1.

Univariate analysis
a. Mode

Statistics

Are you a
Starbucks What's your
customer? gender?

N Valid 62 62

Missing 0 0
Mode 1 1
The mode is the most frequent value. For the first variable 1=yes and for the gender 1=female resulting
most of the customers are female users.

b. Frequencies, percentages, and cumulative percentages

Are you a Starbucks customer?

Cumulative
Frequency Percent Valid Percent Percent

Valid Yes 36 58.1 58.1 58.1

No 26 41.9 41.9 100.0

Total 62 100.0 100.0

What's your gender?

Cumulative
Frequency Percent Valid Percent Percent

Valid Female 34 54.8 54.8 54.8

Male 27 43.5 43.5 98.4

Prefer not to say 1 1.6 1.6 100.0

Total 62 100.0 100.0

Starbucks has been preferred mostly by female with a number of responses of 34 meaning a 54.8%. The
98.4% consists from female and male and the rest of the 1.6% prefer not to say the gender.
c. 1 pie chart (for a variable of your choice) and edit it

d. 1 bar chart (for a variable of your choice) and edit it and transpose its values
 e. Interpret the results

2. Univariate analysis
a. Mode, Quartiles, Median

Statistics
How often do you buy from Starbucks?

N Valid 36

Missing 26
Median 3.00
Mode 3
Percentiles 25 2.00

50 3.00

75 3.00

The mode is the most frequent value, meaning the most of the people are neutral when it comes to
when it comes to buy a Starbucks product.

The median is the middle value of an ordered (ascending or descending) array of data. The number 3 is
dividing the options in equal parts, meaning the 4 means neutral.

The quartiles divide the array of data in 4 equal parts at numbers 2,3 and 3

b. Frequencies, percentages, and cumulative percentages

How often do you buy from Starbucks?

Cumulative
Frequency Percent Valid Percent Percent

Valid daily 2 3.2 5.6 5.6

weekly 10 16.1 27.8 33.3

monthly 24 38.7 66.7 100.0

Total 36 58.1 100.0

Missing System 26 41.9
Total 62 100.0

Frequencies – the most frequent answer in proportion of 24 with a percent of 38.7 are the monthly
buyers. Meaning most of the people buy just once a month from Starbucks.

The cumulative percentage of 33.3 are people who either buy daily and weekly from Starbucks.

Out of 62 answers 26 are missing because they don’t buy not even once a month.

c. Interpret the results

3. Univariate
a. Frequencies, percentages, and cumulative percentages

How would you rate the taste of Starbucks coffee?

Cumulative
Frequency Percent Valid Percent Percent

Valid 1.0 1 1.6 2.8 2.8

3.0 4 6.5 11.1 13.9

 4.0 10 16.1 27.8 41.7

5.0 11 17.7 30.6 72.2

6.0 5 8.1 13.9 86.1

7.0 5 8.1 13.9 100.0

Total 36 58.1 100.0

Missing System 26 41.9
Total 62 100.0

Frequencies – the most frequent answer in proportion of 11 with a percent of 17.7% values the taste of
Starbucks as neutral. Meaning most of the people have a neutral opinion of the Starbucks taste of
coffee.

The cumulative percentage of 41.7% are people who rated with a low grade the taste of the Starbucks
coffee.

b. Measures of location: Mode, Median, Mean

Statistics

On a scale from
1 to 10 how
How would you likely are you to
rate the taste of recommend
Starbucks Starbucks to a
coffee? friend?

N Valid 36 36

Missing 26 26
Mean 4.806 6.917
Median 5.000 7.000
Mode 5.0 7.0
Std. Deviation 1.3695 2.1696
Variance 1.875 4.707
Range 6.0 9.0
Percentiles 25 4.000 5.000

50 5.000 7.000

75 6.000 8.750
The mode is the most frequent value; for the second variable 7 is the most frequent meaning that is high
recommended for other people.

The median is the middle values of the variable, answers up to 7 equal with the answers below 7.

The mean represents the average results of the variable.

 c. Measures of variability: Range, Quartiles, Interquartile Range, Variance and Standard Deviation

Statistics

On a scale from
1 to 10 how
How would you likely are you to
rate the taste of recommend
Starbucks Starbucks to a
coffee? friend?

N Valid 36 36

Missing 26 26
Mean 4.806 6.917
Median 5.000 7.000
Mode 5.0 7.0
Std. Deviation 1.3695 2.1696
Variance 1.875 4.707
Range 6.0 9.0
Percentiles 25 4.000 5.000

50 5.000 7.000

75 6.000 8.750

d. Add a Histogram chart and edit the chart for 1 scale variable
e. Add a boxplot chart for 1 scale variable
 f. Interpret the results

4. Bivariate analysis
Bivariate Analysis: Analyze the cross tabulation between 2 nominal or ordinal variables. Interpret the

results based on the Observed, Percentages – Row, Column, Total values that you will obtain in your

output.

Are you a Starbucks customer? * How old are you? Crosstabulation

Count

How old are you? Total

 <20 20-35 35-40 40-60

Are you a Starbucks Yes 2 9 21 4 36

customer?
No 4 5 13 4 26
Total 6 14 34 8 62

9 people between 20 and 35 years old are Starbucks customers

The total of 26 are not Starbucks customers

The total of 6 are both customers and no customers of Starbucks with the age under 20 years old

5. Bivariate analysis
Examine the relationship between 2 nominal variables, using chi-square. Analyze

this relationship based on direction (sense), intensity and significance. Interpret the results.

Chi-Square Tests

Asymptotic
Significance (2-
Value df sided)

Pearson Chi-Square 2.135a 3 .545

Likelihood Ratio 2.119 3 .548
N of Valid Cases 62

a. 4 cells (50.0%) have expected count less than 5. The minimum

expected count is 2.52.

6. Bivariate analysis
Analyze and interpret the results of the correlation between 2 scale variables based on:

a. Direction (sense)

b. Intensity &amp;

c. Statistical significance of the relationship

d. Scatterplot

Correlations
 On a scale from
1 to 10 how
How would you likely are you to
rate the taste of recommend
Starbucks Starbucks to a
coffee? friend?

How would you rate the taste Pearson Correlation 1 .783**

of Starbucks coffee? Sig. (2-tailed) .000

N 36 36
**
On a scale from 1 to 10 how Pearson Correlation .783 1
likely are you to recommend Sig. (2-tailed) .000
Starbucks to a friend? N 36 36

**. Correlation is significant at the 0.01 level (2-tailed).

We can conclude that between the 2 variables there is a positive (+0.780), a strong relationship (|
0.780|>0.5) and a statically significant relationship (sig.(2-tailed) = .000<0.05)
7. Measures of shape
Chose two scale variables, display their frequency distribution (histogram) and

compare their shape with a normal distribution through the Skewness and Kurtosis values. Interpret the

results.

Statistics
 On a scale from
1 to 10 how
How would you likely are you to
rate the taste of recommend
Starbucks Starbucks to a
coffee? friend?

N Valid 36 36

Missing 26 26
Mean 4.806 6.917
Std. Deviation 1.3695 2.1696
Variance 1.875 4.707
Skewness -.264 -.509
Std. Error of Skewness .393 .393
Kurtosis .414 .057
Std. Error of Kurtosis .768 .768

A skewed distribution coefficient of -.264 highlights a negative skewed distribution in which the values
are on the right side and a long tail on the left.
8. Charts & Split file
Split your database based on nominal or ordinal variable. The, develop a clustered bar chart
according to 3 variables of your choice. Interpret the results.
9. Multivariate analysis
Add a regression analysis based on a minimum of 3 scale variables. Interpret the results based on the
Regression output.

Coefficientsa

Standardized
Unstandardized Coefficients Coefficients

Model B Std. Error Beta t Sig.

1 (Constant) 1.387 .488 2.846 .007

On a scale from 1 to 10 how

likely are you to recommend .494 .067 .783 7.338 .000
Starbucks to a friend?

a. Dependent Variable: How would you rate the taste of Starbucks coffee?
10. ANOVA
Perform a One-Way ANOVA procedure by using a scale variable as dependent variable and a nominal
one as factor. Interpret the results, Contrasts included.

Coefficientsa

Standardized
Unstandardized Coefficients Coefficients

Model B Std. Error Beta t Sig.

1 (Constant) 1.387 .488 2.846 .007

On a scale from 1 to 10
how likely are you to
.494 .067 .783 7.338 .000
recommend Starbucks to a
friend?

a. Dependent Variable: How would you rate the taste of Starbucks coffee?

11. Cluster analysis

To obtain a segmentation of your respondents, develop a TwoStep Cluster Analysis based on relevant
nominal and scale variables from your database. Interpret the segments you discover.
12. Cluster analysis
Develop a TwoStep Cluster Analysis based on only of scale variables from your database. Interpret the
segments you discover.