You are on page 1of 19

CENTRE FOR LOGISTICS AND TRANSPORT STUDIES

FACULTY OF SOCIAL SCIENCES


UNIVERSITY OF PORT HARCOURT
PORT HARCOURT, RIVERS STATE

A REPORT ON

THE STATISTICAL PACKAGE FOR SOCIAL


SCIENCES (SPSS) WORKSHOP

BY

NNOROM, KELVIN UGOCHUKWU


B.Sc./Ed. (UNN) PGD, M.Sc. (UNIPORT)
[G2020/Ph.D./CELTS/FT/]

COURSE TITLE: (SPSS) WORKSHOP


COURSE CODE: LAT 902.1
LECTURER’S NAME: PROF. OSI S. AKPOGHOMEH

MAY, 2022
Executive Summary

This report provides a brief summary of the workshop organized by the Centre for

Logistics and Transport Studies, at the University of Port Harcourt. The Centre

organized a two days workshop on “Statistical Package for Social Sciences (SPSS)”

from May 6th to May 7th 2022 from 04:30 P.M. to 06:00 P.M. in the first day, and

from 09:00 A.M. to 02:00 P.M. in the second day via visual classroom in the

University. It was a visual/online lecture. The workshop brought together over 12

participants from the university. A well-known resource person from the respective

field handled the workshop. All of the participants who attended the workshop were

the new Ph.D. students at the Centre for Logistics and Transport Studies, University

of Port Harcourt. The workshop provided an interactive atmosphere between the

resource person and the participants. The workshop details are as under:

SPSS WORKSHOP SCHEDULE

Time 06/05/2022 07/05/2022

4:30 P.M. – 6:00 P.M. Introduction to SPSS, Starting -


SPSS.

9:00 A.M. – 2:00 P.M. - Creating Variables in


SPSS, Data and its types,
Entering Data into SPSS,
Performing Statistical
Analysis on Correlation
Analysis (Spearman’s
Rank Order Correlation,
and Pearson Product
Moment Correlation),
Regression Analysis,
ANOVA, and Chi-Square
Analysis using SPSS,
Interpreting the output.

  2  
Table of Contents

Cover Page 1

Executive Summary 2

Table of Contents 3

Introduction 4

Workshop Sessions 5

Conclusions 6

References 8

Appendix 9

  3  
Introduction

Possession of research skills is necessary for students pursuing postgraduate studies

(Kimani & Simba, 2017; Doyle, 2017). Research skills are also crucial at the

workplace now that we are in the “BIG DATA” age. At a macro level, improved

research skills imply increased research and innovation, all of which are important for

achievement of Vision 2030. Linking research education at postgraduate level with

industry needs through practical training is crucial in demonstrating the academic-

industry nexus (Ogolla & Kimani, 2016).

However, many fear statistics, a course that is required for most of college,

undergraduate and postgraduate degree courses. Doyle (2017) in an attempt to assess

the college students’ attitudes and perceptions towards statistics noted that statistics

can often be seen as a daunting course, especially for those who feel that mathematics

is not their strongest subject. Students begin to dislike the course before even starting

and this can carry on throughout the entire course. Similar opinions have been

suggested in Lester (2016); Kimani, Guyo and Rotich (2017); Kimani, Iravo and

Waititu (2017) and in Kimani and Simba (2017). To address the anxiety and fear of

statistics, pedagogical approaches/modes of delivery have been suggested to include

both theory and practical training.

In its pursuit to promote research, the Centre for Logistics and Transport Studies,

University of Port Harcourt organized a 2 – Day Workshop on Statistical Package for

Social Sciences (SPSS) from 6th of May to 7th of May 2022. The workshop centered

on the application of the commonly used non-parametric statistical methods with the

aid of SPSS software. These include, the correlation analysis (Spearman’s Rank Order

Correlation, and Pearson’s Product Moment Correlation), regression analysis, analysis

  4  
of variance (ANOVA), and Chi-Square statistics. The workshop aims to improve the

research skills & competence amongst the researcher by strengthening their

theoretical knowledge as well as by giving them the Hands-on training with SPSS,

report writing etc. The aim of the workshop was to enable the participants to operate

the SPSS software on their own and understand when and what type of statistical

method to use. The workshop objectives were:

• To provide hands-on training in the applications of statistical software (SPSS)

in research data analysis.

• To acquire expertise in identifying appropriate statistical tools for different

types of research objectives.

• To understand the rationale behind the application of different statistical

techniques.

The key highlights of the course were:

• Introduction to SPSS

• Hands on Training in SPSS

• Application of Statistical Tools

• Interpretation of SPSS outputs

Workshop Sessions

Mr. Chiemezie Nwangburuka handled the course. He first of all acquainted the

participants with the aims and objectives of the course as well as the relevance of the

class. He also served as the resource person for the workshop. The workshop was in

two sessions. The first session took place on the first day (06/05/2022) and it was

strictly on the theoretical aspect of the statistical analysis. The session focused on the

  5  
introduction to SPSS and the non- parametric tests for testing hypothesis. The session

was just the theoretical aspects of the workshop. He then had one to one session with

the participants and addressed their questions and queries.

The sessions of the second day (07/05/2022) focused on the practical aspects of the

workshop, which was on non-parametric tests of hypothesis using SPSS. Mr.

Chiemezie Nwangburuka explained about correlation and regression analysis. In the

last session he elaborated on analysis of variance (ANOVA), and Chi-square using

SPSS. By the help of Zoom presentations, YouTube video clip shows and discussion:

the participants explored the practical examples given for each of the five statistical

methods covered at the workshop.

The detailed lecture note with the questions and the SPSS output with the

interpretation for each of the statistical tests and associations using SPSS are in the

appendix for better understanding.

Conclusions

The report concludes that there was widespread interest by all the post-graduate

students to learn how to analyze data by use of SPSS software. There is a need for

continued training in statistical software’s. The hands-on training methodology is

effective in improving understanding of SPSS data entry, descriptive statistics,

regression and correlation analysis. The training is relevant to students course work

especially in relation to understanding the Logistics Research methods course and

also in carrying out research work. Students would attend future levels of SPSS

trainings if organized. The training workshops are useful for imparting vision 2030

relevant skills. This is because ingraining statistical research skills to students ensures

  6  
that they are competence to carry out research that would change outcomes in

logistics sector, agricultural sector, health sector, education sector, energy sector, just

to mention but a few. Quality research implies that productivity is improved once the

research output is disseminated and implemented by households. An increase in

productivity is one of the sources of economic growth, according to economic

literature. Economic growth of over 10% per annum as stipulated under Vison 2030

may be achieved this way.

The report recommended that undergraduate and post-graduate students in public and

private universities should be trained on SPSS since it is relevant to their course work.

  7  
References

Doyle, D. A. (2017). Ugh...Statistics! College students’ attitudes and perceptions

toward statistics. Honors in the Major Theses. 165.

Kimani, J.G., & Simba, F. (2017). Effect of practical SPSS training on students’

research competence; A Survey of Jomo Kenyatta University of Agriculture

and Technology Mombasa campus postgraduate students. African Journal of

Education and Practice, 2 (2) No.1, 1 – 15.

Kimani, J.G., Guyo, W., & Rotich, G. (2017). Improving research competence

through practical SPSS training among postgraduate students in Jomo

Kenyatta University of Agriculture and Technology (JKUAT) CBD Campus.

Journal of Education and Practice, 1(1), 19-34.

Kimani, J.G., Waititu, A., & Iravo, M. (2017). Linking practical SPSS training to

research competence among postgraduate students at Jomo Kenyatta

University of Agriculture and Technology (JKUAT), Westland’s Campus.

American Journal of Education and Practice, 2(1), 43 – 56

Lester, D. (2016). Predicting success in psychological statistics courses.

Psychological Reports, 118(3), 772-777. doi:10.1177/0033294116647687

Ogolla, K. & Kimani, J.G. (2016). Bridging the relevance gap between thematic

trends and industry relevance of master of business administration projects:

The double helix of university – Industry relation. DBA Africa Management

Review, 6 (4, 6), 11-32

  8  
Appendix

Lecture Note for the Workshop

USING SPSS FOR CORRELATION ANALYSIS


Correlation analysis is a statistical technique used to discover if a linear relationship

exists between two variables or datasets. It also measures the strength of the

relationship. It is important to note that correlation coefficient lies between -1 and +1.

Example
The table below shows the distribution of fertilizer input (x) in kilograms and crop

yield (y) in tons in Uniport Experimental farm between 2011 and 2020.

Year Road traffic accident Number of persons injured


2011 12 15
2012 15 20
2013 14 21
2014 12 18
2015 16 17
2016 14 28
2017 15 35
2018 9 30
2019 17 40
2020 10 30

a. Calculate the Spearman’s Rank Correlation Coefficient for the dataset and test for

its significance using SPSS. Is there any significant relationship between fertilizer

input (x) and crop yield (y)?

b. Calculate the Pearson Product Moment Coefficient for the dataset and test for its

significance using SPSS. Is there any significant relationship between fertilizer input (x)

and crop yield (y)?

  9  
State your hypothesis

H0: There is no significant relationship between fertilizer input and crop yield
between 2011 and 2020.

General Decision Rule

Reject the null hypothesis (H01) if the p-value obtained is less than the critical level of

alpha (0.05). Otherwise accept.

Steps

1. Open SPSS workspace

2. Under the variable view enter the variable names of the study “fertilizer input”

and “Crop yield”

3. Under the data view enter the corresponding data for each variable name

4. On the menu bar of SPSS go to analyse correlate Bivariate

5. Move the variables names on the left to the variable section to your right

6. Tick on the appropriate type of correlation and click on OK

7. Interpret the output displayed.

Output

Correlations
fertilizer input crop yield
Spearman's rho fertilizer input Correlation Coefficient 1.000 .123
Sig. (2-tailed) . .736
N 10 10
crop yield Correlation Coefficient .123 1.000
Sig. (2-tailed) .736 .
N 10 10

  10  
Correlations

fertilizer input crop yield


fertilizer input Pearson Correlation 1 .130
Sig. (2-tailed) .721
N 10 10
crop yield Pearson Correlation .130 1
Sig. (2-tailed) .721
N 10 10

a. Interpretation for Spearman Rank Correlation Coefficient


The Spearman Rank Correlation coefficient value obtained is 0.123. This figure

indicates a very weak positive correlation. In testing for significance, a p-value of

0.736 was obtained. This value is greater than the critical level of α = 0.05. This

implies that there is no significant relationship between fertilizer input and crop yield.

b. Interpretation for Pearson Product Moment Coefficient


The Pearson Product Moment Coefficient value obtained is 0.130. This figure

indicates a very weak positive correlation In testing for significance, a p-value of

0.721 was obtained. This value is greater than the critical level of α = 0.05. This

implies that there is no significant relationship between fertilizer input and crop yield.

  11  
USING SPSS FOR REGRESSION ANALYSIS
Regression analysis is a statistical technique used to estimate how a dependent or

outcome variable depends on one or more independent or predictor variables.

The regression model is given as y = α +β1x1+ β2x2+……..+ βnxn + ε (1)

Where:

y= the dependent variable

α= the intercept

x1 – xn = Independent variables

β1 – β3 = the parameter estimates or regression coefficients

ε= the error term

Example

A logistics student hypothesizes that level of urbanization (y) can be explained by per

capita income (X1) and number of non-primary activities (X2). To test this conjecture,

the data below were collected:

Y(level of urbanization) X1(per capita income) X2 (non-primary activities)


2 3 4
2 5 7
4 3 12
5 8 9
7 8 11
6 7 9
8 9 12
10 12 13
9 14 16
12 8 16

  12  
a. Find α, β1 and β2

b. Fit the regression model

c. Interpret the model

State your hypothesis

H0: There is no significant effect or influence of per capita income and number of

non-primary activities on level of urbanization

Y = a + b1(per-capita income) + b2(non-primary activities)

General Decision Rule

Reject the null hypothesis (H01) if the p-value obtained is less than the critical level of
alpha (0.05). Otherwise accept.

Steps

1. Open SPSS workspace

2. Under the variable view enter the variable names of the study “level of

urbanization”, “per capita income” and “Non-primary activities”

3. Under the data view enter the corresponding data for each variable name

4. On the menu bar of SPSS go to analyse Regression Linear

5. Move the variable names on the left to the appropriate section on the right and
click on OK

7. Interpret the output displayed.

  13  
Output

a
Coefficients
Standardized
Unstandardized Coefficients Coefficients

Model B Std. Error Beta t Sig.


1 (Constant) -2.101 1.696 -1.239 .255

Per capita income .265 .217 .280 1.220 .262

Non-primary activities .602 .203 .681 2.967 .021


a. Dependent Variable: Level of urbanisation

Interpretation

a. α = -2.101, β1 = 0.265, β2 = 0.602,

b. Level of urbanisation = -2.101 + 0.265 (Per capita income) + 0.602 (Non-

primary activities)

c. From the model we conclude that for every unit increase in per capita income,

level of urbanisation will increase by 0.265 units, similarly for every unit

increase in Non-primary activity, level of urbanisation will increase by 0.602

units.

  14  
USING SPSS FOR ANOVA ANALYSIS
Analysis of Variance (ANOVA) is a statistical technique that is used to compare

means of several populations. It also looks at the difference or variation between

several populations.

Example

An agriculturalist is interested in determining if varying the nitrogen content in a

brand of fertilizer affects the yield parameter of cassava crop. The result of his study

is as follows

Levels of Yield per hectare

Nitrogen

A 17 10 18 15 10

B 22 25 30 20 15

C 26 30 32 24 33

D 31 32 25 28 30

E 16 17 20 19 10

Is there any significant difference in the yield parameter due to the level of

nitrogen? H0: There is no significant difference in the yield parameter due to the

level of nitrogen

General Decision Rule

Reject the null hypothesis (H01) if the p-value obtained is less than the critical level of

alpha (0.05). Otherwise accept.

Steps

  15  
1. Open SPSS workspace

2. Under the variable view enter the variable names of the study “level of nitrogen”,
“Yield per hectare”

3. Since there are different categories in the level of nitrogen, you assign values to

different levels of Nitrogen under the values column. Assign 1 = A, 2 = B, 3 = C, 4 =

D and 5 = E

4. Under the data view enter the corresponding data for yield per hectare

4. On the menu bar of SPSS go to analyse Compare means One-way


ANOVA

5. Move the variable on the left to the appropriate section on the right and click on
OK

7. Interpret the output displayed.

Output

ANOVA
Yield per hectare

Sum of Squares df Mean Square F Sig.


Between Groups 980.800 4 245.200 14.630 .000
Within Groups 335.200 20 16.760
Total 1316.000 24

Interpretation

The result obtained implies that there is a significant difference in the yield of cassava

crop as a result of the varying levels of nitrogen because the p-value (0.000) obtained

is less than the critical value of alpha (0.05).

  16  
USING SPSS FOR CHI ANALYSIS
Chi-square test is a statistical test used to examine the differences between categorical

variables. It also looks at the association between two categorical values

simultaneously.

Example

In a certain university campus, it is desired to test the independence of state of origin

of students and their marital status. Is there any relationship between marital status

and origin?

Marital State of Origin


Status
Enugu Anambra Ebonyi Abia Rivers

Single 17 10 18 15 10

Married 22 25 30 20 15

Divorced 26 30 32 24 33

H0: There is no significant relationship between marital status and state of origin

amongst university students.

Steps

1. Open SPSS workspace

2. Under the variable view enter the variable names of the study “marital status”,

“State of origin”

  17  
3. Since we have two categories (Marital status) and (State of Origin) you assign values

under the values column. Assign 1 = Single, 2 = Married, 3 = Divorced, and for the

(State of Origin) you assign 1= Enugu, 2 = Anambra, 3 = Ebonyi, 4 = Abia and

5 = Rivers.

4. Create another variable name called “frequency”

5. Under the data view enter the corresponding data for each category and
“frequency”.

6. Since our data is in form of counts or frequencies we weight the dataset of the
“frequency”.

To do this, go to “Data” on the SPSS menu, navigate to “weight cases” select the

variable you want to weigh which is “frequency” and click on Ok.

7. On the menu bar of SPSS go to analyse Descriptive statistics Crosstabs

8. Move the variable on the left to the appropriate section on the right and click on

“statistics” and check Chi-square and click on continue

9. Interpret the output displayed.

Chi-Square Tests

Asymptotic
Value df Significance (2-sided)
a
Pearson Chi-Square 7.430 8 .491
Likelihood Ratio 7.458 8 .488
Linear-by-Linear Association 1.249 1 .264
N of Valid Cases 327

a. 0 cells (0.0%) have expected count less than 5. The minimum expected count is 12.42.

  18  
Interpretation

The result obtained implies there is no significant relationship between marital status

and state of origin amongst university students because he p-value (0.491) obtained is

less than the critical value of alpha (0.05).

  19  

You might also like