Digital natives or digital immigrants? : about university teachers’ use of educational technologies

(1)

Digital natives or digital immigrants?

About university teachers’ use of educational technologies

Author: Lisa A. Ottolander Student number: 10375309

Date of submission and version: 22 June 2018, final version

Qualification: MSc. Business Administration – Digital Business Track Institution: University of Amsterdam – Amsterdam Business School First supervisor: mr. prof. dr. P.J. van Baalen

(2)

2

Statement of originality

This document is written by Lisa Ottolander, who declares to take full responsibility for the contents of this document.

I declare that the text and the work presented in this document are original and that no other sources other than those mentioned in the text and its references have been used in creating it.

The Faculty of Economics and Business is solely responsible for the supervision of completion of the work, not for the contents.

(3)

3

supporting evidence that age and age groups (i.e. digital natives and digital immigrants) significantly predict digital nativity, but not (self-reported) use of educational technology tools, a statement that has been debated in previous research. This research found no significant gender differences in digital nativity among university teachers, which supports findings from previous research that gender is not an influencing factor in digital nativity or related constructs, such as digital competence.

Key words: digital nativity, digital natives, digital immigrants, higher education,

(6)

6 1. Introduction

Continuing technological change will both enable and require new types of learning not just during childhood but throughout people’s lives. More than ever, people of all ages should be enabled to continuously update their skills to engage with the environment and the rest of the world (Vereniging van Samenwerkende Nederlandse Universiteiten, 2017, p. 20).

This excerpt is taken from the Digital Society Research Agenda of the Vereniging van Samenwerkende Nederlandse Universiteiten (VSNU; Association of Universities in The

Netherlands), which was published in November 2017. It illustrates the need for new forms of

education focused on enabling people to acquire the skills they need to live and work in a digital society. As part of this research agenda, all fourteen Dutch universities have committed to one common goal: “In ten years’ time, The Netherlands will have become a leader in the field of digitisation in society.” (VSNU, 2017, p. 5).

On a larger scale, the European Union (EU) has also published several reports on the same issue. Already in 2006, the European Parliament and the Council published a

recommendation on key competences for lifelong learning in which they state that digital competences are one of the eight essential competences for lifelong learning and successful participation of European citizens in society. The call to action for the development of a digitally competent workforce is clear and the need for education that fits the labor market is evident.

Universities have a responsibility to address the development of digital competences, not only for students, but also for teachers (Maderick, Zhang, Hartley and Marchand, 2015). Young people get trained in university to enter the labor market and digital competences are essential for surviving in a digital world (Adams Becker, Cummins, Davis, Freeman, Hall Giesinger and Ananthanarayanan, 2017). Teachers are role models for their students and if teachers have to teach students how to develop their digital skills, then surely they must master these skills themselves.

It has been argued that students crave the use of digital technologies in the classroom, but that it is the teachers who are unable to embed the effective use of digital technologies in their courses (Li, 2007). But is it really that black and white?

In the past, researchers have argued that the students of the digital generation are inherently digitally competent (Oblinger, 2006; Prensky, 2006) and should even lead the way into our 21st_{century education system (Prensky, 2006). Moreover, the student as the “digital} native”, a term coined by Prensky (2001a), is supposed to be more digitally competent than

(7)

7 their teachers, or the “digital immigrants” (Prensky 2001a; 2001b). If the digitally native students are inherently digitally competent then this requires the use of appropriate strategies for learning and teaching (Sorgo, Bartol, Dolnicar and Boh Podgornik, 2017). This sounds like a troubling matter, but can we truly speak of “digital natives” and “digital immigrants”?

Many researchers have argued that this is not the case (Kirschner and Merriënboer, 2013; Jones, Ramanau, Cross and Healing, 2010) and some have even argued that students are not in fact the digital natives they are expected to be, demonstrating only basic use and

knowledge of ICT (Margaryan, Littlejohn and Vojt, 2011). Moreover, in the digital native versus digital immigrant debate, age seems to be the defining factor. If you are born in or after 1980, you are a digital native if you are born before, you are not (Jones et al., 2010). If age is indeed the defining factor, there is no possibility of bridging the gap between the digitally native student and its digitally immigrant teacher. This is problematic because in education, teachers will (almost) always be older than their students. In sum, there seems to be a discrepancy in the beliefs about the level of digital nativity of students and teachers.

In this research, the focus is on university teachers. A perspective that has long been sidelined by a predominant focus on learners and elementary, middle and high school

environments. This is surprising, because universities are the places where students are taught their ‘final skills’ before they enter the labor market, and their teachers are the adults making this final impact on transferring the knowledge concerning these skills.

This research consists of an evaluation of the literature on the digital native versus digital immigrant distinction, that explains the theoretical binary that lies at the foundation of this thesis; second, the concept of comfort and confidence in using computers for educational purposes is discussed in terms of barriers to technology uptake. The literature review

concludes with seven hypotheses that address the knowledge gaps concerning this topic. The hypotheses are visualized in a conceptual model. This research further consists of a method section where the measures, population, sample and respondent characteristics, as well as the data preparation are described. Consecutively, the results section provides the statistical analyses used to test the hypotheses and the results of these analyses. It concludes with the conceptual model including the effect sizes that were found. Finally, the discussion section provides the theoretical and practical implications of this research and concludes with the limitations of this research and recommendations for future research.

(8)

8 2. Literature review

This chapter first provides an overview and analysis of the key literature on the concepts of digital natives and digital immigrants. Second, it explores the construct of comfort and confidence with relation to using technology in educational settings.

2.1. Digital nativity

The term “digital native”, coined by Prensky (2001a), is used to describe a young person that moves around in a ubiquitous digital environment. High intensity interaction with digital technologies has dominated the digital native’s development from childhood onwards, which has made the digital native inherently digitally savvy.

According to Prensky (2001a; 2001b), digital natives have almost become an entirely different species with a modified brain that, as he poses, is superior to its antonym the “digital immigrant”. The digital immigrant, conversely, is characterized by an older person who lacks digital skills and understanding of digital technologies. The immigrant is positioned as inferior to the digital native and Prensky (2001a; 2001b) suggests that the immigrant should adjust itself to the needs of the digital native.

Much criticism on this digital divide has since been voiced. Kirschner and

Merriënboer (2013) base their research on an extensive review of the literature surrounding the digital native, and find “overwhelming evidence” (p. 173) that the digital native does not actually exist at all. They find that learners do not have the superior abilities to deal with ICT that are ascribed to them. Where others, such as Jones et al. (2010) propose that the digital native does not exist in the radical form that Prensky (2001a; 2001b) proposes. They argue that what has been called the Net Generation (people born after 1980/1983) is in fact a heterogeneous generation in which a lot of variety exists regarding the “use and appreciation of new technologies” (Jones et al., 2010, p. 722), which was further divided between age groups and gender. Finally, Li and Ranieri (2010) argue that the so-called digital natives do not possess the digital competences that are ascribed to them and that there are large

variations in the digital performance of students that seems to be impacted by factors such as gender, age and school and that the personal use of digital technologies on a daily basis does not result in good or even excellent digital competency.

On the contrary, a study by Li (2007) of 15 secondary mathematics and science teachers and 450 students of two rural and two urban Canadian high schools, suggests a view

(9)

9 that does not correspond with the finding by Kirschner and Merriënboer (2013) and Jones et al. (2010) that the (pure) digital native does not exist. Li’s (2007) viewpoint is more in line with Prensky’s view (2001a; 2001b; 2006) that the digital native does indeed exist in the form of the student. Li (2007) finds in her study that students are, almost inherently, strong

supporters of technology use in the classroom, and that it is the teachers that are unwilling to adopt digital technologies for educational purposes (El-Senaidi, Lin and Poirot, 2009). Li (2007) goes on to argue that there is a “strong dissonance between teachers’ and students’ views on technology.” (p. 391), that, if not addressed, will lead to disharmony of the educational system. She advocates for future research, which could aid the design of a technology-enhanced educational system that serves both the goals of the teachers and the students and thus creates harmony in the (inevitable) future of technology-enhanced educational systems.

Jones et al. (2010) bring nuance to the debate by portraying the digital native as part of a heterogeneous generation. They find that the students in their sample, which is comprised of university students from five universities in England, showed large variations in technology use. Jones et al. (2010) find that the university students are divided into minorities that use different technologies to a more or lesser extent. They conclude that the “participation and generational homogeneity predicted by the Net Generation or Digital Native inspired literature” (p. 731) is not realistic, and that the so-called Net Generation (a generation of digital natives) does in fact not exist.

From a more theoretical perspective, Bayne and Ross (2007) argue that we need to be careful in using the binary of digital native vs. digital immigrant, since they view it as overly simplistic and judgmental. They argue that the digital native occupies a dominant position in the discourse regarding the digital divide, whereas the digital immigrant occupies a

subordinate position. The authors advocate a more critical view of the discourse that otherwise would “over-determine our future understanding of the complex relationships between teacher, learner, technology and higher education.” (Bayne and Ross, 2007, p. 5).

Another discussion has revolved around the narrowing down of the generational cohort that comprises the Net Generation. Jones et al. (2010) give an overview of different authors that have tried to define this generation and although there is some variation in the year that different authors choose as the birth year of the digital native, 1980 is a good average. That means that the oldest digital native in 2018 is 38 years old.

(10)

10 In sum, many researchers have argued about the existence of the digital native and the homogeneity of this generational cohort, providing different claims. Some argue that digital natives are inherently digitally competent by growing up in a digital age, whereas others pose that digital natives are less digitally savvy, or competent, than they are expected to be and that there is much variation in digital natives’ use of technology. Yet others have proposed that the discourse surrounding the digital divide between the native and the immigrant should be more nuanced, because the digital native’s perspective is being overemphasized in the debate.

Up to now, only two studies have been undertaken to explore how to measure the concept of digital nativity. The first study has aimed to develop and validate an initial Digital Natives Assessment Scale (DNAS) and the result is a “self-report instrument designed to measure students’ perception of the degree to which they are digital natives.” (Teo, 2013, p. 51). The DNAS was developed and validated by a total sample of 1,018 students attending three different secondary schools in Singapore, aged 13 to 16.

The final DNAS contains four factors. These factors consist of: growing up with technology (α = .89); being comfortable with multitasking (α = .91); relying on graphics for communication (α = .87); and thriving on instant gratifications and rewards (α = .87). Cronbach’s Alpha for all four factors was well above the accepted threshold of α = .70, indicating relatively high internal consistency.

However, the scale developed by Teo (2013) also has several limitations. One of these limitations is that the sample was taken from secondary school students in Singapore, which limits the generalisability to a larger population of different ages and cultural backgrounds. Teo (2013) suggests that future research should, among other things, focus on different age groups (i.e. under 13 or over 17 years old), gender, school levels, and cultures. Age is especially interesting since much research indicates that age is in fact not the defining factor of digital nativity.

In 2016, Teo, Yurdakul & Ursavas, published a successive research paper on the DNAS demonstrating the extended reliability of the scale for a different sample consisting of pre-service teachers in Turkey. Five hundred fifty-seven respondents from a range of

academic backgrounds with an average age of 20.53 years filled out the DNAS questionnaire. The findings of the study provide support that the DNAS can also be used for older

participants - be it though that the pre-service teachers still fall in the generational cohort of digital natives born after 1980 - and participants from a different cultural background. The study also “examined possible differences in digital nativity by gender, age, years of computer

(11)

11 use, and perceived technology competence” (Teo et al., 2016, p. 1240). The results show that there are no significant gender or age differences among the respondents when it comes to their perceived level of digital nativity. Furthermore, the study shows that respondents who “have used the computer for a longer period of time and those who perceived themselves to be more competent in using the computer” reported significantly higher scores on the DNAS, indicating that these respondents felt more like digital natives than the other respondents “who have used the computer for a shorter period of time and those who perceived themselves to be less competent in using the computer” (Teo et al., 2016, p. 1240). Teo (2013) and Teo et al. (2016) found that age does not significantly contribute to differences in digital nativity, which provides interesting input for future research into this much debated topic.

2.2. Comfort and confidence in using computers for educational purposes

Adoption of new technologies in education have been studied extensively, both theoretically and empirically, and the body of literature surrounding this topic has been growing since the introduction of the personal computer since the 1980s and has grown exponentially since the dawn of the Internet in the 1990s.

Much research has focused on the process of how people come to adopt new technologies. Several influential theoretical models have been developed, such as the Technology Acceptance Model (TAM) (Davis, Bagozzi and Warshaw, 1989) that aims to explain the behavioural intention to use technology, where intent is the predictor of actual use. Besides the TAM, other models such as the Diffusion of Innovation (DoI) and the five-step hierarchical model of technology diffusion (El-Senaidi et al., 2009) have had substantial influence. Besides these theoretical models, there have been numerous empirical studies that focus on the (perceived) barriers people experience when adopting new technologies.

Barriers to technology adoption have been classified in different ways, but most commonly as internal-external, and individual-institutional (Nikolopoulou and Gialmas, 2015). Internal barriers are, for example, negative attitudes towards technology, a lack of confidence (i.e. low self-efficacy) in using technology, but also resistance to change. External barriers are outside the influence of an individual (or group of individuals), and can be, for example, low institutional administrative or technological support, and a lack of resources.

Especially in educational environments, the experienced barriers to technology adoption can be relatively large and, moreover, technology adoption in education moves rather slowly, especially in universities (Watty, McKay and Ngo, 2016).

(12)

12 One of the internal factors, that contributes heavily to experiencing barriers to

technology adoption, is a lack of confidence among teachers in using technology (BECTA, 2004; Dawes, 2000; Russell and Bradley, 1997). Based on an extensive literature review, the conclusion from the BECTA report (2004) was that teachers who do not feel confident in their technology usage, are more likely to avoid working with computers in their classrooms and Wood, Mueller, Willoughby, Specht and Deyoung (2005) found that higher levels of comfortableness with technology predicted higher technology integration in the classroom.

Confidence and comfort in using computers for educational purposes is thus an important factor in predicting technology use and integration in classrooms. This internal barrier has received some attention in the literature, but has mainly been assessed in empirical studies of elementary, middle and high schools learner. Technology use and adoption in universities, and especially among university teachers, has largely remained outside the scope of previous research. Moreover, the connection between how digitally native teachers

perceive themselves to be and how they rate themselves on confidence and comfort using computers for educational purposes has not been researched before. If teachers perceive themselves as highly confident and comfortable, there would logically be a positive relationship with educational technology use as well. The relationships investigated in the literature are summarized in table 1 on the next page.

The research question central in this thesis is twofold: “Is there a significant

relationship between university teachers’ digital nativity and their comfort and confidence in using computers for educational purposes and what is the effect of their comfort and

(13)

13 Table 1. Summary of relationships examined in the literature

Relationship Independent

variable

Dependent variable

Source Sample Finding

Gender Digital nativity Teo, Yurdakul and Ursavas, 2016

557 pre-service teachers enrolled in a Turkish university

No significant differences in digital nativity based on gender Digital competence Li and Ranieri, 2010 317 ninth-grade students from eight Chinese middle schools

Digital competence levels did not differ by gender

Age Digital nativity Teo et al., 2016 557 pre-service teachers in a Turkish university

No significant relationship between Age and DN Prensky, 2001a - Argues that digital natives

are people who were born into a digital world and are therefore inherently skilled users of technology Age Technology usage Salaway, Caruso and Nelson, 2008 27,254 American undergraduates

Age significantly predicted usage of digital communication and collaboration tools Age Digital competences Li and Ranieri, 2010 317 ninth-grade students from 8 Chinese middle schools

Digital competence levels depended on age, where the youngest students performed significantly better than the older students

Age group Technology usage Jones, Ramanau, Cross and Healing, 2010 596 English first-year university students

There were strong age related variations in using new technologies Confidence and comfort Technology adoption Wood, Mueller, Willoughby, Specht and Deyoung, 2005 54 Canadian elementary and secondary school teachers Comfortableness with computers significantly predicted integration of computers in the classroom BECTA, 2004 Literature review and

survey completed by 170 teachers and other practitioners attending a conference on ICT in education

Confidence and comfort in using computers in the classroom is a significant predictor of technology uptake in educational settings

Dawes, 2001 140 British teachers, ICT coordinators and administrators in 2 secondary and 12 primary schools

Lack of confidence and comfort in using ICT in education is an important barrier to ICT uptake in classrooms

Russell and Bradley, 1997

350 Australian primary and secondary school teachers who filled in an open ended

questionnaire

Confidence in using computers was essential to use computers effectively in classrooms Teaching experience Beliefs and perceptions about ICT in education Jimoyiannis and Komis, 2007

1165 Greek primary and secondary education teachers

Teaching experience was strongly associated with teachers’ beliefs and perceptions about ICT

(14)

14 2.3. Hypotheses and conceptual model

From the literature it appears that the concept of digital nativity and related concepts, such as digital competence, technology adoption and usage, but also the effects of age (group) and gender on these variables, have largely been examined using student and teacher samples from elementary, middle and high schools. Therefore, it is interesting to focus on the

university teacher’s perspective in this thesis.

The relationship between gender and digital nativity and related constructs that has been examined in previous research, shows that there are in fact no gender differences. The first hypothesis tests whether these findings hold true when a sample of older respondents is used. However, because previous research has indicated both significant and non-significant relationships between age and digital nativity (and related constructs), this relationship is assessed with age as control variable.

H1 = There are no significant gender differences in digital nativity, after controlling for age.

Age has thus been linked to digital nativity in previous research, but the effects found for this relationship have been inconsistent, the second and third hypotheses seek to test whether there is a relationship between age and digital nativity and whether there is a significant difference in digital nativity between digital natives and digital immigrants.

H2 = There is a significant relationship between age and digital nativity.

H3 = There are significant differences in digital nativity between digital natives and digital immigrants.

Age has been examined as a factor influencing the use of digital tools. Previous research has found significant relationships between these two variables, but this research is largely limited to student samples. The fourth hypothesis tests whether there is relationship between age and use of educational technology tools among university teachers.

(15)

15 Age is further divided into two categories: digital natives (people born in or after 1980) and digital immigrants (people born before 1980). Previous research has found a

significant relationship between different age groups and the use of different technologies, but the study was limited to a sample of first-year university students. The fifth hypothesis is thus as follows.

H5 = There are significant differences in use of educational technology tools between digital natives and digital immigrants.

The relationship between digital nativity and confidence and comfort in using

computers for educational purposes has not been examined before. However, this relationship could provide an interesting insight that - if a positive relationship exists – could indicate that hiring digitally native teachers could be a way to increase confidence and comfort among staff in using computers for educational purposes, which in turn could lead to higher use of

educational technology tools, as hypothesized in hypothesis 6. Based on logical reasoning it could be expected that the relationship between digital nativity and confidence and comfort is moderated by age, in such a way that this relationship is stronger for younger teachers born in or after 1980, who have grown up with technology and are described in the literature as also possessing the other attributes of digital natives (Teo, 2013).

H6a = There is a significant positive relationship between digital nativity and comfort and confidence in using computers for educational purposes.

H6b = There is a significant positive relationship between digital nativity and comfort and confidence in using computers for educational purposes and this relationship is moderated by age in such a way that this relationship is stronger for teachers of a younger age (i.e. for teachers born in or after 1980).

H7 = There is a significant positive relationship between confidence and comfort in using computers for educational purposes and use of educational technology tools, after controlling for teaching experience.

In hypothesis 7, teaching experience is included as control variable, because previous research has indicated that teaching experience influences teachers’ beliefs about ICT for

(16)

16 educational purposes. Beliefs about ICT in education and use of educational technology tools are not the same concepts, but it could be expected that the two are interrelated. If teachers have positive views of ICT for educational purposes, then they could also be expected to use these technologies more in their teaching practice. According to research by Jimoyiannis and Komis (2007), teachers with little teaching experience (1-10 years) and much experience (>30 years) are positive about ICT in education, whereas teachers with 11-15 years, 21-25 years, and 26-30 years of teaching experience have negative beliefs. Finally, teachers with 16-20 years of experience are neutral to positive about ICT in education.

Below, the conceptual model with the corresponding hypotheses is depicted.

Figure 1. Conceptual model

4. Method 4.1. Measures

Digital nativity was one of the main variables, which was measured using a validated

questionnaire (Teo, 2013; Teo et al., 2016). The questionnaire consisted of 21 items divided across four factors: growing up with technology (GUWT); being comfortable with

multitasking (Multi); relying on graphics in communication (GIC); and thriving on instant gratification and rewards (IGR). Respective examples of items were “I use the Internet everyday”; “I am able to surf the Internet and perform another activity comfortably”; “I use pictures more than words when I wish to explain something”; and “I wish to be rewarded for everything I do” (Teo, 2013).

All items were measured on a 7-point Likert scale ranging from 1 “strongly disagree” to 7 “strongly agree”. Digital nativity was calculated by summing the scores (1-7) for every

Age Confidence &

Comfort Digital Nativity Teaching experience Use of technology H1 H6a H7 Gender H6b H2 H5 Birth group H4 H3

(17)

17 item for every respondent. This resulted in a total score of minimum 21 and maximum 147 for every respondent, where a score closer to 147 indicates a higher level of digital nativity. Based on the score range (21-147), a categorization of the scores was made up as follows: a score of 21-62 indicated a low level of digital nativity; a score of 63-104 indicated a mediocre level of digital nativity; and a score of 105-147 indicated a high level of digital nativity.

Confidence and comfort in using computers for educational purposes (CC) was one of

the other main variables. CC was measured using a validated scale consisting of 9 items (Hogarty, Lang and Kromrey, 2003). All items were measured on a 5-point Likert scale, consistent with the original validating study. Examples of items were “I am comfortable using my computer during classroom instruction”; and “the computer enhances my teaching”. Respondents were asked to indicate their level of agreement on a scale ranging from 1 “completely disagree” to 5 “completely agree”.

For each respondent the summed score of all items was taken as the total score CC. The summed scores could thus range from minimum 9 to maximum 45, where a score of 9-20 indicated a low level of CC; a score of 21-32 a mediocre level of CC; and a score of 33-45 a high level of CC.

Use of educational technology tools (ETT), was measured twofold. In the

questionnaire, respondents were first asked to indicate whether or not they used a variety of 16 ETTs (e.g. e-learning modules, digital learning environment, screen recording software, interactive whiteboards, etc.). If a respondent indicated no, zero points were awarded. If a respondent indicated yes, one point was awarded. This resulted in a total summed score ranging from 0 to 16 for every respondent. The overarching variable was computed and labelled “summed score of ETT use” (SSETT).

Second, if the respondents answered “yes”, they were shown the next question which asked about the frequency of use of that specific ETT (1 = never, 2 = rarely, 3 = occasionally, 4 = frequently, and 5 = always). Only if a question was answered positively, the respondent would see the next question asking about the frequency of use. If a question was answered negatively, the respondent would not see the next question asking about frequency of use, but immediately skipped to the next question asking about their use (yes or no) of another one of the 16 ETTs. This resulted in a summed score of minimum 0, when a respondent did not use any of the 16 ETTs and thus did not see any of the follow-up questions about frequency of use, and maximum 96, when a respondent used all 16 ETTs and responded with “always” to

(18)

18 each follow-up question. The overarching variable was computed and labelled “summed score of frequency of ETT use” (SSFETT). To get a final image of someone’s use of the 16 ETTs, both scores for SSETT and SSFETT were summed and a new overarching variable was computed for all respondents and labelled “overall ETT use” (OETT). This new variable had a range of minimum 0 and maximum 112, where a score of 0-37 indicated low overall use of ETTs, a score of 38-75 indicated mediocre overall use of ETTs, and a score of 76-112 indicated high overall use of ETTs.

Age was measured by birth year rather than current age. Each respondent was asked to

fill out their birthyear in a format of YYYY (e.g. 1985). Later a new variable was computed for age, which represented the current age based on the birth year the respondent filled out (i.e. 33 instead of 1985). The reason respondents were asked about their birthyear rather than their age, was because the digital native-digital immigrant divide is based on birthyear. This provided to possibility to group all respondents with a birthyear prior to 1980 in the category of digital immigrants, and the respondents with a birthyear in or after 1980 in the category of digital natives.

Finally, all remaining variables that were used to form the conceptual model were:

gender (1 = male, 2 = female); function (0 = PhD, 1 = Postdoc, 2 = Lecturer, 3 = Assistant

Professor, 4 = Associate Professor, 5 = Full Professor); and teaching experience (0 = 0-5 years, 1 = 6-10 year, 2 = 11-15 years, 3 = 16-20 years, 4 = more than 20 years).

4.2. Population and sample

The population of this research consisted of teaching staff at the University of Amsterdam. This population (last measured in 2016) consisted of 297 professors, 245 associate professors, 773 PhD candidates, and 548 lecturers. This is a total of 1.863 staff members with (potential) teaching responsibilities, although in reality not all (associate) professors, PhDs and lecturers have teaching responsibilities. However, the researcher requested more recent numbers from the UvA personnel administration that would also distinguish between teaching and non-teaching staff. However, the UvA personnel

administration was not able to provide more recent numbers or numbers where a distinction could be made between teaching staff and non-teaching staff.

The type of sampling used was non-probability sampling. A convenience sample of Economics and Business teachers was used due to the fact that the researcher had good access

(19)

19 to the teaching staff of the Faculty of Economics and Business (FEB). According to the online accessible University of Amsterdam factbook, the FEB employed 39 professors, 33 associate professors, 46 PhD candidates, and 62 lecturers in 2016 (last point of measurement). This results in a total of 180 teaching staff, although in reality not all professors, associate professors, and PhD candidates have teaching responsibilities. Unfortunately, there was no possibility to receive more recent figures in which a distinction could be made between teaching and non-teaching staff. For the sake of this research, the sample thus consists of 180 teaching staff.

First, the questionnaire was distributed by the secretariat of the Amsterdam Business School (ABS) by e-mail to all ABS staff at this faculty, the e-mail contained a short

description of the research and a link to the online questionnaire. Upon request, the ABS secretariat provided the information that the mailing list, to which the questionnaire was sent, consisted of approximately 200 ABS staff members. Two weeks later, the link to the survey was also included in the digital teachers’ newsletter that was sent to all FEB teaching staff. Finally, the researcher e-mailed a total of 377 FEB teachers, whose e-mail addresses were listed on the UvA website, with a final reminder to fill out the survey earlier distributed to them by the ABS secretariat and / or the teachers’ newsletter. This e-mail received 75

“address not found” automatic replies and five automatic replies that the teacher did not work at the UvA anymore. This means that the reminder e-mail reached a total of 297 FEB

teachers. The average reach of these two e-mails and the digital newsletter was thus 225 FEB teachers, which is more than the official sample size of 180. However, since the last point of measurement of staff members was 2016, it is possible that this number has increased by 2018.

4.3. Respondents

The questionnaire received 69 responses in total, of which 64 were usable. Five were not usable due to substantial incompletion. The response rate, based on the sample size of 180, therefore was the number of usable questionnaires (64) divided by the total sample (180) multiplied by 100, which results in a response rate of 35.56%, which is a relatively high response rate. Furthermore, twelve teachers indicated they were interested in receiving the results of this research, which indicates that there is interest in this research topic among the FEB teachers.

Frequencies showed that of the 64 respondents, 52 were male (81%) and 12 were female (19%). The average age of the respondents was 45 years (minimum 25, maximum 74).

(20)

20 Of all respondents (64), 37 (58%) belonged to the category of digital immigrants (people born before 1980), and 27 (42%) to the category of digital natives (people born in or after 1980). Average teaching experience was 11 to 15 years and the average number of courses taught per teacher was three. On average, the respondents devoted 26% to 75% of their work at

university on teaching. Of the 64 teachers participating in this study, 11% held a PhD

position; 5% indicated they were postdoctoral researchers; 16% were lecturers; 30% assistant professors; 20% associate professors; and 18% full professors. The full demographics table is included in Appendix section 1, table 1.1.1.

4.4. Research design and instrument

The research design of this research was cross-sectional and the chosen instrument for data collection was an online self-complete questionnaire build in Qualtrics. The

questionnaire was preceded by a short description of the research in both Dutch and English. The questionnaire consisted of 69 closed questions with predetermined answer options, except question 1 which asked for the birthyear of the respondent. All questions required a response (i.e. response was forced) before the respondent could proceed to the next question. However, not every respondent answered the same set of questions, this was due to the fact that the number of questions shown to the respondent depended on the answers the respondent provided (see section 4.1. about the measures).

The time it took to complete the questionnaire was 5 to 10 minutes. The questionnaire was accessible online for anyone who was sent the URL and the questionnaire was active for a period of three weeks and four days. This research design and method provided a good opportunity to reach a relatively large group of potential respondents within the limited time frame of a master’s thesis. The questionnaire was tested on twenty respondents from the population and was found to be adequate before it was officially distributed.

4.5. Data preparation

Of the 69 responses, five were deleted listwise due to incompletion. The questionnaire did not include any counter indicative items, so recoding was not necessary.

Twenty-two Respondents answered “other, namely…” to the question “What function do you have?” and therefore did not choose from one of the three predetermined options (assistant professor; associate professor; full professor). From their answers three categories could be distilled: PhD student (8); postdoctoral researcher (3); and lecturer (8). One

(21)

21 was included in the group of full professors. In the end, three extra categories (PhD, postdoc, and lecturer) were created for the variable “function”. Where 0 was “PhD”, 1 “Postdoc”, 2 “Lecturer”, 3 “Assistant professor”, 4 “Associate professor”, and 5 “Full professor”.

Age was measured by birthyear only, due to the fact that digital immigrants can be classified as born before 1980 and digital natives as born in or after 1980. All respondent were asked to fill out their birthyear in the format YYYY (e.g. 1985). However, for easier

interpretation of the spread of age in the dataset, a new variable (age) was created by subtracting the birthyears of the respondents from 2018 (the current year). Age thus represented the current age of the respondents.

In order to know how many of the respondents belonged to the different categories of digital native and digital immigrant, a new variable (birth group) was computed by coding all respondents with a birthyear before 1980 as 1, and respondents with a birthyear in or after 1980 as 2.

The scores for digital nativity and confidence and comfort in using computers for educational purposes were both summed and created as new variables (DNTOT and CCTOT) to create the total scores of the respondents. These scores could range from 21 to 147 for digital nativity, and from 9 to 45 for confidence and comfort.

All scales used in the questionnaire were adopted from previous research in their original form. In the original studies, Cronbach’s alpha for all scales was well above the >.7 threshold. Reliability for all scales was computed again using the data of this research. The results are shown in table 2 on the next page.

(22)

22 Table 2. Correlation matrix of variables, including mean, standard deviation, and Cronbach’s alpha

M SD 1 2 3 4 5 1. Digital nativity 101.61 16.90 1 GUWT 31.22 3.69 (.59) Multi 30.61 7.54 (.77) GIC 17.17 6.96 (.89) IGR 22.61 4.94 (.75) 2. Confidence & Comfort 36.22 6.76 .30* 1 (.92) 3. Age 45 12.42 -.30* -.18 1 4. Teaching experience 2.00 1.48 -31* -.02 .84** 1 5. Use of technology 31.81 6.76 .17 .40** -.08 .07 1

Note: N = 64, *Correlation significant at the 0.05 level (2-tailed), **Correlation significant at the 0.01

level (2-tailed), Cronbach’s alpha between brackets

Cronbach’s alpha for all scales, except one (GUWT; growing up with technology), were above the accepted threshold of .7, showing good to high internal consistency. The reliability of the GUWT scale was slightly too low with a Cronbach’s alpha of .59, which indicated that the scale did not measure different aspects of the same attribute. Corrected item-total correlations showed that two items (GUWT1: I use the Internet every day and GUWT3: when I need to know something I search the Internet first) had Perason correlation coefficients smaller than .3. These items thus did not correlate properly with the overall GUWT scale. When rerunning the reliability analysis without these two items, Cronbach’s alpha increased to .63, which was little improvement from .59. However, the scale was validated in two previous studies (Teo 2013; Teo et al., 2016) where larger samples were used. In economics and business research, a Cronbach’s alpha of minimum .6 can sometimes still be accepted to continue with the analysis. Deleting two items from the scale, would result in a scale of only three items, which, in comparison to the other digital nativity scales, was too few. For these reasons, the scale was maintained in its original form (Cronbach’s alpha was almost .6 with .59) and was used for the rest of the analysis.

From the descriptives it was also visible that the average score on digital nativity (101.61 on a scale of 21 to 147) could be categorized as mediocre. Furthermore, the average confidence and comfort level was high (36.22 on a scale of 9 to 45), but the self-reported use of educational technologies was rather low (31.81 on a scale of 0 to 112).

(23)

23 The table shows several significant correlations. The strongest significant correlation was +.84 between age and teaching experience. This relatively strong positive relationship was easily explained, since the older someone becomes, the more experience someone will have. The average age in this sample was 45, and the average teaching experience is 11 to 15 years, this shows that the large correlation between the two variables is only logical. This was therefore not an interesting correlation to examine further.

The first interesting significant correlation was between confidence and comfort and use of technology. The correlation of +.40 with a significance level of .01 showed a moderate positive relationship. This indicated that as confidence and comfort increases, technology use increases as well and the other way around.

The remaining three significant correlations between 1) confidence and comfort and digital nativity (+.30), 2) age and digital nativity (-.30), 3) and between teaching experience and digital nativity (-.31) all showed weak relationships. This suggested that 1) there is a tendency such that, the more confident and comfortable someone reports him- or herself to be, the more digitally native he or she is and vice versa; 2) that there is an indication that the older someone is, the less digitally native he or she is; and 3) that the more teaching experience someone has, the less digitally native he or she is.

These conclusions were in line with the expectations based on the literature, since in the literature age has been proposed as one of the main defining factors of either belonging to the category of digital native or digital immigrant, where people born before 1980 are

classified as digital immigrants. The third conclusion was also relatively straightforward, since the more teaching experience someone has, the older they are likely to be and thus how less digitally native they tend to be. The weak positive relationship between confidence and comfort and digital nativity also seemed logical. Based on normal reasoning, it could be expected that the more digitally native someone is, the more confident and comfortable someone would be with using computers for educational purposes. However all these

relationships are relatively weak, which indicates that there are other variables that contribute to these relationships.

(24)

24 5. Results

In this section the statistical procedures for testing the hypotheses are described and the results are provided. Below, the descriptives for the most important concepts are given in table 3. All supporting evidence of the testing of the assumptions is provided in the appendix, section 1.1. through 1.8.

Table 3. Summary of descriptives of main concepts

M SD Min Max Age 45 12.42 25 74 Birth group 1.42 .50 1 2 Gender 1.19 .39 1 2 Digital nativity 101.61 16.90 53 137 Confidence and comfort 36.22 6.76 15 45 Teaching experience 1.98 1.58 0 4 Use of educational technology tools 31.81 12.59 10 65 Note: N = 64 5.1. Hypothesis 1

In order to determine whether any significant gender differences in digital nativity were present after controlling for age, a one-way ANCOVA was run. Before an ANVOCA was run, the corresponding assumptions had to be considered.

As assessed by visual inspection of a scatterplot with superimposed fit lines at both subgroups, a linear relationship between age and digital nativity was established for both men and women. There was homogeneity of regression slopes since the interaction term was not statistically significant, F(1,60) = .09, p = .76. Standardized residuals for gender were normally distributed, as assessed by Shapiro-Wilk’s test (p > .05). There was

homoscedasticity, as assessed by visual inspection of the standardized residuals plotted against the predicted values. There was homogeneity of variance, as assessed by Levene’s test of homogeneity of variance (p = .40). There were no outliers in the data, as assessed by no cases with standardized residuals greater than ±3 standard deviations.

Estimates showed the means and standard deviations adjusted by the covariate (age) for both men and women on their levels of digital nativity. Digital nativity levels were higher for women (M = 109.02, SE = 4.81) than for men (M = 99.90, SE = 2.24). Results are reported below in table 4. However, the difference in digital nativity levels between male and female university teachers was not statistically significant, F(1,61) = 2.86, p = .10. The calculated

(25)

25 effect size of Cohen’s d was .75, indicating a large effect. This could mean that the non-significant difference that was found between men and women in digital nativity scores is perhaps only not significant because the sample size used in this research is too small, which could mean that a significant difference potentially could be found using a larger sample.

Table 4. Unadjusted and adjusted gender means and variability for digital nativity with age as a covariate Unadjusted Adjusted N M SD M SE Female 12 111.50 12.68 109.02 4.81 Male 52 99.33 17.02 99.90 2.24 Total 64 101.61 16.90

Note: N = number of participants, M = Mean, SD = Standard Deviation, SE = Standard Error

5.2. Hypothesis 2

A linear regression was run to determine whether there was a significant relationship between age and digital nativity. Before the regression could be executed, the corresponding assumptions were tested. First, linearity was established by visual inspection of a scatterplot with superimposed fit line. There were no outliers and homoscedasticity was confirmed by visual inspection of the standardized residuals versus standardized predicted values plot. Furthermore, residuals were normally distributed based on the inspection of a histogram with superimposed normal curve and the normal probability plot.

Results of the regression analysis showed that age accounted for 8.9% of the variation in digital nativity with adjusted R2 = 7.4%. Age significantly predicted digital nativity, F(1,62) = 6.04, p = .017. The regression equation was: digital nativity = 119.85 + (-.41*age).

This means that for every year someone gets older, digital nativity decreases by .41 (95% CI, -.74 to -.08). A Pearson’s correlation coefficient of -.41 can be considered a medium effect. The null hypothesis that age does not significantly predict digital nativity could thus be rejected and there was evidence that hypothesis 2 is supported. The results of the regression are reported in table 5.

Table 5. Summary of simple linear regression analysis of age predicting digital nativity

Confidence Interval B

Variable B SE β p Lower Upper

Intercept 119.847 7.694 .000 104.467 135.222

Age -.405 .165 -.298 .017 -.735 -.076

Note: N = 64, dependent variable = digital nativity, B = unstandardized

regression coefficient, SE = Standard Error, β = standardized coefficient,

(26)

26 5.3. Hypothesis 3

In order to determine whether there were significant differences in scores on digital nativity between digital natives and digital immigrants, an independent samples t-test was run. Before running the test, the corresponding assumptions were checked.

There were no outliers in the data, as assessed by inspection of a boxplot. Scores on digital nativity were approximately normally distributed, as assessed by inspection of the Q-Q plots. There was homogeneity of variances for digital nativity scores for digital natives and digital immigrants, as assessed by Levene’s test for equality of variances (p = .12). There were 37 digital immigrants and 27 digital natives. Digital natives scored higher on digital nativity than digital immigrants (M = 108.04 ± 13.33 versus M = 96.92 ± 17.83). Mean digital immigrants’ score on digital nativity was thus -11.12 (95% CI, -19.26 to -2.97) lower than the mean digital natives’ score. This difference was statistically significant, t(62) = -2.73, p = .008. Cohen’s d was calculated which resulted in an effect size of d = .65, which can be considered a medium to large effect. The null hypothesis can thus be rejected and there is evidence that hypothesis 3 is supported. The difference in mean scores on digital nativity for digital natives and digital immigrants are visually depicted in figure 2.

Figure 2. Bar chart displaying group means on digital nativity scores for digital natives and digital immigrants

(27)

27 5.4. Hypothesis 4

In order to determine whether there was a significant relationship between age and use of educational technology tools, a simple linear regression was carried out. Linearity was

confirmed by visual inspection of a scatterplot with superimposed regression line. There were no outliers and there was homoscedasticity as assessed by visual inspection of a scatterplot of standardized residuals versus standardized predicted values.

The prediction equation was: use of educational technology tools = 35.43 (-.08*age). Age did not statistically significantly predict use of educational technology tools, F(1,62) = .39, p = .53, accounting for .6% of the variation in use of educational technology tools with adjusted R2 = -.01. A Pearson’s correlation coefficient of r = -.08 and a non-significant

finding indicate that there is no relationship between age and use of educational technology tools. The null hypothesis could thus not be rejected. The results of the regression are reported in table 6.

Table 6. Summary of simple linear regression of age predicting use of educational technology tools

Intercept 35.43 5.99 .00 23.47 47.40

Age -.08 .13 -.08 .53 -.34 .18

Note: N = 64, dependent variable = use of educational technology tools, B = unstandardized regression coefficient, SE = Standard Error,

β = standardized coefficient, p = significance, confidence interval at 95%

In order to determine whether there were significant differences between digital natives and digital immigrants in use of educational technology tools, a Mann-Whitney U test was performed. The Mann-Whitney U test was used instead of an independent samples t-test, because two assumptions of the independent samples t-test were failed. First, there was a genuine unusual value (outlier) in the dataset for one respondent on overall use of educational technology tools (OETT), and second, the scores for OETT were not normally distributed for digital natives (they were, however, normally distributed for digital immigrants), exhibiting positive skew on a normal Q-Q plot. The outlier referred to a male assistant professor born in 1980 who had an OETT score of 65 out of 112, which was the highest score after a score of 57 for a female associate professor born in 1981.

Distribution of the OETT scores for digital natives and digital immigrants were similar, as assessed by visual inspection of a population pyramid. The median OETT score

(28)

28 was not statistically significantly different between digital natives (mean rank = 31.26) and digital immigrants (mean rank = 33.41), U = 466, z = -.456, p = .649. This non-significant finding combined with a calculated effect size of r = -.057, indicates that the effect is non-existent. The conclusion thus was that there are no statistically significant differences in use of educational technology tools between digital natives and digital immigrants. The null hypothesis could thus not be rejected.

5.6. Hypothesis 6a and 6b

To assess the relationship between digital nativity and confidence and comfort in using computers for educational purposes, a linear regression was carried out. All assumptions for carrying out a regression analysis were met. Linearity was established based on visual inspection of a scatterplot with superimposed regression line. There were no outliers and homoscedasticity was established based on visual inspection of a plot of standardized residuals versus standardized predicted values. Normal distribution of residuals was

established based on visual inspection of a histogram with superimposed normal curve and a normal probability plot. There seemed to be some negative skewness based on inspection of the histogram, however with a sample size of >15, this research is not be affected by slight non-normality.

Results showed that digital nativity accounted for 8.8% of the variation in comfort and confidence in using computers for educational purposes with adjusted R2_{= 7.3%.}

Furthermore, digital nativity significantly predicted confidence and comfort, F(1, 62) = 6.00,

p = .017. One higher level of digital nativity leads to a 0.12 (95% CI, .05 to .50) increase in

confidence and comfort. An effect size of r = .12 indicates a small effect. Hypothesis 6a is thus partially supported.

Hypothesis 6b, which assesses the moderating effect of age on this relationship, was assessed using PROCESS v2.16.3. First the independent variable, digital nativity, and the moderator, age, were standardized. Then the standardized variable for digital nativity was added as the predictor and confidence and comfort as the outcome variable. The standardized variable for age was entered as the moderator. From the non-significant interaction effect (c3 = 1.18, p = .283) it appeared that no moderation was taking place. Thus, the effect of digital nativity on confidence and comfort in using computers for educational purposes did not depend on a teacher’s age. Hypothesis 6b was thus not supported. The results of the regression are reported in table 7.

(29)

29 Table 7. Summary of simple linear regression analysis of digital nativity predicting confidence and comfort in using computers for educational purposes

Intercept 24.147 4.998 .000 14.157 34.137

DN .119 .049 .297 .017 .022 .216

Note: N = 64, dependent variable = confidence and comfort in using

computers for educational purpose, DN = digital nativity,

B = unstandardized regression coefficient, SE = Standard Error,

β = standardized coefficient, p = significance, confidence interval at 95%

To determine whether there was a significant positive relationship between confidence and comfort in using computers for educational purposes and the use of educational

technology tools, after controlling for teaching experience, a multiple linear regression was run.

Before the regression was run, the corresponding assumptions were tested. Linearity between the dependent and independent variables collectively was established upon

inspection of the studentized residuals against unstandardized predicted values plot. Linearity between the continuous independent variable (confidence and comfort) and the dependent variable (use of educational technology tools) was also established based on visual inspection of a partial regression plot. Homoscedasticity was assessed by visual inspection of the

studentized versus unstandardized predicted values plot. There was no multicollinearity as assessed by the values of the correlations of the independent variables which were all below .70. There were no outliers in the data, as confirmed by inspection of the studentized deleted residuals that showed no residuals greater than ±3 standard deviations. Furthermore, the data did not contain any cases with high leverage as assessed by inspection of the leverage values. There were no Cook’s distance values above 1, therefore none of the cases have especially high influence on the analysis. The residuals were relatively normally distributed as assessed by the histogram with superimposed normal curve, which showed only slight positive skew, however the P-P plot showed that all data points were distributed more or less along the diagonal line, which proves normality.

The results showed that R2 for the overall model was 15.7% with an adjusted R2 of 12.9%. Confidence and comfort in using computers for educational purposes and teaching experience combined, statistically significantly predicted the use of educational technology tools, F(2,61) = 5.67, p = .006. When teaching experience was held constant, confidence and comfort in using computers for educational purposes statistically significantly predicted use of

(30)

30 educational technology tools. The regression equation was: use of educational technology tools = 4.82 + (.74*confidence and comfort), p = .001. An effect size of r = .74 indicates a large effect and together with the statistical significance of the relationship between comfort and confidence in using computers for educational purposes and use of educational

technology tools this indicates there is strong evidence that hypothesis 7 is supported. Teaching experience alone, when confidence and comfort were held constant, did not statistically significantly predict use of educational technology tools (use of educational technology tools = 4.82 + (.17*years of teaching experience, p = .863). An effect size of r = .17 indicates a small effect size. It is thus not likely that a significant relationship between teaching experience and use of educational technology exists in reality. The results of the regression are reported in table 8.

Table 8. Summary of multiple regression analysis of confidence and comfort in using computers for educational purposes predicting the use of educational technology tools, after controlling for teaching experience

Intercept 4.821 8.285 .582 -11.746 21.388

CC .736 .219 .395 .001 .299 1.174

TE .163 .938 .020 .863 -1.712 2.038

Note: N = 64, dependent variable = use of educational technology tools,

CC = confidence and comfort in using computers for educational purposes, TE = teaching experience, B = unstandardized regression coefficient, SE = Standard Error, β = standardized coefficient, p = significance; confidence interval at 95%

5.8. Conceptual model with effect sizes

The conceptual model with the effect sizes of the corresponding hypotheses are depicted in figure 3 on the next page.

(31)

31 Figure 3. Conceptual model with effect sizes

Note: Significant at the 0.05 level (2-tailed), **Significant at the 0.01 level (2-tailed), NS_{= not}

significant

6. Discussion 6.1. General discussion

The first hypothesis, that stated that there is no significant difference in digital nativity between male and female university teachers, was supported. This is in line with previous research that did not find gender differences either, but in one of these studies, the dependent variable was not digital nativity, but digital competence among ninth-grade Chinese middle school students (Li and Ranieri, 2010) with an average age of 15 years. Because gender differences in digital nativity have only been assessed once before, using the Digital Natives Assessment Scale (DNAS) (Teo et al., 2016), where the focus was on pre-service teachers enrolled in a Turkish university with an average age of 20.53 years, this finding is rather unique in its context of in-services university teachers from a Dutch university with an average age of 45 years. The results of this research thus indicate that gender differences in digital nativity are non-existent in a sample of Dutch adult university teachers, and that this finding is in line with previous research that found no gender differences in digital nativity or digital competence among school-aged children and adolescent pre-service teachers. It also shows that cultural educational context, whether Dutch, Chinese or Turkish, apparently has no influence on gender differences in digital nativity or digital competence.

The second hypothesis, that stated there is a significant relationship between age and digital nativity, was also supported, albeit the results showed a relatively weak relationship. However, some criticism has been voiced in previous research that looking at age as predictor

Age Confidence &

Comfort Digital Nativity Teaching experience Use of technology H1 (d = .75NS_{) H6a (r = .12*)} H7 (r = .74**) Gender H6bNS H2 (r = -.41*) H5 (r = -.06NS₎ Birth group H4 (r = -.08NS₎ H3 (d = .65**)

(32)

32 of digital nativity is too simplistic (Jones et al., 2010), and this is still true, as proven by the weak relationship between age and digital nativity that has been demonstrated in this research. However, to regard age as not contributing to this relationship, would be incorrect.

Interestingly, Teo (2013) and Teo et al. (2016) who piloted and tested the DNAS found no differences in digital nativity based on age. This may be contributed to the fact that both studies used samples were the mean age of the respondents corresponded with the

generational cohort of digital natives, born in or after 1980. However, the mean age in this research was 45 years, and a slight minority (42%) of the respondents belonged to the generational cohort of digital natives, whereas 58% of the respondents belonged to the generational cohort typically referred to as digital immigrants. This research thus provides a more nuanced picture of the influence of age on digital nativity.

The third hypothesis, which stated that there are significant differences in digital nativity between digital natives and digital immigrants was also supported. Differences in digital nativity based on birth group have never been assessed before, therefore this finding is unique. The age related findings of this research thus show that age significantly predicts digital nativity, where an increase in one year of age corresponds with a decrease of -.41 in digital nativity level, and that digital natives score significantly higher on the DNAS than digital immigrants.

However, the fourth and fifth hypotheses, which stated there is a significant

relationship between age and age group (digital natives and digital immigrants) and use of educational technology tools, were not supported. This counters previous research where significant relationships were found between both age and age groups and the use of digital technologies (Jones et al., 2010; Salaway et al., 2008). This previous research was carried out among English first-year university students and American undergraduates, whereas the respondents of this research were university teachers from a Dutch university. The mean age differences between these samples are thus quite large. However, it is surprising that

significant relationships were found among younger respondents in previous research and not among the older respondents in this research. The weak effect sizes that resulted from testing the fourth and fifth hypotheses moreover indicate that the relationships also are not likely to exist in reality, regardless of the relatively small number of respondents used in this study to test these hypotheses. The difference in these findings could result from the fact that in 2008 and 2010 using ICT in education was still very much rising, whereas now, the use of

technology tools in education, which also includes informal communication tools such as Skype, WhatsApp and Slack, has spread out much more. Use of educational technology tools

(33)

33 has normalized, also among older generations, even though it has not reached its full potential yet. However, this could be the reason why age and age group differences in using

educational technologies were not found. However, to confirm this, research using a larger and more representative sample of university teachers should be used.

Hypothesis 6a, which stated that there is a significant positive relationship between digital nativity and comfort and confidence in using computers for educational purposes was supported, whereas hypothesis 6b, which stated that this relationship was moderated by age, was not. The effect of digital nativity on confidence and comfort has not been studied before, therefore, this contribution to the research surrounding digital natives, education and use of technology is unique. What is more interesting, is that how digitally native a teacher is, at least in some part predicts his or her confidence and comfort in using computers for

educational purposes and this is especially interesting given that hypothesis 7, which states that there is a significant positive relationship between confidence and comfort and use of educational technology tools after controlling for teaching experience, was also supported.

Hypothesis 6b was not supported, which is interesting given the fact that age is does have a significant effect on digital nativity, but that it does not moderate the relationship between digital nativity and confidence and comfort in such a way that the relationship between digital nativity and confidence and comfort is stronger for younger teachers (i.e. teachers born in the generational cohort of the digital native). The result that age did not seem to moderate this relationship, could however, be biased due to the fact that younger teachers were slightly underrepresented in the sample (27 respondents were born in or after 1980 and 37 respondents were born before 1980). For a more nuanced view, the ages in the sample should be more equally distributed.

Finally, hypothesis 7, which stated that there is a significant positive relationship between confidence and comfort and use of educational technology tools after controlling for teaching experience, was supported. This is in line with previous research that indicates that how confident and comfortable a teacher is in using computers for education, is an important factor influencing technology adoption and integration in the classroom. The effect size found in this research was also quite large, and based on previous research there is thus strong evidence that confidence and comfort in using computers for educational purposes is indeed an important factor influencing technology uptake among teachers. It is interesting to explore this relationship further, especially due to the fact that although the average level of

confidence and comfort in using computers for educational purposes measured in this study was relatively high, the self-reported use of educational technology tools was still quite low.

Digital natives or digital immigrants? : about university teachers’ use of educational technologies