|
The |
|
Evan Roberts, Steven Ruggles, Lisa Y Dillon, |
|
Full Text (6072 words) |
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Copyright HELDREF
PUBLICATIONS Spring 2003
The North Atlantic
Population Project (NAPP) brings together late-nineteenth-century
complete-count census microdata from This article provides
a brief history of the project, a discussion of the main issues involved in
creating the database, and an overview of methodological and research
opportunities the completed database will provide scholars. Occupational classification
is the most complex and time-consuming aspect of the project and has received
much of our attention so far. The major issues in classifying occupations in
the NAPP database are discussed in a companion article by Roberts et al. (pp.
89-96 in Part Two of this issue). Background The NAPP will
incorporate complete, machine-readable versions of late-nineteenth-century
population censuses of The NAPP is made
possible by the availability of complete machine-readable transcriptions of
nineteenth-century census enumerators' manuscripts. The Church of Jesus
Christ of Latter-day Saints (LDS), in collaboration with local genealogical
societies, laboriously transcribed and entered into databases the 1881
censuses for In In 2000, participants
from each country met in Collaborators The NAPP is a
collaborative project composed of research teams from each of the five
countries included in the database. These five countries are fortunate to
have extraordinarily rich collections of surviving individual-level census
data, as shown in table 1. The
The NAPP database will
include the seven censuses marked as "N" in table 1: it will
combine the three LDS-transcribed censuses of Great Britain, Canada, and the
United States with two censuses each from Iceland and Norway.6 National
research teams have prepared-or are in the process of preparing-these data
for public use. The NAPP collaborators have pieced together funding for data
preparation from numerous sponsors to support the painstaking tasks of data
cleaning and coding. This project does not fund these basic data preparation
activities; rather, it is intended to coordinate those activities to produce
a harmonized database. The NAPP involves researchers
from the Universities of Montreal and Ottawa in Canada, the University of
Essex in Great Britain, Statistics Iceland, the Universities of Bergen and
Tromso in Norway, and the University of Minnesota in the United States. Each
of the research teams brings extensive experience with large-scale
nineteenth-century census data; most of the participants are key researchers
on projects to create microdata samples and complete-count data sets listed
in table 1.7 Coordinating work in five countries on two continents is
challenging. We communicate extensively through Web-based tools, e-mail,
telephone, and annual meetings of all collaborators. The first two
meetings-held in Minneapolis in November 2001 and in Grundarfjor[eth]ur,
Iceland, in October 2002-focused on developing common coding schemes for
occupations and other key variables. It can also be difficult to reach
agreement when a large group is concerned, but the collaboration is guided by
the philosophy that decisions should be reached by consensus and if that is
not possible, then a majority decision is required. Harmonizing the NAPP
Database Harmonization of these
North Atlantic censuses is eased because of a high degree of comparability in
the original enumeration processes across countries. This comparability was
due primarily to the formal as well as informal collaboration of North
American and European statisticians and to the development of international
standards in forums such as the International Statistical Congress and the
International Statistical Institute. For example, Joseph Kennedy, the United
States census superintendent, became a national statistical expert in part
through his interactions with European statisticians (Anderson 1988, 46, 60).
Enumeration practices in the United States as well as across the Atlantic
informed the design of the nineteenth-century Canadian censuses (Curtis 1994,
2000; Worton 1998).
Table 2 describes the
data files to be included in the North Atlantic database. With respect to
their structure, organization, and available information, the seven censuses
are remarkably comparable. Each census enumerated individuals, and in each
case, those individuals are grouped into households, which were defined in
similar terms in each country. All countries defined a household as "a
group of people sharing a common place of residence." There is a core
set of variables common to virtually all data sets, including relationship of
each individual to the household head, age, sex, marital status, occupation,
and birthplace. Geographical precision for place of residence varies across
countries, but it is possible to identify all places with populations over
5,000 for each country. Across the five countries, we identify approximately
25,000 communities. These common core variables will allow us to construct a
variety of new variables describing community and neighborhood
characteristics, household composition, socioeconomic status, and family
interrelationships.
Most of the censuses
were taken on a de jure basis, under which individuals who were temporarily
absent from home-such as migrant workers and travelers-were to be enumerated
at their usual place of residence. The exception is the British census, taken
under a de facto rule, which specified that no one who was present on census
night at a particular address could be left out of the tally and that no
person absent from home could be written in. From 1870 onward, persons in
Norway and Iceland were to be enumerated both at their usual place of
residence and at the place they stayed on census day, and enumerators
identified both temporary visitors and absent household members. These
variations in enumeration rules pose only minor compatibility problems, but
investigators of some household composition and migration issues will have to
be aware of them. There are two main
tasks necessary to create the harmonized NAPP database. First, we must
develop common classification systems for occupation, birthplace, family
relationship, and group-quarters type that balance the goals of international
comparability, retention of detail, and ease of use. Second, we must build a
consistent set of constructed variables describing household composition,
family interrelationships, urban and metropolitan residence, and other
geographic characteristics. The most challenging
task is the development of harmonized coding systems to allow comparison
across countries. At present, much of the data arc in the form of alphabetic
character strings, representing the transcription from the original source
material. These strings are too unwieldy for statistical analysis; to make
the data usable, we must develop numeric codes for each variable. For
variables such as age, sex, marital status, and religion, this process is
straight-forward, because the variation in values for these variables across
countries is not great. Other variables are more complicated; for example, we
estimate that there are unique strings for 2 million occupations, 1 million
birthplaces, and 30,000 family relationships. To translate from
character strings into numeric codes, we must construct data dictionaries
that assign a numeric code to each alphabetic variation that occurs in the
data. This work is difficult enough in the context of a single country; for a
project of this scale, it requires a team of expert coders working in close
cooperation, sharing coding decisions continuously. We have created a set of
merged dictionaries of unprecedented scale that include the alphabetic
strings from 90 million persons in all five countries. This is uncharted
territory. Until recently, constructing such a dictionary would have
necessitated assembling experts with appropriate language skills and
historical knowledge about each country in a single location, a prohibitively
expensive task. Using Internet technology, however, we are able to distribute
the task, thus working on the dictionaries in multiple countries
simultaneously. For a description of this coding process for occupation, the
most complicated variable, see the Roberts et al. companion article (pp.
89-96 in Part Two of this issue). When feasible, the
coding is compatible with the IPUMS-USA and IPUMS-International coding
systems. We also follow the same IPUMS principle of retaining detail while
providing harmonized codes suitable for comparative research. Composite
coding systems allow us to provide information available across all censuses
in a least-common-denominator code and information unique to one country (or
set of countries) in a subsequent "detail" code. We also preserve
country-specific variables with information or coding systems unique to a
particular country; these include, for example, religion (only available in
Canada and Norway) and occupations coded to the 1881 British classification
scheme. This sensitivity to the particular conventions of each country's data
is important for maintaining productive collaboration and will maximize the
usefulness of the finished database. In addition, researchers will be able to
download the raw data and analyze, replicate, or alter the coding systems
applied by the NAPP team. We will create a
number of constructed variables to facilitate the analysis of household
composition, family interrelationships, and own-child fertility. These
household- and person-level variables are detailed in table 2. Among the most
important of these are pointer variables that indicate the location within
each household of each individual's own spouse, mother, and father, as well
as number of own children.8 These pointer variables allow researchers to
attach the characteristics of immediate kin (spouse's birthplace, father's
occupation, children's characteristics, etc.) without the need to resort to
programming (Ruggles 1995). We will add the variables required for basic
own-child fertility analysis: number of own children present, number under 5
years of age, age of eldest child, and age of youngest child. We also plan to
construct a set of variables to describe the composition of families and
households, replicating the most commonly used historical and contemporary
classification systems, as well as household characteristics such as the
number of persons, families, children, married couples, and servants in each
household. Finally, we will create technical variables to aid in data
management and analysis, such as record type, serial number, and
group-quarters residence. Documenting and
Disseminating the NAPP Database In addition to data preparation, the project
involves documenting the comparability of census enumeration instructions,
procedures, and definitions across the five countries, creating
machine-understandable metadata compliant with the Data Documentation
Initiative (DDI) metadata standard, and developing efficient Web-based
software to optimize and simplify access to the entire database. The creation of
integrated documentation is central to the project, and the development of
these materials will be a collaborative enterprise. We are working on
comprehensive documentation for each of the censuses in the database,
including facsimiles of census forms and enumerator instructions. When the
original documentation is in another language, we plan to translate the most
essential material into English. When foreign-language material is extensive,
however, we will provide the full text in the original language with a brief
English-language translation. The documentation system will also describe all
procedures undertaken to generate the harmonized database, including the
actual computer code, data dictionaries, and a textual description of the
data manipulation process. All documentation will be compliant with the DDI
metadata standard (see Block and Thomas, pp. 97-101 in Part Two of this
issue). Few users will be
interested in analyzing the entire data set of 90 million cases. Therefore,
we will disseminate the database through a data extraction system that
integrates access to metadata and microdata and allows users to carry out
substantial manipulation of the data. The data access system for the North
Atlantic database will be based on a similar system currently under
development as part of the IPUMS-International project (see Ruggles et al., pp.
60-65 in Part Two of this issue). Web sites in Great Britain, Canada, Norway,
and the United States will mirror the NAPP data access system and provide
unrestricted access for academic researchers. Applications for
Complete-Count Data The availability of information
on entire populations will open important new avenues of research. The
following discussion describes some new methodological approaches that will
be possible with complete-count data. Study of small
dispersed population subgroups. Many small population subgroups-defined by
race, ethnicity, occupation, or even age-can be studied only with very high
density data sets. For example, the availability of complete-count data will
allow the study of the indigenous populations of Canada, Norway, and the
United States. The 1881 Canadian census microdata include over 60,000
aboriginal persons; these cases would include sufficient indigenous women of
childbearing age to allow in-depth fertility analysis. Similarly, existing
sample data are inadequate to study immigrant groups in detail. A substantial
proportion of the Icelandic population emigrated to Canada and the United
States, but there are insufficient Icelandic cases in any of the existing
samples to allow quantitative analysis. This is also true for the Irish
population in England. The number of Norwegians in North America is larger,
but there are still not enough for detailed analysis using a sample database.
The new database will also allow, for the first time, comparative study of
particular age groups such as centenarians. It will also have adequate cases
to compare specific occupations (e.g., sailors, fishermen, and sons of
farmers) across all five nations. Community studies. The
community study has been one of the most fruitful analytical approaches in
both history and sociology. Most sample data lack sufficient cases to enable
researchers to examine particular localities. Because the new database
includes the entire population, it will allow scholars to extract customized
data sets focusing on particular communities. The international dimension of
the database will allow investigators to undertake comparative community
studies. For example, researchers could compare patterns in a Minnesota
Norwegian community with the sending community in Norway. Investigators could
also examine cross-border Canadian-U.S. or English-Scottish communities, to
compare economic and social patterns in neighboring areas distinguished only
by political and institutional structures. There is a large demand for local
historical statistical data, and the North Atlantic database will immediately
become an essential tool for community historians of all sorts. Detailed
personal identifiers for individuals-including full name, age, sex,
occupation, marital status, and birthplace-create the potential for
researchers to link NAPP data with other contemporaneous sources of basic
demographic information, such as ship lists, school and college attendance
records, friendly society (benefit-society) membership lists, as well as
birth, death, marriage, and emigration registers. Even historians who make
little use of quantitative analysis will be able to quickly and painlessly
locate their study subjects in the manuscript census. Longitudinal analysis.
Perhaps the greatest limitation of the existing samples is that they are
cross-sectional snapshots and do not allow one to trace individuals across
time. This problem will be alleviated by the new database. As shown in table
1, for each country there are machine-readable samples or complete-count data
sets for multiple years. Thus, it will be possible to create a series of
linked samples; in the case of Canada, for example, individuals in the 1871,
1891, and 1901 census samples can be linked with the complete-count 1881
census. Thus, researchers will be able to construct three linked Canadian
samples, covering 1871-1881, 1881-1891, and 1881-1901. In Norway, Iceland,
and Great Britain-where there are multiple complete censuses-there is
potential to link individuals across more than a single pair of census years.
Researchers will even be able to link some individuals across countries; for
example, from Norway in 1865 to Canada in 1881 and back to Norway in 1900. Linked census data
hold the promise of finally resolving some of the longest-running debates in
nineteenth-century social history. Past studies of social and geographic
mobility were not entirely conclusive because of their exclusion of migrants
and their small sample size. Scholars will be able to gauge the extent of
social and geographic mobility, analyze the interrelationship of geographic
and economic movement, and assess trends and differentials in social mobility
far more reliably than heretofore (Thorvaldsen 1995). In addition, the linked
samples will allow investigation of questions regarding family formation and
dissolution. For example, they will allow scholars to answer several
controversial questions surrounding the formation of multigenerational
households in the nineteenth century (Ruggles 1994, 2001). Multilevel analysis.
In recent years, analyses of the effects of local context on individual
behavior have proven exceedingly valuable tools for research in historical
sociology (see Elman 1998; Kramarow 1995; Ruggles 1997a, 1997b). New methods
of multilevel analysis will increase the effectiveness of this approach
(DiPrete and Forristal 1994). A key problem for such nineteenth-century
research, however, is that the method requires independent variables
tabulated for small geographic units, and such data are scarce before the
twentieth century. The new North Atlantic database will allow creation of a
wide variety of contextual variables-such as racial or ethnic composition,
female labor-force participation, and occupational structure-at many
geographic levels, including the neighborhood and the enumeration district. Geographic information
systems. Geographer's are ordinarily unable to tap the power of microdata.
The existing nineteenth-century microdata files are samples, so when they are
used for small areas they provide insufficient precision for reliable
mapping. Although some relatively high density samples are available for the
period since 1970, those microdata files suppress detailed geographic data.
Therefore, geographers are forced to rely on complete-count aggregate data
that usually provide only basic summary statistics for small areas. The North Atlantic
census database will provide full geographic detail for every individual in
the population. Digitized small-area boundary files are already available for
Great Britain in 1881 and are in preparation for nineteenth-century Canada,
Norway, and the United States (Gregory and Southall 1998; Thorvaldsen 1997;
see also Fitch and Ruggles, pp. 41-51 in Part One of this issue).9 Therefore,
a large scholarly investment in nineteenth-century geographic information
systems already exists. What is lacking is a fine level of geographic detail
in social, economic, and demographic characteristics. The North Atlantic
census database will allow scholars to marry existing geographic boundary files
to population characteristics, thus creating a powerful new analytic tool.
Such fine geographic analysis will be especially potent in the analysis of
topics such as early suburban development and racial and ethnic residential
segregation (see, for example, Gardner 1998). Substantive research
areas. A cross-section encompassing the entire population of the North
Atlantic world in the late nineteenth century will address many of the major
themes of history, economics, demography, and sociology. The following four
examples describe some of these key substantive issues: 1. The North Atlantic
database will allow in-depth analysis of the processes of industrialization.
The first Industrial Revolution may have begun in rural England, but by the
late nineteenth century the entire North Atlantic world was involved in
manufacturing, production of raw materials, or both. The North Atlantic
database will allow unprecedented opportunities to explore economic
structures within and between each nation during this critical transitional
period. For the first time, we will have consistently coded individual-level
occupational data available for nineteenth-century censuses from multiple
countries. This will allow comparative analysis at the level of persons,
families, communities, or regions, as well as allow investigation of the
geographic organization of economic activity. Thus, the NAPP database will
allow scholars to address international and North American regional
differences in the scale and location of manufacturing activity, comparing,
for example, the distribution of particular industrial workers in large urban
centers versus smaller towns and villages. 2. The database will
also contribute to understanding of the fertility transition. At the time
these censuses were taken, each of the North Atlantic countries was
witnessing the impact of widespread deliberate fertility limitation.10 The
North Atlantic database will allow study of differential fertility patterns
in this critical period of demographic transition, to assess the importance
of such factors as occupational class, ethnicity, region, literacy, local
economy, size of locality, and family structures. Past comparative analyses
of the European fertility transition have relied on national-, regional-, or
parish-level aggregate vital statistics (Coale and Watkins 1986). This
approach has two major disadvantages. First, aggregate vital statistics do
not allow direct measures of child spacing or stopping behavior; only the
level of fertility can be considered. Second, the aggregate approach does not
allow control of individual-level socioeconomic characteristics. Study of
this elemental shift in population structure has the potential to enhance our
understanding of ongoing demographic change in the contemporary developing world
(Woods 2000). 3. Research on the
family will also benefit from the common set of constructed variables in the
NAPP database. These variables aid in the analysis of family and household
composition and will thus allow consistent comparisons across all five
countries. It will enable investigators to assess the impact of local context
on family systems through multilevel analysis and so for the first time
permit analysis of the effects of individual-level factors, local economic
conditions, regional inheritance systems, and national characteristics on the
nineteenth-century family. 4. Finally, the North
Atlantic database will constitute an outstanding resource for the study of
migration history. It will allow close and consistent comparisons of
occupational structure, marriage patterns, fertility, and family composition.
Researchers will be able to identify and compare specific sending and
receiving communities. In some instances, it will even be possible to follow
individual migrants across the Atlantic and back again. The NAPP database
will also facilitate the study of cross-border populations.11 In combination
with new machine-readable ship lists and emigration registers, the database
will open a new window on the implications of international population flows. Conclusion This article has
outlined the ongoing project to create the North Atlantic Population Project
database. Capitalizing on three decades of experience in creating census
samples, the NAPP applies them to a complete-count database including data
from five nations. Harmonizing five databases of this scale poses a range of
methodological and practical challenges. The potential payoff, however, is
great; we are confident that the twin benefits of complete-count data and the
ability to compare countries will have substantial benefits for social
science research. The NAPP will be compatible with existing and forthcoming
time-series of census data for all those countries contributing data. We hope
it will spur future efforts to increase the sample density of existing census
samples and promote comparative and transnational research.
|
|
|
Reproduced with
permission of the copyright owner. Further reproduction or distribution is
prohibited without permission.
|
Subjects: |
|
|
Locations: |
|
|
Article types: |
Feature |
|
ISSN/ISBN: |
01615440 |
|
Text Word Count |
6072 |
|
|
|