Analytics Individual Case Analysis Consulting Report

Drout Advertising research Project

Attention: Ms. Jamie Drout

Prepared by
June 7th, 2020

1
Table of Contents

1- Executive Summary ...................................................................................................................................... 3

2- Overview Case Analysis ................................................................................................................................ 3
3- Scope of Consulting Work and Report ....................................................................................................... 3
4- Introduction ................................................................................................................................................... 3
5- Analysis and Findings ................................................................................................................................... 4
5.1- Summarize the numerical data using descriptive statistics measures, find proportions for
categorical variables, examine correlations, and use PivotTables as appropriate to compare average
values. Compute confidence intervals for means and proportions and analyze the sampling errors,
possibly suggesting larger sample sizes to obtain more precise estimates. .................................................. 4
6- Overall Case Analysis Findings Discussion .............................................................................................. 11
7- Overall Case Analysis Recommendations ................................................................................................. 11
8- Overall Case Analysis Conclusion(s) ......................................................................................................... 11
9- References .................................................................................................................................................... 11

2
 1- Executive Summary

- This is a continuation of the first Analysis Consulting Report turned in on May 17th, 2020. Refer
to that report for this section.

2- Overview Case Analysis

This is a continuation of the first Analysis Consulting Report turned in on May 17th, 2020. Refer
to that report for this section.

3- Scope of Consulting Work and Report

This is a continuation of the first Analysis Consulting Report turned in on May 17th, 2020.
Refer to that report for the first Scope of Consulting Work done.

Part II
o Summarize the numerical data using descriptive statistics measures, find proportions for
categorical variables, examine correlations, and use PivotTables as appropriate to compare
average values. Compute confidence intervals for means and proportions and analyze the
sampling errors, possibly suggesting larger sample sizes to obtain more precise estimates

4- Introduction

This is a continuation of the first Analysis Consulting Report turned in on May 17th, 2020. Refer to
that report for this section.

3
 5- Analysis and Findings
5.1- Summarize the numerical data using descriptive statistics measures, find proportions for
categorical variables, examine correlations, and use PivotTables as appropriate to compare
average values. Compute confidence intervals for means and proportions and analyze the
sampling errors, possibly suggesting larger sample sizes to obtain more precise estimates.

Using descriptive statistics through Data Analysis in excel for numerical data, five out of the
eleven categories analyzed. The table below summarizes descriptive statistics measures for all numerical
data. This table shows that the Ad Frequency and Stereotype variable have almost the same
characteristics with the correlation of 0.9999.

According to Chen, 2020 “Skewness refers to distortion or asymmetry in a symmetrical bell

curve, or normal distribution, in a set of data”. According to Kenton, 2019, Kurtosis describe the
distribution as well “Whereas skewness differentiates extreme values in one versus the other tail, kurtosis
measures extreme values in either tail.” For either of the variables, these two measures are not zero,
which can be interpreted as a non-normal distribution. Also, when comparing the mean and median, we
can determine that most of the values are distributed to the right section for the five variables. Kurtosis
is over three in Ad Frequency, Stereotype, and Spending. Considering that Empowerment and Age are
less than three, their distributions is platykurtic

Descriptive Statistics:
Age Ad Stereotype Empowerment% Spending
Frequency
Mean 29.267 45.457 43.381 0.054 649.924
Standard Error 1.153 12.657 12.180 0.006 78.705
Median 24 12 10 0.03 400
Mode 23 10 10 0 500
Standard Deviation 11.811 129.695 124.808 0.062 806.491
Sample Variance 139.505 16820.731 15577.046 0.004 650427.340
Kurtosis 2.181 45.579 44.473 2.169 10.471
Skewness 1.797 6.310 6.237 1.378 2.911
Range 49 1100 1050 0.3 4980
Minimum 19 0 0 0 20
Maximum 68 1100 1050 0.3 5000
Sum 3073 4773 4555 5.63 68242
Count 105 105 105 105 105
Confidence Level (95.0%) 2.286 25.099 24.153 0.012 156.076
4
When combining gender, spending, and frequency ad, we can perceive in figure 1 some disruption in
the data most likely due to one of our first findings in the initial report where the Male sample was
misrepresented. For better results, the values should come closer.

Figure 1: Box plot for gender spending and Ad Frequency

Education, as one of the variables, is evaluated as follow:

Education Frequency Relative Cumulative Cumulative

Frequency Frequency Percentage
Associate Degree 6 6% 6 6%
Bachelor’s degree 44 42% 50 48%
Doctorate Degree 10 10% 60 57%
High school diploma 3 3% 63 60%
J.D. 2 2% 65 62%
Master’s degree 16 15% 81 77%
Some undergraduate 24 23% 105 100%
courses
Total 105 100% 430

5
Figure 2 confirms what our first report states that the majority of the sample has Bachelor’s Degrees.

Figure 2: Education Level Count

Also combining Education and spending variable in a box plot shows some disturbance with
Outliers, in the Bachelor’s Degrees, Some undergraduate course, and Doctorate, which make harder to
predict how much these educational levels will spend those products. J. D seems consistent; however,
only two responses were provided. (Figure 3)

Figure 3: Educational Level and Spending

6
The Income Level has the following characteristics, being the most representative of the sample at a low
level. The count of the 0 to $10,000 income level is 24 representing the 23% of the total (Figure 4).

Income Level Frequency Relative Cumulative Cumulative

Frequency Frequency Percentage
$0 to < $10,000 24 23% 24 23%
$10,000 to < $20,000 21 20% 45 43%
$20,000 to < $30,000 15 14% 60 57%
$30,000 to < $40,000 14 13% 74 70%
$40,000 to < $50,000 3 3% 77 73%
$50,000 to < $60,000 9 9% 86 82%
$60,000 to < $70,000 5 5% 91 87%
$70,000 to < $80,000 4 4% 95 90%
$80,000 to < $90,000 1 1% 96 91%
$90,000 to < $110,000 5 5% 101 96%
$110,000 to < $130,000 3 3% 104 99%
$150,000 or more 1 1% 105 100%
Grand Total 105 100% 958

Figure 4: Income Level Count

The same analysis was done for Transform and Reinforcing variables as the following tables show.
7
 Transform Frequency Relative Cumulative Cumulative
Frequency Frequency Percentage
Disagree 1 1% 1 1%
Somewhat Disagree 1 1% 2 2%
Neutral 6 6% 8 8%
Agree 30 29% 38 36%
Strongly Agree 32 30% 70 67%
Somewhat Agree 35 33% 105 100%
Grand Total 105 100.00% 224

Reinforcing Frequency Relative Cumulative Cumulative

Frequency Frequency Percentage
Trivial 3 3% 3 3%
Limited 5 5% 8 8%
Drastic 35 33% 43 41%
Influential 62 59% 105 100%
Grand Total 105 100% 159

The correlation between Age and Spending is 0.0547 (Figure 5). Our first report also found that the
majority of respondents were in their 20's. Therefore, the correlation is positive but very weak and not
conclusive. Increasing the sample of respondents over 30 years old can help to find better conclusions.

Age vs Spending
$6,000.00

$5,000.00

$4,000.00

$3,000.00

$2,000.00

$1,000.00

$0.00
0 10 20 30 40 50 60 70 80

Figure 5: Age vs Spending

8
The confidence Interval are as follow with a 95% confidence level.
-Average Age will be between 27 and 32 years old
-Average Ad Frequency will be between 20 and 71
-Average Stereotype Ad will be between 19 and 68
-Average Empowerment Ad will be between 4% and 7%
-Average Spending on these products will be between $493 and $806
Age Ad Frequency Stereotype Empowerment Spending
%
Lower 26.981 20.358 19.228 4% $ 493.85
Higher 31.552 70.556 67.534 7% $ 806.00

To conduct the same analysis with Variables Reinforcing and Transform the scales was changed to
numbers as follow:
Transform Reinforcing
Strongly Agree 6
Drastic 5
Somewhat Agree 5
Influential 4
Agree 4
Limited 3
Neutral 3
Somewhat Disagree 2 Trivial 2
Somewhat Disagree 1 None 1
Disagree 0

Descriptive Statistics:
Reinforcing Transform
Mean 4.23 4.83
Standard Error 0.07 0.10
Median 4 5
Mode 4 5
Standard Deviation 0.67 1.06
Sample Variance 0.45 1.12
Kurtosis 1.90 2.95
Skewness (0.89) (1.13)
Range 3 6
Minimum 2 0
Maximum 5 6
Sum 444 507
Count 105 105
Confidence Level (95.0%) 0.129423396 0.205189
Lower 4.10 4.62
Upper 4.36 5.03

9
t-Test: One-Sample:

Reinforcing Transform
Mean 4.23 4.83
Variance 0.45 1.12
Observations 105 105
Hypothesized Mean 4.83 4.23
df 104 104
t Stat (9.22) 5.78
P(T<=t) one-tail 0.00 0.00
t Critical one-tail 1.66 1.66
P(T<=t) two-tail 0.00 0.00
t Critical two-tail 1.98 1.98

The correlation between these two variables is 0.435483, and the variances when comparing their
means are insignificant.

Sample Error for the Variables are:

Sample Error
Age 1.15266
Ad Frequency 12.6569
Stereotype 12.18
Empowerment% 0.00607
Spending 78.7054
Reinforcing 0.06527
Transform 0.10347

The main recommendation is to increase and diversify the sample to gather more data and being able
to arrive at better conclusions.

10
 6- Overall Case Analysis Findings Discussion
This is a continuation of the first Analysis Consulting Report turned in on May 17th, 2020. Refer to
that report for this section.

7- Overall Case Analysis Recommendations

This is a continuation of the first Analysis Consulting Report turned in on May 17th, 2020. Refer
to that report for this section.

8- Overall Case Analysis Conclusion(s)

This is a continuation of the first Analysis Consulting Report turned in on May 17th, 2020. Refer
to that report for this section.

9- References
Chen, J. (2020, January 29). Learn About Skewness. Retrieved June 6, 2020, from
[Link] refers to distortion or,varies
from a normal distribution.
Kenton, W. (2020, January 29). Kurtosis. Retrieved June 6, 2020, from
[Link]

11