7
Empirical Evidence of LMS Adoption
in the Middle East
7.1 Introduction
This chapter explains the process of data preparation, results,
and preliminary data analysis. The prime focus of this
chapter is the appropriateness of the data obtained about data
analysis. This chapter presents the preliminary data results
and the statistical methods applied in data analysis. In this
chapter, the personal profile of the participants and the
descriptive data analysis are discussed. The descriptive data
analysis of the core variables (such as effort expectancy,
performance expectancy, social influence, facilitating conditions, hedonic motivation, and habit) and moderating
variables (experience, age, and cultural dimensions) is discussed. In the last section, the reliability, correlation, factor
analysis, and regression analysis are discussed.
7.2 Descriptive Statistics
This section provides a descriptive analysis of the participants. The main attributes of the personal profile of the
participants include nationality, age, years of LMS experience, department, teaching rank of instructors, and educational level. The participants for this study include the Arab
and non-Arab instructors from various educational institutions in the Eastern Province of Saudi Arabia. The survey
questionnaire was sent to the population of 2005 instructors.
The number of received responses was 396. However, 22
responses were incomplete, wrongly filled surveys, and the
participants were not the teaching staff; therefore, those
responses were discarded. Hence, the number of valid
responses was 374, as shown in Table 7.1. After data entry,
an SPSS 22/AMOS statistical package was used for the
analyses, such as descriptive analysis, Cronbach’s alpha test,
and factor analysis. The entire population of this study
consisted of male instructors. Therefore, the ‘gender’ data
were removed from the research. If in the future, male and
female instructors teach in Saudi universities, then gender
data will be added to the study.
7.3 Quantitative Data Collection
and Analysis
The steps involved in the data analysis are given in Fig. 7.1.
Step 1: Investigation for Potential Biases
One of the most important steps in starting data analysis was
checking the database for potential biases. The collected data
were inspected for non-response bias and a common method
bias.
• Non-response Bias: This bias relates to the probability that
the participants who responded differ from those who did
not, which does not allow to conclude the whole sample. As
a result, the item non-response provides incomplete information and affects the reliability of the results. To avoid
non-responses bias, one of the solutions suggested by the
literature is the reduction of non-response itself.
• Common Method Bias: The collected data from the
same instructor with the same instrument can be a cause
of common method bias. When more variables or
constructs of the same survey questionnaire are measured from the same instructor, there are chances that
participants may respond in the same fashion and in a
similar direction. Subsequently, the two constructs may
correlate with each other and may lead to a wrong
conclusion. According to Podsakoff, MacKenzie, Lee,
& Podsakoff (2003) and Podsakoff & Organ (1986), this
kind of bias impacts the research results and needs to be
controlled.
In this study, the researcher used Harman’s single-factor
test, to address this issue, as recommended by Podsakoff
et al. (2003). In this regard, the factor analysis (EFA) was
conducted using SPSS with the options: in the extraction
process, use several factors to extract ‘1’; in the rotation
method, the selected method was ‘none’. After running this
test, one factor emerged that explains 32.4% of the variance,
© Springer Nature Switzerland AG 2021
R. A. Khan and H. Qudrat-Ullah, Adoption of LMS in Higher Educational Institutions of the Middle East,
Advances in Science, Technology & Innovation, https://doi.org/10.1007/978-3-030-50112-9_7
59
Empirical Evidence of LMS Adoption
in the Middle East
7.1 Introduction
This chapter explains the process of data preparation, results,
and preliminary data analysis. The prime focus of this
chapter is the appropriateness of the data obtained about data
analysis. This chapter presents the preliminary data results
and the statistical methods applied in data analysis. In this
chapter, the personal profile of the participants and the
descriptive data analysis are discussed. The descriptive data
analysis of the core variables (such as effort expectancy,
performance expectancy, social influence, facilitating conditions, hedonic motivation, and habit) and moderating
variables (experience, age, and cultural dimensions) is discussed. In the last section, the reliability, correlation, factor
analysis, and regression analysis are discussed.
7.2 Descriptive Statistics
This section provides a descriptive analysis of the participants. The main attributes of the personal profile of the
participants include nationality, age, years of LMS experience, department, teaching rank of instructors, and educational level. The participants for this study include the Arab
and non-Arab instructors from various educational institutions in the Eastern Province of Saudi Arabia. The survey
questionnaire was sent to the population of 2005 instructors.
The number of received responses was 396. However, 22
responses were incomplete, wrongly filled surveys, and the
participants were not the teaching staff; therefore, those
responses were discarded. Hence, the number of valid
responses was 374, as shown in Table 7.1. After data entry,
an SPSS 22/AMOS statistical package was used for the
analyses, such as descriptive analysis, Cronbach’s alpha test,
and factor analysis. The entire population of this study
consisted of male instructors. Therefore, the ‘gender’ data
were removed from the research. If in the future, male and
female instructors teach in Saudi universities, then gender
data will be added to the study.
7.3 Quantitative Data Collection
and Analysis
The steps involved in the data analysis are given in Fig. 7.1.
Step 1: Investigation for Potential Biases
One of the most important steps in starting data analysis was
checking the database for potential biases. The collected data
were inspected for non-response bias and a common method
bias.
• Non-response Bias: This bias relates to the probability that
the participants who responded differ from those who did
not, which does not allow to conclude the whole sample. As
a result, the item non-response provides incomplete information and affects the reliability of the results. To avoid
non-responses bias, one of the solutions suggested by the
literature is the reduction of non-response itself.
• Common Method Bias: The collected data from the
same instructor with the same instrument can be a cause
of common method bias. When more variables or
constructs of the same survey questionnaire are measured from the same instructor, there are chances that
participants may respond in the same fashion and in a
similar direction. Subsequently, the two constructs may
correlate with each other and may lead to a wrong
conclusion. According to Podsakoff, MacKenzie, Lee,
& Podsakoff (2003) and Podsakoff & Organ (1986), this
kind of bias impacts the research results and needs to be
controlled.
In this study, the researcher used Harman’s single-factor
test, to address this issue, as recommended by Podsakoff
et al. (2003). In this regard, the factor analysis (EFA) was
conducted using SPSS with the options: in the extraction
process, use several factors to extract ‘1’; in the rotation
method, the selected method was ‘none’. After running this
test, one factor emerged that explains 32.4% of the variance,
© Springer Nature Switzerland AG 2021
R. A. Khan and H. Qudrat-Ullah, Adoption of LMS in Higher Educational Institutions of the Middle East,
Advances in Science, Technology & Innovation, https://doi.org/10.1007/978-3-030-50112-9_7
59