Search

Because highly technical expositions of significance tests are readily available, sociologists will proceed from an elementary and non-technical statement that emphasizes the meaning and interpretation of the tests. First and foremost, a test of significance is a formal procedure for making a decision between two hypotheses about some characteristic of a population parameter on the basis of knowledge obtained from a sample statistic of that population. Either the null hypothesis or the alternative hypothesis may be derived from the substantive theory in question, though it is conventional that the theory being tested that is, the researcher's hunch, or the possibility he thinks most interesting is identified with the alternative hypothesis. This means that null hypotheses are usually set up for the express purpose of seeing if there is empirical warrant for their rejection or nullification so that the alternative hypothesis can be accepted hence the term null hypothesis. Testing for significance involves a comparison of the difference between the sample statistic and the parameter specified by the null hypothesis with a theoretically determined sampling distribution. This comparison allows estimation of how often such a difference would occur if difference were due to random errors in the sample selection process sample error.

The article features the Urban Institute, a research institution responsible for policy analysis of issues involving sociology and social science in the U.S. One of the favorite topics of the U.S. administration is the extent to which economists have begun to supplant the lawyers at the right-hand of power. While many sociologists may yet aspire to the left-hand, it may be useful to consider some of the signs in the economists' favor. These include not only the extent to which economists are involved in government itself, but also the degree to which the economists have encircled the government with independent research institutes providing policy analysis. The Urban Institute represents a distinctive new genre on the social science scene and it is now beginning to seek out supplementary sociologists for its staff and projects. Founded in 1968 as one of the last domestic initiatives of the administration of President Lyndon Johnson. The institute is involved in research on income maintenance, income distribution, and housing. Other current research concerns of the institute include, forecasting demand for urban transportation systems, evaluating alternatives in generating and allocating public school revenue.

To do or not to do a test of significance that is a question that divides men of good will and sound competence. Only if the sociologist can apply experimental treatments independently of prior states of his units of observation is he able to conduct a true experiment. Because of democratic ideology and a host of other considerations, true experimentation is unusual in sociology. The inability to use true experimentation results in a lack of control over a wide variety of largely unknown and un-numerable potential influences. These uncontrolled variables throw doubt on any finding that a difference observed between two sub samples results from the variable used by the researcher in classifying his sample of observations into one or the other of the two sub samples. Accordingly, the empirical sociologist is confronted with the necessity to identify and to find justification for rejecting as many of these potential influences as he can. The effect of a pre-test in increasing or decreasing the respondent's sensitivity or responsiveness to the experimental variable, thus making the results obtained for a pre-tested population unrepresentative of the effects of the experimental variable for the pre-tested universe from which the experimental respondents were selected.

This article discusses the measurement of sociological productivity. One of the focuses of empirical investigations in the sociology of sociology has been publication output. In reading the literature on the productivity of sociologists, however, one becomes aware of a lack of research continuity. While numerous data have been reported, they have been not cumulative but, rather, ambiguous and often contradictory. One of the major obstacles to linking the results of the various studies together is the disparate procedures that have been employed to measure productivity. One measure with potential as an index of publication quantity was developed by Glenn and Villemez in research concerned with the productivity of American Graduate departments of sociology. The Glenn-Villemez Comprehensive Index (GVCI) was designed to cover a wider range of publications than most previously published indexes had covered. As originally formulated, the GVCI included all articles in twenty-two journals of sociology and allied fields. By means of a mailed questionnaire to a sample of professors of sociology Glenn and Villemez derived a system of weighting the various types of publications. A number of arguments can be made for the use of the GVCI as a measure of productivity. One is that it is composed of a broad range of journals and circumscribes most general sociological work as well as important specialty areas within discipline.

Presents news briefs on several social science and sociology organizations in the U.S., as of April 1972. Information on a workshop sponsored by the Committee on Basic Research in Education of the National Research Council; Discussion on the collaboration between the National Opinion Research at the University of Chicago and the Roper Public Opinion Research Center at Williams College for a social science project; Details of the support of the Law Enforcement Assistance Administration on research projects dealing with crime and delinquency prevention and rehabilitation.