SPSS Tutorial 2

1
SPSS Tutorial (2)

• Research Methods Data Analysis

2
Assignment example

• Methodology (about 200 words)
• Results (about 400 words)
• Discussion (about 400 words)
• Executive summary (max 200 words structured
  into bullet points)
• APPENDIX including a small selection of tables,
  graphs and statistical output
• NOTE the following examples of the final report
  are purely imaginary!!! They are not based on
  actual results!!!

3
Methodology section

• State the objective
• Briefly describe the methodology you chose
• What does the methodology do?
• Why is it useful for your objective

4
Methodology (example)

• The objective of this study was to test whether
  being vegetarian influences the amount spent in
  GCS stores.
• An independent sample t-test was chosen as the
  methodology to investigate such hypothesis.
• The t-test allows to compare two means from
  different consumer groups and test the null
  hypothesis that the two means are equal. If the
  probability of the t-statistics falls below a
  threshold level (set at 0.05, i.e. 5) then the
  null hypothesis is rejected in favour of the
  alternative.
• The t-test is based upon the normal distribution
  of the target variable (I.e. the amount spent)
  within each of the groups. If the sample size is
  reasonably large (gt40-50 units) it is possible to
  exploit the normal approximation. 118 words

5
Results

• Show/summarise the most relevant output of your
  SPSS analysis
• Describe and comment statistically (i.e. not in
  marketing terms) such output
• Illustrate any limit/problem that might have
  emerged from your application

6
Results (example)

• The t-test was carried out to check whether the
  following customer characteristics led to
  statistically significant differences in the
  group means
• Vegetarian
• Use coupon
• Gender
• Table 1 summarises the output of the Independent
  Samples T-test for the 3 above customer
  characteristics
• Table 1 here
• Table 1 shows that the t-statistic for the
  vegetarian characteristic has a p-value of 0.77.
  As this p-value is above 0.05, the null
  hypothesis of equal means can not be rejected.
  This means the vegetarian factor is not
  influencing the amount spent. However, the mean
  comparison hypothesis test does not take
  explicitly into account the potential influence
  of other disturbing factors (e.g. store size).
  Partial correlation and regression analysis could
  give further information in that direction, but
  for the objective of this study and given the
  very high p-value we can confidently assume that
  the t-test result are reliable.

7
Discussion

• Now interpret the result you have presented in
  the previous section under an operational
  (marketing) perspective, leave out technicalities
  and focus on the main findings.

8
Discussion (example)

• This study showed that being vegetarian is not an
  influential factor in determining the amount
  spent, while there are significant differences in
  terms of gender and the use of coupon. More
  specifically, table 2 gives the average amount
  spent for male/female and user/non users of
  coupon. It looks that the amount spent by men is
  significantly higher, and also the use of coupon
  lead to a higher expenditure.
• This could lead to a strategy for increasing the
  amount spent as follows
• Create an advertising campaign to attract males
  into the GCS chain
• Improve the distribution of coupon (after an
  accurate cost/benefit analysis)

9
Executive summary

• Now just extract a bullet point list summarising
  the key passages of your study.

10
Executive summary (example)

• The objective of the study is to investigate what
  factors influence the amount spent
• We used hypothesis testing (independent samples
  t-test) as a methodology
• Other methodologies (ANOVA, partial correlations)
  could give further indication
• Being vegetarian is not a relevant factor
• Gender and use of coupon are relevant factor
• An male-targeted advertising strategies and the
  calibration of the distribution of coupon could
  increase the profits for GCS

11
Task A

• Examine the relationship between the amount spent
  and the following customer characteristics
• Being male/female
• Being vegetarian
• Shopping for himself / for himself and others
• Shopping style (weekly, bi-weekly, etc.)

• Potential methods
• Battery of hypothesis testing Analysis of
  variance
• Correlation / Regression Analysis

12
Task B

• Examine the relationship between the amount spent
  and the following customer characteristics
• Hypothesis the average amount spent in
  health-oriented shop is higher than those of
  other shops. True or false?
• Test the same hypothesis accounting for different
  shop sizes

• Potential methods
• Battery of hypothesis testing Analysis of
  variance
• Partial correlation (accounting for size)
• Regression Analysis

13
Task C

• Find a relationship between the average amount
  spent per store and the following store
  characteristics
• Size of store
• Health-oriented store
• Store organisation

• Potential methods
• Transform the customer data set into a store
  data set
• Battery of ANOVA
• Correlation / Regression Analysis

14
Task D

• Hypothesis is the amount spent by those that use
  coupon significantly higher?
• What is the most effective way of distributing
  coupons
• By mail
• On newspapers
• Both

• Potential methods
• Recode the variable into 1not using coupon and
  2using coupon
• Hypothesis testing
• Analysis of variance

15
SPSS basics

• Opening SPSS files
• Defining variables
• Restructuring data
• Saving data
• The output window
• Cross-tabulation
• Graphs

16
Variable view
17
Data view
18
Case summaries

• Analyze / Report / Case summaries
• Select target variable(s)
• Select grouping variable(s)
• Include additional statistics

19
Variable(s) you are interested in
Grouping variables
Do not limit/display cases
Click here to choose the statistics you need
20
Output window
21
Categorising variables

• Transform/categorize variables
• Select variable
• Choose number of categories

22
Computing new variables

• Transform/Compute
• Choose expression
• Define if category

23
Define a numeric expression to compute it
Name the new variable

• Define a condition

24
If you want to work with stores as rows

• Data / Aggregate

Aggregating variable
Name the new file
25
Descriptive statistics in SPSS

• Click on Analyze / Descriptive Statistics /
  Frequencies
• Select the variable you are interested in
• Select the STATISTICS you are interested in

26
(No Transcript)
27
First select the variable
Then choose the statistics
28
SPSS output
29
Charts descriptive stats
Charts
30
(No Transcript)
31
Grouped statistics in SPSS

• Click on Analyze / Custom Tables / General tables
• Select the target variable(s) as rows
• Select the grouping variable(s) as column
• Choose the statistics you want to be computed

32
Tick here to have the statistics in the table
Grouping variable
Select the statistics you want
33
Output
34
Hypothesis testing in SPSS

• One-sample test (value of the mean in the
  population)
• Analyze / Compare Means / One sample test

Click on OPTIONS to choose the confidence level
35
Output
T-Test
p value
The null hypothesis is not rejected (as the
p-value is larger than 0.05)
36
Test on two means (independent samples)

• Analyze / Compare means / Independent samples
  t-test

Specify which groups you are comparing
37
Output
The null hypothesis is rejected(as the p-value
is smaller than 0.05)
38
ANOVA in SPSS

• Analyze / Compare means / One-way ANOVA

39
ANOVA dialog box
40
ANOVA output
41
Correlation and covariance in SPSS
Choose between bivariate partial
42
Bivariate correlation
Select the variables you want to analyse
Require the significance level (two tailed)
Ask for additional statistics (if necessary)
43
Bivariate correlation output
44
Partial correlations
List of variables to be analysed
Control variables
45
Partial correlation output
- - - P A R T I A L C O R R E L A T I O N C
O E F F I C I E N T S - - - Controlling for..
SIZE STYLE AMTSPENT USECOUP
ORG AMTSPENT 1.0000 .2677
-.0116 ( 0) ( 775) (
775) P . P .000 P
.746 USECOUP .2677 1.0000 .0500
( 775) ( 0) ( 775)
P .000 P . P .164 ORG
-.0116 .0500 1.0000 ( 775)
( 775) ( 0) P .746 P
.164 P . (Coefficient / (D.F.) / 2-tailed
Significance) " . " is printed if a coefficient
cannot be computed
Partial correlations still measure the
correlations between two variables, but eliminate
the effect of other variables, i.e. the
correlations are computed on consumers shopping
in stores of identical size and with the same
shopping style
46
Bivariate regression in SPSS
47
Regression dialog box
Dependent variable
Explanatory variable
Leave this unchanged!
48
Regression output
Statistical significance Is the coefficient
different from 0?
Value of the coefficients
49
Multivariate regression in SPSS

• Analyze / Regression / Linear

Simply select more than one explanatory variable
50
Output
51
How good is the model?

• The regression model explain less than 19 of
  the total variation in the amount spent