MBA-OE-01: STATISTICAL MODELLING USING SPSS

PRACTICAL NO. 1 - (Data Input In SPSS)

SPSS (Statistical Package for the Social Sciences) is a statistical analysis
software developed by IBM Corporation. It is widely used in social
sciences, business, and other fields to analyze and manipulate data. The
software provides tools for data management, descriptive and inferential
statistics, regression analysis, and graphical presentation of data. SPSS is
user-friendly and has a graphical user interface, which makes it easy to use
for both novice and experienced users. It is commonly used in academic
research, market research, and other data analysis tasks.

Step 1: First of all, make a google form using your Gmail.

Fig.1
Step 2: Then download the responses in the Excel sheet. Next, do the
measurements (i.e. replace the options given in the google form
questionnaire with the numeric data.)
source: Ms excel Fig.2

source: Ms excel Fig.3

Step 3: Next, copy all the numeric data so replaced from the excel sheet
and copy it in the data view of SPSS.
source: SPSS Data view Fig.4
Step 4: Next, go in the variable view and remove all the decimals, change
the alignment to the center, and select the measure according to the
questions (i.e. Nominal, Ordinal, Scale). Also, change the variable names
(that one can do by naming the questions in a separate excel sheet or in the
same excel sheet and then copying the new names in the variable view)

source: SPSS Data view Fig.5

Now the data is ready for further operations in SPSS
 PRACTICAL NO.2 – (Preparing Cross Tables and Charts)

Step 1: Open SPSS.

Step 2: Select the data.
Step 3: Now, go to Analyze > Descriptive Statistics > Crosstabs.
Step 4: Now, the Crosstab will display various options as shown in the figure
given below:

source: SPSS Data view

Fig.6

source: SPSS Data view Fig.7

Step 5: Now, select the variables to run the Crosstab. Next, go to Statistics >
Chi-square > Continue.
Step 6: Click on OK.
And the result will appear on the output sheet.

source: SPSS Data view Fig.8

source: SPSS Data view output Fig.9

source: SPSS Data view output Fig.10

source: SPSS Data view output Fig.11

 source: SPSS Data view output Fig.12

source: SPSS Data view output Fig.13

PRACTICAL NO.3: (Showing the results of Uni-variate Analysis on
SPSS including normality test)
Step 1: Open SPSS.
Step 2: Select the data.
Step 3: Next, go to Analyze > Descriptive statistics > Explore. A small dialogue box with
numerous options will appear on your screen.
Step 4: Next, select the variables to test the Normality.
Step 5: Now, from Statistics & Plots, select the desired options and click on Continue >
OK. And the output will appear on your screen.

source: SPSS Data view Fig.14

Interpretation:
Here the responses are less therefore we will check the value of Shapiro wilk,
which is .872 which is greater than 0.05, hence the data is normally
distributed.
 PRACTICAL NO.4: (Analyzing and interpreting the results of t-test)

Step 1: Open SPSS and select the data.

Step 2: Next, go to Analyze > Compare Means > one sample t-test. A small dialogue box
will appear on your screen.
Step 3: Now, select the variables for t-test. Click on the options, select the confidence
interval at 95% level of significance. Next, click Continue > OK. The output will be visible
on your screen.

source: SPSS Data Fig.

source: SPSS Data Fig.

source: SPSS Data output Fig.

ource: SPSS Data output Fig.

Interpretation: Here the significance value 0.00 which is smaller than 0.05, this implies rejection of
null hypothesis and acceptance of alternative hypothesis.
TWO INDEPENDENT SAMPLE T-TEST
Step 1: Open SPSS and select the data.
Step 2: Next, go to Analyze > Compare Means > Two independent sample t-test. A small
dialogue box will appear on your screen.
Step 3: Now, select the variables for t-test. Click on the options, select the confidence
interval at 95% level of significance. Next, click Continue > OK. The output will be visible
on your screen.

source: SPSS Data output Fig.

Interpretation: Here the significance value is greater than 0.05, hence the acceptance of null
hypothesis is there i.e. there is no significant difference between two population means.

TWO DEPENDENT SAMPLE T-TEST

Step 1: Go to Analyze > Compare Means > paired sample t-test. A small dialogue box will
appear on your screen.
Step 2: Transfer one in variable 1, then transfer dependent variable in variable 2
Step 3: Click on the options, select the confidence interval at 95% level of significance.
Next, click Continue > OK. The output will be visible on your screen.
source: SPSS Data output Fig.

Interpretation: Here the value of p is greater than 0.05, hence null hypothesis is accepted i.e. there is
no significant difference between two variables.
 PRACTICAL NO.5 – (Explaining the commands of applying correlation
and regression in SPSS and interpreting the results)

Step 1: Open SPSS and select the data.

Step 2: Now, go to Analyze > Correlate > Bivariate. And a small dialogue box
will appear on your screen.
Step 3: Select the variables and click on OK. The output demonstrating
correlations will appear on your screen.

source: SPSS Data Fig.

source: SPSS Data Fig.

source: SPSS Data Fig:
Interpretation: Here correlation between EI and OS is .413. Hence there is
significant relation between two factors i.e. there is significant relationship
b/w emotional intelligence and occupational stress.

For Regression,
Step 1: Go to Analyze > Regression > Linear.
Step 2: A small dialogue box will appear. Select the dependent and
independent variables
Step 3: Click on OK. The desired output will be visible on your screen.

source: SPSS Data Fig.

source: SPSS Data Fig.

source: SPSS Data output Fig.

Interpretation: In ANOVA table, the significant value is smaller than 0.05.

hence the data is significant and F value is greater than 1 i.e. 7.595 which is
good.
 PRACTICAL NO. 6 – (Using demo spreadsheet on SPSS, explain the
process and output of Factor Analysis)

Step 1: Open your data in SPSS.

Step 2: Tap on Analyze in menu bar. Go to dimension reduction > Factor
analysis.
Step 3: A small dialogue box will appear. Now, transfer the required variables
in variables tab for analysis.
Step 4: Then go to Descriptive tab and select initial solution in Statistics.
Step 5: Select Coefficients, KMO and Bartlett’s in Correlation matrix.
Step 6: In extraction, Select Correlation matrix in Analyze.
Step 7: In Rotation, we select none in method and continue. In scores, Select
Display factor score coefficient matrix and continue. Then click on OK. The
output will be visible on your screen.

source: SPSS Data Fig.

source: SPSS Data Fig.

source: SPSS Data Fig.

source: SPSS Data Fig.

source: SPSS Data Fig.

source: SPSS Data output Fig.

source: SPSS Data output Fig.

 source: SPSS Data output Fig.

source: SPSS Data output Fig.

Interpretation:
 The value of KMO came out greater than 0.6 which is sufficiently good
& stte that our samle is adequate one.
 The bartlets test of spherecity shown the significant value .000 , which
is smaller than .05. Hence the our null hypothesis is rejected &
alternative is accepted i.e. There is a significant multicollinearity among
the variables.
 In anti image table of correlation if diagonal values are greater than 0.6
than that statements should be taken & rest should be excluded.
 Check the communalities table, all the values should be greater than 0.5
which in our case is greater.
 At last if we see in the rotated component matrix three factors came out
as shown below.
 PRACTICAL NO. 7 – (Performing one-way and two-way ANOVA on
SPSS)
ONE-WAY ANOVA
Step 1: Open SPSS and Load your Data.
Step 2: Go to Analyze > Compare Means > One-Way ANOVA.
Step 3: Move the variable representing the outcome (dependent variable) to the “Dependent
List” box. And, move the variable representing the grouping factor (independent variable)
to the “Factor” box.
Step 4: You can specify additional options such as post hoc tests (e.g., Tukey, Bonferroni)
and effect size measures.
Step 5: Click “OK” to run One-Way ANOVA.

source: SPSS Data Fig.

source: SPSS Data Fig.

Interpretation of Results:
The value of levene’s test is .712 & significant value is .198 which is greater than 0.05;
hence the null hypothesis is accepted i.e. there is no significant difference between the
means of variables.
 Now homogeneity of variance requirement is met hence we can check for the ANOVA
value,
 In this case, significant value foe ANOVA table came out to be .376 which is greater
than 0.05 hence null hypothesis is accepted i.e. there is no significant difference between
the means of variables.
 If value for ANOVA significance will be less than .05 then we can say there is a
significant difference among atleast one variable.
 To check that difference we will see the post-hoc tukey’s value & will determine
which one is different on basis of significant value.

TWO-WAY ANOVA
Step 1: Open SPSS and Load your Data.
Step 2: Go to Analyze > General Linear Model > Univariate.
Step 3: Move the dependent variable to the “Dependent Variable” box. And,
move the independent variables (Factors) to the “Fixed Factors” box.
Step 4: You can specify additional options such as post hoc tests, interaction
plots, and effect size measures.
Step 5: Click “OK” to run Two-Way ANOVA.

source: SPSS Data Fig.

source: SPSS Data Fig.

source: SPSS Data output Fig.

PRACTICAL NO. 8 – (Conducting Cluster analysis on SPSS and
interpreting the results)
Step 1: Open SPSS and Load your Data.
Step 2: Go to Analyze > Classify > Hierarchical Cluster or K-Means Cluster.
Step 3: Choose the variables you want to use for clustering and move them to
the “Variables” box (these variables should be continuous or ordinal in
nature).
Step 4: For hierarchical clustering, you can specify the method (e.g., Ward’s
method, single linkage, complete linkage) and distance measures (e.g.,
Euclidean distance). And, for K-means clustering you need to specify the
number of clusters you want to generate.
Step 5: Click “OK” to run the cluster analysis.

source: SPSS Data Fig.

source: SPSS Data Fig.

source: SPSS Data Fig.

source: SPSS Data Fig.

source: SPSS Data Fig.
Interpretation:
We need to focus on the dendrogram chart. From this we can analyse there are many
clusters.
Also, after it we can see the cluster in or data
GO to analyze >>>> classify>>>> K means cluster analysis
source: SPSS Data Fig.

source: SPSS Data Fig.

source: SPSS Data Fig.

Interpretation: the number of cases in cluster is 2 after using K means cluster

analysis
 PRACTICAL NO. 9 - (Conducting Discriminant analysis on SPSS and
interpreting the results)
Step 1: Open SPSS and Load your Data.
Step 2: Go to Analyze > Classify > Discriminant.
Step 3: Choose the variables you want to use as predictors (Independent
variables) to discriminate between groups.
Step 4: Select the grouping variable (dependent variable) representing the
groups you want to predict.
Step 5: You can specify additional options such as method (e.g., Mahalanobis,
Canonical), prior probabilities, and cutoff scores.
Step 6: Click “OK” to run the discriminant analysis.

source: SPSS Data output Fig.

source: SPSS Data output Fig.

 PRACTICAL NO. 10 – (Analysing the results of
questionnaire with the help of SPSS)

Here's a guide to help you analyze the data effectively:

-Data Preparation:
*Ensure your data is properly formatted in SPSS. Each column represents a question or
variable, and each row represents a respondent's response.
*Check for missing data and handle it appropriately (e.g., imputation, deletion).
*Assign variable labels and value labels to make the data more understandable.

-Descriptive Statistics:
*Use descriptive statistics to summarize the main characteristics of your data.
*Go to `Analyse > Descriptive Statistics > Frequencies` to get frequency distributions for
each variable.
*Review frequencies, percentages, means, and standard deviations to understand the
distribution of responses.

-Reliability Analysis:
*If your questionnaire includes multiple items measuring the same construct (e.g., a Likert
scale), assess the internal consistency using reliability analysis.
*Go to `Analyse > Scale > Reliability Analysis` and select the items measuring the same
construct.
*Review Cronbach's alpha coefficient to assess the reliability of the scale. A value above
0.7 is generally considered acceptable.

source: SPSS Data output Fig.

source: SPSS Data output Fig.

source: SPSS Data output Fig.

source: SPSS Data output Fig.

-Comparative Analysis:
*Use comparative analysis techniques to compare responses between different groups or
conditions.
*For categorical variables, use crosstabulation (`Analyse > Descriptive Statistics >
Crosstabs`) to examine relationships between variables.
*For continuous variables, conduct independent samples t-tests (`Analyse > Compare
Means > Independent Samples T-Test`) or non-parametric tests if assumptions are violated.

-Exploratory Factor Analysis (EFA):

*If your questionnaire aims to measure multiple constructs, conduct exploratory factor
analysis to identify underlying factors.
*Go to `Analyse > Dimension Reduction > Factor` and select the items representing each
construct.
*Review factor loadings and interpret the underlying factors.
-Interpretation and Reporting:
*Interpret the results of your analysis in light of your research objectives and hypotheses.
*Provide descriptive statistics, reliability coefficients, comparative analysis results, and
factor analysis findings.
*Discuss the implications of your findings and any limitations of the study due to the small
sample size.
*Make recommendations for future research or practical applications

-Additional Analyses:
*Depending on the nature of your questionnaire and research questions, consider other
analyses such as regression analysis, ANOVA, or logistic regression if applicable.
*Choose the appropriate statistical tests based on the types of variables and research
objectives.
Caution:
-Interpret results cautiously, especially with a small sample size, as findings may not be
generalizable to larger populations.
-Consider the limitations of small sample sizes, such as reduced statistical power and
increased variability.
-Validate findings with additional studies or replication.

By following these steps and considering the cautionary notes, you can effectively analyze
the results of your questionnaire using SPSS, even with a small sample size.