Available via license: CC BY-SA 4.0
Content may be subject to copyright.
The Indonesian Journal of Computer Science
www.ijcs.net
Volume 13, Issue 6, December 2024
https://doi.org/10.33022/ijcs.v13i6.4580
Attribution-ShareAlike 4.0 International License 9396
Factors Influencing the Usage of Twitter During Presidential Elections in Nigeria
Kayode David Kolawole1, Emmanuel Oyasor2
kolawolekayode@yahoo.com1, emmanualoyasor89@gmail.com2
1 Faculty of Economic and Financial Sciences, Walter Sisulu University, Mthatha, Private Bag X1,
UNITRA, 5117, South Africa.
2 Department of Accounting Science, Walter Sisulu University, Mthatha, South Africa
Article Information
Abstract
Received : 26 Dec 2024
Revised : 29 Dec 2024
Accepted : 30 Dec 2024
This study examined the factors influencing the usage of Twitter during
presidential elections in Nigeria. The population of the study comprises
those under the age of voting in Nigeria that has Twitter account in Nigeria.
Creswell (2012) sampling technique was used to select 576 respondents in
Nigeria who have twitter account and are in voting age. Data was obtained
through questionnaire administration. Structural Equation Modelling
method of multiple regression analysis were used to achieve the objectives
of the study. Results of the regression analysis revealed that factors (such as
compatibility, cost effectiveness, interactivity, trust, ease of use and
networking capability) are significant determinant of usage of Twitter during
presidential elections in Nigeria at 1% and 5% level of significance. The
study concluded that there are factors influencing the usage of twitter during
presidential elections in Nigeria. Therefore, it is advised that politicians
should always seek for information through social media that might improve
their interactions with various populace. Therefore, social media adoption
becomes essential due to its multifaceted and symmetrical nature.
Keywords
Compatibility, Cost
Effectiveness,
Interactivity, Twitter,
Presidential Elections
The Indonesian Journal of Computer Science
https://doi.org/10.33022/ijcs.v13i6.4580 9397
A. Introduction
Over the last decades, social media has become an important source of
information and has attained a trending issue in the world. It has become an ideal
means for communicating with a large audience and freely transmitting ideas,
thoughts, and news [1]. Social media is often used to source ideologically biased
content that supports preexisting opinions, regardless of its validity [2].
The difficulty of detecting and combating new types of propaganda is one
that social media platforms are currently experiencing. For instance, in "disguised
propaganda," sources are purposefully obscured to conceal the propagandist's true
identity [3]. In addition, propagandist pays for space in media houses or sources to
broadcast their political adverts as reliable news pieces [4]. Social media has many
positive impacts on society, but it may also be used for evil purposes, such as
propaganda and the creation of phony identities to influence people and sway
public opinion. Some of these illegal online actions target election contexts
deliberately because they provide the best opportunities to have the most impact
on political trends [5]. Propaganda modifies information so as to actively sway
people's opinions and further a planned objective [6]. Broadly speaking,
propaganda is the "dissemination of information facts, arguments, rumours, half-
truths, or lies to affect public opinion.
It is crucial to remember that propaganda also refers to real (factual)
information packaged in a way to sway people's opinions, discredit competing
ideas, or mobilize the public. Disinformation and misinformation relate to
purposely and unintentionally erroneous information, respectively [7]. In order to
have the most impact, propaganda employs psychological and rhetorical strategies
that are meant to be imperceptible. According to [8], propaganda employs
psychological and rhetorical strategies that are meant to be imperceptible in order
to have the greatest impact. Because of this, nefarious propaganda news sources
have demonstrated their ability to have a significant impact on the mindset of
people in a society. People were more likely to reduce their natural barrier of
critical thinking when disinformation was disseminated under the pretense of
news because it gave the impression that the material was reliable [9]. As a result,
it is crucial to prevent or minimize propaganda during the presidential election.
Detecting fake news entails identifying the possibility that a specific news article is
false [10]. A fake news detection system is important so as to assist in the
identifying and filtering of deceptive news. Hence, this study tends to examine the
factors influencing the usage of Twitter during Nigeria’s presidential election.
B. Literature Review
1. Democratic Participant Theory
A political and media philosophy known as the "democratic-participant
hypothesis" contends that democratizing access to stakeholders would improve
the media's effectiveness [11]. Since achieving independence in 1960, the majority
of Nigeria's governmental administration has been under military authority. There
were still concerns about the freedom of expression and the ability to make one's
own political judgments as the country transitioned to democracy. Voters avoided
open elections as a result because they may be physically harmed by opposition
parties sponsored by the military or by opponents with a history of violence.
The Indonesian Journal of Computer Science
https://doi.org/10.33022/ijcs.v13i6.4580 9398
Because voters in Nigeria may now voice their thoughts and make decisions
from the comfort of their own homes thanks to the rise of social media, the
country's political landscape has been transformed as a consequence. They could
take part in conversations on the benefits and drawbacks of certain candidates. As
a direct consequence of this, encouragement to vote was directed at younger
people. For the purpose of reducing the risk of violence and abuse, the voting age
of millennials was historically low or nonexistent. According [11], the democratic
participation theory is an application of the social responsibility theory that
maintains that the media should disseminate different thoughts and perspectives
of people or groups regardless of their political leanings. This theory holds that the
democratic participation theory is an application of the social responsibility
theory.
2. Twitter and Presidential Campaign
Effective communication is essential to the practice of politics. Politicians
utilize communication strategies that are intended to persuade others about a
certain subject or cause. The capacity for effective communication has always been
seen as a highly desirable quality in political candidates. This is of the utmost
importance in Nigeria, where political leaders are obligated to address the most
pressing issues currently being discussed in the public sphere, such as terrorism,
unemployment, and other related issues [12]. They make use of specialized
communication channels in order to convey a predefined message to a specific
target audience in order to acquire information that may influence the political
behavior of their audience. The message being sent and the demographic that it is
aimed at will, to a significant extent, determine which communication medium will
be the most successful in terms of persuading votes [13].
As was said before, the introduction of the Internet as a new means of mass
communication has stoked competition amongst the old forms of media, which
include radio, television, newspapers, and magazines [14]. It is now abundantly
obvious that social media may be used to organize support for political causes.
Campaign strategists now rely heavily on social media as a tool for moving the
conversation forward and organizing political support since social media has
matured into a significant political instrument. Twitter has emerged as a
significant tool for political campaigning in recent years, particularly in the United
States.
C. Methodology
The population of the study is the users of Twitter in Nigeria. However, the
sample size was determined by [15]. “Sample Size - Infinite Population (where the
population is greater than 50,000)
……………………………(1)
SS = Sample Size
Z = Z-value(1.96 for a 95 percent confidence level)
P = Percentage of population picking a choice, expressed as decimal (0.5)
C = Confidence interval, expressed as decimal (.04 = +/- 4 percentage points)
The Indonesian Journal of Computer Science
https://doi.org/10.33022/ijcs.v13i6.4580 9399
SS = 600
……………………… (2)
Note: Calculate the sample size using the infinite population formula first.
Then, use the sample size derived from that calculation to calculate a sample size
for a finite population.
Example:
…………………….(3)
New SS = 575.631
New SS ≈ 576
Therefore, a questionnaire was developed using google form and was filled
by five hundred and seventy-six respondents. The nature of data is primary data
which was obtained through structured questionnaire. The method of data
analysis is Structural Equation Modelling using Partial Least Square (PLS) 4.0 were
used to examine the objectives of the research.
Model Specification
The model of the research is stated in its functional form as:
UTP= f (COM, COST, INT, TRU, EASE, NETC)…………………………………….…(4)
Therefore, in its econometric form; the model for this study becomes
UTP = β0 +β1COM+ β2COST+ β3INT+ β4TRU+ β5EASE+ β5NETC +μ…...(5)
Where: β0 = Constant
UTP= usage of Twitter during presidential elections in Nigeria
COM= Compatibility
COST= Cost effectiveness
INT= Interactivity
TRU= Trust
EASE= Ease of Use
NETC= Networking Capability
The Indonesian Journal of Computer Science
https://doi.org/10.33022/ijcs.v13i6.4580 9400
D. Data Presentation and Interpretation of Result
Table 1. Descriptive Statistics and Normality Test
Mean
Media
n
Min
Max
Standar
d
Deviatio
n
Excess
Kurtosi
s
Skew
ness
Number of
Observation
s Used
Compatibil
ity
3.896
4.000
1.00
0
5.00
0
0.653
1.464
-
0.623
384.000
Usage of
Twitter
4.068
4.000
1.00
0
5.00
0
0.729
2.161
-
1.077
384.000
Cost
Effectivene
ss
3.701
4.000
1.00
0
5.00
0
0.723
-0.345
0.191
384.000
Interactivit
y
3.914
4.000
1.00
0
5.00
0
0.600
3.559
-
0.983
384.000
Trust
3.846
4.000
2.00
0
5.00
0
0.649
0.805
-
0.525
384.000
Ease of Use
3.701
4.000
1.00
0
5.00
0
0.708
-0.348
0.238
384.000
Networking
Capability
3.948
4.000
2.00
0
5.00
0
0.759
-1.088
0.015
384.000
Source: Field Survey (2024)
The descriptive statistics and normality tests for factors influencing the
usage of Twitter during presidential elections in Nigeria provide a comprehensive
overview of the central tendencies and distribution of data across variables. The
mean values are above 2.5 which is the average of the responses, indicating that
most respondents relatively favor the questions towards agreement than
disagreement. The standard deviations which are all below 1 suggest moderate
variability in the responses of the respondents. Skewness values are negative for
all variables showing a negatively skewed normal distribution as they are all
within -3 and +3. Excess kurtosis is generally positive and shows a peaked
distribution, indicating a few extreme cases. The data is also verified for normality
here since all results are within -3 and +3. The implication of these findings is that
the data is valid and normally distributed, hence it can be used for further analysis.
Measurement Model Assessment
The Indonesian Journal of Computer Science
https://doi.org/10.33022/ijcs.v13i6.4580 9401
Figure 1. Factors Influencing the Usage of Twitter During Presidential Elections in
Nigeria
Source: Field Survey (2024)
The structural model depicted in Figure 1 demonstrates the relationships
between compatibility, cost effectiveness, interactivity, trust, ease of use,
networking capability and usage of Twitter during presidential elections in
Nigeria. The model clearly depicts that the outer model has strong weight on the
latent variables with the values higher than 0.5. The model also makes it easier to
understand the relationships between the variables. The visual representation of
the data allows for a more intuitive understanding of the relationships between the
variables, which can be helpful in decision-making and strategizing regarding
factors influencing the usage of Twitter during presidential elections in Nigeria.
Table 2. Validity and Reliability Statistics
Cronbach's
Alpha
Composite
Reliability
Average Variance
Extracted (AVE)
Compatibility
1.000
1.000
1.000
The Indonesian Journal of Computer Science
https://doi.org/10.33022/ijcs.v13i6.4580 9402
Usage of Twitter
1.000
1.000
1.000
Cost
Effectiveness
1.000
1.000
1.000
Interactivity
1.000
1.000
1.000
Trust
1.000
1.000
1.000
Ease of Use
1.000
1.000
1.000
Networking
Capability
1.000
1.000
1.000
Source: Field Survey (2024)
The reliability and validity statistics, including Cronbach's Alpha and
Composite Reliability are all above 0.7. This confirms that the constructs used in
this study are robust and reliable. High Cronbach's Alpha values indicate strong
internal consistency, suggesting that the items used to measure factors influencing
the usage of Twitter during presidential elections in Nigeria reliably reflect these
constructs. Additionally, the Average Variance Extracted (AVE) for all constructs
are above 0.5, indicating excellent convergent validity, meaning that the constructs
explain a large portion of the variance among the items. This high level of validity
and reliability implies that the study's measurements are well-constructed and
accurately reflect the underlying compatibility, cost effectiveness, interactivity,
trust, ease of use, networking capability and their impact on the usage of Twitter
during presidential elections in Nigeria. The implication is that the constructs are
sufficiently precise and reliable for analyzing the relationships, which strengthens
the study's conclusions about these relationships.
Table 3. Discriminant Validity
Compa
tibility
Usage of
Twitter
Cost
Effective
ness
Intera
ctivity
Tr
ust
Ease
of Use
Networking
Capability
Compatibili
ty
1.000
Usage of
Twitter
-0.050
1.000
Cost
Effectivenes
0.623
-0.030
1.000
The Indonesian Journal of Computer Science
https://doi.org/10.33022/ijcs.v13i6.4580 9403
s
Interactivity
0.835
-0.022
0.529
1.000
Trust
0.245
0.418
0.135
0.274
1.0
00
Ease of Use
0.574
-0.097
0.812
0.442
0.1
04
1.000
Networking
Capability
0.241
0.110
0.589
0.202
-
0.0
27
0.543
1.000
Source: Field Survey (2024)
The discriminant validity results show clear distinctions between the
constructs. The diagonal values represent the square root of the AVE, while the off-
diagonal values represent the correlations between the variables. The diagonal
values are all greater than the off-diagonal values, indicating adequate
discriminant validity for the variables. Usage of Twitter during presidential
elections in Nigeria having a high self-correlation and much lower correlations
with other variables. This suggests that while certain factors may overlap or
influence each other, they have unique impacts on Twitter during presidential
elections in Nigeria.
Multicollinearity Test
Table 4. Variance Inflation Factor
Compa
tibility
Usage of
Twitter
Cost
Effective
ness
Intera
ctivity
Tr
ust
Ease
of Use
Networking
Capability
Compatibili
ty
4.147
Usage of
Twitter
Cost
Effectivenes
s
3.817
Interactivity
3.416
Trust
1.093
The Indonesian Journal of Computer Science
https://doi.org/10.33022/ijcs.v13i6.4580 9404
Ease of Use
3.153
Networking
Capability
1.658
Source: Field Survey (2024)
The VIF results for all the construct variables indicate that multicollinearity
is not a concern within this model, as all values are below the commonly accepted
threshold of 5. This suggests that each independent variable contributes uniquely
to explaining variations in the usage of Twitter during presidential elections in
Nigeria, without significant overlap with other variables. The relatively low VIF
values of the constructs, imply that these factors provide distinct insights into the
usage of Twitter during presidential elections in Nigeria that are not confounded
by other factors influencing the usage of Twitter. The implication here is that the
factors influencing the usage of Twitter, independently influence the usage of
Twitter during presidential elections in Nigeria without being distorted by
multicollinearity. This enhances the reliability of the regression coefficients and
supports the robustness of the model in identifying distinct predictors of the usage
of Twitter during presidential elections in Nigeria.
Structural Model Assessment
Test of Hypothesis: There are no Significant Factors influencing the Usage of
Twitter during Presidential Elections in Nigeria
Table 5. Path Coefficient
Original
Sample
(O)
Sample
Mean
(M)
Standard
Deviation
(STDEV)
T Statistics
(|O/STDEV|)
P Values
Compatibility-> Usage of Twitter
0.030
0.028
0.090
3.338
0.005
Cost Effectiveness -> Usage of
Twitter
0.001
0.003
0.077
2.014
0.009
Interactivity -> Usage of Twitter
0.074
0.079
0.094
7.791
0.000
Trust -> Usage of Twitter
0.479
0.481
0.059
8.076
0.000
Ease of Use -> Usage of Twitter
0.247
0.250
0.076
3.267
0.001
Networking Capability -> Usage
of Twitter
0.280
0.282
0.066
4.225
0.000
Source: Field Survey (2024)
The bootstrap path coefficient analysis depicted in table 5 was conducted to
test the null hypothesis that there are no significant factors influencing the usage
The Indonesian Journal of Computer Science
https://doi.org/10.33022/ijcs.v13i6.4580 9405
of Twitter during presidential elections in Nigeria. A look at these path shows that
the relationship is statistically significant. The p-values are less than the
conventional significance level of 0.05 and the T statistics are greater than 1.96,
suggesting strong evidence to reject the null hypothesis. Therefore, compatibility,
cost effectiveness, interactivity, trust, ease of use and networking capability
significantly affects the Usage of Twitter during Presidential Elections in Nigeria.
However, compatibility, cost effectiveness, interactivity, trust, ease of use
and networking capability have a positive significance with Usage of Twitter
during Presidential Elections in Nigeria. Overall, compatibility, cost effectiveness,
interactivity, trust, ease of use and networking capability have a partial significant
effect on Usage of Twitter during Presidential Elections in Nigeria. Hence, the null
hypothesis is not totally rejected not accepted. This implies that compatibility, cost
effectiveness, interactivity, trust, ease of use and networking capability
significantly affect Usage of Twitter during Presidential Elections in Nigeria. This is
consistent with the works of [16] and [17].
Table 6. Coefficient of Determination
R Square
R Square Adjusted
Usage of Twitter
0.258
0.246
Source: Field Survey (2024)
The R² value for Usage of Twitter during Presidential Elections in Nigeria is
0.258, indicating that 25.8% of the variance in Usage of Twitter during Presidential
Elections in Nigeria can be explained by compatibility, cost effectiveness,
interactivity, trust, ease of use and networking capability. This moderate level of
explanation suggests that while compatibility, cost effectiveness, interactivity,
trust, ease of use and networking capability contribute to understanding Usage of
Twitter during Presidential Elections, a significant portion of the variance is
influenced by other factors not captured in this model, such as macroeconomic
conditions etc.
Table 7. F Square
Compa
tibility
Usage of
Twitter
Cost
Effective
ness
Intera
ctivity
Tr
ust
Ease
of Use
Networking
Capability
Compatibili
ty
0.000
Usage of
Twitter
Cost
0.000
The Indonesian Journal of Computer Science
https://doi.org/10.33022/ijcs.v13i6.4580 9406
Effectivenes
s
Interactivity
0.002
Trust
0.083
Ease of Use
0.026
Networking
Capability
0.004
Source: Field Survey (2024)
The effect size, often denoted as f-square is depicted in table 7, this
measures the magnitude of the relationship or impact of independent variables on
a dependent variable in statistical analysis. This study assesses the effect sizes of
various latent variables on "Usage of Twitter during Presidential Elections". All the
independent variables all have a value above 0.02 except trust which is considered
small effect size. Hence, this suggests all the variables indicate a moderate effect
size, indicating that they all have a noticeable impact on Usage of Twitter during
Presidential Elections. In other words, changes or differences in any of the
variables can explain moderately the variability in Usage of Twitter during
Presidential Elections in Nigeria.
E. Conclusion and Recommendations
Social media networks made it possible for users to read news, share it, and
discuss significant events in addition to interacting with one another. With the use
of mobile devices, social media has enabled a large number of individuals to have
access to the Internet. The study concludes that there are factors influencing the
usage of twitter during presidential elections in Nigeria. Therefore, it is advised
that politicians should always seek for information through social media that
might improve their interactions with various populace. Consequently, social
media adoption becomes essential due to its multifaceted and symmetrical nature.
F. References
[1] J., Pastor-Galindo, P., Nespoli, F. Gómez Mármol, and G., Martínez Pérez. The
not yet exploited goldmine of OSINT: Opportunities, open challenges and
future trends, IEEE Access 8, p. 10282–10304, 2020.
[2] E. Shearer, and A. Mitchell. News use across Social Media Platforms in 2020:
Facebook stands out as a regular source of news for about a third of
Americans. Pew Research Center.
https://www.journalism.org/2021/01/12/news-use-across-social-media-
platforms-in-2020, 2021.
[3] J., Farkas, and C., Neumayer. Mimicking news: How the credibility of an
established tabloid is used when disseminating racism, Nordicom Review,
vol. 41 n. 1, p. 1-17, 2020.
[4] Y., Dai, & L., Luqiu. Camouflaged propaganda: A survey experiment on
political native advertising. Research & Politics, vol. 7 n. 3, p. 1–10, 2020.
The Indonesian Journal of Computer Science
https://doi.org/10.33022/ijcs.v13i6.4580 9407
[5] L., Stephan, S., Laura, G., David, H., Ralph, W., Jim, E., Stefanie, R.R.E. O., Cailin,
K., Anastasia, L. S., Philipp, B. Yannic and L., Mark. Technology and
Democracy: Understanding the influence of online technologies on political
behaviour and decision-making, Tech. rep., EU Joint Research Center, 2020.
doi:10.2760/709177.
[6] A., Cruz, G., Rocha and H., Cardoso. On Sentence Representations for
Propaganda Detection: From Handcrafted Features to Word Embeddings.
Proceedings of the 2nd Workshop on NLP for Internet Freedom: Censorship,
Disinformation, and Propaganda, Hong Kong, China, p. 107–112, 2019.
[7] K., Born and N., Edgington. Analysis of philantrhopic opportunities to
mitigate the disinformation/propaganda problem. Retrieved from
https://www.hewlett.org/wp-content/uploads/ 2017/11/Hewlett-
Disinformation-Propaganda-Report.pdf, 2017.
[8] A., Barron-Cedeno, I., Jaradat, J., Giovanni, D., Martino, and P., Nakov. Proppy:
Organizing the news based on their propagandistic content. Information
Processing & Management, vol. 56 n. 5, p. 1849–1864, 2019.
[9] G., Martino, S., Cresci, A., Barrón-Cedeño, S., Yu, R., Pietro, P., and Nakov, P. A
survey on computational propaganda detection, in Proceedings of the
twenty-ninth international conference on international joint conferences on
artificial intelligence, 4826–4832, 2021.
[10] V., Rubin, Y., Chen, N., and Conroy. Deception Detection for News: Three
Types of Fakes . Retrieved from
https://asistdl.onlinelibrary.wiley.com/doi/epdf/10.1002/pra2.2015.14505
2010083, 2016.
[11] J., Johnson, and W. A., Johnson. Explication of theoretical foundation of
communication policy. Arabian Journal of Business and Management Review
2013.
[12] N. D., Morah, and O. Omojola. Nigeria, Media and Economic Effects of
Terrorism. In O. Omojola (ed), Communication Aspects of Conflicts and
Terrorism: A Focus on Nigeria and Some Multi-ethnic Societies. Lagos: Corel
Serve Publishing, p. 139-159, 2011.
[13] N. C., Ezeh, N. A., Chukwuma, and N. V. Enwereuzo. Internet mediatization:
New opportunity for women in politics? in J. Wilson and Gapsiso, N. (eds)
Overcoming Gender Inequalities through Technology Integration.
Pennsylvania (USA): IGI Global, p. 211-224, 2015.
[14] N. C., Ezeh, N. A., Chukwuma, and O. B. Okanume. Internet and women self -
empowerment: opportunities, strategies, and challenges. International
Journal of Social Sciences and Humanities Review. Vol. 17 n. 1, p. 203-212,
2017.
[15] J. W., Creswell. Educational research: Planning, conducting, and evaluating
quantitative and qualitative research (4th ed.). Boston, MA: Pearson, 2012.
[16] N., Okorie, T., Oyedepo, S., Usaini, and O. Omojola. Global news coverage on
victimization and challenges of Roma migrants from Romania: An
experiential study. 3rd International Conference on Creative Education (ICCE
2017) Location: Kuala Lumpur, Malaysia. 2017.
[17] A. F., Segun-Alalade, O. A., Alalade, A., Olasore, A. O., Iyilade and P.O. Popoola.
Factors Influencing the Choice of Social Media Usage among Secondary
