Project 3: The criminality narrative
B. Jones
June 2, 2025
Overview
A common trope in the immigration debate is that undocumented immigrants commit, at high rates, violent crimes. Therefore, the supposition is that migrants who are deported are migrants who have committed serious criminal infractions. This idea is prevalent in political rhetoric surrounding the issue of deportation. But is the claim consistent with the actual data? This is the basis of this short Project 3. This assignment is asking you to analyze real-world data on deportations in the United States between the years 2003 and 2024. The data you access records annual ICE removals (deportations) based on what ICE records as the “Most Serious Criminal Conviction” for someone who is deported. The following information is from TRAC (Transactional Records Access Clearinghouse) and describes what the classification levels mean:
“Seriousness Level of MSCC Conviction. ICE classifies National Crime Information Center (NCIC) offense codes into three seriousness levels. The most serious (Level 1) covers what ICE considers to be”aggravated felonies.” Level 2 offenses cover other felonies, while Level 3 offenses are misdemeanors, including petty and other minor violations of the law. TRAC uses ICE’s “business rules” to group recorded NCIC offense codes into these three seriousness levels.”
Essentially what this loosely means is that “Level 1” convictions are the most serious and “Level 3” convictions are generally minor legal infractions. In addition to Levels 1-3, there is a fourth category called “None” denoting that the deportee had no criminal convictions. Review the Patler and Jones article, especially the section on the criminality narrative.
## Year President All None
## Min. :2003 Length:22 Min. : 56882 Min. : 19495
## 1st Qu.:2008 Class :character 1st Qu.:178148 1st Qu.: 85446
## Median :2014 Mode :character Median :238765 Median :106426
## Mean :2014 Mean :248987 Mean :122287
## 3rd Qu.:2019 3rd Qu.:356423 3rd Qu.:165287
## Max. :2024 Max. :407821 Max. :253342
## Level1 Level2 Level3 Undocumented
## Min. : 9819 Min. : 3846 Min. : 11045 Min. :10100000
## 1st Qu.:38484 1st Qu.: 9056 1st Qu.: 34978 1st Qu.:10500000
## Median :46743 Median :17480 Median : 63186 Median :11050000
## Mean :46534 Mean :15601 Mean : 64541 Mean :11015455
## 3rd Qu.:57148 3rd Qu.:20342 3rd Qu.: 90950 3rd Qu.:11375000
## Max. :75590 Max. :29436 Max. :130251 Max. :12200000
## ER_Non
## Min. : 4018
## 1st Qu.:28563
## Median :41647
## Mean :38980
## 3rd Qu.:50230
## Max. :71686Task 1
The following is a line plot of the four levels of criminality (Levels 1-3 and None). First add proper labels to each axis and give a main title. Next, provide a thorough interpretation of the plot that is non-mechanical and substantive. If you were conveying the information from this plot to an audience interested in understanding deportation, what would you say? This task is worth 100 points.
Task 1 answer goes here
Task 2: Regression
For this task you will create three new variables from existing ones in the data set.
First, create a new variable called “minor” that sums all deportations associated with no criminal conviction (“None”) and Level 3 convictions. These are the deportations associated with minor or no criminal activity.
Second, compute the percentage of all deportations that are “minor” deportations (i.e. \(100 \times \frac{None + Level~3}{None + Level~1 + Level~2 + Level~3}\)). Call this variable “percent_minor.”
Third, center the variable using 2014 as the basis year (how to do this will be discussed in class). Name this variable “time.”
Fourth, estimate a linear regression model using the variable percent_minor as the dependent variable and the variable time as the independent variable. Provide an interpretation of the regression results including presenting the results visually using plot_model. What do we learn about the criminality narrative based on these results. This task is worth 100 points.
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 55.01 71.82 74.61 73.68 77.02 86.63#Center Year from 2014
##
## -11 -10 -9 -8 -7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 7 8
## 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
## 9 10
## 1 1Regression
##
## Call:
## lm(formula = minortotal ~ Year, data = reasons)
##
## Residuals:
## Min 1Q Median 3Q Max
## -19.0062 -1.9510 0.9342 2.9694 12.4761
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -17.09483 446.25280 -0.038 0.970
## Year 0.04508 0.22163 0.203 0.841
##
## Residual standard error: 6.595 on 20 degrees of freedom
## Multiple R-squared: 0.002065, Adjusted R-squared: -0.04783
## F-statistic: 0.04138 on 1 and 20 DF, p-value: 0.8409Task 2 answer goes here
Task 3: Presidential differences
Are there differences in criminality levels of deportees by President? This is the question you will answer here. To do this, create a factor-level variable denoting each President. In the data set, there is a variable called “President” and records each president as: “Bush1”, “Bush2”, “Obama1”, “Obama2”, “Trump”, “Biden.” Estimate a regression model treating this factor-level variable as the indpendent variable and “percent_minor” as the dependent variable. What do the results show? Provide an interpretation of the regression results, including a plot of the regression model. This task is worth 100 points.
Making factor
##
## Bush1 Bush2 Obama1 Obama2 Trump Biden
## 2 4 4 4 4 4Regression model
##
## Call:
## lm(formula = minortotal ~ PresFactor, data = reasons)
##
## Residuals:
## Min 1Q Median 3Q Max
## -16.1117 -1.7610 -0.6386 2.6331 15.5058
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 67.026 4.686 14.304 0.000000000156 ***
## PresFactorBush2 8.834 5.739 1.539 0.143
## PresFactorObama1 9.314 5.739 1.623 0.124
## PresFactorObama2 6.397 5.739 1.115 0.281
## PresFactorTrump 7.955 5.739 1.386 0.185
## PresFactorBiden 4.098 5.739 0.714 0.485
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 6.627 on 16 degrees of freedom
## Multiple R-squared: 0.1939, Adjusted R-squared: -0.05797
## F-statistic: 0.7699 on 5 and 16 DF, p-value: 0.585Plotting the model
Task 3 answer goes here
Task 4: Total deportations by President
In this task, we are asking the question: do total deportations vary across Presidencies. There is a basis for this question. President Obama has often been called the “deporter-in-chief” because of the number of deportations that occurred during his presidency, especially the first term. Also, President Trump, in his first administration promised to increase deportations. Are any of these claims valid? Estimate a regressiom model treating the total number of deportations as the dependent variable and the presidential factor variable as the independent variable. Provide a substantive interpretation of the regression results as well the plot of the regression model. This task is worth 100 points.
Regression model and plot
##
## Call:
## lm(formula = All ~ PresFactor, data = reasons)
##
## Residuals:
## Min 1Q Median 3Q Max
## -86997 -33504 250 26966 115451
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 168379 38462 4.378 0.000468 ***
## PresFactorBush2 101881 47106 2.163 0.046044 *
## PresFactorObama1 227705 47106 4.834 0.000183 ***
## PresFactorObama2 123022 47106 2.612 0.018891 *
## PresFactorTrump 64590 47106 1.371 0.189250
## PresFactorBiden -73855 47106 -1.568 0.136482
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 54390 on 16 degrees of freedom
## Multiple R-squared: 0.8124, Adjusted R-squared: 0.7538
## F-statistic: 13.86 on 5 and 16 DF, p-value: 0.00002467
Task 4 answer goes here
Task 5: Broken stick regression
For this task, first create a diagnostic plot of all deportations by year. Based on inspection of the plot, how many piecewise functions do you think would best fit these data? Following this, estimate a regression function using a spline function with a polynomial of order 1 and the number of splines equal to what your diagnostic plot suggests. Comparing a model with 2 or 3 degrees of freedom, which model best describes the data? This question is worth 50 points.
Diagnostic plot (labels are not needed)
Spline function
