Page:American Journal of Sociology Volume 10.djvu/713

 REVIEWS 697

An Introduction to the Theory of Mental and Social Measure- ments. By EDWARD L. THORNDIKE. New York: The Science Press, 1904. Pp. xii + 212.

Professor Thorndike has prepared this book primarily as an aid in doing statistical work of the sort required in laboratories of experi- mental psychology. While he says in the preface that he " has had in mind the needs of students of economics, sociology, and education, possibly even more than those of students of psychology, pure and simple," it is not probable that many teachers of these other subjects will think the book so well adapted to their needs as one or two others that might be mentioned. This remark is no disparagement of the merits of Professor Thorndike's manual in its own field. The kind of help needed by a student beginning to handle statistics is a few practical hints referring to the particular material with which he is dealing. When a discussion of statistical method is made so general as to be applicable to the gathering, tabulating, and pre- senting of figures of every sort, it necessarily becomes of little help to the tyro attempting to manipulate statistics of any particular kind. Therefore an elementary manual of statistical method in general would not be a good book for beginners. Professor Thorndike does not make the mistake of writing such a book. On the contrary, his discussion seems to be admirably adapted to meet the concrete puzzles that a young psychologist will meet. But because of this very adaptation to the needs of students of one sort, the book is not well adapted to students whose puzzles are of a somewhat different character. The principles and technique developed may be, as Professor Thorndike says, " applicable to all the sciences which study variable phenomena," but a student who is beginning his struggle with figures ought not to have laid upon him the additional burden of interpreting and applying a book on psychological measurements to his problems in economic or sociological measurements. On the contrary, a student who has acquired a considerable degree of skill in using statistical material of any kind will find profit in looking over this discussion.

In its special field the book is worthy of a man who is a teacher as well as a psychologist. It begins simply, and by affording abun- dant material for the student to practice what the text preaches gradually develops in him capacity to master the more difficult later chapters. While the writer makes a point of keeping within the comprehension of young students, it may be noted that he has once