMENT CYCLEs 101 series which the present study analyzes in the same manner as the Harvard Economic Service studies material from the commercial field. The number of juveniles leaving Boston schools for the labor market has declined at a marked rate since 1915, despite the increase in population. The decline was at the same rate for all series except that of reissues of educational certificates where school at- tendance is compulsory. Whether this decline is due primarily to protective legislation or to the growing reluctance upon the part of employers to engage this type of labor because of its inefficiency and greater turnover, OT whether the decline is due to a recoPg- nition by parents of the increasing need for training and education is hard to say. There is a pronounced cyclical fluctuation in these juvenile employment series varying somewhat with the different series. The children who are changing jobs respond with a better ebb and flow than do the children who are first entering the labor market. The children who are mentally retarded and enter the labor market did not hold their own in com- petition with mentally more alert children when the latter were attracted to the labor market by the in- cidents growing out of the war. The dip during the war years in the curve showing the certificates issued to children who needed further schooling and the rise in the other curves lends color to this conclusion. Children who are mentally retarded do not show as much seasonal variation as do the others. They tend to leave school at the earliest possible moment after reaching the age prescribed by law for102 JUVENILE EMPLOYMENT CYCLEs compulsory school attendance. This may be due either to the futility with which their teachers and parents view further schooling or to the drive of economic necessity which affects disproportionately the families from which these children come. The children who are mentally less alert re- spond more sluggishly to changes in economic con- ditions than do those who are mentally more alert. Extremely high coefficients of correlation are reached with both the Federal Reserve Board Index and Day’s Index as correlatives, a value which com- pares very favorably with Dr. Berridge’s results based upon senior employment data. It is interesting to note that the timing re- lationship between the series reflecting children just entering the labor market differs from that of children already in the labor market who are changing pos- itions. In the former case the relationship is a syn- chronous one or a position of lead in favor of the juvenile series for both the production and the em- ployment index, while for the children changing jobs there is a tendency to lag. For children first entering the labor market a production series, i.e., Day’s Index, is a better cor- relative while an employment series, i.e., Federal Reserve Board Index, is a better correlative for chil- dren already in the labor market and changing their positions. Certain of these timing relationships suggest that we have here reflected the circumstance that children can afford and do take chances more readily in changing their positions than do their elders. Children 16 to 21 years of age, already in theJUVENILE EMPLOYMENT CYCLES 103 labor market, register changes in the economic cir- cumstances of the market place more sensitively than do either younger children who have been in the labor market or children of the same age group who are just entering the market. The younger children do not stand up well in the competition of the market place with older chil- dren, a circumstance which is rapidly causing employers to manifest a growing reluctance to employ these younger “hands”’ because of the high labor cost in- cident to such employment which arises from a number of different sources. The possibility of creating a representative national juvenile employment index is perhaps clearly demonstrated by comparison of similar data from Worcester, Massachusetts.APPENDIX DUCATIONAL CE 104 )D ISSUE _ - RE S + * RTIFICATE , / I. TOTAL E TABLE JUVENILE EMPLOYMENT CYCLES gees CEN ae ae tye he (2 «WA So NY => CY) - CO N - x S | ae } =) ~ — ' | | eer at ate eet | OMT a 3 OG wy Oo 102 Dy . Saves IOV; 6 OO re Mt 10 AS) ON Gh oN oe - ; = : | : valet ee POA Rese + Bee HH oe wf) Tm cc c) Ton — * 0 co —_— SO Vo Mae os RS: OO as AN! Oo Ao CQ) | ; : crn ! oe —_— oO : ; tal a(n a fi Me | gh lee : ’ ' : ' — ¢ MA =o COTTA AA SCO eee om | i z = a N mM —v =~ © uD S io OG ; i ' ; a - a a - _ - ~~ - io ~~ A] CN xX CY. CY. HO SO CY) ve ° CY NVivo i MO © OC * SC S & ome Cy . . . - : — 7) me me Tp l | | | | | | sie ie. —— a. a — _ an maid vr ~ A] = =P ~ - ’ > — ae ; a — a — | im GON) ea CNL Oe OO). C0 MO. Ose aos S AX ~ , Mee dade ee eee ce Se ) het St CQ oS OOF Nee So 1M —_ aD uD ] YD rt CY = CO CQ | — : aaa a | 4k j ; j | | late a ake ee ei | ia LN — YD = NI = i Z YD YD = sx N — Oo mm Tm — YD DL CY. vt rat N uD ——— { : . . . . . ) | ' ' aie oto} | +04 4+ 4+ 4+ ee ; -“-<« = —— ——_ i LL XC — YD CY) — D rt CY) = SS Co 73 naa © i © © i © -— CY) oO — — : ~ ON a oo — — — “i _ — ~] : : : _— + i a. -4 co oF aed —- + ++ + ~~ we . . = = * ae ~ aan — Nn : ion i+ + +++ + + + + z CY) N i Oo — 79 oS N N o> i Shia cm SS os AS 4 + SN @ = en sa1+ + + + + + | | + +4 eee ee ES ce eS Oe ee ae Sue te a mal : : ~~ Oo SoS = - GO m= N — + - . : a, — QI mS | — N 7 ee (aa | : = kK © o 5 > ~ + ~ . ran La = — oo — "e> — 5 o _ ® D = vw ~ — . ® - ~ © > => — nm rn ~ > SD > = oe ay io =e os — _ 4) ~ 3 _~ —_ — — on — al ae ~ 7, f=1) i SUE ~ ’ le] ’ ‘ a TIFICATE, + 4 iI SDUCATIONAL [| TABLI _ 1916 >| 19] () 12 ‘> +) Uy >] 7 ed 4<) “yey + )e) OY 1 O4 >] } f 45 <> . ; { ‘)e> 2? ONS > oA Nov emberREISSUED S s 4 EDUCATIONAL CERTIFICATE IIT. TABLE 106 School Attendance Compulsory Standard Units JUVENILE EMPLOYMENT CYCLEs Re tea coe ae GO| BO). Os © N fal : — ; * > > . Terman cl a vl. cls a ee = aon | & =, © o> = N XD co oD N oe co lm YD © uw I~ N i YD Coo © > ul a Ne 1; iLse te ese ahs el irr aun oS = OPO te ANS Tere ee eS SN MO om oO! hU6CeOlUN US UU coe Cr . . . . ° . ° . . . on | — = — ~~ Se ee a se ee Boas been oD on co — LH Oi 16" aa cS: SON ON) Oo SN CS) GO) eH 8 Ml es ae nn ~|]+ +.°+ + + + «+ + + + + i ad Kr = — ie CO N us N _ ld = Be eo Tw YD oY = YD ee io © N rm o> Or | N _ See Po | kite InGOr O° ea OS SN Oo CO FO) aCe — | SD Le > H ~S © et QO Se ceo on i= ; — : { Saeed tk) ee | and So So To ayes ae > S ot ae D> a | GN Oo a = a oO o> I~ Ton at) cY) = CQ si ° ; : ; ; ; | | 4 ae |. - o> sie : A - ar uray cS: eH mats Si eee 9 =e aN 5S HA & Ie: Be. (00) (00pm Salialie a | feo es aks ee NI o ae S LD o> Lx NI S Sh = oo I~ L us _ _ —_ YD — Oo nN re © — : - . . a — _— ae | eee - + | | sai =H = —N x CO — N =H N a) CO mic — oOo + ~=s - A = oo a . . . ° . . . . . a | cata ett | | Ipooar ams | | | See Sm S SS eS Aaa — . ° ° ° . . . —=j+4+4+++i/]/++ !/ ++ xX il —_ © io, io _ ._ N NI N uo IniIis® > ON xe; 2on 1! Of cre — : . . . . . on — —_ _ mea eete cametes ot che ate ee ate nies eat — oat a — ~ February March August September January April May June July October November December1924 D . A ‘4 ISSUI 1923 “ - RE 1922 S 7 4 RTIFICATE ird Units { Ck 1 ‘ INT and t Q \ EMPLOY ME TABLE IV ry ‘ + wu ‘ 4 J a) = —_— — | — Aa ses mg > a ~~ + uw” ~ ON ~ -_ = a _ — 2 a ee ~ ~ ot '~ ——_ t~ — a x. ~ + + = ! > a ——_ uy _ os + + oo” —_ _ 4 4 . —= ant -+ — —_ . * — ~— ——— . Tm cc : a = ~— - — —_— = } > eR —_ Ow ~ — — oad - a = ~ co —_—— co “ -_ _ , = _ — JUVENILE EMPLOYMENT os ™~ . ~ — _— = i ne » + 36 ‘ — ~~ — = Y c ; L N C: = —| Te —_— i * J ieee ae ~~ ae ] ** .* < en Sf = — — = im ‘ ~- . ~_ im ~~ a — . S ~ —_— aa ] <‘e _ ; zi ~ “= t ~ -_ ——_ im — > - —>- > — a -- * * = J ** —_— — => aot es : = .* - oe a =—* —* on ~ — — i“ —~— — —+ x f . ¥ . = * _ ; + +- —+- 6 Rg 7 -_— a rh oe — = _ * + . -_ | | ' ~ —_ —_ oe ~ =~ = Tm ul oY. ° a . April May June oO 2 « | l September October July N x ~* = YD — | | | YD -~ ci ——_ = —_~ al —__ ; a ~ > Ps _~ -* oN —_ + : ie CS i — _ o— _—_ — — = ic _— ‘ > > — ~~ = —_ e ] — hes ; $$$ <_—x— N _ ~—— 2 it~ 74 + -—— — a ag ee — ~- NI CY uty apa ipa == t~ C ~ _-_~ ~ ‘ * os ‘ > ’ : —_ ee —— + 0 uf) -_ Y. ™N I _ ~]RTIFICATES i ry ince Volunt DUCATION AL School Attend + k a W a ~ oo - + 7 : - TABLE \V 108 “ ‘ ird Units und St JUVENILE EMPLOYMENT CYCLES wet) Ne Tey BH — 19 } + * “ * ce. - _ oa, ~ 7 ed ~* — 4 \. ] N — Oo om ro a -- ok = ae HL : ' i ' ' ' 4 < j « ~ - : < un CS CY) N = om KL iD rm YD o> — S ON) oO —_— -— “= — ae (\] - = ' ee al | | | | | | | fat Poe Be oe Gat ale | —_ 5 =r —_ AY ae — A — _™ Al — a a . _— a : ne - — a oS IS - STN ta GaN «oo MOM OD tS es ewe ' ‘ : - ps4 = —_— = ' ' | ' ; ; ode - axa _ on + — i — r | ' saa: a fl ye ae a“, a — A] — = ~_— — ‘ LH ; in -_ = _ « os — — — « ~ — ~~ a = _s > am naam q ; - — —- ~* N io CO —_- an CY N a ON , >» yay _ -_ —— ; i | } T — T — oa ite i ; | — —_ ™] —_~ — % A] o— ~ Sf A) —_— ~ a ~ Jn. —— an. a a _- - = oe ~~ ) ; - - ] ‘ . ae - at. —_ — ci = t~ * + Ni . N | N _ ee | | 1 ; | ! ' ' a | | a aE or ! MII —_ - a .-, aad 7 beers x = = = ae aks ~ S A : x | a . + -* - —- — ~_ 7 i — —_— CQ) | a : ; ie coe : | ' | ; : a. iN — ~ Ms — ' ; j j i | ' | —— + ae te 7 ae as —+— | { -- — - a - - = _ ie ca : ce AL N 4 ~ =) if.) - — —— onm ~ ~ —_ a ~ ~ —_ a —_ at. at. =. — re ~ ] [ OS 1 a a . : : : oan | — — _— _ — ; : | ' | ' ; + + -+- —+- - —— —— —— a oo : : a aa = — —_ — —_ _— —_ - mi wD a N I~ o> : = YD YD N CO earl : ; Sel rt ett N d ' ! | i : ' usr] =a | am + — aearall | 6 i - a we . _ a) RS . 2 oe -) = al Bag -o - ey a e ~ a . a — ./ — © @ = cd Di : > «3 . . _— Ne > - cal _ = = ae — nr &0 ~ +— ‘Ee I - — =< 5 ~ = — ~ ne c) © © * 4 —— — SS nn <_< ¥ — = _ a oo a, ~~ — cs 7 x XK = = ~ 2 ~ =. TIFICATES R ’ - DUCATIONAL CE School Attend I: W ~ - NE TABLE VI. ince Compulsory . ‘ indard Units ‘ ‘ t S 1926 25 1923 | 1924 | 19 99 1915 | 1916 | 1917 | 1918 | 1919 | 1920 | 1921 | 19 1914 ‘> () ede) ‘<)> he) t O5 L6 ry ‘ ‘ Janu — / recy «sa ry ‘ 1 Febru ~~ ede) QQ | (U 0 36 arch M JUVENILE EMPLOYMENT CYCLEs es *) ) 61) April iy) LV + I July June — ~~. . . . on | star ‘+ = = = [~ eo ~ pa SF. ‘eo oN re —_ —_— —_— es ; rel) eel i ) — ———e | [~ (“] [ ~ I~ {- A“) “ _ -— ] | - a J e : ° ! ‘ 4 . ) ' — a ——_—-——— | NI io eS > I ~~ = = ~ Ye i~ ~? a nN +- +— —+— a te i~ c= aD y i~ - at. a . + + pa —+— — an’ , Yo ~~ a - - . ~« Ns xs A | 4 = axa : [~ -— =T io im ‘ = as — e 4 *« —_ ; }. aoe ons aoe Seance By is bd ail “f OJ - - ° — —_— — —_— j 7 i —: rm I~ LD > = ~~ =— ‘ AL. va ao ! | pa oe EL = ~~ "I —, _ — a. —— -_ ~~ © yond ' : : ; oS = = “7! “ N N a ] . | ; — >= ' > CO aes e _s a. w ‘oe CO rt — -_ — : + -- —p~— + > 5 2 ) { CE | | | fone be Sus ~ ~ ~ oon ~ mt 2 _ = @ = a — “s "~ ~ © ™ ~~ - > = of ~~ ja a ~ -_ ma CJ os _ — = , | a = at © —_ = _ I~ _ L eS — CO DS . on ! ! } fr | Th sis ' + + lia oe le oa oleae XK N SD SS — i I~ S oD nD — GSA SO oe 89) Ss 1 ee — 2 — on a i+ + | + 4+ 4+ 4+ 4+ 4+ 4+ 4 ~/1 Oo D S I~ oo oD iD _ oo iD oo So N = = cS i i =H — oD N N N i + -- -+- sr OO ot I ee ne ee eS eee aim NaH tO OO ODE OA 2 S on — -_—— —_ —_— — —_ — — | | | ! ; | ; ' ' — co CL = i = oe =f © i OG _ Siem (Oo oN NN Co Oo ON Oe US ON On | NX -- — ete ete ee eee ie i ee et OL iO = N co LH La _ A S on i alte Co S © st Oo 20 Qt st (Or we =~ == _ } —— ; es: He de oe | LN S > = io > I~ = — ~ oo > co | ~~ S nD — — NY = nN AY N o> -—_ . . . — eee ON AN et oe = hele Come oe ek ee ee eee | © io N aS io it~ N = = oS Ss = ee [Pee J _ ie N Co oe YD CY o rel or 3 ; oO — Sei ad ee — oD K N c To = N = I~ SS To Oo | a N N _ — _ ™N ee, —N oO N oe wae aie coat lea eae i+ + + + + | + + + a N D N YD N N NN 20 = =P So in | oD = = = “sD *@ Lo co a S NI on i =— ao — — —- os _— — _— — | | ) : : | | : +- =: 2 oui tl Sy ' i | < ® o° oe a Be = gy et” oS = So ea 3 ev e o — _ i ° =n > © — o = = cj ou A tay el PN ee ed ee Oo AO ei te a ly OS OO eeeSUED Dp , REI ¥ uD ~ RTIFICATI / _ AL CI a DUCATION AND E NT _ . EMPLOY ME TOTAL s VITI. TABLE ard Units and st JUVENILE EMPLOYMENT van vend i i t~ uf) x i . : eo '~ o> x x N ay fD co - or . : . on —_— -_ . . - ea a ak = =i he en aie ath - ° . ' : * . ~ i i a a nl e- Oee eee InN | A a — _ NI = N i XD us ' ’ — | | | ' _ — CY. ce '™ — Ta — — CS CN m7 | x oo oD N XL ™~ LH >) a co. oN oO) ‘ . ° . . Cn = = = me | lt = i. + + + + + H+ | ; = ™ 1 — CY. vt = t~ N — N el A | OS <7" XK oO © N N mm YD © it~ © cy . . ° ° . . . GN I t : ' — —+— oe —- as — Ga N SO Te) . 7 Co To —_— iD xD NI _ —_ a © ee Jo D YD N oO i i LD n -- , ‘ ' Sle eal ale ee eal ; i eels BFS 068 =" OO eS ee — 7 CO 7 _ —_— _— — D FI \ Ge oY oO . e - . . . . ° on ae _ — — — -_ — _ ania lee —4+— —+— -+— aa aon oe + oe e -_ oO a i — Y. aa — ™N io ~ — a = <7" = x Co =—_= -_— |; = St, pe “ —_ -— i~ l~ ~ = S = mul Ge oo) SO) ot OT = i OO et | el \ => : — quand — — — _— —_— } : ’ : ; Ihe : : ; ~ —+ -~— --— — — -_— —— _ — a <-> io . = ule - =: aS A i~ ™N — . * ] —_ ™N _ a ~ \ - j — ~~. -_* -_ - = _— | ’ ' Fon = ~~ -+ -+ + a +e —— -- a, —_ — _ - - a oe =_ or -- ~~ ed —_ ~~ — = — -* —_— : -- ~« wn oD co - ul afd =~ uf? = N © = N _- . : Co oe — _ —_ _ - ms -- rt 1 ' : = | } : ‘ ' | ; : : - — N i Ni OQ = | . SO 72 4D re : e Con ° a —- _ — ~~ _ ! | : “i = 7” - me | > a Y ap > as “ = ( > ol = ina a = we. a ey — — — Vd 7 =~ Sans “wi _— > © " n +—_ = > Ps _ = — co a —= - = . CS dq | rs ~ 5 os _ — 5 : S oO vo bor. a —_ _ tend — = = ~ - o = —_ — < i = — < F2 ~ a 1PLEASE RETURN TO ALDERMAN LIBRARY DUE DUE