Search

This article discusses aspects of mortality, inequality and social theory. If health and longevity depend essentially on nutrition and the material level of living, at least before circa 1900, then mortality rates furnish an acceptable measure of living conditions. Change level of living, particularly nutrition, and the death rate will change too. There is a great difference between population response and individual response, and research techniques geared to the latter may fail to predict or explain the former, population response. Mortality offers an excellent illustration. In epidemiology as well as in much sociological work a strategy is followed which rests on variation among numerous units at a given time. These units, typically individuals, are classified by some index of environment. Thus mortality can be studied in different social classes. To measure inequality, one should take mortality, a measure of the probability of death, in some privileged stratum as the numerator and the overall risk for the entire population as the denominator, thus forming a relative risk or risk ratio. A first objection to the method is that one might equally have computed the difference as the ratio, subtracting population risk from the risk within the most-exposed stratum instead of dividing the latter by the former. In terms of risk ratios for death, inequality would appear to be increasing, in terms of survival ratio to be increasing.

The article analysis short-run variations in marital fertility in Sweden during the period 1830-1879 in order to explore different models of fertility determination before the secular decline of fertility took place, and to use fertility data to demonstrate a method of studying sociological trends and oscillations. One of the main determinants of aggregate marital fertility, the total number of legitimate births or the legitimate birth rate, is the number of marriages, and fluctuations in that number. It will be shown that if this were the sole determinant the result would be a very high degree of smoothness in the response, fertility. Auto-correlations would be high, much higher than actually observed. So a simple inspection of the response series and the correlogram is sufficient to reject this explanation, even if there were no direct statistical information on the annual number of marriages. But as direct information on marriages is available, this may be used to compute the expected level of aggregate marital fertility assuming a constant fertility schedule by duration of marriage. This turns out to follow the smooth pattern predicted, whilst observed fertility shows much more pronounced short-term variations.

In Western society the process of fertility decline is often regarded as an innovation process. The assumptions behind this approach seem rather questionable, and the diffusion lags or gradients of limited importance. Both Swedish and other European data are used as illustrations. It is suggested that the decline be treated within the wider sociological perspective of a time-consuming adjustment or change process, not necessarily starting from a position of completely uncontrolled fertility within marriage. The situation in to-day's high fertility populations is briefly discussed against this background. [ABSTRACT FROM AUTHOR]

This paper deals with random walk models and their possible application to psychological and sociological data. Such data may present a combination of apparent regularity and great difficulty of causal explanation. Random walk, implying the summation of individually small shocks or effects could account for both properties. As a special case a binary structure consisting of two interacting entities, each also influenced by an exogenous factor, is considered and ways of looking at the behavior of such systems suggested. Conventional techniques of analyzing variability rely heavily on stratification, e.g., by social class, as a means of determining the effects of environment. This method often performs rather poorly, and the outcome is put in the same perspective, and related to different conceptions of environment. Finally, developmental data and developmental correlations are examined from the same point of view. [ABSTRACT FROM AUTHOR]

In this paper the author stresses the importance of the time factor as a fundamental variable in the process of attitude change, as shown in national voting trends in Sweden, Norway, Great Britain, and the United States. More specifically, he asks: Why do these Voting trends tend to take the shape of smooth time series? He suggests alternative reasons, which he labels the "stimulus" and the "response" hypotheses, and devotes particular attention to an explanation and elaboration of the latter. [ABSTRACT FROM AUTHOR]