Data Analysis & Interpretation 1.1: Selecting a Research Question

In this first week of the first course we are required to choose a real-world dataset, either provided by WU or one of our own, and develop some research questions and hypothesis.  The following is the breakdown of my dataset choice and research agenda.

STEP 1: Choose a data set that you would like to work with.

I have chosen a dataset not provided by WU.  I have chosen three datasets that I will merge into one.  I am calling this dataset “US Census and Crime 2015”.  The individual datasets are as follows:

STEP 2. Identify a specific topic of interest

My specific topic of interest is crime rate.  My initial area of interest is the relationship between poverty and crime.

STEP 3. Prepare a codebook of your own (i.e., print individual pages or copy screen and paste into a new document) from the larger codebook that includes the questions/items/variables that measure your selected topics.)

The primary variables of interest from these datasets are Violent Crime and Property Crime from the “US crime by state and county 2015” dataset and TotalPop and Poverty from the “US Census data by state and county 2015” dataset.

I will calculate the Violent Crime Rate and the Property Crime Rate by dividing Violent Crime and Property Crime by TotalPop and multiplying by 100,000.  I will then use these rates with Poverty to explore their correlation.

I will make the full codebook available once I’m done typing it up.  Until then, below are screenshots of the relevant fields or variables I’ll be using.






STEP 4. Identify a second topic that you would like to explore in terms of its association with your original topic.

I would like to use Total Employees from “Full-time law enforcement employees by state and county 2015” as my secondary topic related to my primary topic of crime rate.

STEP 5. Add questions/items/variables documenting this second topic to your personal codebook.

I am interested in exploring the relationship between crime rate and full-time law enforcement employees.

Do counties with higher crime rates typically have more full-time employees?

What relationship, if any, is there between full-time employees and population?

STEP 6. Perform a literature review to see what research has been previously done on this topic. Try to find multiple sources, and take note of basic bibliographic information.

Search terms:

“poverty and crime rate”

Unemployment, Inequality, Poverty and Crime: Spatial Distribution Patterns of Criminal Acts in Belgium, 2001–06
Marc Hooghe Bram Vanhoutte Wim Hardyns Tuba Bircan
The British Journal of Criminology, Volume 51, Issue 1, 1 January 2011

Previous research has indicated that various deprivation indicators have a positive impact on crime rates at the community level. In this article, we investigate the impact of deprivation indicators on crime in Belgian municipalities (n = 589) for the period 2001–06. A spatial regression analysis demonstrates that unemployment figures have a strong and significant impact on crime rates, and this effect is stronger than the effect of income levels. Income inequality has a significant positive impact on property crime rates but a negative impact on violent crime. Crime is heavily concentrated in the urban centres of Belgium, but we also observe some important regional variations. Demographic structure was not related to crime levels, while spatial analysis shows there is a spill-over effect to neighbouring communities for property crime, but not for violent crime. We close with some theoretical and policy considerations on the relation between unemployment and crime.

Poverty, Income Inequality, and Violent Crime: A Meta-Analysis of Recent Aggregate Data Studies
Ching-Chi Hsieh, M. D. Pugh, First Published September 1, 1993

In the late 1970s and early 1980s, several important reviews of the literature failed to establish a clear consensus on the relationship between economic conditions and violent crime. The research presented here applies the procedures of meta-analysis to 34 aggregate data studies reporting on violent crime, poverty, and income inequality. These studies reported a total of 76 zero-order correlation coefficients for all measures of violent crime with either poverty or income inequality. Of the 76 coefficients, all but 2, or 97 percent, were positive. Of the positive coefficients, nearly 80 percent were of at least moderate strength (>.25). It is concluded that poverty and income inequality are each associated with violent crime. The analysis, however, shows considerable variation in the estimated size of the relationships and suggests that homicide and assault may be more closely associated with poverty or income inequality than are rape and robbery.

“full time law enforcement employees and crime rates”

Police levels and crime rates revisited: A county-level analysis from Florida (1980–1998)
Author links open overlay panelTomislav VKovandzicJohn JSloan

Research on the police–crime relationship generally shows police levels have little impact on crime rates. Two recent studies [Criminology 34 (1996) 609; American Economics Review 87 (1997) 270.] presented evidence that prior police–crime studies were methodologically flawed and found that increased police levels reduced crime. Using county-level data collected from Florida for the period 1980–1998 and a multiple time series (MTS) design, this study revisited the police–crime relationship. for a sample of large cities, the study found that increased police levels reduced most types of crime at the county level. Similar results have now been reported in three recent studies using similar research designs but different units of analysis and time periods. Due to this, prior research showing no relationship between police levels and crime should be reconsidered.

STEP 7. Based on your literature review, develop a hypothesis about what you believe the association might be between these topics. Be sure to integrate the specific variables you selected into the hypothesis.

Hypothesis 1: As poverty increases so does the violent crime and property crime rates.  In other words, there is a positive correlation between poverty and crime rate.

Hypothesis 2: As crime rates increase so does the number of full-time law enforcement employees.  In other words, there is a positive correlation between crime rate and law enforcement employment.

I have other pressing questions that I don’t believe can be answered by this dataset alone, which consists of only one year, 2015.  These other questions would require historical data in order to analyze trends.  These questions are as follows:

  1. Is poverty necessarily the cause or influence of crime or could an increase in crime be a cause or an influence in an increase in poverty?  After all, an increase in crime is likely to encourage relocation for those who are able to relocate.  Likewise, businesses are likely to be hurt by an increase in crime and therefore may be forced to close or relocate, thereby reducing the number of jobs.
  2. Does an increase in law enforcement employment lead to a decrease in crime rate over time?

Again, I don’t expect to answer these questions with this dataset or throughout this course.  I simply include them here for consideration.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s