MODELING SPATIAL AND FAMILY FACTORS IN THE TRANSMISSION OF WEALTH | Grant

SOUTH CAROLINA RESEARCH FOUNDATION

Grant: $226,333 - National Institutes of Health - Sep. 22, 2009

No votes have been cast for this award yet

Join the conversation: Post a comment about this award

Award Description: MODELING SPATIAL AND FAMILY FACTORS IN THE TRANSMISSION OF WEALTH Specific Aims Our goal is to specify conditions under which men were able to accumulate wealth between 1850 and 1870 in a native born US population of related individuals. To do this, we will use three new Bayesian methods for estimating how families of origin, migration histories and occupational changes affected household wealth over time. SA1. To estimate the respective contributions of family wealth and current spatial location to household wealth at three time points when the census provides wealth and household information (1850, 1860, 1870). Method 1: Bayesian hierarchical spatial modeling using WINBUGS can model the wealth of particular types of relatives (fathers and brothers, for example) and spatial autocorrelation at the same time. Thus one can tell how much of a households wealth was associated with spatial effects and how much to the wealth of these relatives. We will include the length of time the family has been in the location as well as the occupation of the father as covariates. SA2. To analyze the extent to which men could change their wealth by moving or changing occupations over the 20 year period. This will include investigating contextual effects by studying whether rich and poor households fared differently depending on the opportunities available where they were located. These opportunities will be measured by county level economic measures and a GIS containing soil quality, elevation, and transportation networks. Method 2: Longitudinal Bayesian models will be used over the three time points to assess the degree to which changes in occupations and in locations may have affected wealth accumulation. SA3. To analyze the spatial clustering of wealth in this population at the three time points and how this changed over time, focusing especially on how wealth clustering was affected by migration and family processes, such as the death of parents, as well as occupations. We will examine how the wealth of farmers was affected by soil quality and whether sons who left farming from such clusters retained their wealth. Method 3: Bayesian spatial cluster models analyze the aggregation of wealth and explain it by the subsequent analysis of the covariates. This involves the use of special Bayesian cluster models but is different from Method 1 in that the clustering is directly modeled with associated covariates. The dataset is based upon 9 published genealogies each of which traces the descendents of a founder who arrived in Massachusetts from England before 1650. The population increased rapidly and after 1650 most migrants from Europe came to the middle colonies. So this population was largely a closed group which, after the Revolution, pioneered New York and the Middle West. During the period of study, they became the Yankees and were living across the US North. The history of this sample mirrors that of the native born living in the areas of the US North to which this population subsequently spread (see Matthews 1962) . The genealogies link relatives in the male line and provide migration histories going back 8 to 10 generations. The manuscript censuses of 1850, 1860 and 1870 provide information on their wealth, household composition, location, and occupations. We will analyze a set of migration streams of the 20 years, which we think were also wealth trajectories.

Project Description: As defined in the Award Description field.

Jobs Summary: None created (Total jobs reported: 0)

Project Status: Not Started

This award's data was last updated on Sep. 22, 2009. Help expand these official descriptions using the wiki below.