Lisa Dillon

The North Atlantic Population Project An Overview

Abstract The North Atlantic Population Project (NAPP) brings together complete-count census data from late-nineteenth-century Canada, Great Britain, Iceland, Norway, and the United States into a single harmonized database. When released in 2005, the final version of the database will include the records of nearly 90 million people. The project will consistently code all variables across the different countries, while still retaining important national variation in census questions and responses. The authors provide a brief history of the project, discuss the main issues involved in creating a harmonized international census database, and outline the methodological and research opportunities the completed database will provide for scholars.

BEST PRACTICES WITH LARGE DATABASES ON HISTORICAL POPULATIONS

Since the late 1960s researchers who transform routinely generated primary sources into machine-readable data have produced numerous methodological articles and book chapters detailing the process of data creation and describing how the peculiarities of primary sources can affect interpretation of the data. However, in such articles, the best practices for creating large databases have usually been implicit rather than explicit. Here we introduce a comprehensive list of best practices for the creation of large databases on historical populations, drawing upon the experiences of the Historical Sample of the Netherlands (HSN), the Canadian Families Project, the Integrated Public Use Microdata Series (IPUMS), and other projects. The following guidelines represent a revised version of the protocol formulated on the occasion of the HSN Workshop on Large Databases: Results and Best Practices, held in Amsterdam on 17-18 May, 2001. All the participants

Occupational classification in the North Atlantic Population Project

Abstract The North Atlantic Population Project (NAPP) is a complete-count data set of late-nineteenth-century censuses from Canada, Great Britain, Iceland, Norway, and the United States. One of the projects most challenging tasks is the coding and classification of 2 million distinct responses to occupational questions. Using the Historical International Standard Classification of Occupations (HISCO) as the basis for their classification scheme, the authors have adapted it to address particular problems applicable to the NAPP occupational data—the inconsistent specification of tasks, industry, and employment status by census respondents; variation among the NAPP countries in the level of occupational detail provided; and spatial and temporal variation in the language used to describe occupations. Compared with HISCOs classification scheme, the NAPP system reduces the overall number of codes, introduces new codes, and retains more detail from vaguely specified occupations.

The consequences of sibling rivalry on survival and reproductive success across different ecological contexts: a comparison of the historical Krummhörn and Quebec populations

Abstract This article investigates the relationship between additional siblings and the probability of offspring survival, marriage, and fertility across the historical populations of the St Lawrence Valley in Quebec (1670–1799) and the Krummhörn region in Germany (1720–1874). Both populations existed in agriculturally based economies, but differ in important ways. The Krummhörn population faced a fixed supply of land, which was concentrated amongst a small number of farmers. Most individuals were landless agricultural workers who formed a relatively competitive labor supply for the large farmers. In contrast, individuals in Quebec had access to a large supply of land, but with far fewer available agricultural workers, and had to rely on their family to develop and farm that land. Results indicate that more siblings of the same gender were generally associated with increases in mortality during infancy and childhood, later ages of first marriage, and fewer numbers of children ever born. For mortality and age at first marriage, the effects of sibling formation appear strongest in the Krummhörn region. Notwithstanding these observed differences, the general consistency and robustness of the sibship effect across the different ecological and economic contexts is our most interesting result. In addition, through side-by-side comparison of across-family and within-family analyses, we argue that sibling competition – or sacrifice – is manifested as an internal familial dynamic, but is obscured in non-fixed effects models by a broader trend of family cooperation. Through this comparison we are able to reconcile family solidarity and sibling competition/sacrifice as coexisting phenomena. Results are robust to inclusion of covariate interactions with time, inclusion of indicators for high levels of extrinsic risk, estimation of shared frailty models, alternative methods of dealing with ties in the dataset, including recomposed families in the dataset, excluding individuals whose death dates are ‘heaped’, and excluding individuals born to large families.

Pandemic Paradox: Early Life H2N2 Pandemic Influenza Infection Enhanced Susceptibility to Death during the 2009 H1N1 Pandemic

ABSTRACT Recent outbreaks of H5, H7, and H9 influenza A viruses in humans have served as a vivid reminder of the potentially devastating effects that a novel pandemic could exert on the modern world. Those who have survived infections with influenza viruses in the past have been protected from subsequent antigenically similar pandemics through adaptive immunity. For example, during the 2009 H1N1 “swine flu” pandemic, those exposed to H1N1 viruses that circulated between 1918 and the 1940s were at a decreased risk for mortality as a result of their previous immunity. It is also generally thought that past exposures to antigenically dissimilar strains of influenza virus may also be beneficial due to cross-reactive cellular immunity. However, cohorts born during prior heterosubtypic pandemics have previously experienced elevated risk of death relative to surrounding cohorts of the same population. Indeed, individuals born during the 1890 H3Nx pandemic experienced the highest levels of excess mortality during the 1918 “Spanish flu.” Applying Serfling models to monthly mortality and influenza circulation data between October 1997 and July 2014 in the United States and Mexico, we show corresponding peaks in excess mortality during the 2009 H1N1 “swine flu” pandemic and during the resurgent 2013–2014 H1N1 outbreak for those born at the time of the 1957 H2N2 “Asian flu” pandemic. We suggest that the phenomenon observed in 1918 is not unique and points to exposure to pandemic influenza early in life as a risk factor for mortality during subsequent heterosubtypic pandemics. IMPORTANCE The relatively low mortality experienced by older individuals during the 2009 H1N1 influenza virus pandemic has been well documented. However, reported situations in which previous influenza virus exposures have enhanced susceptibility are rare and poorly understood. One such instance occurred in 1918—when those born during the heterosubtypic 1890 H3Nx influenza virus pandemic experienced the highest levels of excess mortality. Here, we demonstrate that this phenomenon was not unique to the 1918 H1N1 pandemic but that it also occurred during the contemporary 2009 H1N1 pandemic and 2013–2014 H1N1-dominated season for those born during the heterosubtypic 1957 H2N2 “Asian flu” pandemic. These data highlight the heretofore underappreciated phenomenon that, in certain instances, prior exposure to pandemic influenza virus strains can enhance susceptibility during subsequent pandemics. These results have important implications for pandemic risk assessment and should inform laboratory studies aimed at uncovering the mechanism responsible for this effect. IMPORTANCE The relatively low mortality experienced by older individuals during the 2009 H1N1 influenza virus pandemic has been well documented. However, reported situations in which previous influenza virus exposures have enhanced susceptibility are rare and poorly understood. One such instance occurred in 1918—when those born during the heterosubtypic 1890 H3Nx influenza virus pandemic experienced the highest levels of excess mortality. Here, we demonstrate that this phenomenon was not unique to the 1918 H1N1 pandemic but that it also occurred during the contemporary 2009 H1N1 pandemic and 2013–2014 H1N1-dominated season for those born during the heterosubtypic 1957 H2N2 “Asian flu” pandemic. These data highlight the heretofore underappreciated phenomenon that, in certain instances, prior exposure to pandemic influenza virus strains can enhance susceptibility during subsequent pandemics. These results have important implications for pandemic risk assessment and should inform laboratory studies aimed at uncovering the mechanism responsible for this effect.

The Programme de recherche en démographie historique: past, present and future developments in family reconstitution

Abstract The Programme de recherche en démographie historique (Historical Demography Research Programme) (PRDH), founded in 1966 and based at the Département de Démographie of the Université de Montréal, has since its inception featured a central project, a family reconstitution database of Quebec’s Catholic population from 1621 to 1799 named the Registre de la population du Québec ancien (Population Register of Historic Quebec) (RPQA). This article, which marks the fiftieth anniversary of the project, explores the development of the RPQA over the five decades in the context of similar international databases, explains the current state of the database as well as our record linkage methodology, describes an important collaboration now underway to build a larger Quebec historical data infrastructure, outlines new and renewed international collaborations, and summarizes research conducted using these data as well as future research possibilities. The particular geographic context, historical development and manageable colonial population size of Quebec favoured family reconstitution of the whole colony from the beginning of the project. Today, the RPQA comprises 438,193 individual biographies and 74,000 family files encompassing up to nine generations. To reconstitute families, we must identify and incorporate into the database all demographic events, including those whose existence can only be inferred through other sources. Future efforts to link nineteenth-century parish acts will need to deal with large case counts, mixed Catholic–Protestant marriages, and increased geographic and social mobility. The integration of complementary data will provide information on household co-residence, occupations, help track the destinies of mixed-religion persons and persons outside nuclear families and provide additional points of observation.

Mining microdata: Economic opportunity and spatial mobility in Britain and the United States, 1850–1881

For almost two centuries social theorists have argued that the fundamental difference in social structure between Europe and North America arises from greater economic and geographic mobility in North America. We study social mobility in three countries across two generations using machine learning techniques to create panels of individuals linked between censuses thirty years apart (1850-1880, 1880-1910). This paper reports on a preliminary analysis of social mobility between 1850 and 1880, finding that mobility was markedly higher in the United States and Canada, compared to Great Britain.

Women and the dynamics of marriage, household status, and aging in victorian canada and the united states☆

This article compares the life course transitions and household statuses of Canadian and American women and men in late nineteenth-century Canada and the United States. Using a set of integrated census data from 1871 Canada and the United States in 1880, the article suggests that household status differences between the two nations centered on gender. Canadian and American men timed or experienced their own transitions into and out of marriage and household headship at similar ages and to a similar extent. Demographic and economic differences between Victorian Canada and the United States, however, produced distinctions in Canadian and American womens life course transitions and household status: for Canadian women, older ages at first marriage, and the prolongation of the duration of the status, spouse of the household head. For their part, American elderly women more frequently lived as single and widowed heads of households than did their Canadian counterparts.

Using the Canadian Censuses of 1852 and 1881 for Automatic Data Linkage: A Case Study of Intergenerational Social Mobility

This chapter discusses the issues of missing and uncertain data in the Canadian census sample of 1852 within the context of automatic linkage with the complete census of 1881. The resulting linked sample from these two censuses was created to provide an opportunity to study intergenerational social mobility in Canada between fathers (in 1852) and sons (1881). We discuss the accuracy and representativeness of the automatically generated links and show how the use of marriage registers can be helpful in order to verify the results of the automatic linkage. Our verifications suggest that most of the links are accurate. However, the linked sample is not representative of some subgroups of the studied population, since some attributes favoured while others hindered the fact of being automatically linked from 1852 to 1881. Finally, based on our efforts of manual linkage between the BALSAC marriage registers and the automatically linked census sample for the verification of the latter, we present some considerations about the great research potential of linking census and parish register data in Quebec.

Epidemic cycles and environmental pressure in colonial Quebec: BRUCKNER et al.

Research on historical populations in Europe finds that infectious disease epidemics appear to induce predictable cycles in age‐specific mortality. We know little, however, about whether such cycles also occurred in less dense founder populations of North America. We used high‐quality data on the Quebecois population from 1680 to 1798 to examine the extent to which age‐specific mortality showed predictable epidemic cycles. We further examined whether environmental pressures—temperature, lack of precipitation, or crop failure—may have set the stage for the emergence of epidemics.