Following a review of the history and sources of socioeconomic indexes for occupations, we estimate a new set of indexes for 1990 Census occupation lines, based on relationships between the prestige ratings obtained by Nakao and Treas in the 1989 General Social Survey and characteristics of occupational incumbents in the 1990 Census. We also investigate theoretical and empirical relationships among socioeconomic and prestige indexes, using data from the 1994 General Social Survey. Many common occupations, especially those held by women, do not fit the typical relationships among prestige, education, and earnings. The fit between prestige and socioeconomic characteristics of occupations can be improved by statistical transformation of the variables. However, in rudimentary models of occupational stratification, prestige‐validated socioeconomic indexes are of limited value. They give too much weight to occupational earnings, and they ignore intergenerational relationships between occupational education and occupational earnings. Levels of occupational education appear to define the main dimension of occupational persistence across and within generations. We conclude that composite indexes of occupational socioeconomic status are scientifically obsolete.

ObjectivesMissing laboratory data is a common issue, but the optimal method of imputation of missing values has not been determined. The aims of our study were to compare the accuracy of four imputation methods for missing completely at random laboratory data and to compare the effect of the imputed values on the accuracy of two clinical predictive models.DesignRetrospective cohort analysis of two large data sets.SettingA tertiary level care institution in Ann Arbor, Michigan.ParticipantsThe Cirrhosis cohort had 446 patients and the Inflammatory Bowel Disease cohort had 395 patients.MethodsNon-missing laboratory data were randomly removed with varying frequencies from two large data sets, and we then compared the ability of four methods—missForest, mean imputation, nearest neighbour imputation and multivariate imputation by chained equations (MICE)—to impute the simulated missing data. We characterised the accuracy of the imputation and the effect of the imputation on predictive ability in two large data sets.ResultsMissForest had the least imputation error for both continuous and categorical variables at each frequency of missingness, and it had the smallest prediction difference when models used imputed laboratory values. In both data sets, MICE had the second least imputation error and prediction difference, followed by the nearest neighbour and mean imputation.ConclusionsMissForest is a highly accurate method of imputation for missing laboratory data and outperforms other common imputation techniques in terms of imputation error and maintenance of predictive ability with imputed values in two clinical predicative models.

No abstract

No abstract

We describe a method for producing annual estimates of the unauthorized immigrant population in the United Sates and components of population change, for each state and D.C., for 1990 to 2010. We quantify a sharp drop in the number of unauthorized immigrants arriving since 2000, and we demonstrate the role of departures from the population (emigration, adjustment to legal status, removal by the Department of Homeland Security (DHS), and deaths) in reducing population growth from one million in 2000 to population losses in 2008 and 2009. The number arriving in the U.S. peaked at more than one million in 1999 to 2001, and then declined rapidly through 2009. We provide evidence that population growth stopped after 2007 primarily because entries declined and not because emigration increased during the economic crisis. Our estimates of the total unauthorized immigrant population in the U.S. and in the top ten states are comparable to those produced by DHS and the Pew Hispanic Center. For the remaining states and D.C., our data and methods produce estimates with smaller ranges of sampling error.