TX_1~ABS:AT/ADD:TX_2~ABS:AT 20 http://journals.cihanuniversity.edu.iq/index.php/cuesj CUESJ 2022, 6 (1): 20-25 ReseaRch aRticle The Full Factorial Design Approach to Determine the Attitude of University Lecturers toward e-Learning and Online Teaching due to the COVID-19 Pandemic Nozad H. Mahmood1*, Dler H. Kadir2, Hawkar Q. Birdawod3 1Department of Business Administration, Cihan University Sulaimanyia, Kurdistan Region, Iraq, 2,3Department of Statistics, Salahaddin University, Erbil, Kurdistan Region, Iraq, 3Department of Business Administration, Cihan University-Erbil, Kurdistan Region, Iraq ABSTRACT The purpose of this study was to determine faculty members’ attitudes toward online learning in Kurdistan region universities. The study examined the biographic and personal characteristics of the lecturer toward e-learning. The data were collected among faculty members at Cihan University-Sulaimaniya, and to analyze the data, a full factorial design with five main factors at two levels and no central points was applied for this specific purpose. The study’s findings indicated that there was no significant relationship between gender and lecturer attitude toward e-learning. In comparison to teachers with an MSc degree, those with a PhD have a more negative attitude toward e-learning. Furthermore, full-time faculty members have a greater positive effect on teachers’ attitudes than part-time lecturers. Likewise, the results indicate that lecturers who earned their most recent education degrees outside of Iraq have a more favorable attitude. Similarly, lecturers in the sciences are more favorable to e-learning than those in the arts and social sciences. In addition, the findings demonstrated that the interaction factors (Gender) and (Education Degree) have a negative effect on lecturers’ attitudes when they are combined. Besides that, the interaction of factors (Country of Last Education Degree) and (Faculty Member Types) improves attitudes toward e-learning. Based on the results, it is suggested that academic staff receive e-learning training to deepen their knowledge and understanding of such a modern teaching system. There is also a need to enhance factors related to positive attitudes toward e-learning among university lecturers. The findings of this study are necessarily significant to both teachers and educational organizations in Kurdistan Region universities. Keywords: e-Learning, attitudes, COVID-19, faculty members, factorial design INTRODUCTION One of the most influential sectors of human life that were stopped due to the coronavirus in most countries was the university education system. According to reports by UNESCO, numerous governments around the world have closed their educational institutes for the short term to control the spread of the COVID-19 Pandemic. Approximately 1.725 billion college students globally have been suffering from the closure of colleges due to the coronavirus pandemic. In line with the UNESCO monitoring report, 192 countries have applied for nationwide closures.[1] Staying in quarantine at home and avoiding daily physical interactions with others is one way to control the severity of this virus and prevent its rapid spread. That is why most countries and universities have decided to use online teaching as an alternative educational methodology.[2] In most developed countries, universities offer not only face-to-face instruction but also distance learning options for students who prefer to study at home rather than on campus. Based on the effects of the COVID-19 disease, a new education system was mandated by the Ministry of Higher Education for the Kurdistan Region universities, which was a blended education system, or a mixed education system, in which the practical lectures are taught face-to-face on the campus, but the theoretical ones are online.[3,4] When compared to developed countries, developing countries such as Iraq and the Kurdistan Region face several Corresponding Author: Nozad H. Mahmood, Department of Business Administration, Cihan University Sulaimanyia, Sulaimanyia, 46001, Kurdistan Region, Iraq. E-mail: Nozad.mahmood@sulicihan.edu.krd Received: December 26, 2021 Accepted: January 24, 2022 Published: February 10, 2022 DOI: 10.24086/cuesj.v6n1y2022.pp20-25 Copyright © 2022 Nozad H. Mahmood, Dler H. Kadir, Hawkar Q. Birdawod. This is an open-access article distributed under the Creative Commons Attribution License (CC BY-NC-ND 4.0). Cihan University-Erbil Scientific Journal (CUESJ) Mahmood, et al.: Using Factorial Design to Determine the Attitude of University Lecturers Towards Online Teaching 21 http://journals.cihanuniversity.edu.iq/index.php/cuesj CUESJ 2022, 6 (1): 20-25 obstacles when applying for online education. It might be due to the lack of having a strong internet connection and insufficient skills in the use of information technology.[5,6] Likewise, learning how to use a new modern platform and an advanced application is still a weird thing, even at universities in developing countries.[7,8] The problem statement for this project is students’ dissatisfaction with the online system, as well as teachers’ performance in using it. The main research questions that guided the study are: 1. What are teachers’ attitudes toward e-Learning, In general? 2. Is there any association between each the factors (Education Degree, Types of Faculty Members, the Country of the Last Education’s Degree, Gender, Stream of Education), and Teachers’ Attitude Toward e-Learning? 3. Is there any interaction effect between the factors towards e-learning? The Objective of the Study The essential goals of this research are to examine teachers’ attitudes toward e-learning in Kurdistan Region universities and find out the factors that influenced their attitudes toward e-learning during the COVID-19 Pandemic. The Significance of the Study The importance of this research is to increase the desire of teachers for the use of an online education system that may, in the next few years, replace the current system of on-campus education. The Hypothesis of the Study The first hypothesis (H1): there are no statistically significant differences in the attitudes of lecturers toward e-learning implementation, whether they are male or female. The second hypothesis (H2): There are no statistical differences in lecturers’ attitudes toward e-learning based on the type of faculty members. The third hypothesis (H3): There are no statistical differences in lecturers’ attitudes towards e-learning based on their education degrees. The fourth hypothesis (H4): There are no statistical differences in lecturers’ attitudes toward e-learning based on the last education degree in a country. The fifth hypothesis (H5): there are no statistically significant differences in the attitudes of lecturers toward e-learning implementation, whether their stream of education is art or science. LITERATURE REVIEW Lecturer attitudes are extremely crucial for successfully establishing and implementing an educational method in the teaching and learning process. On the other hand, teachers are the central figures in all formal education, and their attitudes toward e-learning have a significant impact on their decisions to accept or reject the system. In other words, the success of e-learning in education is highly dependent on the lecturers’ attitudes toward it.[9,10] Effectively initiating and implementing instructional innovation in e-Learning processing at the university is highly dependent on lecturers’ attitudes. Thus, understanding lecturers’ personal characteristics is a prerequisite for implementing an e-learning system.[11] Academic staff demographic and personal characteristics, such as educational degrees, gender, educational stream, faculty type, and country of last educational degree, may influence lecturers’ favorable attitudes toward the e-learning system. Several previous studies have examined the demographic and personal characteristics of teachers in relation to e-learning. There was a study that investigated the factors of gender in e-learning and discovered that gender has no influence on teachers’ attitudes. In other words, there is no statistically significant difference in attitudes regarding e-learning between men and women.[8,12] In addition, the literature links teachers’ attitudes to their personal characteristics. Demographic characteristics, including gender, education level, and age, had no significant effect on their attitudes toward online learning. Teachers’ nationality and marital status, on the other hand, have significant attitudes toward e-learning.[13] University teachers in the arts have negative attitudes toward e-learning, whereas lecturers in the sciences have positive attitudes.[14] Methodology The general purpose of this study is to examine lecturer support and attitudes toward e-learning in terms of several personal variables and to determine the direction of their attitudes and the level of correlation between these variables. The questionnaire consists of two main sections. The first section contains the demographic profile of the participants, such as gender, education degree, types of faculty members, last education degree country, and stream of education, while the second part includes 15 statements that describe overall attitudes toward e-learning, and a five-point Likert scale was used to rate each item from strongly disagree to strongly agree. Then, JMP statistical software was used to analyze the data to apply a full factorial design. Since there are five main factors, and each factor has two levels with a response variable, we used a 2k factorial design. which included (five main factors) indicates five main effects, and a 5 2 10 � � � � � � � ��two factor�interactions). Three-factor interactions and above are very rarely significant effects.[15] Description and Selection of Levels To conduct a 25 full factorial design, all factors must be set to two levels, the low and high levels. The factors used in the experiment and their associated levels are given in the following Table 1. The Planning Matrix and Experiment Results The planning matrix and experiment results for 25 full factorial design with 32 runs, and without any center point get the following raw data. Then, a randomized order was used to execute test runs of the design Table 2. Mahmood, et al.: Using Factorial Design to Determine the Attitude of University Lecturers Towards Online Teaching 22 http://journals.cihanuniversity.edu.iq/index.php/cuesj CUESJ 2022, 6 (1): 20-25 Data Analysis We need to achieve a model between the response variable (University lecturers’ attitudes toward e-learning) and the five factors to clarify their power of relationships and the effects which influenced the e-learning during the COVID-19 pandemic. The results of Table 3 show that R-Square is 0.967806, meaning that the factors altogether explain a large amount of 96.78% of the variance in response variable (university lecturers towards e-learning), and it is evidence that the model is appropriate. Furthermore, Root Mean Square Error shows a quite low 0.22727. Figure 1 is a half-normal plot of parameter that employed to show the effect of main factors on the Y response (university lecturers’ attitudes toward e-learning). All the factors that lie along the blue line (Lenth’s Pseudo Standard Error) are negligible. In contrast, the large effects, or significant factors are far from the blue line. The most significant effects based on this analysis show the main effects of D (last Education degree’s country), E (Stream of Education), C (Types of Faculty Table 1: Factors and their levels Factors Low (-1) High (+1) A: Gender Female Male B: Education Degree MSc degree PhD Degree C: Types of Faculty Members Part-Time Full-Time D: Last Education degree’s country Inside (Iraq) Outside (Other Countries) E: Streams of Education Art Science Table 2: Planning matrix and experiment results # Pattern A B C D E Y 1 −−−−− −1 −1 −1 −1 −1 2.44673 2 +−−−− 1 −1 −1 −1 −1 1.98352 3 −+−−− −1 1 −1 −1 −1 1.54322 4 ++−−− 1 1 −1 −1 −1 2.05211 5 −−+−− −1 −1 1 −1 −1 3.10284 6 +−+−− 1 −1 1 −1 −1 2.19841 7 −++−− −1 1 1 −1 −1 2.273441 8 +++−− 1 1 1 −1 −1 2.31429 9 −−−+− −1 −1 −1 1 −1 3.49831 10 +−−+− 1 −1 −1 1 −1 3.40167 11 −+−+− −1 1 −1 1 −1 3.39617 12 ++−+− 1 1 −1 1 −1 3.38519 13 −−++− −1 −1 1 1 −1 3.73804 14 +−++− 1 −1 1 1 −1 3.51055 15 −+++− −1 1 1 1 −1 3.68208 16 ++++− 1 1 1 1 −1 3.44734 17 −−−−+ −1 −1 −1 −1 1 2.95147 18 +−−−+ 1 −1 −1 −1 1 2.22851 19 −+−−+ −1 1 −1 −1 1 2.11037 20 ++−−+ 1 1 −1 −1 1 2.4 21 −−+−+ −1 −1 1 −1 1 3.26823 22 +−+−+ 1 −1 1 −1 1 2.77876 23 −++−+ −1 1 1 −1 1 2.67552 24 +++−+ 1 1 1 −1 1 2.974158 25 −−−++ −1 −1 −1 1 1 4.32626 26 +−−++ 1 −1 −1 1 1 4.59298 27 −+−++ −1 1 −1 1 1 3.84214 28 ++−++ 1 1 −1 1 1 4.62367 29 −−+++ −1 −1 1 1 1 4.26693 30 +−+++ 1 −1 1 1 1 4.60214 31 −++++ −1 1 1 1 1 4.72248 32 +++++ 1 1 1 1 1 4.48045 Table 3: Summary of Fit R-Square 0.967806 Root Mean Square Error 0.22737 Mean of Response 3.213062 Observations 32 Figure 1: Half-normal probability plot for the factors and their interactions Table 4: Parameter estimate Mahmood, et al.: Using Factorial Design to Determine the Attitude of University Lecturers Towards Online Teaching 23 http://journals.cihanuniversity.edu.iq/index.php/cuesj CUESJ 2022, 6 (1): 20-25 Members), and it looks like there are some interactions effects between the factors. The above Figures 2 are the visualization of confidence intervals (CI) that can be applied to the F-test by using a leverage plot. The first panel shows the overall test of the model which explains the linearly relationship between actual values and the predicted values of response Y. Likewise, the other panels are the main factor effects for A, B, C, D, and E, respectively. Each panel displays the mean as a horizontal line (blue line), a regression line (red line), and a 95% CI. Since, the significance of each effect is indicated by the CIs and the mean, when the CI crosses the mean line, indicating a significant effect. As the result, there are some significant effects of the main factors (B), (C), (D), and (E), whereas factor A (gender) is insignificant. Y = −0.092898 (B)+ 0.1641668 (C) + 0.7567132 (D)+ 0.3396924 (E) − 0.1163893 (A*B) + 0.1226638 (D*E) (1) According to the results in Table 4, the main factors of D, E, C, and B are statistically significant due to their small P-values which are less than. On the other hand, since the P-value of the main factor A (Gender) is equal to 0.5082 and it is larger than, indicates that gender is statistically insignificant or shows no effect. Furthermore, there is a meaningful interaction between factors D and E which shows a positive effect on the response Y. Likewise, there is a significant interaction between factors A and B toward a positive direction. Consequently, the regression model equation becomes as below: In the linear regression equation (1), factor (B: Education Degree) has a negative effect on the (University lecturers’ attitudes toward e-learning), indicating that teachers with a PhD degree 9.22898% have a more negative attitude towards e-learning than holders of MSc degree. As well, factor (C: Types of Faculty Members) has a positive impact on the teachers’ attitudes. In other words, full-time lecturers 16.411668% have more positive towards e-learning than part-time lecturers. Furthermore, factor (D: Last Education Degree’s Country) has a positive influence on the teachers’ attitudes towards e-learning, and the results illustrated that the lecturers whose last Response Y A Leverage Plot Factor Names Y: University lecturers’ attitudes toward e-learning A: Gender B: Education Degree C: Types of Faculty Members D: Last Education Degree’s Country E: Stream of Education B Leverage Plot C Leverage Plot D Leverage Plot E Leverage Plot Figure 2: Leverage plots important factors Mahmood, et al.: Using Factorial Design to Determine the Attitude of University Lecturers Towards Online Teaching 24 http://journals.cihanuniversity.edu.iq/index.php/cuesj CUESJ 2022, 6 (1): 20-25 Table 5: Goodness-of-Fit Test Shapiro-Wilk W Test W Prob