Introduction

\label{introduction}
Among the many challenges and commitments associated with the European Higher Education Area, of particular note is a methodological focus that seeks to transform education systems solely based on ’teaching’ to systems based on ’teaching and learning’. In other words, the impulse toward a high-quality international Higher Education system should be grounded in an educational approach in which teaching and learning methods aim to transform the student into an active agent in their learning process, as opposed to merely being a receptacle. This transformation process should be interactive and has three basic principles at its core: (1) greater involvement and aut\ref{references}onomy on the part of the student; (2) the use of more active methodologies such as practical case studies, problem-based learning, tutoring, seminars, and so on; and (3) modernization of the role of teaching staff, who need to act as agents capable of creating learning environments that stimulate students (University Coordination Council, 2005).
Given these basic principles, the use of Information and Communications Technologies (ICTs) is a fundamental element in the successful application of university curricula. More specifically, universities are increasingly implementing or considering the use of, collaborative online learning tools based on Web 2.0, such as forums, wikis and google applications (apps), to support project work (Cheung & Vogel, 2013) and students are configuring their individual PLEs (Personnal Learning Environments) (Dabbagh & Kitsantas, 2012; Marín & de Benito, 2011; Yen, Tu, Sujo-Montes, & Sealander, 2016). Such ICTs enable learners to develop their work beyond the physical classroom and offer a new means of information-exchange between teachers and students. Furthermore, they combine the major advantages of online learning (virtual classrooms, discussion forums and helpful links) with the scope to access a ’traditional’ lecturer–tutor who can supervise all of the student’s activities and tasks. In short, ICTs lend themselves to a valid teaching and learning model.
Las plataformas convencionales de e-learning basadas en ICTs se enfocan y estructuran en cubrir las necesidades institucionales o del profesorado; whereas PLEs (Personal Learning Enviroments) se basan en la integración de diversas prácticas y fuentes de información para resolver necesidades de aprendizaje individuales (Frias & Barrio, 2013). PLEs represents a more flexible approach to use digital technology in education because it focuses on the students’ personal needs instead of institutional course designs or instructor needs, as generally occurs in Learning Management Systems (LMS), Moodle, etc. En concreto, los PLE no están establecidos en una única plataforma sino en una serie de herramientas que proporcionan funciones diferentes según las preferencias del usuario.
El uso de PLE basados en herramientas 2.0 suponen un uso de las herramientas sociales que la mayoría de estudiantes ya utilizan actualmente (Barrio-García, Arquero, & Romero-Frías, 2015). Many studies indican un impacto positivo de los servicios 2.0 en educación para desarrollar habilidades clave como: el análisis crítico, selección de información relevante, trabajar colaborativamente o compartir conocimiento (Ajjan & Hartshorne, 2008). Nevertheless, the complex scenario created requires the teachers” creativity and flexibility to incorporate these novelties into formal setting (Barrio-García et al., 2015)
Por tanto, para evaluar el éxito potencial de cualquier diseño educational basado en PLEs, se hace necesario entender las variables que afectan las actitudes y el nivel de aceptación de las nuevas tecnologías para el aprendizaje (Teo, 2010). Understanding the mechanisms by which students accept and use these technologies is therefore essential.
The acceptance and use of new technologies by individuals have been widely studied in the last two decades, particularly via the Technology Acceptance Model (TAM) developed by Davis (1986), which seeks to identify the antecedent variables of ICT use intention. However, in the context of university teaching and learning, there are other variables of major importance in student behavior – beyond those captured in the TAM – such as the subjective norms defined by the lecturer in their particular working environment or the social image those students using these tools project.
Despite the broad corpus of literature on ICT use intention, based on the TAM model, there are certain contexts in which more extensive knowledge is required – such as in the case of Higher Education – where it has been identified by the literature that there are other variables of major importance that have not, to date, been considered in relation to the TAM. (see external variables in ANNEX 1). Higher Education is undergoing a modernization process in which PLEs are taking on an increasingly prominent role (Castañeda, Dabbagh, & Torres-Kompen, 2017). In this scenario, actions to encourage PLE use among students are essential to the successful implementation of university curricula.
According to Cheung & Vogel (2013), Web-based platforms for supporting collaborative learning should integrate applications based on Web 2.0 features; these features promote interaction and collaborative learning among participants. Students not only learn by participation, but they also see how other students work in the Web 2.0 environment. However, although many educational institutions already use PLE in their teaching, there are few studies to date that have identified the factors explaining student acceptance of specific PLE-type environments tools such as Google Apps (Dabbagh & Kitsantas, 2012).
Focusing on the acceptance of Google apps as a PLE-type environments tools, the present study examines the intention to use Web-based platforms for supporting collaborative learning that integrate applications based on Web 2.0 features, such as the five components of Google apps: Google Docs, Google Forms, Google Sites, Google Group Forums and Google Drive for sharing. The aim of the present work is, therefore, to better understand how Web-based platforms (google application as a PLE-type environments tools) for supporting collaborative learning use intention is formed, in the context of university student learning and other aspects such as innovation adoption. The work considers antecedent variables from the TAM, namely perceived usefulness and perceived ease of use (perceived usefulness and perceived ease of use, respectively), subjective norms (SN) and social image (IM).

Theoretical background

\label{theoretical-background}
Our theoretical background is focused on: 1) Personal Learning Environments based on ICT´s and Web 2.0 technologies, 2) Application of the Technology Acceptance Model (TAM) and education from the point of view of the social effect of using collaborative online learning tools, 3) our extension on the model of acceptance.

PLE-type enviroments based on ICT´s and Web 2.0 technologies

\label{ple-type-enviroments-based-on-icts-and-web-2.0-technologies}
Los PLEs son plataformas de aprendizaje efectivas basadas en Web 2.0 cada vez más usadas entre los estudiantes. Los PLEs Permiten el desarrollo de espacios y experiencias para el aprendizaje social y personal, mediante el desarrollo de habilidades para el self-learning. Siendo especialmente útiles al estar construidas desde abajo hacia arriba por estudiantes, comenzando con el establecimiento de metas personales, el manejo de información, la construcción de conocimiento individual y social, y finalmente hacia el aprendizaje en red (Dabbagh & Fake, 2017).
The New Media Consortium’s 2012 K-12 Horizon Report describes PLEs as a process or pedagogical approach that is individualized by design, centred around each user’s goals, and customized using a personalized collection of distributed and portable tools and resources to support formal and informal learning as well as one’s ongoing social and professional activities. Although not wedded to a particular technology, PLEs are primarily facilitated by cloud-based Web 2.0 technologies (ICTs) and services designed to help students create, organize, and share content, participate in collective knowledge generation, and manage their own meaning-making (Dabbagh & Kitsantas, 2012).
Los PLE-type environments apoyan el aprendizaje adaptativo y el aprendizaje personalizado, enabling learners to define, develop and configure learning spaces and experiences for themselves and for the audience they choose, using components such as a personal profiler, a content aggregator, a recommender, a progress tracker, and the ability to identify learning goals and link to social networks around shared goals (Castañeda et al., 2017).
En este sentido algunos estudios apoyan la consideración del entorno proporcionado por google mediante sus google apps como un PLE-type enviroments muy adecuado para su uso por alumnos (Dabbagh & Kitsantas, 2012; Marín & de Benito, 2011).
On the basis that PLE-type environments have demonstrated their suitability in supporting the learning process, it is important to acknowledge that the degree to which they are implemented successfully in the Higher Education context will depend on the student’s intention to use them. It is therefore vital to study the antecedent variables that may contribute to achieving greater PLE-type environments use intention among students. The present study focuses in particular on the intention to use Google apps, by the benefits highlighted by Cheung & Vogel (2013).
According to Cheung & Vogel (2013), Web-based platforms for supporting collaborative learning should integrate applications based on Web 2.0 features, such as the five components of Google apps: Google Docs, Google Forms, Google Sites, Google Group Forums and Google Drive for sharing. These features promote interaction and collaborative learning among participants. Students not only learn by participation, but they also see how other students work in the Web 2.0 environment. This environment promotes a ’meaningful discourse’ to support the learning activities, where knowledge is constructed through collaborative facilities provided by Google apps.
By using Web 2.0 collaborative tools, the students become producers and consumers of information. Different members of the group can then assess the content they generate, their evaluations being registered on the Google Site and monitored by the lecturer. Hence the effective use of group interactions based on a collaborative platform offers major advantages for group learning. In turn, the study of student attitudes toward the collaborative platform is of particular relevance (Cheung & Vogel, 2013; Soller, 2001).

Application of the TAM and the social effect of using collaborative online learning tools

\label{application-of-the-tam-and-the-social-effect-of-using-collaborative-online-learning-tools}
The TAM, developed by Davis (1989), is one of the models most widely drawn-upon to explain individuals’ acceptance of new technologies. Since it was first published, the TAM, together with subsequent models based on it, has enjoyed significant attention and strong empirical support (Venkatesh & Bala, 2008). TAMs derive from the theory of reasoned action developed by Ajzen and Fishbein (1980), according to which beliefs are influenced by attitudes, which lead to intentions and certain types of behavior. The theory of reasoned action is a general theory that seeks to explain and predict virtually all types of human behavior and is premised on the importance of individual beliefs. In the context of technology acceptance, this theory was used in an attempt to establish the factors that condition users toward innovation, behavioral intention, and intensity of use of a given system (Ajzen & Fishbein, 1980; Fishbein & Ajzen, 1975).
The literature review demonstrates that TAMs comprises a set of key variables, namely: Perceived usefulness (PU), Perceived Ease of Use (PEOU) and Use Intention (UI), together with variables that extend their predictive power. The original TAM endeavors to explain at least 40% of the variance in individual use intention for new technologies. The subsequent extensions to the original are denominated TAM2 (Venkatesh & Davis, 2000) and TAM3 (Venkatesh & Bala, 2008). The TAM holds that perceived usefulness, and perceived ease of use determine an individual´s attitude toward their intention to use an innovation, with the intention serving as a mediator to the actual use of the system (Mohammadi, 2015). Perceived usefulness is also considered to be affected directly by perceived ease of use (Mohammadi, 2014).
Various studies have demonstrated the validity of this model across a wide range of new technologies (e.g. Moon & Kim, 2001), and have affirmed that the model has acceptable predictive validity to measure ICT use. Examples of studies on the use of different ICTs include: email acceptance (Gefen & Straub, 1997; Karahanna & Straub, 1999; Karahanna & Limayem, 2000; Huang, Lu, & Wong, 2003); the Internet (Agarwal & Prasad, 1998; Agarwal & Karahanna, 2000; Sánchez-Franco & Roldán, 2005); search engines (Morris & Dillon, 1997); websites (Lin & Lu, 2002; Van der Heijden, 2003); and online sales (Chen, Gillenson, & Sherrell, 2002; O’Cass & Fenech, 2003). The model’s validity has also been demonstrated for predicting online purchase intention (Van der Heijden, Verhagen, & Creemers, 2003). The TAM has also been widely used in the e-learning sphere, specifically to evaluate students’ use intention for different technologies. Annex 1 provides an overview of relevant studies spanning 2003–2015, all of which examine technology acceptance in education, based on the Internet. Y más concretamente in the personal learning environments (Barrio, 2015).
The TAM has been used in different situations, contexts and cultures, for example, within the education sector. From the perspective of teaching staff, the TAM has also been used: by an educational portal (Pynoo et al., 2012); to examine teacher behavior when adopting mobile phone messages as a parent–teacher communication medium (Ho, Hung, & Chen, 2013); to determine use intention for a learning management system among secondary school teachers (De Smet, Bourgonjon, De Wever, Schellens, & Valcke, 2012); in Higher Education (Schoonenboom, 2012, 2014); and to analyze the usage of the geographic information system among geography teachers (Lay, Chi, Hsieh, & Chen, 2013). The TAM has also been used to assess podcasting acceptance on campus (Lin, Zimmer, & Lee, 2013)(see Annex 1).
From the student perspective, the TAM has also been empirically applied in the following works: among business administration students (Escobar-Rodriguez & Monge-Lozano, 2012); investigate e-learning systems such as Moodle in the university context (Islam, 2013); study user acceptance of YouTube for procedural learning (Lee & Lehto, 2013); establish attitudes toward learning in augmented reality environments (Wojciechowski & Cellary, 2013); and measure user acceptance of collaborative technologies (such as Google apps) (Cheung & Vogel, 2013), the role of integrating user interface and personal innovativeness into the TAM for mobile learning (Joo et al., 2014); integration of the TAM and the Information Systems success model (Mohammadi, 2015); and intention to use of a PLE based on Web 2.0 tools (Barrio-García et al., 2015)(see Annex 1).
The TAM has therefore proved itself to be useful in explaining the acceptance of e-learning technologies – a fact reflected in the continued academic interest surrounding it. The present study, therefore, draws on the key variables that explain technology acceptance in e-learning environments (these being perceived usefulness, perceived ease of use and use intention). However, it is important to bear in mind that the findings of some studies demonstrate that the relationships established via the TAM are sometimes inconsistent. This is the case with the ‘perceived ease of use’ variable, which presents a direct but inconsistent impact in the acceptance phase that may become non-significant in subsequent decisions regarding use (Davis, 1989; Karahanna & Straub, 1999). This inconsistency may be due to the TAM being applied in different scenarios in which other variables may be influencing the relationships between the variables of perceived usefulness, perceived ease of use and use intention. This reaffirms the need to capture more fully the variables that ultimately influence use intention via perceived usefulness and perceived ease of use in the sphere of application of the present work.
Some studies have used both the original and the extended version of the TAM to explore the usefulness of the model in new scenarios, such as that of teaching and learning, examining students’ acceptance of virtual learning environments (see Annex 1). Within the learning context, the literature highlights that the variables of the processes of social influence, such as Subjective Norms and Social Image, contribute to largely explaining use intention via the variables of perceived usefulness and perceived ease of use. These variables and their relationships have previously been justified in the context of technology use in the education field by Toots and Idnurm (2001), quoted in Kuskaya-Mumcu & Kocak Usluel, 2010); and these variables are recognized by the literature as being among the most influential on the intention to use a technology (Venkatesh & Davis, 2000; Venkatesh & Bala, 2008).
According to Venkatesh and Bala (2008), subjective norms have a direct influence on perceived usefulness. This influence is produced via the processes of identification and internalization. In identification, the individual believes that using the technology in question will improve their social status within the group of reference; while in internalization they incorporate the beliefs of the social reference group (regarding the technology) into their own belief system (Venkatesh & Davis, 2000).
Van Raaij and Schepers (2008) studied the acceptance of a virtual learning environment in China, using the extended model TAM2 (Venkatesh & Davis, 2000). Their results suggested that perceived usefulness has a direct effect on use intention in virtual learning environments and that perceived ease of use and subjective norms have an indirect effect on use intention, via perceived usefulness. These same results were found in the work of Rejón-Guardia, Sánchez-Fernández, and Muñoz-Leiva (2013), who included this variable in their assessment of use intention for micro-blogging social networks for teaching and learning purposes.
::TAM and the further development of the model have been criticised as being technology-oriented instead of user-oriented and some author’s don’t see a value of using and adapting TAM while the underlying framework (e.g. see Kreijns, K., Vermeulen, M., Kirschner, P. A., Buuren, H. V., & Acker, F. V. (2013).
::Adopting the Integrative Model of Behaviour Prediction to explain teachers’ willingness to use ICT: a perspective for research on teachers’ ICT usage in pedagogical practices. Technology, Pedagogy and Education, 22(1), 55-71).

Social Image and Subjective norms as extension of TAM

\label{social-image-and-subjective-norms-as-extension-of-tam}

Hypotheses and proposed research model

\label{hypotheses-and-proposed-research-model}
To understand the adoption of collaborative technologies, the TAM is used to describe individual users’ acceptance of information systems (Lee, Kozar & Larsen, 2003). Following on from, and developing, this model, the two key variables that have been found to both determine use intention and also predict the acceptance of an innovation are perceived usefulness and perceived ease of use (e.g. Castañeda, Muñoz-Leiva & Luque, 2007; Davis, 1989; Davis & Wiedenbeck, 2001; Gefen, Karahanna & Straub, 2003; Muñoz, 2008; Sánchez-Franco & Roldán, 2005). Hence the models developed out of the TAM suggest that both the acceptance and use of a technology are determined by users’ beliefs deriving from its perceived usefulness and perceived ease of use.
Perceived usefulness was originally defined by Davis (1989, p. 985) in the context of the workplace as “the degree to which a person believes that using a particular system would enhance his or her job performance”. More specifically, the worker will expect to achieve a positive performance when using a system that presents a high level of perceived usefulness. As regards the analytical context of the present work, perceived usefulness may be defined as the extent to which the user (student) of the micro-blogging social network considers the information they derive from it to offer a series of benefits that would otherwise be difficult to obtain. Perceived usefulness is the only variable to have repeatedly demonstrated itself to be adequate for determining the development of affect and use intention in technological contexts (Davis, 1989; Karahanna & Straub, 1999).
Meanwhile perceived ease of use is “the degree to which a person believes that using a particular system would be free of effort” (Davis, 1989). Taking this definition, the present study considers perceived ease of use to be the extent to which the student believes using micro-blogging social networks to be effortless. Selim (2003) researched the use and acceptance of course websites, based on three variables: perceived usefulness of the courses; perceived ease of use of the website; and usage. His results showed that there is a significant relationship between perceived usefulness and ease of use when determining the usage of a given website.
The relationship between perceived ease of use and perceived usefulness has also been analyzed, as have their effects on user behavior – with the literature on new technologies providing extensive support to these effects (Venkatesh & Bala, 2008). The results show that perceived ease of use has a direct effect on perceived usefulness and a positive effect on technology use intention, both of these relationships being directly and indirectly moderated by perceived usefulness (Davis, 1989). It is thought that perceived usefulness is one of the most important factors to influence the acceptance of a website (Moon & Kim, 2001; Chen et al., 2002; Featherman & Pavlou, 2003; Sánchez-Franco & Roldán, 2005; Bhattacherjee & Premkumar, 2004; Castañeda et al., 2007; Venkatesh & Bala, 2008).
In view of this scenario the following hypotheses are proposed:
Hypothesis 1: Perceived ease of use has a direct and positive effect on perceived usefulness.
Hypothesis 2: Perceived ease of use has a direct and positive effect on use intention.
Hypothesis 3: Perceived usefulness has a direct and positive influence on use intention.
Another new sphere of empirical application for the TAM is that of the use of Google apps within the Spanish Higher Education context. In response to this need, the present work analyzes two variables that may play an important role, namely the subjective norms that are defined in the working environment of the teaching staff and the social image projected by the individuals using these Google apps.
Subjective norms may be defined as the degree to which an individual perceives that people who are important to them think the individual should use an information technology system, or not (Fishbein & Ajzen, 1975; Venkatesh & Davis, 2000; Venkatesh & Bala, 2008). Such norms are therefore associated with social pressure that influences behaviors (in the present study, the behavior being to use an information technology system). Opinions regarding the adoption of technology may differ depending on the social group. Hence the teaching staff may regard adoption as improving the student’s learning, while the latter may be receptive to its adoption thanks to being a digital native who is quite comfortable with new technologies (Ajjan & Hartshorne, 2008). The present study considers subjective norms to directly influence perceived usefulness, as in the use of Google apps in teaching the teacher makes an important judgment which prescribes – and thus generates a perception of the importance of – using these tools to foster learning and support student success. What is more, many students expect the teacher to support them in integrating ICTs within traditional classroom learning to improve the learning process and render it more effective (Salaway & Caruso, 2008) (quoted in Chen et al., 2010).
Meanwhile, Van Raaij and Schepers (2008) and Rejón-Guardia, Sánchez-Fernández, and Muñoz-Leiva (2013) found empirical evidence suggesting that perceived ease of use and subjective norms have an indirect effect on use intention, via perceived usefulness.
It is therefore proposed that:
Hypothesis 4: Subjective norms have a direct and positive influence on perceived usefulness.
Hypothesis 5: Subjective norms have a direct and positive influence on perceived ease of use.
Turning to social image, Moore and Benbasat (1991) define this as “the extent to which an individual perceives that the use of an innovation will improve their status in a social system”. Roger (1983, p. 215) (quoted in Moore & Benbasat, 1991, p. 195) posits that “a person’s principal motivation for adopting an innovation is their desire to achieve social status”. Within this context, the relevance of social image as a differentiated concept has been debated (Holloway, 1977; Tornatzky & Klein, 1982) (quoted in Moore & Benbasat 1991, p. 195). For this reason, the present study also considers social image as an antecedent variable of perceived usefulness (Venkatesh & Bala, 2008). Within the Higher University teaching context of the present work, social image represents the degree to which the student perceives that using Google apps will improve their status among their peers. The following hypotheses are therefore proposed:
Hypothesis 6: Social image has a direct and positive influence on perceived usefulness.
Hypothesis 7: Social image has a direct and positive influence on perceived ease of use.
Venkatesh and Bala (2008) further propose that subjective norms have a positive influence on social image – a relationship that is also addressed in the present study, the proposed hypothesis being that:
Hypothesis 8: Subjective norms have a direct and positive influence on social image.
Figure 1 shows the proposed research model for acceptance and use of Google apps.
Figure 1. Proposed conceptual model

Method

\label{method}

Sample

\label{sample}
An empirical study was undertaken among university students who had been invited to use Google apps for one of their course assignments. A pilot study was first conducted among 30 individuals to test the validity of the items. More specifically, the pilot evaluated whether the items were clearly understood and that each item positively correlated to the set of items belonging to the same scale. The scales chosen for the questionnaire have previously been used successfully in the literature (Venkatesh & Davis, 2000; Moore & Benbasat, 2001, Venkatesh & Bala, 2008; Escobar-Rodriguez & Monge-Lozano, 2012; Pynoo et al., 2012; Cheung & Vogel, 2013; Rejón-Guardia et al., 2013; Schoonenboom, 2014). The fieldwork was carried out from 8–23 January 2015, to coincide with the end of the assignment in question. Prior to data collection, ethics approval was sought, and Obtained, using the procedures laid down by the ethics committee at the University of Granada (Spain) [(http://investigacion.ugr.es/pages/etica]). Sampling was undertaken based on all of the students on a distribution management program. Some 309 questionnaires were obtained, and once those that were incomplete had been eliminated the final sample comprised 267 individuals. This represents a response rate of 86.4% of the students.
The questionnaire included items of a demographic nature and the aforementioned variables. From the total sample, 41.2% were males (n = 110) and 58.8% were females (n = 157) with a mean age of (M = 22.40, SD = 2.42) (Table 1). Based on the number of responses received and for a confidence interval of 95%, in the case of estimations at a proportion in which p=q=.5, and on the basis of simple random sampling, the sampling error was +-5.6%.

Measures

\label{measures}
The variables included in the research model (Figure 1) were adapted from earlier studies and based on the TAM (Davis, 1989) and subsequent models developed out of it, as reflected in TAM3 (Venkatesh & Bala, 2008). Perceived usefulness and perceived ease of use were both measured using the scale developed by Davis (1989). The variable ‘use intention’ (UI) was measured using the scale produced by Bhattacherjee (2001). The ’subjective norms’ (SN) variable was measured using Taylor and Todd’s (1995) scale. All of these scales have previously been applied in other works such as those of Venkatesh and Davis (2000), Venkatesh and Bala (2008), Escobar-Rodriguez and Monge-Lozano (2012), Pynoo et al. (2012), Cheung and Vogel (2013), Rejón-Guardia et al. (2013), and Schoonenboom (2014). Finally, social image (IM) was measured using a scale adapted from that of Moore and Benbasat (2001), which has been used in other empirical applications of extended TAMs (Venkatesh & Bala, 2008). In the present work, the final questionnaire covered 18 items measuring the five variables under study, using a 7-point Likert scale on which 1 equaled ”entirely disagree” and 7 equaled ”entirely agree” (see Appendix I).

Results

\label{results}
The proposed structural model was estimated using AMOS 18. Since the Chi-square test of multivariate normality of the variables included in the proposed model was significant, the maximum likelihood approach was considered appropriate for estimating the model, combined with the bootstrap method. A valid reference is the value of Normed Chi-square, which in the present case gave a value of 2.02 and was within the limits recommended by the literature (Kaiser, 1974; Nunnally, 1978; Hair, Anderson, Tatham & Black, 1999). As regards the overall fit of the model, the GFI gave a value of .96, above the value recommended in the literature, and the RMSEA value was acceptable, below the recommended limit. The incremental fit measurements CFI (.97) and NFI (.94) were also acceptable (see Table 2). In its totality, the fit of the model was considered to be acceptable (see Table 3).
The measurement scales reflect the composition of all the latent variables when their validity and reliability can be confirmed (Hair et al., 1999). To assess this, the individual reliability coefficient (R2) of each of the items had to be considered (Table 2). These indicators achieved values greater than, or very close to, the minimum acceptable level of .50 (Hair et al., 1999), except for item PU4, belonging to the variable Perceived of Usefulness; item PEOU1, belonging to the variable perceived ease of use and item SN4, belonging to the subjective norms latent variable. It was decided that these items should be retained, to avoid reducing the predictive capacity of the model.
The existence of a suitable level of internal consistency for each of the variables was verified by measuring composite reliability and variance extracted. In all cases the values were acceptable, being either greater than or very close to the reference values of .70 for composite reliability and .50 for variance extracted (Hair et al., 1999) (Table 1). The results indicate that the set of scales used in the study was adequate for measuring each of the latent variables in the proposed research model.
It was demonstrated that the variability of the values for perceived use and perceived ease of use were explained by 22% and 56%, respectively. It was possible to explain 74% of use intention, in line with the individual’s perception of the tool’s ease of use and usefulness, the latter having a much greater direct effect on use intention. Similarly, social image was particularly relevant as an external variable, explaining 43% of variance in perceived usefulness. Social image was explained by 75% of subjective norms – a result that is in line with other previous works such as those of Cellary (2013) and Cheng (2014) and significantly higher than the explanatory power of classical models such as TAM1.
Returning to the hypotheses under examination, Figure 2 shows the standardized coefficients that constitute the basis for the empirical support given to each of the proposed hypotheses. The following aspects are particularly notable:
H1 proposes that perceived ease of use has a direct and positive influence on perceived usefulness. The results show a statistically significant relationship (p<.05). Furthermore, the effect detected is quite marked (β=.69). Therefore, there is statistical support for this hypothesis. H2 proposes that perceived ease of use has a direct and positive influence on use intention (β=.15). The results show a statistically significant relationship (p<.10) therefore there is empirical support for this hypothesis. H3 proposes that perceived usefulness has a direct and positive influence on use intention. The results reveal a statistically significant relationship, with a significance level of p<.05 (β=.74). Furthermore, the effect is found to be notable, and therefore there is empirical support for this hypothesis.
Regarding the hypothesis pertaining to the extended model, H4 proposes that subjective norms have a direct and positive influence on perceived usefulness. The results show a statistically nonsignificant relationship (p=.64) and hence there is no statistical support for this hypothesis. H5 proposes that subjective norms have a direct and positive influence on perceived ease of use. The results indicate a statistically significant relationship (p<.10), with a standardized coefficient of β=.32. Therefore, there is empirical support for this hypothesis. H6 proposes that social image has a direct and positive influence on perceived usefulness. According to the results, there is a statistically significant relationship, with a significance level of p<.05 (β=.43). Furthermore, the effect is found to be notable, and therefore there is empirical support for this hypothesis. H7 proposes that social image has a direct and positive influence on perceived ease of use. The results reveal that there is empirical support for this hypothesis (p<.10). H8 proposes that subjective norms have a direct and positive influence on the social image. The results show a statistically significant relationship (p<.05) with a standardized coefficient of β=.75, therefore, there is statistical support for this hypothesis. In terms of relevant indirect effects, it is worth noting the indirect effect of subjective norms on use intention, via perceived ease of use (β=.33), and the indirect effect of the social image on use intention, via perceived usefulness (β=.36).
Figure 2. Results of the research model.

Conclusions and recommendations

\label{conclusions-and-recommendations}
The main contribution of the present study is its application of the TAM to explain use intention for new technologies (Google apps) in a new context, that of Higher Education. Although the variables involved in use intention for different ICTs have already been studied in the literature, via the TAM, the present work explains use intention in an original approach, centered on Google apps, and proposes an extended TAM that includes causal relationships between five variables. Three of these (perceived use, perceived ease of use, and use intention) have previously been analyzed to a significant degree in TAMs. The two remaining variables derive from the external variables proposed in successive extensions of the TAM. More specifically, the present study evaluates two additional cause–effect relationships (subjective norms and social image), proposing that these have an influence on the intention to use Google apps in the university context. This extension constitutes a relevant contribution to the TAM as, to date, no other works have evaluated the effect of variables of a more social nature on the intention to use socially visible collaborative working tools such as Google apps.
The university environment has differentiating characteristics that set it apart from other contexts in which TAM has been applied, such as an orientation toward methodologies that facilitate collaborative working beyond the classroom. For such methodologies to work effectively, it is vital to foster use intention (in this case, for Google apps) among the student population. TAMs offer an insight into the key variables behind use intention for ICTs.
The present analysis provides a robust, parsimonious model in which the majority of relationships receive empirical support. TAMs are shown to be suitable for measuring acceptance of collaborative e-learning tools (Ngai, Poon, & Chan, 2007; Sanchez & Hueros, 2010). More specifically, it is demonstrated that both perceived usefulness and perceived ease of use have a positive and significant influence on use intention, and that perceived ease of use has a positive and significant influence on perceived usefulness. These results are in line with earlier findings from the literature (e.g. Selim, 2003; Saadé & Bahli, 2005; Ngai et al., 2007; Teo, Lee, Chai, & Wong, 2009; Liu, Chen, Sun, Wible, & Kuo, 2010; and Sanchez-Franco, 2010).
The present study also proposes other external variables for the TAM that may influence use intention for Google apps, namely subjective norms and social image. The results show that subjective norms have a positive and significant influence on perceived ease of use, and this relationship contributes to the indirect effect of subjective norms on the intention to use Google apps. These results are in keeping with the previous literature that examines this variable in relation to use intention for ICTs (e.g. De Smet et al., 2012; Pynoo et al., 2012; Rejón-Guardia & Sánchez-Fernández, 2013; Ho et al., 2013; and Cheung & Vogel, 2013).
Meanwhile, the second of these two proposed external variables, social image, is an original proposal for the literature. The results show that social image has a positive and significant effect on perceived usefulness. This indicates that social image exerts an indirect effect on use intention for Google apps, mediated by the perceived usefulness variable. Furthermore, it is found that subjective norms have a positive influence on the social image variable.

Practical implications

\label{practical-implications}
The results of the present work have a series of practical implications for the Higher Education sector. First, understanding the variables associated with adoption of Google apps will be of interest to those universities whose teaching staff wish to encourage students to adopt these tools and new methodologies for undertaking their assignments.
In light of the present findings, teaching staff can be aware that when endeavoring to encourage the use of Google apps there are certain variables they can address with their students in advance, so as to generate greater use intention for the ICTs and therefore ensure greater success in the methodological application they are proposing. Two such variables are perceived ease of use and perceived usefulness, both with a direct influence on intention to use Google apps. Knowing this, the lecturer can stimulate adoption of these tools by planning ahead and taking steps to convey how they can be of value (ease of use and usefulness) to the students when carrying out coursework. It is also important that the lecturer understands the importance of emphasizing ease of use as this will have a positive influence on the perceived usefulness of Google apps.
Second, it is also identified in the present study that the subjective norms of the working environment (in this case, Higher Education) and the social image of students who use Google apps have an indirect influence on use intention (via the variables ’perceived usefulness’ and ’perceived ease of use’). The results show that teaching staff can encourage students to use Google apps by showing them how following the instructions on how to use the apps in the assignment will contribute to achieving a good result. They can also show examples to illustrate to the students the high level of visibility and acceptance, on the social level, that using Google apps can generate.

Limitations and future lines of research

\label{limitations-and-future-lines-of-research}
The results of the present work will be of interest to the specialist literature on acceptance models for new technologies and to the Higher Education sphere. However, the work also presents certain limitations that need to be borne in mind, but that may point to potential future lines of research. One of the main limitations of the study is that of sample size and origin (undergraduates of a single Spanish university), hence generalizations on the basis of the results should be treated with caution. Possible lines of research for the future could be to apply the proposed model in the context of students at different stages of their degree program, with different levels of experience in using ICTs, or to different geographical areas, to achieve more generalizable results.
Other variables could also be included in the model that may influence the extent to which students use Google apps for their assignments, such as the fun that users may experience during use, or their performance level in the assignment in question.

References

\label{references}
Ajjan, H., & Hartshorne, R. (2008). Investigating faculty decisions to adopt Web 2.0 technologies: Theory and empirical tests. The Internet and Higher Education , 11 (2), 71–80. https://doi.org/10.1016/j.iheduc.2008.05.002
Ajzen, I., & Fishbein, M. (1980). Understanding attitudes and predicting social behavior . Englewood Cliffs, N.J.: Prentice-Hall.
Barrio-García, S. del, Arquero, J. L., & Romero-Frías, E. (2015). Personal Learning Environments Acceptance Model: The Role of Need for Cognition, e-Learning Satisfaction and Students’ Perceptions. Journal of Educational Technology & Society , 18 (3), 129–141. Retrieved from http://www.jstor.org/stable/jeductechsoci.18.3.129
Castañeda, L., Dabbagh, N., & Torres-Kompen, R. (2017). Personal Learning Environments: Research-Based Practices, Frameworks and Challenges. Journal of New Approaches in Educational Research , 6 (1), 1–2. https://doi.org/10.7821/naer.2017.1.229
Cheung, R., & Vogel, D. (2013). Predicting user acceptance of collaborative technologies: An extension of the technology acceptance model for e-learning. Computers & Education , 63 , 160–175. https://doi.org/10.1016/j.compedu.2012.12.003
Dabbagh, N., & Fake, H. (2017). College Students’ Perceptions of Personal Learning Environments Through the Lens of Digital Tools, Processes and Spaces. Journal of New Approaches in Educational Research , 6 (1), 28–36. https://doi.org/10.7821/naer.2017.1.215
Dabbagh, N., & Kitsantas, A. (2012). Personal Learning Environments, social media, and self-regulated learning: A natural formula for connecting formal and informal learning. The Internet and Higher Education , 15 (1), 3–8. https://doi.org/10.1016/j.iheduc.2011.06.002
Davis, F. D. (1986). A technology acceptance model for empirically testing new end-user information systems: Theory and results . Massachusetts Institute of Technology. Retrieved from https://www.researchgate.net/profile/Fred_Davis2/publication/35465050_A_technology_acceptance_model_for_empirically_testing_new_end-user_information_systems__theory_and_results_/links/0c960519fbaddf3ba7000000.pdf
Davis, F. D. (1989). Perceived Usefulness, Perceived Ease of Use, and User Acceptance of Information Technology. MIS Quarterly , 13 (3), 319–340. https://doi.org/10.2307/249008
De Smet, C., Bourgonjon, J., De Wever, B., Schellens, T., & Valcke, M. (2012). Researching instructional use and the technology acceptation of learning management systems by secondary school teachers. Computers & Education , 58 (2), 688–696. https://doi.org/10.1016/j.compedu.2011.09.013
Escobar-Rodriguez, T., & Monge-Lozano, P. (2012). The acceptance of Moodle technology by business administration students. Computers & Education , 58 (4), 1085–1093. https://doi.org/10.1016/j.compedu.2011.11.012
Fishbein, M., & Ajzen, I. (1975). Belief, attitude, intention and behaviour: An introduction to theory and research . Addison-Wesley.
Ho, L.-H., Hung, C.-L., & Chen, H.-C. (2013). Using theoretical models to examine the acceptance behavior of mobile phone messaging to enhance parent–teacher interactions. Computers & Education , 61 , 105–114. https://doi.org/10.1016/j.compedu.2012.09.009
Islam, A. K. M. N. (2013). Investigating e-learning system usage outcomes in the university context. Computers & Education , 69 , 387–399. https://doi.org/10.1016/j.compedu.2013.07.037
Lay, J.-G., Chi, Y.-L., Hsieh, Y.-S., & Chen, Y.-W. (2013). What influences geography teachers’ usage of geographic information systems? A structural equation analysis. Computers & Education , 62 , 191–195. https://doi.org/10.1016/j.compedu.2012.10.014
Lee, D. Y., & Lehto, M. R. (2013). User acceptance of YouTube for procedural learning: An extension of the Technology Acceptance Model. Computers & Education , 61 , 193–208. https://doi.org/10.1016/j.compedu.2012.10.001
Lin, S., Zimmer, J. C., & Lee, V. (2013). Podcasting acceptance on campus: The differing perspectives of teachers and students. Computers & Education , 68 , 416–428. https://doi.org/10.1016/j.compedu.2013.06.003
Marín, V., & de Benito, B. (2011). A design of a postgraduate course on Google Apps based on an Institutional Personal Learning Environment (iPLE). Retrieved from http://journal.webscience.org/652/
Moon, J.-W., & Kim, Y.-G. (2001). Extending the TAM for a World-Wide-Web context. Information & Management , 38 (4), 217–230. https://doi.org/10.1016/S0378-7206(00)00061-6
Pynoo, B., Tondeur, J., van Braak, J., Duyck, W., Sijnave, B., & Duyck, P. (2012). Teachers’ acceptance and use of an educational portal. Computers & Education , 58 (4), 1308–1317. https://doi.org/10.1016/j.compedu.2011.12.026
Schoonenboom, J. (2012). The use of technology as one of the possible means of performing instructor tasks: Putting technology acceptance in context. Computers & Education , 59 (4), 1309–1316. https://doi.org/10.1016/j.compedu.2012.06.009
Schoonenboom, J. (2014). Using an adapted, task-level technology acceptance model to explain why instructors in higher education intend to use some learning management system tools more than others. Computers & Education , 71 , 247–256. https://doi.org/10.1016/j.compedu.2013.09.016
Soller, A. (2001). Supporting Social Interaction in an Intelligent Collaborative Learning System. International Journal of Artificial Intelligence in Education (IJAIED) , 12 , 40–62. Retrieved from http://telearn.archives-ouvertes.fr/hal-00197321
Teo, T. (2010). Development and validation of the E-learning Acceptance Measure (ElAM). https://doi.org/10.1016/j.iheduc.2010.02.001
Venkatesh, V., & Bala, H. (2008). Technology Acceptance Model 3 and a Research Agenda on Interventions. Decision Sciences , 39 (2), 273–315. https://doi.org/10.1111/j.1540-5915.2008.00192.x
Venkatesh, V., & Davis, F. D. (2000). A theoretical extension of the technology acceptance model: Four longitudinal field studies. Management Science , 46 (2), 186. Retrieved from http://proquest.umi.com/pqdweb?did=52438017&Fmt=7&clientId=54638&RQT=309&VName=PQD
Wojciechowski, R., & Cellary, W. (2013). Evaluation of learners’ attitude toward learning in ARIES augmented reality environments. Computers & Education , 68 , 570–585. https://doi.org/10.1016/j.compedu.2013.02.014
Yen, C.-J., Tu, C.-H., Sujo-Montes, L., & Sealander, K. (2016). A Predictor for PLE Management: Impacts of Self- Regulated Online Learning on Students’ Learning Skills. Journal of Educational Technology Development and Exchange (JETDE) , 9 (1). https://doi.org/10.18785/jetde.0901.03