# Police attack: citizen attacking policeman with fits vs independent variables

Updated on January 17, 2018

## Introduction

For my dependent variable, I chose POLATTAK: CITIZEN ATTACKING POLICEMAN WITH FISTS. The question that is being asked for POLATTAK variable is “If yes or not sure: would you approve of a police officer striking a citizen who was attacking the police officer with his fists?” the answer response is; 1. Yes, 2. NO, 0.IAP, 8.DK, 9.NA. I chose this variable because it shows a way to measure people’s attitude towards police and indicates the acceptance of violence towards police. I defined the acceptance of violence towards police as “authorities losing respect due to the police-self-defense being violence”.

My first independent variable is AGE: AGE OF RESPONDENT, the question and answer for this variable are “What is your age?” the answers ranged between the ages of 18-89. The AGE variable would be a good variable for this paper to be able to see the different ages and their responses. The younger they are the less likely they would respond “No” to approving of a police officer striking a citizen who was attacking the police officer first. The older the respondent is the more likely for them to respond “Yes” to the same question. I think the younger you the more likely you are not to want to respect the authorities because younger people tend to not like to be told what to do because older people have more experience in life and know how important it is to follow authority figures. I think it is interesting that each respondent gives different responses because of the difference in their age, those that are barely getting to know the adult life and those who have lived longer should see people whose attitude towards police officers have changed over a period towards more violence.

My second independent variable is RACE: RACE OF RESPONDENT, the question and answer for this variable is “What race do you consider yourself?” the answer response is; 1. WHITE, 2. BLACK, 3. OTHER. I believe that if the respondent's race is black they would answer “No” on approving of a police officer striking a citizen who was attacking the police officer first. Racial tension between police officer’s and citizens could affect their opinion of police violence.

My last independent variable would be USEMEDIA: CONTACTED IN THE MEDIA TO EXPRESS VIEW, the question and answer for this variable is “Here are some different forms of political and social action that people can take. Please write down, for each one, whether you have done any of these things in the past year, whether you have done it in the more distant past, whether you have not done it but might do it, or have not done it and would never, under any circumstances, do it. Contacted or appeared in the media to express your views”, the answers for the response are; 1. Have done it in the past yr., 2. Have done it in the more distant past, 3. Have not done it but might do it, 4. Have not done it and would never do it, 0. IAP, 8. CAN’T CHOOSE, 9. NO ANSWER. I believe that people who are more likely to contact or appear in the media are more likely to approve a police officer to strike at a citizen who was attacking the police officer with his fists, just because they can change what really happen by saying the citizen was doing no harm to the officer and the officer just decided to strike on a citizen for no reason. I find it interesting that when citizens use media will only help promote more violence towards officers because they can change the situation and story to tell society the story they want to tell since the media viewers were not there when the violence first took place. When people are answering the question on the variable USEMEDIA I believe that those that have used the media to express their view will not approve with the idea of an officer striking a citizen, when the citizen attacked the officer with his fist, but would still answer “no” to the dependent question just so they can use the media against the officer.

## Data and Methods

The data used for this analysis is the General Social Survey (GSS). Since 1972, the GSS has been monitoring societal change and studying the growing complexity of American society. The GSS aims to gather data on contemporary American society in order to monitor and explain trends and constants in attitudes, behaviors, and attributes; to examine the structure and functioning of society in general as well as the role played by relevant subgroups; to compare the United States to other societies in order to place American society in comparative perspective and develop cross-national models of human society; and to make high-quality data easily accessible to scholars, students, policy makers, and others, with minimal cost and waiting. The GSS contains a standard core of demographic, behavioral, and attitudinal questions, plus topics of special interest. Among the topics covered are civil liberties, crime and violence, intergroup tolerance, morality, national spending priorities, psychological well-being, social mobility, and stress and traumatic events. Altogether the GSS is the single best source for sociological and attitudinal trend data covering the United States.

The target population of the GSS is adults (18+) living in households in the United States. The GSS sample is drawn using an area probability design that randomly selects respondents in households across the United States to take part in the survey. Respondents that become part of the GSS sample are from a mix of urban, suburban, and rural geographic areas. Participation in the study is strictly voluntary. However, because only about a few thousand respondents are interviewed in the main study, every respondent selected is very important to the results. The GSS has a response rate of over 70 percent above that of other major social science surveys and 40-45 percentage points higher than the industry average.

The survey is conducted face-to-face with an in-person interview by NORC at the University of Chicago. The survey was conducted every year from 1972 to 1994 (except in 1979, 1981, and 1992). Since 1994, it has been conducted every other year. The survey takes about 90 minutes to administer. As of 2014, 30 national samples with 59,599 respondents and 5,900+ variables have been collected. The sample size for 2014 was n = 3842.

## Results

Descriptive Statistics

My dependent variable is POLATTAK: CITIZEN ATTACKING POLICEMAN WITH FISTS, and the main question for this variable is “if yes or not sure: would you approve of a police officer striking a citizen who was attacking the officer with his fists?” In table 1 we can see that 88.1% of the respondents said yes and 11.9% of the respondents said no to the question of if it is okay for a police officer to strike on a citizen who was attacking the police officer. We can see that there is a higher percentage of the “yes” response; it can be interpreted as 88.1% of the respondents say it is okay for a police officer to strike on a citizen who was attacking the officer with their fists as self-defense, and that anyone has the right to strike back if they are in any kind of danger.

## Table 1: Frequency Table for approving a citizen attacking a policeman

FREQUENCY
PRECENT
VALID PERCENT
CUMULATIVE PERCENT
Valid YES
2241
58.3
88.1
88.1
NO
303
7.9
11.9
100.0
TOTAL
2544
66.2
100.0

MISSING IAP
1264
32.9

DK
33
.9

NA
1
.0

TOTAL
1298
33.8

TOTAL
3842
100.0

## TABLE 2: MEASURE OF TENDENCY OF APPROVING A CITIZEN ATTACKING A POLICEMAN

N VAILD
2544
MISSING
1298
MEDIAN
1.00
MODE
1

The first independent variable is AGE: AGE OF RESPONDENT, and for this variable, the researcher was trying to figure out the age of the respondent, which can be between the ages of 18-89 years old. In Table 3 it shows the mean of the variable is 49.84, which means that in the average age of the respondent would fall in the age either of 49 or 50 years old. The standard deviation was 17.048, and the standard of deviation means the square root of the average of the squared deviation from the mean. Then I had the value for this variable changed so it can be easier to read in chunks of categories; the old values were just each number of age from 18-89 and for the new values I had categorized it as shown in table 4: 0. 18-25, 1. 26-35, 2. 36-45, 3. 46-55, 4.56-65, 5. 66-75, 6. 76-85 and 7. 86-89 years and older. I felt it would be easier to interpret and describe the variable by separating the responses in categories. In the recode for age it shows that in the category of 0 got 6.8% of respondents say that they are in the age of 18-25, 18.2% of the respondents fell in the age group of 26-35, 17.8% of the respondents were in the age group of 36-45, 18.8% of the respondents were in the age group of 46-55, 18.8% of the respondents answered that they were in the group of 56-65, 11.6% of the respondents replied that they were in the age group of 66-75, 6.1% were in the age group of 76-85 and lastly 1.9% of the respondents were in the age group of being 88-89 or older. The age group that had the most respondents would be a category of 4 and 5 which are 56-65 and 66-75 years of old. The mean for the recode age would be 2.93 which means that most of the average of the respondents would say what age group they would fall in would be 3 which its value would be 46-55 years old. My hypothesis is that AGE and POLATTACK do have a positive relationship together, I think that the relationship would be when the AGE of the respondent increase the more likely they would respond “yes” to POLATTACK. Age does influence the decision one makes in support of police officer defending himself from being attacked by a citizen. My null hypothesis is that there is no effect on POLATTACK by AGE.

## TABLE 3: Descriptive Statistics For Recoding Age

N STATISTIC
MINIMUM STATISTIC
MAXIMUM STATISTIC
MEAN STATISTIC
STD. DEVIATION STATISTIC
SKEWNESS STATISTIC
SKEWNESS STD. ERROR
RECODE AGE
3818
.00
7.0
2.9327
1.73844
.213
.040
VALID N
3818

## TABLE 4: MEASURE OF TENDENCY OF RECODE AGE

STATISTICS

N VAILD
3818
MISSING
24
MEAN
2.9327
MEDIAN
3.000

The second independent variable is RACE: RACE OF RESPONDENT and the responses are; 1. WHITE, 2. BLACK and 3. OTHER. In this variable of the race of the respondent, it presented that 75.7% of the respondents are considering themselves are white, 15.2% of the respondents have responded that they are black and lastly that 9.0% of the respondent consider themselves in the other category as shown in table 5. I can see that there are more people who responded that they are white, in the frequency table it is shown that it is 2910; I can say that either that more people of the race of white were asked this question more than the other groups or more people do consider themselves as white. My hypothesis is that blacks are more likely to not support on POLATTACK than whites, the null hypothesis is that there is no impact on POLATTACK based on RACE.

## TABLE 5: FREQUENCY TABLE FOR RACE

FREQUENCY
PERCENT
VALID PERCENT
CUMULATIVE PERCENT
VALID WHITE
2910
75.7
75.7
75.7
BLACK
585
15.2
15.2
91.0
OTHER
347
9.0
9.0
100.
TOTAL
3842
100.0
100.0

N VAILD
3842
MISSING
0
MEDIAN
1.00
MODE
1

## TABLE 7 FREQUENCY TABLE OF DEALING DRUG ADDICTION

FREQUENCY
PERCENT
VALID PERCENT
CUMULATIVE PERCENT
VALID TOO LITTLE
1064
27.7
57.5
57.5
562
14.6
30.4
87.9
TOO MUCH
223
5.8
12.1
100.0
MISSING IAP
1924
50.1

DK
56
1.8

TOTAL
1993
51.9

TOTAL
3842
100.0

## TABLE 8: MEASURE OF TENDENCY OF DRUG ADDICTION

N VAILD
1849
MISSING
1993
MEDIAN
1.00
MODE
1

For the last independent variable, would be USEMEDIA: CONTACTED IN THE MEDIA TO EXPRESS VIEW, for this variable it is basically asking if they contacted the media to express their view, and the responses for this variable are: have done it the past year, have done it in the more distant past, have not done it but might do it and have not done it and would never do it. In table 9, the respondents have answered very differently to each other as shown in the percentages, 3.8% have said that they have done it in the past year, 7.5% was that they have done it in the more distant past, 38.9% said that they have not done it but might do it and 49.8% of the respondents have said that they have not done it and would never do it. I think it is interesting to find that 49.8% of the respondents would not contact the media to express their views. I would not have expected the numbers to be high because I would have considered that in the year 2014 more respondents would have wanted to contact the media to express their views, or at least that they have not done it but might do it. My hypothesis is that there is a negative relationship between USEMEDIA and POLATTCK would be that the increase of the likelihood of someone contacting the media to express their view will decrease the respondents answer to yes of the police officer to protect themselves. My null hypothesis is that a number of people who will not express views do not influence POLATTACK.

## TABLE 9: FREQUENCY TABLE FOR USE OF MEDIA

FREQUENCY
PERCENT
VALID PERCENT
CUMULATIVE PERCENT
VALID HAVE DONE IT IN THE PAST YR
47
1.2
3.8
3.8
HAVE DONE IT IN THE MORE DISTANT PAST
93
2.4
7.5
11.3
HAVE NOT DONE IT BUT MIGHT DO IT
482
12.5
38.9
50.2
HAVE NOT DONE IT AND WOULD NEVER DO IT
618
16.1
49.8
100.0
TOTAL
1240
32.3
100.0

MISSING IAP
2578
67.1

CANT CHOOSE
22
.6

2
.1

TOTAL
2602
67.7

TOTAL
3842
100.0

N VALID
1849
MISSING
1993
MEDIAN
1.0
MODE
1

## Bivariate Statistic

For the first cross table, would be Citizen attacking police officer with fists dependent on the recode of age, as shown in Table 1. 82.9% of those 18-25 thought it was okay for the officer to strike a citizen who was attacking the police officer with his fists, 85.4% of the respondents who are in the age category of 26-35 had responded yes, 86.8% of the respondents who are in the age category of 36-45 had respondent yes, 91.4% of the respondents in the age group of 46-55 had respondent yes, 90.1% of the respondents who are in the age group of 56-65 had respondent yes, 89.8% of the respondent who are in the age group of 66-75 had responded yes, 88.5% of the respondents who are in the age group of 76-85 had responded yes, and 83.7% of the respondents who are in age of the group of 86-89 and older had responded yes of approving that police officer to strike on a citizen who was attacking the police officer with his or her fists. In the crosstab, there is a small sample size in the category of the ages between 86-89 and older which makes it harder to justify the outcome. As the age of the respondents increase the increase up to the age of 45-55 there is an increase of approving of police officers striking on a citizen who was attacking the police officer with their fists. The chi-square in table 2 shows that the p-value is .019 which means that there is a relationship between each variable since it is less than .05, because it meets the 95% confidence level, therefore I have rejected my null hypothesis.

## TABLE 1: CROSSTABLE FOR AGE AND POLATTACK

18-25
26-35
36-35
46-55
56-55
66-75
76-85
86-89 AND OLDER
TOTAL
CITIZEN ATTACKING POLICEMAN WITH FITS
YES COUNT
145
393
393
424
436
254
139
41
2225

% WITHING RECODE AGE
82.9%
85.4%
86.8%
91.4%
90.1%
89.8%
88.5%
83.7%
88.1%

NO COUNT
30
67
60
40
48
29
18
8
300

% WITHIN RECODE AGE
17.1%
14.6%
13.2%
8.6%
9.9%
10.2%
11.5%
16.3%
11.9%
TOTAL
COUNT
175
460
453
464
484
283
157
49
2525

% WITHIN RECODE AGE
100.0%
100.0%
100.0%
100.0%
100.0%
100.0%
100.0%
100.0%
100.0%

## TABLE 2: CHI-SQUARE FOR AGE AND POLATTACK

VALUE
DF
ASYMPTOTIC SIGNIFICANCE ( 2-SIDED)
PEARSON CHI-SQUARE
16.764^A
7
.019
LIKELIHOOD RATIO
16.545
7
.021
LINEAR-BY-LINEAR ASSOCIATION
6.188
1
.013
N OF VALID CASES
2525

A.0 CELLS (0.0%) HAVE EXPECTED COUNT LESS THAN 5. THE MIIMUM EXPECTED COUNT 5.82

The second cross table will be showing variables of POLATTACK and RACE, as shown in the table 3. 92.2% of the respondents who are white responded yes, 77.7% of the respondents who are black responded yes, 71.3% of the respondents who are in the other category responded yes. 7.8% of the respondents who are white responded no, 22.3% of the respondents who are black responded no, 28.8% of the respondents who are in the other category responded no. People who are in the white category are more likely to approve for the police officer to strike on a citizen who had attacked the officer with their fists than the category of blacks. There is a relationship between whites and the approval of an officer striking because it reaches the 10% percentile compared to blacks. In table 4 it will show the chi-square, the p-value of these two variables is .000, this does show that there is a significance within my variables of POLATTACK and RACE, I rejected my null hypothesis.

## TABLE 3: CROSSTABLE FOR RACE AND POLATTACK

RACE OF RESPONDENTS WHITE
BLACK
OTHER
TOTAL
CITIZEN ATTACKING POLICEMAN WITH FISTS
YES COUNT
1785
285
171
2241

% WITHIN RACE OF RESPONDENT
92.2%
77.7%
71.3%
88.1%

NO COUNT
152
82
69
303

% WITHIN RACE OF RESPONDENT
7.8%
22.3%
28.8%
11.9%
TOTAL
COUNT
1937
367
240
2544

% WITHIN RACE OF RESPONDENT
100.0%
100.0%
100.0%
100.0%

## TABLE 4: CHI-SQUARE FOR RACE AND POLATTACK

VALUE
DF
ASYMPTOTIC SIGNIFICANCE (2-SIDED)
PEARSON CHI-SQUARE
133.421^A
2
.000
LIKELIHOOD RATIO
114.512
2
.000
LINEAR-BY-LINEAR ASSOICATION
129.368
1
.000
N OF VALID CASES
2544

A. 0 CELLS (0.0%) HAVE EXPECTED COUNT LESS THAN 5. THE MINIMUM EXPECTED COUNT IS 28.58

In the crosstab of table 5, it will be on the variables on POLATTACK and NATDRUG as shown in table 5.87.9% of the respondents say we are spending too little money on dealing with drug addiction programs responded yes on the approval of police officers striking on a citizen who attacked a police officers with their fists, 89.2% of the respondents said we are about right spending money responded yes, and 83.8% of the respondents said we are spending too much money responded yes. 12.1% of the respondents who said we are spending too little money responded no, 10.8% of the respondents who said we are about right spending money had responded no, and 16.2% of the respondents who said we are spending too much money had said no. The p-value of the shown in the chi-square in table 6 is .292 which is at the 95% confidence level, due to that there is no relationship between these two variables, my null hypothesis is there is no relationship between NATDRUG and POLATTACK, therefore I fail to reject my null hypothesis.

## TABLE 5 CROSS TABLE FOR NATDRUG AND POLATTACK

DEALING WITH DRUG ADDICTION TOO LITTLE
TOO MUCH
TOTAL
CITIZEN ATTACKING POLICEMAN WITH FISTS
YES COUNT
620
331
119
1070

% WITHIN DEALING WITH DRUG ADDICITION
87.9%
89.2%
83.8%
87.8%

NO COUNT
85
40
23
148

12.1%
10.8%
16.2%
12.2%
TOTAL
COUNT
705
371
142
1218

% WITHIN DEALING WITH DRUG ADDICTION
100.0%
100.0%
100.0%
100.0%

## TABLE 6 CHI-SQUARE

VALUE
DF
ASYMPTOTIC SIGNIFICANCE
PEARSON CHI-SQUARE
2.835
2
.242
LIKELIHOOD RATIO
2.679
2
.262
LINEAR BY LINEAR ASSOCIATION
.656
1
.418
N OF VALID CASES
1218

For the last cross table, it will be view on the variables of POLATTACK and USEMEDIA as shown in table 7. 79.4% of the respondents who are in the category of having done it in the past year of contacting the media to express their view have responded yes on the approval for POLATTACK, 92.9% of the respondents in the category of have done it in the more distant past had responded yes, 91.9% of the respondents who are in the category of have not done it but might do it have responded yes, 85.2% of the respondent who is in the category of have not done it and would never do it had responded yes. Then there are the respondent who said no, 20.6% of the respondent who have responded that they have done it in the past year said no on approving that an officer should strike on a citizen who had attacked the officer with their fist,7.1% of the respondents who have done in the more distant past also answered no, 8.1% of the respondents who have said that they have not done it but might do it so no, and then 14.8% of the respondent who said that they have not done it and would never do it said no. As the likelihood of using the media increases, the responds towards approving an officer striking on a citizen also increases. In table 8, it shows that p-value of the two variables is .011 which shows it is significant because it is below the .05, which shows it has reached the 95% confidence level, my null hypothesis is that the number of people who will not express views do not influence POLATTACK, therefore I have rejected my null hypothesis.

## Discussion

I ran ANOVA testing for my variables to see if there is any kind of significance between the independent and dependent variables. The comparison of each group in the age group of 18-25 there was 1.17 of them that responded yes, in the category of 26-35 there was a 1.15 of them responded yes, 36-45 there was a 1.13, 46-55 there was a 1.09, 56-65 there was a 1.10, for the ages of 66-75 there was 1.10, 76-85 there was a 1.11 and 86-89 and older there was a 1.16. The comparison of the means within the age group shows there is not really a big significance of each category. For the first test it would be with the variable of AGE and POLATTACK as shown in table 10, it shows the significance and that would be .019. Since the significance is lower than .05 it means there’s a significance with the two variables, but in the next table it will show what categories are the most effective ones. In table 11, it shows that there is not a significance with the categories, because every p-value is over .05. This proves that the younger a person is the less likely they are to approve of a police officer striking a citizen even if it is in self-defense, and the older a person is the more likely they are of supporting that officer.

The next test that was done would be RACE and POLATTACK, in table 12 is shows the mean comparison between RACE and POLATTACK it shows there is barely a significant between each of the categories. The means for whites was 1.08, blacks 1.22 and for others was 1.29. In the table 13 it shows there is significance between the groups because the significance came out to be .000 which is below .05. The next table 14 of Bonferroni on RACE and POLATTACK, it shows there is a significance between the categories of white with blacks which shows that I was able to reject my null hypothesis. If a police officer is defending themselves, the race of a person witnessing this results in how much that person approves of the officer’s actions. Specifically, white people are more likely to support an officer striking a citizen who attacked the officer with his fists.

The third test of ANOVA, would be on NATDRUG and POLATTACK in the comparison of the mean with these two variables would be that the mean are barely any difference between each category shown in table 15. The means for dealing with drug addiction for too little was 1.12, about right was 1.11 and too much was 1.16. In table 16 it shows the significance to be .243 and since that is over .05 it would be making the variables comparing to each other that they are not significance with each other. With the last table 17 it shows that there no significance with each other which would cause me to fail to reject my null hypothesis. Spending money on drug programs does not have an effect on whether or not it is ok for police officers to defend themselves by striking a citizen with their fists.

For the last test that was done would be for USEMEDIA and POLATTACK, in table 18 it shows the comparison of mean that there is not a significant with any of the categories. The means for contacted in the media to express view was have done it in the past year was 1.21, have done it in the more distant past was 1.07 and have not done it but might do it would be 1.08 and lastly for those have not done it and would never do it was 1.15. In the next table 19 shows there is a significance with between the groups because it is .010 which is under .05. For the next table, it will show if there is a difference within the categories which is table 20, the two categories that shows there is a significance would be “Have not done it but might do it” and “Have not done it and would never do it” which the p-value is .035 which is under the .05 level. People that have not used the media but are likely to do so do have an impact on how a police officer is portrayed when striking a citizen with their fists, mainly by disapproving of the police officer striking a citizen.

## Conclusion

I was only able to find two significance between all four of my variables, only two of my four independent variables were significant which were RACE and USEMEDIA and the other two that were not significant were AGE and NATDRUG. Which was contrary to what I thought prior to my research. Throughout my research for the project was I was really surprised that when society views things most of the time age does have an impact on what one thinks on the question, but surprisingly there was no significance between AGE and POLATTACK. The use of media did not surprise me that there was going to be a significance because people are more likely to use that as way to express their view and be able to see their view on a topic such as POLATTACK. I think that to have a better outcome there could have been more independent variables to be significant with my dependent variable would be using other variables that people do not normally believe would have such an effect on POLATTACK.

6

6

78

0

142