2. Team A will develop the most effective
research method to gather information required
to contact consumers that are owners of
Mercedes-Benz and have been impacted by the
recall of models that have Takata-branded
airbags. The methodologies implemented will
address the research backed up with data that
will be used to determine how many consumers
that received the message for the recall on their
vehicle will actually bring their car in to get fixed.
3. H (0): There is no correlation in consumers
bringing their vehicle in for repairs based on
the number of customers contacted.
H(1): There is a correlation in the number of
vehicles brought in for repairs based on the fact
they were contacted regarding the recall.
4. “How many consumers after contacted a
second time (Independent variable) will bring
their vehicles in for repairs? (Dependent
variable).
5. Hypothesis Statement
The first, null hypothesis is the statement (Ho),
“There is no correlation in consumers bringing
their vehicle in for repairs based on the number
of times the consumer is contacted.” The
alternative hypothesis (Ha) will be defined as,
“There is a correlation in the number of
vehicles brought in for repairs based on the fact
they were contacted a second time regarding
the recall.”
6. The number of recalls have been determined
using a state code and dealer number for
owners of the Mercedes Benz impacted by the
recall in New York, New Jersey, Connecticut,
Pennsylvania, Ohio, Delaware, Washington
D.C., and Maryland. This is a method of
stratified random sampling because it is
advantageous when it can be used accurately
because it ensures each subgroup within the
population receives proper representation
within the sample
7. CONSUMER CONTACTED
ONCE FOR RECALL
CONSUMER CONTACTED A
SECOND TIME FOR RECALL
Mean
14.65765766
Standard Error
0.15368712
Median
15
Mode
16
Standard Deviation
4.85758285
Sample Variance
23.59611114
Kurtosis
-0.22590297
Skewness
0.392420692
Range
22
Minimum
6
Maximum
28
Sum
14643
Count
999
Confidence Level(95.0%)
0.301586974
Mean
14.89333
Standard Error
0.564912
Median
15
Mode
16
Standard Deviation
4.892281
Sample Variance
23.93441
Kurtosis
-0.33608
Skewness
0.276843
Range
22
Minimum
6
Maximum
28
Sum
1117
Count
75
Confidence Level(95.0%)
1.125612
8. Calculating the confidence interval for the
difference of two means is valuable to specify a
range of values important to Team A as we
address the value of a second contact to increase
the number of consumers that bring their cars in
for repairs after the recall of Mercedes Benz with
Takata-branded airbags. According to
"Comparison Of Two Means" (1997), “The
confidence interval for the difference between two
means contains all the values of ( - )(the difference
between the two population means) which would
not be rejected in the two-sided hypothesis test of
H0: = against Ha: , i.e.
H0: - = 0 against Ha: - 0.
9. Reporting and
Results
A paired T-test was
performed to
measure the before
and after results of
the sample data.
The results reveal
t-Test: Paired Two Sample for
Means
17 18
Mean 14.72 14.54909091
Variance 23.79358 23.93462508
Observations 275 275
Pearson Correlation 0.153131
Hypothesized Mean Difference 0
Df 274
t Stat 0.445795
P(T<=t) one-tail 0.328049
t Critical one-tail 1.650434
P(T<=t) two-tail 0.656097
t Critical two-tail 1.96866
10. We are interested in the t Stat calculation
0.445795. The t Critical two-tail value is
1.96866.
T is not in excess of the critical value therefore we
will fail to reject Ho statement,
“There is no correlation in the number of
consumers that bring their cars in for repairs
based of the number of times they are
contacted.”
11. Comparison of two means. (1997). Retrieved from
http://www.stat.yale.edu/Courses/1997-
98/101/meancomp.htm
Safecar National Highway Traffic Safety
Administration (NHTSA). (2014). Retrieved from
http://www-
odi.nhtsa.dot.gov/recalls/recallprocess.cfm
Editor's Notes
A sample becomes necessary to properly and effectively analyze data gathered with the standard confidence level of 95% with a margin of error of either 5% or 2.5%.Each member of the team can reasonably calculate 1,000 parts of strata or the subgroups.
The number of consumers contacted a second time using a true sample method:
True Sample = Sample Size X Population / (Sample Size + Population – 1)
If the confidence interval includes 0 we can say that there is no significant difference between the means of the two populations, at a given level of confidence
We are interested in the t Stat calculation 0.445795. The t Critical two-tail value is 1.96866. T is not in excess of the critical value therefore we will fail to reject Ho statement, “There is no correlation in the number of consumers that bring their cars in for repairs based of the number of times they are contacted.”