Texting and Driving
description
Transcript of Texting and Driving
Texting and DrivingJoanna Curran And Brianna Baer
Texting and Driving •How many teenagers use their phones
while driving •Are there factors that affect if people use
their phones while driving•Is texting while driving actually as
widespread as the news makes it seem??
Distractions While Driving• By observing a national survey
of 900 teens around the country• Teenagers find these attributes
the most distracting for drivers:▫ Instant or text messaging while
driving - 37 percent ▫ [The teen driver's] emotional
state - 20 percent ▫ Having several friends in the
car - 19 percent ▫ Talking on a cell phone - 14
percent ▫ Eating or drinking - 7 percent ▫ Having a friend in the car - 5
percent ▫ Listening to music - 4 percent
Gathering our Data• We observed different surveys
given by Insurance companies on teens texting and driving
• We also conducted a survey of all the different attributes of the surveys we found▫ We sent the questions out in
Penn State Altoona’s and St. Joseph’s University student Facebook groups We collected our data in a
systematic random sample, and used the results of every third person that responded
We used 57 student’s results
Age Vs. Gender• We found that the majority
of the data we collected came from females▫ Females were also the
only two 19 years olds tested in our experiment
• There were more females than males in each aspect of the experiment
Gender
0
4
8
12
16
20
f
4
8
12
16
20
m
17 18 19 20 21Age
Collection 1 Histogram
Collection 1
RowSummary
Gender
m
Gender
f
Age35
17.57141
2217.6818
1
5717.614
1S1 = countS2 = meanS3 = columnproportion
Phone Use While Driving vs. Type of Phone
• The smart phones have higher results ▫ People that own a smart
phone use their cell phone more while driving
• Yet, more people who have regular phones do not use their cell phone while driving
ny
Type_of_Phone
5 10 15 20 25 30r
0 5 10 15 20 25 30s
Frequency of Phone_use_while_driving count
Collection 1 Bar Chart
Collection 1
RowSummary
Column Summary
Type_of_Phone
s
Type_of_Phone
r
n
y
Phone_use_while_driving
80.32
70.21875
170.68
250.78125
251
321
150.263158
420.736842
571
S1 = countS2 = columnproportion
Gender vs. What Activity Used Most on Phones
• We found that females most use their phones to make calls while they drive, consisting of 19 subjects
• The males and females have low results for iPod use while only 2 males and 3 females responded yes
• Males most favor texting with 12 subjects responding yes
• There were 4 subjects that did not apply for this test for they do not use their cell phone while driving
fm
what_activity
6 12 18c
0 6 12 18ipod
0 6 12 18t
0 6 12 18X
Frequency of Gender count
Collection 1 Bar Chart
Collection 1
RowSummary
Column Summary
Gender
m
Gender
f
c
ipod
t
X
what_activity
190.542857
60.272727
30.0857143
10.0454545
100.285714
120.545455
30.0857143
30.136364
351
221
250.438596
40.0701754
220.385965
60.105263
571
S1 = countS2 = columnProportion
Share Car vs. Pay for Own Insurance
• We found that a majority of our subjects do share a car with a parent/family member
• Yet, most of the respondents stated that they do not pay for their own car insurance
• Therefore, we performed a test to see who text while driving without having to worry about paying their car insurance
nx
y
share_car
5 10 15 20 25n
0 5 10 15 20 25y
Frequency of pay_for_own_insurance count
Collection 1 Bar Chart
Collection 1
RowSummary
Column Summary
share_car
y
share_car
n
n
x
y
pay_for_own_insurance
220.666667
150.625
10.030303
20.0833333
100.30303
70.291667
331
241
370.649123
30.0526316
170.298246
571
S1 = countS2 = columnproportion
Pay for Own Car Insurance vs. Cell Phone Use
• The majority of our subjects responded that they DO use their cell phone while they are driving, but they do not pay for their own car insurance
• Those who pay for their own car insurance are less likely to use their cell phone while they are on the road
ny
pay_for_own_insurance
10 20 30n
0 10 20 30x
0 10 20 30y
Frequency of Phone_use_while_driving count
Collection 1 Bar Chart
Collection 1
RowSummary
Column Summary
pay_for_own_insurancepay_for_own_insurance
x y
pay_for_own_insurance
n
n
y
Phone_use_while_driving
80.216216
31
40.235294
290.783784
00
130.764706
371
31
171
150.263158
420.736842
571
S1 = countS2 = columnProportion
Support Laws vs. Behavior Change• A large amount of our
subjects responded that they would not support new laws against cell phone use while driving▫ Although, these same
subjects say that they would change their behavior if they were put out
• A good amount of our subjects also responded that they would support these laws
• Almost all of our subjects stated that they would change their behavior if these laws were enforced
nx
y
behavior_change
4 8 12 16 20n
0 4 8 12 16 20X
0 4 8 12 16 20y
Frequency of support_laws count
Collection 1 Bar Chart
Collection 1
RowSummary
Column Summary
support_lawssupport_laws
x y
support_laws
n
n
X
y
behavior_change
190.826087
00
100.322581
00
00
20.0645161
40.173913
31
190.612903
231
31
311
290.508772
20.0350877
260.45614
571
S1 = countS2 = columnProportion
Analysis and Conclusions• Most teens use their phone in some way while driving
▫ The majority call, many text, and few use a music feature• People are more likely to use their phones in the
afternoon or evening• Most people would not change their behavior if laws
were put in place ▫ however most people support a law banning cell phone
use while driving
1-Proportion Z Interval• Conditions
▫ SRS▫ Np, nq >10 ▫ Pop>10n
▫ Assumed▫ 42,15>10▫ # of teens>540
Conditons met=> norm dist=> 1-prop z int
.73681.645.7368(1 .7368)
57=(.64091,8328)
We are 90% confident that the true proportion of people who use their phones while driving is between 64.091% and 83.28%.
1-Proportion Z TestConditons met=> norm dist=> 1-prop z test
z .7368 .8.8(.2057
P(p<-1.1921)=.1166=-1.1921
We fail to reject the claim because our p-value of .1166 is greater than alpha=.05
We have sufficient evidence that the true proportion of people that use their phones while driving is equal to 80%.
H0 : p .8HA : p .8
Chi-Square Goodness of Fit Test• Conditons
▫ Categorical data ▫ SRS▫ All exp counts>5
▫ Activity on phone is categorical
▫ Assumed▫ All exp counts>5
Conditons met=> chi-square dist=> chi-square GOF test
Chi-Square Goodness of Fit Test• Ho: Distribution of our data for activity on phone
matches the distribution of nationwide’s data • Ha: Distribution of our data for activity on phone
does not match the distribution of nationwide’s data
We reject the claim because our p-value of 2.504 x 10^-14 is less than alpha=.05
We have sufficient evidence that the distribution of our data for activity on phone does not match the distribution of nationwide’s data.
6366.62...14.7)14.74(
22.11)22.1122(
expexp)( 222
2
obs
Chi-Square Goodness of Fit Test
We reject the claim because our p-value of 2.504 x 10^-14 is less than alpha=.05
We have sufficient evidence that the distribution of our data for activity on phone does not match the distribution of nationwide’s data.
142 10504.26366.63 xP
Chi-Square Test for Indepence• Conditions
▫ Categorical Data▫ SRS▫ All exp cell counts>5
▫ Phone use and sharing a car are categorical data
▫ Assumed▫ All exp cell counts>5
Conditons met=> chi-square dist=> chi-square test for independence
Chi-Square Test for Independence • Ho: There is an association between cell
phone use and sharing a car• Ha: There is no association between cell
phone use and sharing a car
74202....579.23
)579.2325(421.18
)421.1817(expexp)( 222
2
obs
Chi-Square Test for Independence
•We fail to reject the claim because our p-value of .3891 is greater than alpha=.05
•We have sufficient evidence that there is an association between cell phone use and sharing a car.
3891.7402.2 xP
Chi-Square Test for Independence• Conditions
▫ Categorical Data▫ SRS▫ All exp cell counts>5
▫ Phone use and paying for insurance are categorical data
▫ Assumed▫ All exp cell counts>5
Conditons met=> chi-square dist=> chi-square test for independence
Chi-Square Test for Independence • Ho: There is an association between cell phone use and
paying for insurance• Ha: There is no association between cell phone use and
paying for insurance
4071....28
)2829(14
)1413(expexp)( 222
2
obs
Chi-Square Test for Independence
We fail to reject the claim because our p-value of .52341 is greater than alpha=.05
We have sufficient evidence that there is an association between cell phone use and paying for car insurance.
5234.4071.2 xP
Our Findings• 1-Prop Z Test
▫ Good test to perform, showed our data was not too far away from the national data
• Chi-Square GOF Test▫ Good test to perform▫ Showed a bias in our data collection (only having data
from teens)• Chi-Square Tests for Independence
▫ Good tests to perform▫ Proved a person is more likely to use their phone if they
do not have to share it with another family member▫ Proved a person is more likely to use their phone if they
do not have to pay for car insurance.
Bias/Error•Mostly females responded•Only teenagers (ages 17-19) had been
able to respond to the survey•Only students attending Saint Joe’s or
Penn State Altoona as freshman next year could respond
•Relied on voluntary response
Personal Opinions• Data
▫ Easy to collect data▫ People are more willing to participate in our survey
than we had expected▫ Surprised our data did not match the distribution of
nationwide’s data▫ Not surprised to find associations in our tests for
independence• Project
▫ Took a long time to put together all of the components (as there were many)
▫ Fun project to research