Investigation of 21st-Century Competencies and E-Learning Readiness of Higher Education Students on the Verge of Digital Transformation

This study aims to determine the 21st-century competencies of higher education students on the verge of digital transformation and their readiness for e-learning and to reveal the relationship between these two variables. Having been carried out by using single and relational screening models, this study was conducted with 2100 students attending higher education and participating in the research voluntarily. E-Learning Readiness Scale consisting of 33 items and 6 factors and 21st-Century Skills Scale consisting of 41 items and 5 factors were used as data collection tools. KolmogorovSmirnov, Mann-Whitney U, Games-Howell and Dunnett T3 tests were used for the analysis of data collected online. As a result of the research, it has been concluded that 21st-century competencies of higher education students are at a high level while their level of elearning readiness is moderate. It has been determined that there are differences in the scores of the scale and its sub-dimensions according to age, gender and internet use status, and there is a statistically significant, positive and weak relationship between 21stcentury competencies of students and their level of e-learning readiness. Received: 4 December 2020 Accepted: 3 March 2021


Introduction
It has been observed in recent years that the use of digital technologies in daily life has gradually increased followed by changes occurring in many areas. In particular, technological tools play crucial roles in the daily lives of individuals defined as "digital natives" by Prensky (2001). The learning and thinking styles of these individuals, who have been surrounded by technology in almost every aspect of their lives, also differ from previous generations (Gümüşoğlu, 2017), leading to differentiation in the demands and needs of individuals. In an era when learning information remains insufficient, the content of the information as well as how it is learned and put into practice in real life has gained Besides, almost all of the universities in Turkey have begun to play a part in the online environment of the educational process by establishing Distance Education and Application Centres. In this way, individuals have been provided with access to education without time and space limitations. The use of information and communication technologies such as electronic media, educational technology, the internet, and computers in the teaching process is defined as e-learning (Arslan, Kutluca & Özpınar, 2011;Contreras & Hilles 2015). Elearning platforms not only increase learner engagement, but also provide the opportunity to develop an individual understanding of course content and to have access to a wider range of resources (Wang, Chen, & Khan, 2014). The introduction of teaching in online environments has brought some problems as well. One of these problems is to what extent students possess readiness for e-learning, which has a different characteristic than face-toface learning. Mosadegh, Kharazi and Bazargan (2011) stated that readiness is an important factor in the successful implementation of e-learning. Similarly, Al-araibi, Naz'ri, Yusoff & Chuprat (2019) emphasized that one of the main reasons for the failure in the implementation of e-learning is the inadequacy of students' readiness.
Higher education institutions are supposed to adapt themselves to the needs of society (Gümüşoğlu, 2017). For, higher education institutions undertake the crucial role of raising qualified individuals needed in society and business life. In this context, individuals are expected to graduate from higher education institutions as individuals with 21st-century skills. The level of 21st-Century skills of individuals in Turkey has been determined in various studies. In the study conducted by Aygün et al. (2016), competence perceptions of prospective teachers regarding 21st-Century skills were determined. Gürültü, Aslan, and Alcı (2019) examined secondary school teachers' competencies of using 21st-century skills in their study. In the study conducted by Bozkurt and Çakır (2016), the 21st-century skill levels of secondary school students were examined in terms of various variables. Cemaloğlu et al. (2019) sought to determine the self-efficacy perceptions of the teachers working in vocational high schools regarding 21st-century skills. Özdemir-Özden et al. (2018) also tried to determine the competence perceptions of prospective teachers regarding 21st-century skills in their study. It has been observed that there are limited studies at the higher education level and these limited studies are mostly aimed at prospective teachers. However, higher education aims to raise qualified individuals not only for a specific field but for all fields. It is The population of the study consists of 11078 students at a public university in Turkey. There are a total of 19 academic units in the relevant university, including 10 Faculties, 3 Colleges and 6 Vocational Schools. However, a total of 6 academic units, including 2 faculties, 1 College and 3 Vocational Schools, which do not have students yet or do not have enough students to meet the level of representation despite being within the university, were not included in the study. Proportional stratified sampling method, one of the probability-based sampling methods, was used in determining the participants in the higher education institution where the study was conducted. The proportional stratified sampling is the representation of subgroups in the population in proportion to their weight in the population (Shaughnessy, Zechmeister, & Zechmeister, 2016). In this context, taking into account the distribution of students in academic units, the population was divided into 13 strata, and a sample group close to the same percentage from each stratum was formed by considering the percentage shares of students in these strata in the population. Accordingly, 2100 students constitute the sample of the research. The sample of the research represents 20% of the population of the research. The distribution of the students who make up the population and sample of the research according to academic units is given in Table 1. in the study are given in Table 2.   (Nunnally & Bernstein, 1994). This situation indicates that the scale items serve to measure the desired properties. to "Completely suitable for me (7)". The lowest score that can be obtained from the scale is 33 and the highest score is 231. If the score is close to 231, it means that there is more readiness for e-learning. The reliability coefficients of the factors of the e-learning readiness scale vary between 0.84 and 0.95. The construct reliability coefficient of the scale was 0.98 and the Cronbach Alpha reliability coefficient was calculated as 0.93. In this study, the Cronbach Alpha reliability coefficient of the scale was calculated as 0.94. The reliability coefficients of the scale factors vary between 0.83 and 0.92. Nunnally and Bernstein (1994) stated that the critical value should be 0.70 to ensure that the measurement results based on the measurement tool are reliable. In this respect, it was concluded that the e-learning readiness scale applied to higher education students produced reliable measurements.

Data Collection and Analysis
Due to the Covid-19 pandemic in the 2020-2021 academic year, when the study was conducted, the data were collected online, as the courses were conducted through distance education at the university in question. For this purpose, the necessary permissions and approvals for the decision of the Ethics Committee dated 18.06.2020 and numbered 49 were Research and Application Centre, the necessary link address for the scales to access the forms prepared in the electronic environment was sent to students' e-mail addresses.
Necessary instructions were given in the link address on which the scale was filled in by the students.
Before analysing the data collected in the digital environment, the reliability of the data was checked. In this context, among the data collected from 2376 students, 276 data, which were not considered valid for reasons such as marking the same items, missing incorrect marking, were not included in the study. In this respect, the number of data, which was 2376 at the beginning, was determined as 2100 after the elimination process.
Calculations were made after 6 negative items in the 21st-century competencies scale were reversed while no change was made in the calculations regarding the items included in the elearning readiness scale with no negative items.
To interpret the scale and its sub-dimensions, an evaluation was made on average scores. For ease of interpretation, three evaluation intervals, which are low, medium and high, were taken as a basis. In the calculation of the score range, the formula (Highest Value -Lowest Value) / Evaluation Interval Number formula was used. The evaluation interval and criteria accepted in the study are presented in Table 3. During the data analysis stage, the SPSS program was used in all statistical analyses.
As a result of the Kolmogorov-Smirnov test (p <.05) applied to the scale and its subdimensions in order to control the assumption of normality of the data, it was observed that they did not show normal distribution. For this reason, the Mann-Whitney U test was used for paired comparisons and the Kruskal Wallis-H test for comparisons more than two. In order to determine the difference between groups, Dunnett T3 for the sample size less than 50 and Games-Howell for the sample size larger than 50 from Post Hoc tests were used.
Pearson Moments Multiplication Coefficient correlation value was examined to determine the relationship between 21st-century competencies and e-learning readiness. The data were analysed with the SPSS program and the significance level was accepted as 0.05.

Findings
The findings of higher education students' 21st-century competencies and their readiness for e-learning are presented below within the framework of research problems.
On what levels are 21st-century competencies and e-learning readiness of higher education students? Descriptive statistics values for the scale and its sub-factors within the scope of the sub-problem are given in Table 4. The arithmetic mean, standard deviation, skewness and kurtosis values are included in the descriptive statistics table. It is understood from Table 4 that higher education students find themselves -century competencies and at a medium -learning. Considering 21st-century competencies in terms of sub-dimensions, it is seen that the sub-dimensions of "information and critical thinking and problem--dimensions are found to be at a moderate level.
When examined in terms of e-learning readiness sub-dimensions, "computer self--efficacy --dimensions were found to be moderate. "Internet self-5.47), "self--dimensions are observed to be at a high level.
Do higher education students' 21st-century competencies and their e-learning readiness differentiate according to the gender variable? Mann-Whitney U test analysis was conducted within the scope of the sub-problem. Whitney U analysis results are given in Table 5.

90
As seen in Table 5, the level of 21st-century competence of higher education students differs by gender [U = 406654,00 p <0.05]. Considering the mean rank, it is seen that the differentiation is in favour of female students. The level of the 21st-century competence of female students is higher than male students. When examined in terms of sub-dimensions, gender is a significant factor in all sub-dimensions of 21st-century competencies. In Considering the e-learning readiness levels of higher education students, it is revealed that there is a significant difference according to gender. [U = 503386.00 p <0.05].
Considering the mean rank, it is seen that the differentiation is in favour of female students.
It is observed that the level of e-learning readiness of female students is higher than that of Do higher education students' 21st-century competencies and their e-learning readiness differ according to the age variable? Kruskal Wallis-H test analysis was performed within the scope of the sub-problem. Kruskal Wallis test analysis results are given in Table 6. As seen in Table 5, the age variable does not cause a significant difference in 21stcentury competence levels of higher education students (X 2 (sd = 3, n = 2100) = 5.64, p> .05), yet it causes a significant difference in "critical thinking and problem-solving" (X 2 (sd = 3, n = 2100) = 8.33, p <.05), "social responsibility and leadership" (X 2 (sd = 3, n = 2100) = 16.00, p < .05) and "career awareness" (X 2 (sd = 3, n = 2100) = 21.38, p <.05) sub-dimensions. The Games-Howell test, which is one of the Post-Hoc tests conducted to determine the difference, shows that the difference in the sub-dimension of "critical thinking and problem-solving" is between the ages of 18-21 and 30+, the difference in the sub-dimension of "social responsibility and leadership" is between the ages of 26-29 and 30+, and the difference in the "career awareness" sub-dimension is in the 30+ age range. In line with these data, the critical thinking and problem-solving competencies of the students between the ages of 18-21 and the social responsibility and leadership competencies of the higher education students between the ages of 26-29 are higher than the students between the age range of 30+.
However, career awareness competencies of higher education students in the age range of 30+ are lower than those of other age groups.
Do higher education students' 21st-century competencies and their e-learning readiness differentiate according to the daily internet usage variable? Kruskal Wallis-H test analysis was performed within the scope of the sub-problem. Kruskal Wallis-H test analysis results are given in Table 6. As seen in Table 6, the variable of daily internet usage causes a significant difference in 21st-century competence levels of higher education students (X 2 (sd = 3, n = 2100) = 9.73, p <.05). It also causes a significant difference in all sub-dimensions except for the "social responsibility and leadership" sub-dimension. As a result of the Dunnett T3 test, one of the Post-Hoc tests performed to determine the difference, it is seen that the difference is caused by the daily internet usage range of 0-2 to 3-5 hours. In line with these data, students who use the internet for an average of 0-2 hours a day, their 21st-century competencies, critical thinking and problem-solving, entrepreneurship and innovation, career awareness competencies are lower than students who use the internet for 3-5 hours a day.
Daily internet usage variable does not cause a significant difference in the e-learning readiness levels of higher education students (X 2 (sd = 3, n = 2100) = 2.05, p> .05). However, the internet self-efficacy (X 2 (sd = 3, n = 2100) = 12.38, p <.05) of the students who use the internet for an average of 0-2 hours per day is lower than the students who use the internet for 3-5 hours daily.
Is there a statistically significant relationship between higher education students' 21st-century competencies and their e-learning readiness? Pearson Correlation analysis was 94 performed within the scope of the sub-problem. Pearson Correlation analysis results are given in Table 7. As can be seen in Table 7, there is a statistically significant, positive, weak relationship (r = .370; p <.05) between higher education students' 21st-century competencies and their level of e-learning readiness. When examined in terms of sub-dimensions, while there is a positive and low (.220 ≤r≤.396; p <.05), there is a very low and positive correlation (r = .118; p <.05) between the motivation for e-learning sub-dimension among the subdimensions of "computer self-efficacy", "internet self-efficacy", "online communication selfefficacy", self-learning", and "learner control". .
There is a moderate positive correlation (.406 ≤r≤.427; p <.05) between "information and technology literacy" and "career awareness" and a weak positive correlation between "entrepreneurship and innovation" and "social responsibility and leadership" (.209 ≤r≤ .370; p<.05). On the other hand, it was found that there is a very weak negative correlation (r = -. 137; p <.05) between the "critical thinking and problem-solving" sub-dimension.

Discussion, Conclusion and Recommendations
The results of this study, which examines 21st-century competencies and e-learning readiness of higher education students on the verge of digital transformation, are presented under the headings within the framework of the research sub-problems.
According to the results of the research, it was determined that 21st-century competence level of the female students is higher than the male students. It was concluded that female students were more competent than male students in terms of information and technology literacy, critical thinking and problem solving, entrepreneurship and innovation, social responsibility and leadership. However, in the career awareness sub-dimension, it has been seen that there is a significant difference in favour of male participants. When the literature is examined, unlike the present study, it is seen that the level of 21st-century competencies does not differ according to the gender variable (Cemaloğlu etc., 2019;Kozikoğlu & Altunova, 2018). In the present study, the fact that 21st-century competencies of female students are higher than those of men may be because women are more open to learning in terms of information and technology literacy, critical thinking and problem solving, entrepreneurship and innovation, social responsibility and leadership and continue their learning habits outside the university. In addition, it is thought that these results may have occurred because there are more departments related to social sciences in the higher education institution where the study is conducted and that these departments are preferred more by female students. Therefore, considering this difference, it is recommended to determine the sample characteristics well and to reconsider the gender variable. Thus, it can be said that the results obtained in terms of 21st-century competencies can be better discussed in terms of gender variable.
Higher education students' levels of readiness for e-learning differ in favour of female students. E-learning readiness of female students is higher than male students. While there is a significant difference in favour of women in the sub-dimensions of computer self-efficacy, internet self-efficacy, online communication self-efficacy, motivation for e-learning, there is a significant difference in favour of male participants in the learning control sub-factor. In the self-learning sub-dimension of the scale, there is no significant difference according to gender. This result obtained from the research differs at this point with the results of similar studies in the literature. In similar studies, it has been concluded that there is no overall gender differentiation (Jackson et al., 2008;Yılmaz, Sezer, & Yurdugül, 2019). This situation is thought to be due to the characteristics of the sample group. In future studies, it is suggested that the data obtained from the sample should be analysed together with different variables (such as technological access, technological infrastructure, use of technology, technical support). As a matter of fact, it is stated that problems such as lack of technological support can negatively affect students' e-learning readiness (Naresh & Reddy, 2015). This result reveals the necessity of various studies to increase students' readiness for e-learning. In this context, institutional measures can be taken by organizing technology-supported elective courses, additional sessions or courses for students.
The age factor does not cause a significant difference in 21st-century competence levels of higher education students. However, critical thinking and problem solving competencies of students aged 18-21, and social responsibility and leadership competencies of higher education students aged 26-29 are higher than students aged 30+. However, career awareness competence of higher education students in the age range of 30+ is lower than students in the other age range. Although there is no direct study in the literature regarding this result, it can be said that higher education students under the age of 30 have higher levels of critical thinking and problem solving, social responsibility and leadership, career awareness. However, it is recommended to future researcher(s) in order to interpret the results obtained from the information about the students' previous learning experiences more properly.
The age factor differs in favour of students under the age of 30 in the level of elearning readiness of higher education students. Computer self-efficacy, internet selfefficacy, online self-efficacy and self-learning sub-dimension levels of higher education students were also higher than students aged 30 and over. When the literature is examined, there is no study directly related to the age variable regarding the state of being ready for elearning. However, it is possible to come across studies on the grade level variable (Çetin, 2008;Hung et al., 2010;Yurdugül & Demir, 2017). Yurdugül and Demir (20107) concluded in their study that students in the 4th-grade of higher education are more ready for e-learning in terms of all components of e-learning readiness compared to the 1st-grade students.
Similarly, in the study conducted by Çetin (2008), the level of computer self-efficacy of upper-class university students was found to be higher than those of lower classes. This situation in the studies may be related to the learning experiences of the students during their time in higher education. However, it is thought that this result, which emerged within the scope of this study, may be more related to the learning experience in higher education.
Therefore, it is recommended that the data to be obtained regarding the age variable in future studies should be presented by considering them together with different parameters.
The variable of daily internet usage causes a significant difference in 21st-century competence levels of higher education students. Students who use the internet for an average of 3-5 hours a day have higher 21st-century competencies and information and technology literacy, critical thinking and problem solving, entrepreneurship and innovation, career awareness competencies than students who use the internet daily for 0-2 hours.
Considering this situation, it is recommended that institutions should take into account the daily internet usage periods and take measures for this in developing the 21st-century competence levels of students. In this context, in line with the suggestion of Orban, Teeling-Smith, Smith, and Porter (2018), online courses and face-to-face lessons can be taught in a way as blended learning. However, the variable of daily internet usage does not cause a significant difference in the level of e-learning readiness of higher education students. The internet self-efficacy of students who use the internet for an average of 0-2 hours per day is lower than the students who use the internet for an average of 3-5 hours per day. In general, it is known that there is a positive relationship between internet usage time and internet selfefficacy (Yenilmez et al., 2011). Therefore, it can be said that this situation is an expected result. However, for what purposes students use the internet and the effect of this on the variables rather than how long they spend on the internet can be revealed by the studies.
There is a statistically significant, positive and weak relationship between higher education students' 21st-century competencies and their level of e-learning readiness. While a weak and positive relationship is found between the 21st-century competencies and the dimensions of computer self-efficacy, internet self-efficacy, online communication selfefficacy, self-learning and learner control, there is a positive and weak relationship between the 21st-century competencies and the motivation for e-learning sub-dimension. There is a positive, moderate-level relationship between students' e-learning readiness and information and technology literacy, career awareness and a low and positive relationship between entrepreneurship and innovation, social responsibility and leadership.
In the 21st century, higher education institutions are one of the most effective institutions in changing the skills expected from individuals and gaining these skills. In this context, it is important to determine to what extent individuals studying at higher education institutions possess these skills. The pandemic process, which emerged at the end of 2019 and affected the world and was an obstacle for face-to-face education, once again revealed the importance of e-learning. It is also important to determine how ready individuals are for distance education processes, both in the normal process and in compulsory situations such as pandemics. Within the scope of this research, 21st-century skills of individuals and their level of e-learning readiness were determined and the relationship between these variables was revealed. Future studies may investigate how to ensure individuals acquire 21st-century skills and how to increase the e-learning readiness of individuals. It is thought that it would be beneficial to carry out similar studies in primary, secondary and high schools to reveal the situation and discuss the measures that can be taken for that.

Ethics Committee Permit Information
Name of the board that carries out ethical evaluation: Siirt University Ethics Committee The date of the ethical assessment decision: 18/06/2020 Ethical assessment document number number: 2020/49