Tattoos and Body Piercings

Tattoos and Body Piercings
Randomization Design
• To randomly select the two classes we surveyed,
we first numbered all the classes during Period 2
alphabetically by teacher, so 1 = Aguilar, 2 =
Armstrong, etc. up until 54 = Younglund. Then,
using a random integer generator on a calculator,
we obtained 2 numbers from 1-54. If the two
numbers were the same, we would regenerate
another random integer set using the calculator.
The first number in the set corresponded to the
class that gets our unbiased survey, while the
second number in the set corresponded to the
class that receives our biased survey.
We Saved Paper!
3 Surveys on one page = minimization of paper!
Introduction: Bias We Introduced
For our project, we evaluated the effects of tattoos and body piercings in the
workplace, at school, or around peers. We wanted to see if there was any bias
towards people with body piercings or tattoos. We introduced bias by adding
additional information to the questions about what other people thought
Sample Size
• Sample Size for Biased Questions
(Aguilar’s class): 37
• Sample size for Unbiased Questions
(Main’s class): 36
• Our sample sizes are big enough because
they are both greater than 30.
Our Survey: Unbiased
1) What is your gender?
a) Male
b) Female
2) What is your age? _____
3) Do you have any of the following? **Earrings do not count as body piercings FOR FEMALES **
a) Body Piercing(s)
b) Tattoo(s)
c) Neither
d) Both
4) What is your religion?
a) Christian
b) Catholic
c) Jewish
d) Muslim
e) Other: _______________
5) How strongly do you think tattoos affect a person's ability to obtain a job?
a) Not affected at all
b) A little bit affected
c) Moderately affected
d) Definitely won't be hired
6) How strongly do you think body piercings affect a person's ability to obtain a job?
a) Not affected at all
b) A little bit affected
c) Moderately affected
d) Definitely won't be hired
7) Do you believe that a person's co-workers or peers would view them any differently if that person got a
tattoo?
a) Yes
b) No
8) Do you believe that a person's co-workers or peers would view them any differently if that person got a
body piercing? a) Yes
b) No
9) Do you believe that in a professional setting, a person should conceal their tattoos?
a) Yes
b) No
10) Do you believe that in a professional setting, a person should conceal their body piercings?
a) Yes
b) No
Our Survey: Biased
1) In a 1999 study by researchers from Emory University and Howard University, it was found that more than
80 percent of the 160 women surveyed were pierced but less than 20 percent were tattooed. Do you believe
females are less likely to get tattoos than males?
a) Yes
b) No
2) Many states require that a person getting a tattoo or piercing, have to be at least 18. But, some states do
allow minors (17 or less) to be tattooed or pierced with parental consent. Do you wish you lived in one of
these states?
a) Yes
b) No
3) Certain religions, for example Judaism, prohibit tattoos and body piercings. Do you think a person's
religion should influence whether or not they get a tattoo or body piercing?
a) Yes
b) No
4) According to a Vault.com survey, 85% of people believe that tattoos and body piercings impede ones
chances of finding a job. Do you agree?
a) Yes
b) No
5) Joe works at a law firm and recently got a tattoo on his right forearm. Suddenly, no one would go to lunch
with him. Do you believe coworkers and peers discriminate towards a person with a tattoo?
a) Yes
b) No
6) Mr. Shay came to class one day with a tongue piercing and ear gauges. His math team went to their
competition a week later and placed 3rd (last year they placed 1st). Do you think his piercings affected his
team's performance?
a) Yes
b) No
7) In a survey, over 42 percent of managers said their opinion of someone would be lowered by that person's
visible body art. Do you believe that in a professional setting, a person should conceal their tattoos?
a) Yes
b) No
8) Mr. Köhn decided to leave Canyon Crest Academy. During his job interview at Geico, he decided not to
remove his piercings and was declined the position. Do you believe that in a professional setting, a person
should conceal their body piercings?
a) Yes
b) No
Why Unbiased Questions 1-3?
• We needed to ask these questions because they are all
potential confounding variables.
• Questions #1 and #2: It has been shown that a person’s age
and gender influence their attitudes about tattoos and
piercings.
• Question #3: Obviously someone with piercings and/or
tattoos will have different feelings towards piercings and/or
tattoos than people without piercings and/or tattoos.
• Because we are trying to generalize our results to all high
school students, we needed to understand the distribution
in our study population of these potential confounding
variables. It is important to control for these factors that
could influence their response.
Biased Question #1
In a 1999 study by researchers from Emory University and Howard
University, it was found that more than 80 percent of the 160 women
surveyed were pierced but less than 20 percent were tattooed. Do you
believe females are less likely to get tattoos than males?
a) Yes
b) No
Biased Question #1
60
50
Percent
The results:
n = 37
15 no ==> 15/37 ==> 40.54%
22 yes ==> 22/37 ==> 59.46%
70
40
30
20
10
0
No
Yes
Answer
For this question, we expected many
more students to answer “yes” because of the
bias in the question.
Unbiased Question #1
What is your gender?
a) Male
Gender
Female
Male
Frequency
17
19
Percentages
47.2%
52.8%
Gender
47%
53%
b) Female
Female
Male
Bias:
There are a few more males in the
sample than there are females. This
could change the data because females
are slightly underrepresented compared
to males.
Biased Question #2
Many states require that a person getting a tattoo or piercing, have to be at least 18.
But, some states do allow minors (17 or less) to be tattooed or pierced with parental
consent. Do you wish you lived in one of these states?
a) Yes
b) No
The Results:
Answer
No
Yes
Don’t Care
Frequency
67%
30%
3%
Biased Question #2
3%
30%
No
Yes
Don't Care
67%
Unbiased Question #2
What is your age? _____
Age
13
14
15
16
17
18
19
Frequency
1
27
6
1
0
0
1
Percentages
2.8%
75%
16.7%
2.8%
0%
0%
2.8%
Age Distribution
Values:
Mean = 14.3333
Median and Mode = 14.0
Standard Deviation = 0.95618
Variance = 0.91429
Thirteen Years Old
Fourteen Years Old
Fifteen Years Old
Sixeen Years Old
Seventeen Years Old
Eighteen Years Old
Nineteen Years Old
Unbiased Question #2: Bias
Unbiased Question #3
Do you have any of the following? **Earrings do not count as body piercings FOR
FEMALES **
a) Body Piercing(s)
b) Tattoo(s)
c) Neither
d) Both
Options
Piercing(s)
Tattoo(s)
Neither
Both
Frequency
2
0
33
1
0%
91.7%
2.8%
Percentages 5.6%
Body Piercing/Tattoo Distribution
Body Piercing(s)
Tattoo(s)
Neither
Both
Unbiased Question #3: Bias
Body Piercing/Tattoo Distribution
Body Piercing(s)
Tattoo(s)
Neither
Both
Religion: #4 Unbiased
Religious Beliefs
0.35
%Occurrence
0.3
0.25
0.2
0.15
0.1
0.05
0
Christian Catholic
Jewish
Series 1 0.3333333 0.3055556 0.1388889
Muslim
0
Agnostic Buddhist
0.1944444 0.0277778
Religion: #3 Biased
• Certain religions, for
example Judaism,
prohibit tattoos and
body piercings. Do you
think a person's religion
should influence
whether or not they get
a tattoo or body
piercing?
Religious Confidence Interval
• P: Our parameter is the proportion of high school
students who are either Jewish or Muslim (p-hat)
• A: SRS-randomization, Normality-n=36>30 by CLT,
Independence-N>10n=360
• T:
• C: We are 95% certain that the population
proportion of Jewish or Muslim high school
students is between 2.59% and 25.19%
• Why this population?
Religion
Religion:
#3 Biased
• We extrapolate the data from the unbiased
survey to estimate similar proportions for the
biased data.
• We know that, generally, Muslims and Jews are
prohibited by their religion to have tattoos or
body piercings.
• The number who said a person’s religion should
be influential (38%) was higher than the upper
limit of the confidence interval for the population
of Jews and Muslims.
Biased Question #4
The Results:
6/37 no  16.22%
31/37 yes  83.78%
Percent
According to a Vault.com survey, 85% of people believe that tattoos and body
piercings impede ones chances of finding a job. Do you agree?
a) Yes
b) No
Biased Question #4
90
80
70
60
50
40
30
20
10
0
Yes
No
Answer
Our results very closely support the national statistic of 85% because 83.78% of our
sample answered yes. We expected this result because of the wording bias.
Unbiased Question #5
5) How strongly do you think tattoos affect a person's ability to obtain a job?
a) Not affected at all
b) A little bit affected
c) Moderately affected
d) Definitely won't be hired
The Results:
19/36 “A little bit”  52.78%
5/36 “Not affected”  13.89%
12/36 “Moderately”  33.33%
** In order to compare these unbiased questions
to #4 biased question, we are assigning answers (a)
and (b) on questions #5 and #6 to the answer “no”
on the biased question and answers (c) and (d) to
answer “yes” on the biased question.
Unbiased Question #5
Unbiased Question #5: Reformatted
80
50
70
40
60
Percent
Percent
60
30
20
50
40
30
20
10
10
0
Not Affected
Little Bit
Moderately
Answer
Definitely Won't
0
No
Yes
Answer
Unbiased Question #6
6) How strongly do you think body piercings affect a person's ability to obtain a job?
a) Not affected at all
b) A little bit affected
c) Moderately affected
d) Definitely won't be hired
** In order to compare these unbiased questions
to #4 biased question, we are assigning answers (a)
The Results:
8/36 “Not affected”  22.22% and (b) on questions #5 and #6 to the answer “no”
on the biased question and answers (c) and (d) to
21/36 “Little bit”  58.33%
6/36 “Moderately”  16.67% answer “yes” on the biased question.
1/36 “Definitely”  2.78%
Unbiased Question #6
Unbiased Question #6: Reformatted
70
60
Percent
Percent
50
40
30
20
10
0
Not Affected
Little Bit
Moderately
Answer
Definitely
90
80
70
60
50
40
30
20
10
0
No
Yes
Answer
Comparing Biased #4 to Unbiased #5 and #6
Percent
Comparing #4 Biased to #5 and #6
Unbiased
90
80
70
60
50
40
30
20
10
0
Biased #4
Unbiased #5
Unbiased #6
No
Yes
Answer
Our wording bias for #4 dramatically affected the results, compared to the
unbiased questions. Between piercings and tattoos, less people thought that
piercings impeded one’s chance of getting a job.
Discrimination by Peers (Tattoos)
Biased
• 5) Joe works at a law firm and
recently got a tattoo on his
right forearm. Suddenly, no
one would go to lunch with
him. Do you believe coworkers
and peers discriminate
towards a person with a
tattoo?
Chart Title
% Occurrence
Unbiased
• 7) Do you believe that a
person's co-workers or peers
would view them any
differently if that person got a
tattoo?
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
Yes
No
Unbiased
0.5
0.5
Biased 0.6756756760.324324324
Discrimination by Peers (Tattoos)
• There were more “yes’s” in the responses to
the biased survey as compared to the
unbiased survey, which is what we expected.
Unbiased
Question #8
Do you believe that a person's coworkers or peers would view them
any differently if that person got a
body piercing?
a) Yes
b) No
The results:
n = 36 students
1 maybe ==> 1/36 ==> 2.77%
20 no ==> 20/36 ==> 55.56%
15 yes ==> 15/36 ==> 41.67%
Biased Question #6
Mr. Shay came to class one day with a
tongue piercing and ear gauges. His
math team went to their competition
a week later and placed 3rd (last year
they placed 1st). Do you think his
piercings affected the team's view of
Mr. Shay and their performance?
a) Yes
b) No
The results:
n = 37 students
31 no ==> 31/37 ==> 83.78%
7 yes ==> 7/37 ==> 18.92%
Unbiased
Question #8
Unbiased Question #8
60
40
Biased Question #6
30
20
10
0
Yes
No
Maybe
Biased Question #6
Answer
Percent
Percent
50
90
80
70
60
50
40
30
20
10
0
Yes
No
Answer
Comparing #6 Biased to #8 Unbiased
90
70
60
Percent
Personalizing the question, for
example using a popular teacher’s
name such as Mr. Shay, biases the
question because students want to
agree with that teacher’s behavior.
This is why we see such a large
number of “no” responses in the
biased question group.
80
50
40
Unbiased
30
Biased
20
10
0
Maybe
No
Answer
Yes
Gender Bias in Unbiased Question #8
Do you believe that a person's co-workers or peers would view them any differently
if that person got a body piercing?
a) Yes
b) No
Female
Male
Yes
41%
47%
No
59%
53%
Analysis of this question by gender did not show evidence of a confounding
effect of gender in this population. Because the majority of females and the
majority of males both answered “no” to question #8, there is not gender bias in
this population. Gender will not bias our results and it is not necessary to adjust
our analyses by gender.
Age Bias in Unbiased Question #8
Do you believe that a person's co-workers or peers would view them any differently
if that person got a body piercing?
a) Yes
b) No
≤ 14 years
15 years
≥ 16 years
Yes
50%
33%
0%
No
50%
64%
100%
As age increases, the proportion of “no” responses to question #8 increases from 50%
in the ≤ 14 year old group to 100% in the ≥16 age group. Because age is a confounding
variable in this population, it would be necessary to adjust for age in all analyses.
Existence of a Current Tattoo/ Body
Piercings Bias in Unbiased Question #8
Do you believe that a person's co-workers or peers would view them any differently
if that person got a body piercing?
a) Yes
b) No
Tattoo/Piercing*
No Tattoo/Piercing
Yes
33%
43%
No
67%
57%
*Excluding ear piercings for females
There does not appear to be evidence of confounding by current tattoos/body
piercings because the majority of each groups answered “no” for this question.
So it is not necessary to adjust these analyses by existence of a current tattoo/
piercing.
Concealing in a Professional Setting
(Tattoos)
Unbiased
• 9) Do you believe that in a
professional setting, a person
should conceal their tattoos?
% Occurrence
Biased
• 7) In a survey, over 42
percent of managers said
their opinion of someone
would be lowered by that
person's visible body art. Do
you believe that in a
professional setting, a person
should conceal their tattoos?
Comparing Results
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
Yes
No
Unbiased0.8055555560.194444444
Biased 0.8108108110.189189189
Concealing in a Professional Setting
(Tattoos)
• Despite the wording bias for this question on
the biased survey, there were about the same
number of people who answered yes for both
surveys.
• Because of this, we assume around 81% of
high school students believe a person should
conceal tattoos, which is much higher than
the 42% of managers in the national survey.
Unbiased
Question #10
Do you believe that in a professional
setting, a person should conceal their
body piercings?
a) Yes b) No
n = 36 students
The results:
1 maybe ==> 1/36 ==> 2.77%
16 no ==> 16/36 ==> 44.44%
21 yes ==> 21/36 ==> 58.33%
Biased Question #8
Mr. Köhn decided to leave Canyon
Crest Academy. During his job
interview at Geico, he decided not to
remove his piercings and was
declined the position. Do you believe
that in a professional setting, a
person should conceal their body
piercings?
a) Yes
b) No
n = 37 students
The results:
14/37 answered no ==> 37.84%
24/37 answered yes ==> 64.86%
Unbiased
Question #10
Unbiased Question #10
70
60
40
30
Biased Question #8
20
10
0
Maybe
No
Yes
Answer
Biased Question #8
70
60
50
Percent
Percent
50
40
30
20
10
0
No
Yes
Answer
Comparing #8 Biased to #10 Unbiased
Comparing #8 Biased to #10
Unbiased
70
60
Percent
50
40
30
Unbiased
20
Biased
10
0
Yes
No
Answer
Maybe
The difference between the unbiased
and biased results do not appear to be
large. It is possible that the
unexpected number of “no’s” was a
result of students identifying and
supporting Mr. Köhn.
Bias We Encountered
• Undercoverage: A problem we discovered was that, of
the two classes we surveyed, one was mostly made up
of 9th graders, while the other was mostly made up of
11th graders.
• Nonresponse: Another problem we encountered was
that some people “chose” an answer that was not one
of our choices, or did not answer the survey at all.
• Response bias: Some students might have not told the
truth, especially since one class was aware that they
had the biased survey. For example, a student in Mr.
Main’s biology class said that she was 19, which is
unlikely.
Conclusion
• For the majority of the survey questions that overlapped on biased
and unbiased surveys, the bias we introduced affected the results in
favor of the statistics and anecdotes we provided.
• While conducting our surveys, Ms. Aguilar’s class realized that they
had a biased survey, which might have affected our answers.
• Confidence Intervals are wide because our sample size is small.
Most likely as the sample sizes increase, the Confidence Intervals
would narrow.
• We learned how wording bias can affect answers, and that it is hard
to exclude all other forms of bias. Something unexpected we
discovered about body piercings and tattoos was that some high
school students actually did have tattoos.
• If we were to repeat this survey, we would exclude all single-grade
classes from our randomization so we would have more variety.
Any Questions?