No abstract

According to a paper by Staff, Personnel Research and Procedures Branch, the Adjutant General's Office (5), the basic idea for the forcedchoice technique was developed by Paul Horst with reference specifically to personality scales. Robert Wherry developed a similar idea while working on personality measurement for the Civil Aeronautics Author-it}', and later the Staff of the Personnel Research Section of the Adjutant General's Office attempted to apply the concept first to the design of personality inventories and later to the problem of rating officers. The latter application resulted in the production of the Army Efficiency Report, which was used for some years for the official rating of officers. More recently the technique has been used by various professional persons and for a number of different purposes. The method of making a forced-choice scale as used by the Personnel Research Section, has been well described by Sisson (3).The general nature of a forced-choice rating scale is best understood in terms of the procedure which is followed in building it. The following are the essential steps involved:1. Descriptions are collected of individuals who are at each extreme of the scale to be measured. If the scale is that of effectiveness as a supervisor, descriptions are obtained of the most effective and the least effective supervisors. This procedure partly defines the scale that is to be measured by the final instrument.2. The descriptions collected are then dissected into a list of small elements. Each one of these elements describes, in essence, a rather specific item of behavior. The complete list should cover all the important aspects of the job, and the number of items covering each aspect should be related in some rational way to the importance of that aspect.3. Two values are determined for each item listed. One value, the discrimination value, indicates the degree to which the item measures the particular characteristic that is being measured. The other value indicates the extent to which individuals tend to rate others too high or too low on the particular characteristic. This latter is sometimes referred to as the preference index or preference value of the item. Both values are determined experimentally.4. The characteristics are then arranged in pairs such that the two members of each pair differ in the extent to which they discriminate. Ideally, one of the characteristics in each pair should have a discrimination value of zero and one 62

MosT of the work undertaken at the present time on the construction of tests is for the purpose of developing instruments useful in connection with some practical problem. Few psychological or educational tests are built for the purpose of testing some hypothesis of broad scientific significance. This trend toward the development of tests as tools for the technician rather than as scientific instruments for the research worker has had important effects on the methodology of test construction. The technician has developed techniques of his own which are somewhat different from those commonly adopted by the scientist. While one is tempted to contrast these techniques as the empirical versus the rational, or in terms of one of the other dualisms of philosophy, it seems better for the purpose of this discussion to refer to them simply as that of the technician and that of the scientist.In the construction of tests, the scientist and the technician start from a common point of agreement, namely, that it is in nearly all cases unjustifiable to assume that a test measures what it is supposed to measure. There are exceptions to this, but they are few. The scientist and the technician also agree that since there is no certainty that a particular type of test will measure a specified variable it is necessary to set up hypotheses concerning the kind of test that will measure the given variable. Hypothesis formation is agreed upon to be an essential step, but the process is fundamentally different for the scientist than for the technician. For the scientist, hypothesis formation is the crucial and usually the most difficult step in his entire working procedure and one which calls for genius and originality. For him, the problem is to set up hypotheses which are rational, that is to say, which are consistent with established at University of Ulster Library on May 28, 2015 epm.sagepub.com Downloaded from

No abstract

Robert Travers' analysis of British conquests in late eighteenth-century India shows how new ideas were formulated about the construction of empire. After the British East India Company conquered the vast province of Bengal, Britons confronted the apparent anomaly of a European trading company acting as an Indian ruler. Responding to a prolonged crisis of imperial legitimacy, British officials in Bengal tried to build their authority on the basis of an 'ancient constitution', supposedly discovered among the remnants of the declining Mughal Empire. In the search for an indigenous constitution, British political concepts were redeployed and redefined on the Indian frontier of empire, while stereotypes about 'oriental despotism' were challenged by the encounter with sophisticated Indian state forms. This highly original book uncovers a forgotten style of imperial state-building based on constitutional restoration, and in the process opens up new points of connection between British, imperial and South Asian history.