A Statistical Cookbook as Novel: “Historical Statistics of the United States”

Cows and methane

Food and statistics collide in this infographic!

My father reads cookbooks as if they were novels, from beginning to end, looking through every corner of every ingredient list, watching the story’s protagonist move past the dangers of broiling until being served up on a platter with roasted carrots and a garnish of kale and a melon slice.

For him, it’s a literary journey. To me, it’s just a cookbook.

That’s sorta how I felt about the library resource entitled “Historical Statistics of the United States”. I thought, “I bet it’s just a pile of statistics.” But I was wrong. The “Historical Statistics of the United States” [hereafter referred to as “HStUS”, pronounced “Hostuss” which makes me hungry] is a resource that is full of essays that transform HStUS from a statistical dump into a rich text about American history, sociology, and economics [I don’t mean to demean statistics by using the word “dump”, but I must confess that I find raw statistical data without context as exciting as rotting vegetables].

Each subtopic explored in HStUS is accompanied by a contextual essay. Each essay provides the following information:

1- Details on how the data was compiled using statistical tools and methods;
2- Descriptions of the subject matter that explain and provide context;
3- How the data informs the bigger picture of change across time.

As an example, let’s look at the essay that accompanies the 15 tables of statistics in the subtopic “Family and Household Composition”.

1- The essay begins with an explanation of the data and the problems associated with data about “Family and Household Composition”:

The Census Bureau published minimal statistics on families and households until 1940, focusing mainly on the size distribution of households. Even for the period since 1940, the official published statistics on the subject are minimal…During the past two decades, new microdata samples of historical censuses have become available. These data are collected in the Integrated Public Use Microdata Series (IPUMS), a coherent national database describing the characteristics of 55 million Americans in thirteen census years spanning the period from 1850 through 1990…

In addition to problems caused by shifting census definitions, statistical series can also be distorted by changes in census processing. In particular, Ruggles and Brower argue that because of errors in tabulation procedures, published statistics on subfamilies from both the census and the Current Population Surveys are unreliable. Accordingly, the statistics relating to subfamilies that are presented here are derived entirely from the IPUMS.

Okay, if you’re like me, that’s not the most thrilling text you have ever read, BUT it does provide a riveting account of how the authors worked the data.

2- The essay then provides a conceptual context in the section entitled “Trends in American Living Arrangments”:

Within the preindustrial family economy, older-generation men exercised control over the means of production. Women and younger-generation men provided labor in exchange for food, shelter, and economic security. The decline of farming, the rise of wage labor, and the growth of mass education fundamentally shifted the balance of power within American families. First, the rise of wage labor among men reduced the importance of agricultural and occupational inheritance by providing opportunity for young men. Second, the rise of wage labor among women curbed the control that husbands and fathers exercised over their wives and daughters…

Hence, the transformation of the economy made the transformation of the family possible. The changes in the family were not, however, purely economic; little would have happened had there not also been profound attitudinal changes. It is not especially useful to debate whether the economic or cultural changes were primary; both were essential.

3- Within this discussion, the essay’s author provides statistical support and links to the relevant data:

The changing marital status of mothers with children under 18 years old is summarized in Figure Ae-C. From 1880, when marital status was first recorded in the census, until 1950, the overall percentage of young children without married mothers declined slightly from 11.6 to 8.8 percent. The percentage of children with divorced or separated mothers more than doubled during this period, but that increase was canceled out by a dramatic decline in the percentage of children with widowed mothers. From 1950 to 1990, however, the percentage of children residing with never-married mothers rose sixteen-fold. Simultaneously, the percentage of children with divorced or separated mothers continued to rise. By 1990, about one quarter of all children was residing with a single mother.

The increase in single parenthood is a consequence of the rapid rise of divorce, separation, and unmarried fertility (Table Ae507–513). The causes of these changes have been vigorously debated. The traditional explanation is that rising female labor force participation weakened marriage. Writing in 1893, Emile Durkheim pointed to the sexual division of labor as a source of interdependence between men and women, producing what he called “organic solidarity.” Durkheim warned that if the sexual division of labor receded, “conjugal society would eventually subsist in sexual relations preeminently ephemeral” (Durkheim 1960 [1893], p. 60).

Maybe this text will never be feted on the New York Times Best Seller List, but I find the essays in HStUS compelling.

For those of us whose brains shut down when they come within twenty feet of a table of numbers, this libary resource can be a valuable tool for staying awake while researching statistical data.

I feel that the essays in HStUS are excellent examples of how to provide contextual information in a discussion of statistical analysis. They would be of benefit in classes concerning history, sociology, as well as statistics.

Below you will find some images that show the assorted topics and sub-topics that are described in HStUS.

“Historical Statistics of the United States” can be found by…
Going to the library’s main website (http://www.library.millersville.edu) >
Scroll to the bottom and click on “Articles and Databases” >
Click on “All Databases by Title” in the right hand column >
Scroll down to the “H” section>
You’ll find the link to “Historical Statistics of the United States” in the middle of that section.

If you have any questions about using this resource, please feel free to contact me at nathan.pease@millersville.edu.

[Nathan Pease is an adjunct Research Librarian at the McNairy Library and Learning Forum on the campus of Millersville University. In his spare time, Mr. Pease digitizes out-of-print vinyl records and plays “European board games” such as Targi, Pandemic, Dominion, among others. He also volunteers and works part-time atLancasterHistory.Org, also know as the Lancaster County Historical Society.]

Citation for the essay excerpts: Ruggles, Steven , “Family and Household Composition” in chapter Ae of Historical Statistics of the United States, Earliest Times to the Present: Millennial Edition, edited by Susan B. Carter, Scott Sigmund Gartner, Michael R. Haines, Alan L. Olmstead, Richard Sutch, and Gavin Wright. New York: Cambridge University Press, 2006. http://dx.doi.org/10.1017/ISBN-9780511132971.Ae.ESS.01

[The image at the top of this blog, “Methane production of cows vs grass” by allispossible.org.uk, is licensed under CC BY-NC 2.0.]