This chapter deals with the development of mental abilities that are measured by intelligence tests. In perhaps no branch of psychology has more progress been made than in the measurement of verbal intelligence. This holds true even though the instruments for measuring intelligence, especially at the lower age levels, are still imperfect and many issues remain unsettled.The intelligence test is designed to measure an individual's ability to cope with situations that call for the exercise of mental processes, his ability to act in accordance with the demands of the situation that confronts him, to comprehend the situation and solve the problems it involves, to learn and to apply past learnings.Numerous definitions have been made of intelligence, including varying emphasis upon the ability to learn, to apply past learnings, to carry on. abstract thinking. Among the operations listed by Thorndike (93) under the heading of intelligence are "a wide variety of operations such as we may call attention, retention, recall, recognition, selective and relational thinking, abstraction, generalization, organization, inductive and deductive reasoning, together with learning and knowledge in general." Other things being equal, according to Thorndike, the more intelligent person is one who not only can master a greater number of tasks and solve problems with greater speed but also is able to perform harder tasks, such as solving a mathematical problem which a lesser intellect never could master or reaching an effective solution to an economic problem that would bewilder a less able person with as much good will and access to pertinent information.According to a definition by Stoddard ( 84), "Intelligence is the 521 * To illustrate what is meant by a correlation coefficient two examples are given below. One example shows the correlation between I.Q. ratings based on two separate tests; the other, the correlation between I.Q. and strength of grip. In these examples, one of the simplest methods of correlation, known as the "rank-difference method,' is used. It can be applied when there are only a small number of cases. The examples below include only seven cases, a number which is large enough for illustrative purposes but too small for ordinary statistical work.The rank difference correlation coefficient is expressed by the symbol p. The formula used in the present case is: 6 X sum of P* *From Thorndike, R. L.: "'Constancy' o£ the I.Q.," Psychological Bulletin (1940), 37, p. 173. Reproduced by permission of the American Psychological Association.