National Longitudinal Survey of Youth 1979

National Longitudinal Survey of Youth (1979 – 2012) is a longitudinal project that follows a sample of American youth born between 1957-64 on various life aspects from 1979 to 2012. The data set provided below is a subset of this database, focusing on variables of 4 main topics: socioeconomic status, employment, education, and marriage. Some recommended statistical analysis techniques to be applied are multiple regression, time series analysis, logistic regression, and ANOVA.

Dataset 1: Individual by Year Level

Download: CSV (39.1MB) STATA (39.1MB)

Dataset 2: Individual Level

Download: CSV (412KB) STATA (527KB)

Please like & share:

National Incident Based Reporting System (NIBRS) Database (2000 – 2014)

These data from NIBRS include the nature and types of specific offenses in the incident, characteristics of victim(s) and offender(s), and types and values of property stolen and recovered.

Dataset 1: Individual Incident Level

Download: CSV (1.03GB) STATA (1.04GB)

Dataset 2: State By Date Level

Download: CSV (8.84MB) STATA (12.0MB)

Please like & share:

Visualizing the Production Function and Cost Curves

Single, static images of data trends aren’t the most effective way to communicate the ways the different elements of an equation or formula contribute to a trend.  This is especially true for introductory economics concepts such as cost curves or the production function. Dynamic, interactive visualizations that allow users to manipulate the variables contributing to a relationship which enables the audience to better understand how equations express trends.

Krit Petrachaianan ‘17 of DASIL programmed a visualization using R that illustrates cost curves and the production function, two core concepts of introductory economics.  DASIL’s visualization allows users to manipulate the different parts of the equations that define cost curves and the production function. For instance, users can manipulate the costs per input (denoted r and w) and the amount of a particular input (denoted K for capital and L for labor). Users can also define the productivity of the firm’s inputs.

Cost curves visualize the costs of producing different levels of output. The total cost of production for a business can be subdivided into fixed and variable costs.  Some costs, such as raw materials and production supplies, change proportionally as more or less of the good or service is produced and are known as variable costs.  Other costs, such as the annual rent or salary of workers, are independent of the level of goods or services a business produces and are known as fixed costs.

The production function shows the relationship between the output produced by a firm from a given amount of inputs (i.e. labor and capital). The productivity of inputs in producing output can vary in three ways: 1) with constant productivity, the additional output produced by a given amount of input is constant as more of the input is used, 2) with diminishing productivity, the additional output produced by a given amount of input declines as more of the input is used, and 3) with increasing productivity, the additional output produced by a given amount of input increases as more of the input is used.

Explore DASIL’s latest R visualization below, as well as in the Graphs section of the Data Visualizations page and in the Economics tab of the DASIL website.


Enter your e-mail address to receive notifications of new blog posts.
You can leave the list at any time. Removal instructions are included in each message.

Powered by WPNewsman

Please like & share:

Mapping State Tax Expenditures to Demonstrate that All Else Really is Equal

Typically, when a business invests in a new piece of equipment, it cannot immediately deduct the full purchase price from its taxable income in the first year.  Instead, according to federal tax regulations, it deducts a percentage of the price in each of 2, 5, or 7 years depending on the type of equipment.  Businesses, of course, would prefer the tax deduction to happen in the first year so they have lower current taxes and therefore increased current cashflow which can be used to make additional investments that will pay off in the future.

In an effort to help small businesses, the federal government has long allowed for all investment costs below a specified threshold by any given firm to be immediately deducted.  This threshold, is specified in Section 179 of the tax code and is generally referred to as the Section 179 allowance.   For example, in 2002, all investment costs below $20,000 could be immediately deducted from taxable income but investment costs beyond $20,000 were subject to normal rules.


Since 2003, the government has worked to encourage business investment by significantly increased this threshold (see figure 1). Interestingly, as the government has increased the threshold, many states have made equivalent alterations to their state tax policies.  Other states have increased their Section 179 allowance some.  Still others have not increased Section 179 generosity at all.  In new research, I attempt to use this state-level variation in Section 179 generosity to estimate how manufacturing investment and employment respond to state Section 179 conformity.

An important step in this research process has been demonstrating that states that do and do not conform to the federal threshold are not substantially different in other ways that would affect investment or employment trends.  One major concern, in particular, is that conforming states might be concentrated in a single region.  If investment and employment is changing in this region for reasons other than Section 179 conformity, then the research design, which compares conforming and non-conforming states, would inappropriately attribute investment and employment effects to state 179 conformity when, in fact, these effects are really due to regional trends.

To allay this concern, I enlisted the help of Bonnie Brooks in DASIL to create an interactive ArcGIS application which shows the evolution of state 179 conformity during the years 2000 to 2011.  From the application, it is immediately apparent that state conformity or non-conformity is not concentrated in any region.  Thus, the ArcGIS application simply and elegantly allays concerns that regional trends may undermine the key assumption in this and all applied microeconometrics research project: that all else really is equal.

To use the map:

  • Drag the second ticker to the beginning of the timeline to start the visualization from the year 2000

Enter your e-mail address to receive notifications of new blog posts.
You can leave the list at any time. Removal instructions are included in each message.

Powered by WPNewsman

Please like & share:

School-to-Prison Pipeline: School Funding

As you may know, education spending in the United States is chronically low, totaling six percent of the total federal funding per year.


chart (2)


According to the U.S. Department of Education, the primary source of funding in the U.S. is intended to be from states and localities (approximately 83 cents per dollar, as of the 2004-2005 school year). The issue with this is the high rate of income inequality in the United States, particularly by location (rural, suburban, or urban). As such, those who live in poorer areas are able to contribute less to students’ education, and school funding, as well as educational quality, suffer. Lack of funding in schools and school districts often coincides with low rates of educational attainment and high rates of school closures (leading to overcrowding and high dropout rates). These issues correlate with higher rates of imprisonment, lending to the “school-to-prison pipeline” that academics and activists have become aware of.

Since funding of schools is intended to be mostly by locality, it is valuable to assess how poverty rates in different localities might indicate a link between school resource allocation and the school-to-prison pipeline.

Without accounting for locality, the average elementary and secondary education funding by local sources is approximately 44% (as of 2011).




When taking locality and poverty rates into account using the same data set, there is a clear trend in the allocation of local funding versus total funding (which includes federal, state, and local funding).




In areas of low poverty, all localities are able to contribute more to schooling, High poverty rates in all localities produce the opposite effect, as one might expect. Interestingly, suburbs and cities with high poverty rates are able to contribute roughly the same percentage to funding, while rural localities are able to contribute a lower percentage to school funding.

Schools that are funded less by the community receive less funding overall and also enjoy less community engagement with the schools and student education. Student educational attainment is lessened by these factors and research shows higher rates of arrest and incarceration among those with lower educational attainment.

These findings suggest that school resource allocation does not consist of a single number across the board, or a certain number of cents out of a dollar. One must take into account locality and poverty rate in determining school funding and understanding how the method of funding schools in the United States privileges certain students over others, creating educational opportunities for some and all but ensuring criminal records for others.

Additionally, funding for prisoners is increasing at a much higher rate than funding for students — nearly three times as quickly. Coupled with the chronic rise in America’s prison population, funding priority at federal and state levels seems to be given to prisons rather than schools. This results in resource-strapped schools, particularly in areas that are already limited in funding (i.e. poor locations, cities, and rural areas). Through this funding structure, students continue to be shuttled through the school-to-prison pipeline from schools to prisons, perpetuating the funding cycle and the pipeline.




The issues of the school-to-prison pipeline in relation to school funding is two-pronged then; the methods through which the majority of school funding is received work against high-need, high-risk areas that already have limited community resources, and the allocation of funding to rapidly growing prisons instead of schools. A re-examination of funding priorities and methods may go a long way in alleviating one aspect of the school-to-prison pipeline and resource inequality for students from kindergarten onward.

Enter your e-mail address to receive notifications of new blog posts.
You can leave the list at any time. Removal instructions are included in each message.

Powered by WPNewsman


Please like & share: