Life‐Course Transitions and the Age Profile of Internal Migration

Despite underlying regularities in the age profile of migration, there is mounting evidence of cross‐national variations in the ages at which migration occurs. Explanations for these differences have variously been sought by reference to cultural, social, and economic factors, and through analysis of reasons for moving. There is also a growing body of work linking migration events to particular transitions in the life course. We set out a conceptual framework that links contextual factors to the age structure of migration through life‐course transitions that act as proximate determinants of the age at migration. We propose metrics to capture the prevalence, timing, and spread of four key life‐course transitions: education completion, labor force entry, union formation, and first childbearing. We then seek to quantitatively establish the link between these indicators and the age and intensity of internal migration at its peak for a global sample of 27 countries. Correlation and factor analysis reveal substantial diversity in the timing and spread of transitions to adult roles, and show that cross‐national differences in the age profile of migration closely parallel variations in the age structure of the life course for over two‐thirds of countries. Migration age profiles are aligned with transitions to adulthood for both sexes but most strongly among women.     

Sexes and statistics

Arthur Geissler's data for the distribution of the sexes in families with eight children, published in 1889, became famous through its use by R. A. Fisher in Statistical Methods for Research Workers in 1925 as an example of the binomial distribution. Geissler compiled data from Saxony for the period 1876-1885 for all family sizes. How did Fisher obtain the figures for his example? And why did he change the sixth significant figure of his value for the variance in his fourth edition of 1932?     

On population‐based measures of agreement for binary classifications

The authors describe a model‐based kappa statistic for binary classifications which is interpretable in the same manner as Scott's pi and Cohen's kappa, yet does not suffer from the same flaws. They compare this statistic with the data‐driven and population‐based forms of Scott's pi in a population‐based setting where many raters and subjects are involved, and inference regarding the underlying diagnostic procedure is of interest. The authors show that Cohen's kappa and Scott's pi seriously underestimate agreement between experts classifying subjects for a rare disease; in contrast, the new statistic is robust to changes in prevalence. The performance of the three statistics is illustrated with simulations and prostate cancer data.     

Enhancing response rates to a smoking survey for enlisted U.S. Navy women.

This study examined the effectiveness of a stepped approach for increasing response rates to a mailed follow-up smoking survey among newly enlisted women in the navy. The effect of the stepped approach on response rates and on the characteristics of respondents was evaluated. Also, costs were estimated for each of the steps to determine their relative benefits. Results showed that the stepped approach was effective in more than doubling the survey response rate among smokers. Reluctant respondents did not differ from on-time respondents in terms of demographics or baseline smoking, although nonrespondents were less educated and heavier smokers than on-time and reluctant respondents. Strategies documented here could well apply to survey efforts with other hard-to-reach populations.     

The politics of concepts: family and its (putative) replacements

The central concern of this paper is that there has been a move within British sociology to subsume (or sometimes, even replace) the concept of ‘family’ within ideas about personal life, intimacy and kinship. It calls attention to what will be lost sight of by this conceptual move: an understanding of the collective whole beyond the aggregation of individuals; the creation of lacunae that will be (partially) filled by other disciplines; and engagement with policy developments and professional practices that focus on ‘family’ as a core, institutionalized, idea. While repudiating the necessity (and indeed, pointing out the dangers) of providing any definitive answer to definitions of ‘family’, the paper calls for critical reflection on the implications of these conceptual moves.     

Model Selection and Regression t-Statistics

It is shown that dropping quantitative variables from a linear regression, based on t-statistics, is mathematically equivalent to dropping variables based on commonly used information criteria.