Taylor Kantner

Data Assignment 3

10/31/14

CRIM 250W- Fall 2014

Mike Herb Section 004

HYPOTHESIS TEST 1

1.

a. IV= DEGREE: If finished 9th-12th grade: Did you ever get a high school

diploma or a GED certificate?

DV= ARREST: Were you ever picked up, or charged, by the police, for any

(other) reason whether or not you were guilty?

b. Hypothesis 1

Among the US adults those with higher degrees of education(graduate) are less likely to have been arrested than those with lower degrees of education(no high school).

c. My reasoning for this hypothesis is that people with higher degrees should be more educated in knowing what not to do and less likely to commit an illegal crime than those who didn’t complete as much schooling. I assume that with a higher degree of education a person would have better choices of occupation and chances for success and therefore less likely to be arrested.

2. Table 1a: Baseline Crosstab- Level of Education by Arrest Charges

p= 0.02

3. Table 1a shows that9% ofUS adults who completed their graduate degree of education have been arrested while 91% of US adults that never completed high school have been arrested. The difference in arrest record is those who have completed degrees of higher education and those who have not is statistically significant (p=.02).

4. The data in Table 1a does support my hypothesis: Among the US adults, those with higher degrees of education (graduate) are less likely to have been arrested than those with lower degrees of education (no high school).

5. a. CV= SEX: male or female

b. Controlling for the sex, US adults, those with higher degrees of education (graduate) are less likely to have been arrested than those with lower degrees of education (no high school).

c. I suspect that men and women who have completed a higher degree of education may not be equally likely to have been arrested so as a control variable I’ll add gender to see if there’s a relationship between degrees of education and arrest records between men and women.

6. Table 1b: Elaboration Crosstab- Degree of Education by Arrest Charges, Controlling for Gender

PANEL A: MALES
Cells contain:
-Column percent
-N of cases / ARREST
1
YES / 2
NO / ROW
TOTAL
DEGREE / 0: LT HIGH SCHOOL / 33.2
384 / 29.3
1,262 / 30.1
1,646
1: HIGH SCHOOL / 49.3
570 / 46.0
1,981 / 46.7
2,551
2: JUNIOR COLLEGE / 3.1
36 / 3.2
138 / 3.2
174
3: BACHELOR / 10.0
116 / 13.4
577 / 12.7
693
4: GRADUATE / 4.4
51 / 8.1
349 / 7.3
400
COL TOTAL / 100.0
1,157 / 100.0
4,307 / 100.0
5,464

P< 0.001

PANEL B: FEMALES
Cells contain:
-Column percent
-N of cases / ARREST
1
YES / 2
NO / ROW
TOTAL
DEGREE / 0: LT HIGH SCHOOL / 28.5
113 / 29.9
1,923 / 29.8
2,036
1: HIGH SCHOOL / 52.9
210 / 52.2
3,358 / 52.2
3,568
2: JUNIOR COLLEGE / 5.3
21 / 3.5
228 / 3.6
249
3: BACHELOR / 10.6
42 / 10.1
649 / 10.1
691
4: GRADUATE / 2.8
11 / 4.3
279 / 4.2
290
COL TOTAL / 100.0
397 / 100.0
6,437 / 100.0
6,834

p= 0.26

7. According to Table 1b Panel A, 33.2% of males who did not complete high school have been arrested and 4.4% of males who have completed higher degrees of education have been arrested. The difference in arrest charges between males who have and who haven’t been arrested is statistically significant (p<.05).

According to Table 1b Panel B, 28.5% of females who did not complete high school have been arrested and 2.8% of females who have completed higher degrees of education have been arrested. The difference in arrest charges between females who have and who haven’t been arrested is not statistically significant (p= .26).

8. Controlling for sex specified the conditional relationship between the IV and DV. Males are more likely to have arrest charges if they never completed high school compared to those males who have completed a higher level of education. For females on the other hand, it is opposite with it being less likely for arrest charges if they have completed a higher degree of education.

9. The data in Table 1a did support my hypothesis that adults with a higher degree of education would be less likely to have any arrest charges, but then when controlling for sex the data showed that it was different based on men and women.

HYPOTHESIS TEST 2

  1. a. IV= GUN:Have you ever been threatened with a gun, or shot at?

DV=RES16: Which of the categories on this card comes closest to the type of place you were living in when you were 16 years old?

b. Hypothesis 2

Among US adults, those who were living in a larger city were more likely to be threatened or shot at with a gun than those who lived in less populated areas.

c. My reasoning for this hypothesis is that cities are more populated and where most of the news of shootings and crime are reported and less likely to occur in smaller population areas such as farms or smaller towns.

  1. Table 2a. Baseline Crosstab- Residence at 16 years of age by Gun Violence

Frequency Distribution
Cells contain:
-Column percent
-N of cases / GUN
1
YES / 2
NO / ROW
TOTAL
RES16 / 3: TOWN LT 50000 / 61.3
1,074 / 68.3
4,914 / 66.9
5,988
6: CITY GT 250000 / 38.7
679 / 31.7
2,279 / 33.1
2,958
COL TOTAL / 100.0
1,753 / 100.0
7,193 / 100.0
8,946

p<0.001

  1. Table 2a shows that those who live in a town less than 50000 and have been threatened or shot at with a gun is 61.3% and for those living in a city 38.7% of people have been threatened or shot at. The difference in residential area at the age of 16 and being threatened or shot at with a gun is statistically significant (p<0.001).
  1. The data displayed did not support my hypothesis since the table actually shows people with residence in a town of less than 50000 were more likely to be threatened with a gun than those in a larger city.
  1. a. CV= RACE: what race do you consider yourself?

b. Controlling for race, US adults, those who were living in a larger city were more likely to be threatened or shot at with a gun than those who lived in less populated areas.

c. My reasoning for this hypothesis is race may make a difference when looking for the area where people are more likely to be threatened or shot at with a gun so I am adding race as a control variable to see if there is a relationship or not.

  1. Table 2b. Elaboration Crosstab- Residence at 16 years of age by Gun Violence, Controlling for Race

PANEL A: RACE(WHITE)
Cells contain:
-Column percent
-N of cases / GUN
1
YES / 2
NO / ROW
TOTAL
RES16 / 3: TOWN LT 50000 / 66.1
932 / 71.2
4,353 / 70.2
5,285
6: CITY GT 250000 / 33.9
479 / 28.8
1,763 / 29.8
2,242
COL TOTAL / 100.0
1,411 / 100.0
6,116 / 100.0
7,527

p<0.001

PANEL B: RACE(BLACK)
Cells contain:
-Column percent
-N of cases / GUN
1
YES / 2
NO / ROW
TOTAL
RES16 / 3: TOWN LT 50000 / 39.5
121 / 51.5
458 / 48.4
579
6: CITY GT 250000 / 60.5
185 / 48.5
432 / 51.6
617
COL TOTAL / 100.0
306 / 100.0
890 / 100.0
1,196

p<0.001

PANEL C: RACE(OTHER)
Cells contain:
-Column percent
-N of cases / GUN
1
YES / 2
NO / ROW
TOTAL
RES16 / 3: TOWN LT 50000 / 58.3
21 / 55.1
103 / 55.6
124
6: CITY GT 250000 / 41.7
15 / 44.9
84 / 44.4
99
COL TOTAL / 100.0
36 / 100.0
187 / 100.0
223

p=0.72

  1. As shown in Table 2b. Panel A 33.9% of whites who lived in a city greater than 250000 have been threatened or shot at with a gun and 66.1% of whites living in a town less than 50000 have been threatened or shot at with a gun. The difference is statistically significant (p<0.001).

As shown in Table 2b Panel B, 60.5% of blacks living in city of more than 250000 have been threatened or shot at and 39.5% of blacks living in a town less than 50000 have been threatened or shot at. The difference is statistically significant (p<0.001).

As shown in Table 2b Panel C, 41.7% of other races living in a city have been threatened or shot at with a gun and 58.3% of other races living in a smaller town have been threatened or shot at with a gun. The difference is not statistically significant (p=0.72).

  1. Controlling for race, specified the relationship between the IV and the DV. The tables for whites and blacks were significant because more whites were threatened with a gun in a town residence and blacks were more likely to be threatened with a gun in a big city, while the other races category had no significance meaning only part of the variable had an effect on the relationship between the IV and the DV.

9. The data in table 2b did not support my hypothesis once I controlled for race since the IV-DV relationship did not remain true for all races when they were tested separately.