THE TERRITORIAL POSSESSIONS.

Not the least surprising of the results of the study is the fact that the educational systems of our territorial possessions, such as Hawaii, the Canal Zone, and Porto Rico, have higher ratings than those of many of the forty-eight states. According to the tables of the report, the schools of Hawaii rank above those of the majority of American states, standing in their index number just above that of Illinois and just below that of Nebraska. The schools of the Canal Zone are in twenty-eighth place, just below those of Kansas and above those of South Dakota. The schools of Porto Rico are in forty-second place, above those of Virginia and below those of West Virginia. Ten Southern states have records poorer than that of Porto Rico. This last showing is the more remarkable in view of the fact that the per capita wealth of Porto Rico is only about $200 per inhabitant. This is about one-fourth that of the poorest of American states, about onetwenty-fifth as much as that in the richest state, and about onetenth as great as that of the average state. When it is taken into consideration that Porto Rico supports her schools from her own revenues, and that she has a large colored population, it is particularly noteworthy that she has a longer school year than any state in the Union and pays higher salaries than most of our Southern states.

LOSSES AND GAINS AMONG THE STATES.

The degree to which the different states have lost or gained in relative rank since 1890 is shown by a table which reflects the tendency of the Western states to move forward and the Eastern states to fall back in maintenance of their educational systems. The table follows:

ECONOMIC STATISTICS APPLIED TO EDUCATION.

The Foundation's report covers a piece of work which has been under way for many months. It applies, to the problems of education, statistical methods that have long been in use in the field. of economics. The product of the investigation is an educational index number compiled by methods similar to those used by the Federal Government in compiling its index numbers for the cost of living, the prices of retail commodities, and the like. The Government's index numbers take into account changes in the

prices of a large number of different elements and combine the results in a single number. By closely similar methods the new educational measurement is worked out, by taking the official data showing the number of children attending school, the amount of training they secure, the progress they make, the amounts expended for buildings and supplies, the salaries paid their teachers, and other similar items, and combining these factors into a single index number which shows the general standing or efficiency of the school system.

The most notable educational change that has taken place during the 50 years covered by the report, is in the attendance in high schools. There are now 100 times as many pupils enrolled in high schools as there were in 1870. The number then was 19,000, and now it is nearly 2,000,000.

The effect of the war is plainly seen in the high school figures for the past three years. The attendance has increased with great rapidity, but the new pupils have been mostly girls. Apparently the prosperity of the country has been such that families have sent their girls to high school, while their brothers have gone to work and taken advantage of the high wages obtainable.

In the matter of teachers' salaries, the states show most diversified practices. The lowest average salary is $25 per month for the state of North Carolina, while the highest is $88 in California, both being based on a year of 12 months.

The author of the report served during the war as chief statistical officer of the army, being attached to the General Staff in Washington, to General Pershing's headquarters staff in France, and later serving as chief statistical officer of the American Commission to Negotiate Peace.

A Comprehensive Plan for Tabulating and Interpreting Educational Statistics

C. L. STAPLES, PH. D., RESEARCH DIVISION, BUREAU OF FOREIGN AND DOMESTIC COMMERCE, WASHINGTON, D. C.

T

WO purposes, in general, actuate educational statistics. Sometimes the tabulations of figures represent the scope, progress and present condition of education in a town, city, state or nation. Sometimes a particular problem or phase of education is under investigation or examination. The figures, then, bear upon and point to a special subject or individual research. In both types of data, since educational matters must be accessible to a larger mass of readers than import and export, economic, vital, health, or even sociological statistics, the prime requisites of educational data must be simplicity and directness. The parent of average intelligence should be able to go to a volume of educational statistics and find the figures he wants. Transmutation of figures, therefore, into conventional scales, indices or coefficients should be in the very small minority of educational tabulations, if at all.

Statistics of school systems are acquired in several ways. The regular, annual, stereotyped blanks sent out by the U. S. Department of Education year after year with slight changes, furnish a large supply. These depict general conditions. The questionnaire, circulated to cover a current educational problem under investigation, brings in new figures and supplements and interprets the available data already on file. Occasionally the nature of the research requires a personal collection of data, or at least a beginning along that line, for a starting point. Always available, of course, are the published reports and pamphlets of both the United States and foreign governments.

The assembly of educational figures may be aided by the mechanical tabulators in current use today. For adding and balanc

ing columns of relatively large quantities, the Burroughs, or better the electrical Dalton machine, is recommended. The advantages of a machine, after the operator has had time to become thoroughly accustomed to its mechanism, are greater accuracy and greater speed. For conversions in the reports of foreign countries, or for ordinary multiplications and divisions and for percentages, the Monroe machine is recommended.

As to methods in tabulation, the place of the totals may be considered briefly. Totals may be placed first, either horizontally at the top, or vertically at the left, instead of at the bottom of a column or at the right. In educational figures, perhaps, the position of the totals should be determined by the motive of the statistical compilation. When the emphasis is on the sum, the totals should meet the eye first; when the details are more under the searchlight, totals should come at the end. Small points, perhaps too slight to mention, are that time is saved by dropping the units' period when manipulating large numbers, and all percentages should be carried to three decimal places, e. g., 39.5 per cent.

"Figures without thoughts never to Heaven go;
Rather they clutter up and die below."

From the "Language of the Counters." "Words without deeds are idle," is a well known adage. "Ideals untransmuted into action tend to make for mental flabbiness," "By their deeds ye shall know them," rings a familiar anthem. And so it is with figures. A page of figures without an interpretation is like a house without paint or finishings. The pointing out what the page of figures means, and the statistical analysis of the figures compiled, form really more than half the justification for the collection of data. It is very important, then, that the various tables should be followed by verbal analysis, giving the drift of the movement in question.

To show tendencies, percentages of increase or decrease are very commonly employed. The various graphs give the meaning of a page of figures almost at a glance. Among the chief forms of graphs may be mentioned the following: the Circle graph, the Horizontal and Vertical Bars, the Picture graph, and the Curve