Research Gap

Computers & Education 88 (2015) 202e214
Contents lists available at ScienceDirect
Computers & Education

journal homepage: www.elsevier.com/locate/compedu
Becoming more specic: Measuring and modeling teachers' perceived

usefulness of ICT in the context of teaching and learning
Ronny Scherer a, *, Fazilat Siddiq b, Timothy Teo c
a
Centre for Educational Measurement at the University of Oslo (CEMO), University of Oslo, Norway
b
Department of Teacher Education and School Research, Faculty of Educational Sciences, University of Oslo, Norway
c
Faculty of Education, University of Macau, China
a r t i c l e i n f o a b s t r a c t
Article history: Studies on teachers' acceptance and use of information and communication technology (ICT) have
Received 22 November 2014 revealed perceived usefulness to be a crucial determinant for integrating ICT in classrooms. In conse-
Received in revised form quence, the present study focuses on teachers' perceived usefulness of ICT for teaching and learning and
25 February 2015
is aimed at describing its structure and relations to self-efcacy, ICT use, and teachers' age. By means of
Accepted 29 May 2015
Available online 6 June 2015
Bayesian analysis, we specied conrmatory factor-analytic and structural equation models to a large-
scale data set of N 1190 Norwegian teachers. Our results supported the hypothesized four-factor
structure of teachers' perceived usefulness of ICT, signifying different facets of ICT-related teaching
Keywords:
Bayesian analysis goals in classrooms. Moreover, it was possible to disentangle general and specic components of the
Measurement construct in nested factor models. In support of existing research, we found positive relations to self-
Perceived usefulness of ICT efcacy and ICT use, but a negative relation to teachers' age. Our study provides evidence on a multi-
Structural equation modeling dimensional conceptualization of teachers' perceived usefulness of ICT for teaching and learning, and
Technology acceptance veries the relations to teacher-related characteristics. Implications for the measurement and modeling
of the construct, and future research directions are discussed.
2015 Elsevier Ltd. All rights reserved.
1. Introduction
The development of students' digital competences has become an important goal in 21st century education (Grifn, Care, & McGaw,
2012; OECD, 2014; Tristan-Lopez & Ylizaliturri-Salcedo, 2014). Recent educational research is therefore concerned with the determinants
of students' digital competences on the one hand (Aesaert, van Nijlen, Vanderlinde, & van Braak, 2014; Fraillon, Ainley, Schulz, Friedman, &
Gebhardt, 2014; Fraillon, Schulz, & Ainley, 2013; Kikis, Scheuermann, & Villalba, 2009); and teachers' integration of ICT in classrooms on the
other hand (Donnelly, McGarr, & O'Reilly, 2011; Herna ndez-Ramos, Martnez-Abad, GarcaPen ~ alvo, HerreraGarca, & Rodrguez-Conde,
2014; Sang, Valcke, van Braak, & Tondeur, 2010; Teo, 2011a). In particular, research has shown that there are a number of factors deter-
mining teachers' ICT integration such as teachers' attitudes towards ICT, ICT experience and skills, self-efcacy, and perceived usefulness of
ICT (Govender & Govender, 2009; Herna ndez-Ramos et al., 2014; So, Choi, Lim, & Xiong, 2012; Teo, 2011b). Of the factors mentioned,
teachers' perceived usefulness of ICT has received special attention in the context of technology acceptance and use: Perceived usefulness is
regarded as the only construct which directly predicts teachers' intentions to use ICT and their attitudes towards computer use, which are
determinants of the actual ICT use (Chien, Wu, & Hsu, 2014; Davis, 1989; Oye, Iahad, & Rahim, 2014; Smarkola, 2008; Teo, 2011a, b; Teo, Lee,
Chai, & Wong, 2009). Moreover, perceived usefulness is the strongest predictor of teachers' intention to use ICT (e.g., Teo, 2009).
The current conceptualizations of perceived usefulness of ICT operationalized the construct at a general level, referring to whether
teachers generally regard ICT as useful to increase their job performance (Venkatesh, Morris, Davis, & Davis, 2003). Hence, when studying
the nature of the construct, most studies assumed unidimensionality and found support by using conrmatory factor analysis (e.g., Chien
et al., 2014; Herna ndez-Ramos et al., 2014; Liaw & Huang, 2013; Teo, 2009). From a conceptual point of view, the unidimensional
* Corresponding author. Centre for Educational Measurement at the University of Oslo (CEMO), Faculty of Educational Sciences, University of Oslo, Postbox 1161 Blindern,
0318 Oslo, Norway.
E-mail address: ronny.scherer@cemo.uio.no (R. Scherer).
http://dx.doi.org/10.1016/j.compedu.2015.05.005
0360-1315/ 2015 Elsevier Ltd. All rights reserved.
R. Scherer et al. / Computers & Education 88 (2015) 202e214 203
conceptualization is critical, as it does not account for the specic goals of teachers' job performance facilitating teaching and students'
learning (Niederhauser & Perkmen, 2010). Since the specic purposes of using ICT for teaching and the specic goals to foster students'
learning are multifaceted (e.g., using ICT for assessment, collaboration, feedback, skill development; Proctor & Marks, 2013; Schroeders &
Wilhelm, 2011; Terzis & Economides, 2011), one may also conceptualize perceived usefulness as multidimensional. In other words, teachers'
perceptions on whether ICT could improve their job performance may not only refer to the use of ICT in general (unidimensionality) but
rather to the usefulness of ICT for specic teaching and learning purposes in classrooms (multidimensionality). In addition to this
conceptualization, one needs to take into account that teaching and learning goals are often overlapping, thus challenging the modeling of
the factor structure (construct overlaps). Consequently, there is a need for a modeling approach that represents the conceptualization of
teachers' perceived usefulness with respect to the specic goals of using ICT in classroom instruction.
In light of this argumentation, the present study proposes a multidimensional conceptualization of perceived usefulness of ICT for
teaching and learning and is aimed at (a) testing the hypothesis of a multidimensional factor structure of the construct and potential
overlaps between the factors; and (b) investigating the relations to external variables (teachers' ICT self-efcacy, ICT use, age) as a step of
construct validation. We use Bayesian analyses to address our aims. Since perceived usefulness is a key element of technology acceptance,
our study provides insights into the nature of the construct, impacting future modeling of technology acceptance.
2. Theoretical framework
2.1. Conceptualizing teachers' perceived usefulness of ICT for teaching and learning
In the context of technology acceptance, researchers are concerned with the factors determining teachers' attitudes towards and their
intention to use ICT (Davis, 1989; Teo, 2009), of which perceived usefulness is one of the most important determinants (e.g., Chien et al., 2014;
Niederhauser & Perkmen, 2010; Oye et al., 2014; Teo, 2011a, b; Teo & Noyes, 2011). Perceived usefulness is part of teachers' belief system and
refers to their individual beliefs that using ICT will help them improve their job performance (Davis, 1989). The construct can be distin-
guished from teachers' general attitudes towards the use of computers, as it does not rely on their positive feelings about using ICT in general
(Ifenthaler & Schweinbenz, 2013; Niederhauser & Perkmen, 2010; Teo & Noyes, 2011). In current research on technology acceptance,
perceived usefulness has been conceptualized broadly, focusing on the potential improvement of teachers' general job performance when
using ICT, without dening the specic aspects of job performance (e.g., Teo, 2011a). Nevertheless, since teachers' job performance in the
context of ICT is complex, particularly in fostering students' competences of accessing, retrieving, evaluating, and communicating digital
information, the question on the specicity of perceived usefulness for teaching and learning arises.
Against this background, conceptualizations of perceived usefulness that are linked to aspects of teaching performance may account for
the specic attainment goals in the curricula, which are students' digital competences. Digital competences are often conceptualized as
constructs that comprise different sub-dimensions or cognitive processes. For instance, Claro et al (2012) refer to ICT skills in information
sourcing, information retrieval, effective communication, collaboration and virtual interaction, evaluation of ICT use with respect to ethics
and social impact. Ferrari (2013) proposed a framework which extended these sub-dimensions by introducing problem solving and creative
skills. On the basis of the many conceptualizations, Aesaert et al. (2014) concluded that digital competences should be regarded as
multidimensional. Moreover, the authors stressed the importance of information processing and communication skills as crucial factors.
These conceptualizations align with the common understanding of what information and communication technology (ICT) refers to:
Toomey (2001) proposes that ICT relates to those technologies that are used for accessing, gathering, manipulating and presenting or
communicating information (in Lloyd, 2005, p. 3). In this sense, ICT does not only include specic hardware but also software applications
and connectivity (e.g., Internet, local networks) to enact these skills.
Given the multidimensionality of digital competences, teachers' use and perceived usefulness of ICT for teaching and learning may
directly refer to the sub-dimensions of digital competences (Ertmer, Ottenbreit-Leftwich, Sadik, Sendurur, & Sendurur, 2012; Hermans,
Tondeur, van Braak, & Valcke, 2008). Antonietti and Giorgetti (2006) pointed out that teachers' use of ICT in classroom instruction should be
oriented towards fostering students' interest and higher-order cognitive activities, along with the optimization of learning processes and
learning outcomes. These considerations on students' digital competences and teachers' use of ICT in classrooms suggest taking a multi-
dimensional perspective of perceived usefulness.
So far, the conceptualization of perceived usefulness of ICT for teaching and learning addressed teachers' positive perceptions of ICT.
However, teachers' perceived usefulness with respect to the problems and obstacles that using ICT may impose in classroom instruction
might differ from their perceptions of the benets of using ICT (Carstens & Pelgrum, 2009). Moreover, research on perception-based
evaluations of behavioral or personality-related constructs indicated that individuals respond differently on items that relate to negative
or positive perceptions (Ibrahim, 2001). In fact, the different response tendencies could create further sources of variability in measurement
models (Preckel, 2014). Hence, in order to cover a broader range of teachers' perceived usefulness of ICT, assessments may also capture
teachers' negative perceptions accounting for the problems and obstacles of ICT in classroom instruction (Player-Koro, 2012).
Taken together, given the multiple teaching purposes of using ICT, a multidimensional and specic perspective differentiating between
factors of perceived usefulness for teaching and learning may represent the construct more appropriately than a unidimensional and general
view. For instance, one may think of focusing on the following aspects of positive usefulness of ICT (Antonietti & Giorgetti, 2006; Ferrari, 2013;
Hern andez-Ramos et al., 2014): Fostering interest and learning, fostering collaboration and communication, fostering information retrieval.
In addition, looking at negative perceptions as causing problems and creating obstacles extends this conceptual perspective. Consequently,
the measurement of perceived usefulness in our study requires specic items that are linked to teachers' positive and negative perceptions,
and represent the multidimensionality of teaching goals and the digital competences that students should acquire in learning with ICT.
2.2. Measuring teachers' perceived usefulness of ICT for teaching and learning
As mentioned previously, the measurement of teachers' perceived usefulness has taken a rather general approach until now. In
particular, the construct is mostly measured by items such as Using technology improves my performance', which rely on a general
204 R. Scherer et al. / Computers & Education 88 (2015) 202e214
conceptualization (Teo, 2011a). Other measures focus on the perceived usefulness of using ICT by linking the construct to the outcomes of
performance. For instance, Oye et al. (2014) administered items such as Using the ICT in my classes would increase my productivity. In
contrast, Chien et al. (2014) took a more specic approach by linking teachers' perceived usefulness of using ICT with teaching and learning
goals such as Using ICT for allowing students to share answers with others. Following this approach and accounting for the complexity of
teachers' job performance, as argued in the previous section, a specic and multidimensional measurement of perceived usefulness of ICT
for teaching and learning is warranted. But differentiating between different factors of the construct requires more complex tests which
capture the factors thoroughly. Against this background, test developers of the International Computer and Information Literacy Study
(ICILS) 2013 addressed this need and proposed a 15-item scale of the construct (Fraillon et al., 2013; Fraillon et al., 2014). The scale contains
items that refer to teachers' perceived usefulness of ICT to foster interest and learning, collaboration and communication, and information
retrieval. In addition, it also addresses problems and obstacles caused by using ICT in classroom instruction. Given that the purposes of using
ICT for teaching and learning are interwoven, the items that are intended to measure a single factor may also measure further factors to
some extent, a characteristic that is often referred to as construct overlap. For instance, the use of ICT for fostering students' interest and
learning is often accomplished by providing collaborative learning opportunities, in which students communicate and help each other
(Anjewierden, Gijlers, Kolloffel, Saab, & de Hoog, 2011; Cheung & Vogel, 2013). Furthermore, fostering students' learning with ICT most often
involves cognitively demanding learning environments which require retrieving and evaluating digital information (Ertmer et al., 2012;
Ferrari, 2013). For instance, the item ICT helps students develop skills in planning in the ICILS 2013 questionnaire (Table 1) may also
contain competences of retrieving digital information, since planning collaborative endeavors is always based on existing information on
the topic or the specic group to work with. This item would consequently load on at least two factors. Hence, when teachers rate items on
the usefulness of ICT for teaching and learning, they may not only rely on a single aspect of usefulness. However, we notice that the
occurrence of cross-loadings may also depend on the specic teacher sample, since teachers' individual perceptions of ICT-related con-
structs may differ (Antonietti & Grigorietti, 2006).
Although a construct overlap is expected from a conceptual point of view, it complicates the modeling of the construct with respect to its
factor structure because the link between items and factors is not perfect and cross-loadings have to be accounted for. Although existing
models such as conrmatory factor analysis are able to account for such cross-loadings for a limited number of items, models such as
exploratory or Bayesian structural equation modeling are more exible and do not suffer from problems of identication (e.g., Marsh, Morin,
Parker, & Kaur, 2014; Muthe n & Asparouhov, 2012a).
2.3. Correlates of teachers' perceived usefulness of ICT for teaching and learning
In research on technology acceptance and ICT integration, the relations between teachers' perceived usefulness and further constructs
have been studied extensively (e.g., Chien et al., 2014; Oye et al., 2014; Teo, 2009, 2011a, b). Consequently, these relations have often been
used as evidence for construct validity. For instance, using the technology acceptance model (TAM), Teo (2011a) found that perceived
usefulness was positively related to teachers' intentions to use ICT (b .22, p < .001). Ifenthaler and Schweinbenz (2013) supported this
nding and showed that the intention to use ICT signicantly predicted the actual integration and use of ICT in classrooms (see also Pynoo
et al., 2011). Looking at these ndings from a psychological perspective, researchers argued that perceived usefulness can be regarded as a
behavioral belief which forms the prerequisite for acting out behavior (Chien et al., 2014).
Another belief that affects teachers' perceived usefulness of ICT refers to their self-efcacy, that is, the beliefs in their competences of
using ICT (e.g., Chien et al., 2014; Govender & Govender, 2009; Kreijns, Van Acker, Vermeulen, & van Buuren, 2013). Research on teacher
effectiveness suggested that self-efcacy, as a motivational construct, affects job satisfaction and the way teachers perceive their job per-
formance (Klassen & Tze, 2014). Liaw and Huang (2013) supported this nding and showed a direct relation between self-efcacy and
perceived usefulness (b .12, p < .05), suggesting that teachers who perceive themselves as competent in using ICT perceive the use of ICT as
useful and vice versa. Niederhauser and Perkmen (2010) found an even stronger relation (r .67, p < .01). From a conceptual perspective,
Table 1
Items measuring teachers' perceived usefulness of ICT for teaching and learning (Fraillon et al., 2013; Fraillon et al., 2014).
Item wordings Label

To what extent do you agree with the following statements on using ICT for teaching and learning?
Usefulness of ICT to foster interest and learning Factor 1
Helps students develop greater interest in learning Interest1
Helps students work at a level appropriate to their learning needs Interest2
Improves academic performance of students Interest3
Usefulness of ICT to foster collaboration and communication Factor 2
Helps students learn to collaborate with other students Collab1
Enables students to communicate more effectively with others Collab2
Helps students develop skills in planning Collab3
Usefulness of ICT to foster information retrieval Factor 3
Enables students to access better sources of information Info1
Helps students to consolidate and process information more effectively Info2
Using ICT in teaching and learning causes problems and creates obstacles Factor 4
Results in poorer writing skills among students Negative1
Only introduces organisational problems for schools Negative2
Impedes concept formation better done with real objects Negative3
Only encourages copying material from published Internet sources Negative4
Limits the amount of personal communication among students Negative5
Results in poorer calculation and estimation skills among students Negative6
Only distracts students from learning Negative7
this correlation can be interpreted as an interaction between different belief systems: self-beliefs and behavioral beliefs (Aubusson, Burke,
Schuck, Kearney, & Frischknecht, 2014; Bong & Skaalvik, 2003; Ertmer et al., 2012). Moreover, this relation stresses the importance of
teachers' perceptions of their competences for their intention to use ICT (Lee & Lee, 2014).
Finally, research has identied a negative relation between teachers' age and the integration of and attitudes towards ICT (e.g., O'Bannon
& Thomas, 2014). Hence, one may assume that the effect of age on perceived usefulness might also be negative and mediated by teachers'
anxiety towards using ICT in classrooms (Mac Callum, Jeffrey, & Kinshuk, 2014).
Since the relations described above have often been investigated in the context of technology acceptance (e.g., Herna ndez-Ramos et al.,
2014; Ifenthaler & Schweinbenz, 2013; Teo, 2009), studying the correlates of perceived usefulness can be used to validate our multidi-
mensional conceptualization of the construct.
3. The present study
In the present study, we focus on teachers' perceived usefulness of using ICT, conceptualized in the context of teaching and learning: First,
we study the factor structure of the construct by testing whether or not our assumptions of four factors (i.e., perceived usefulness of ICT for
fostering interest and learning, collaboration and communication, and information retrieval; using ICT in teaching and learning causes
problems and creates obstacles) and construct overlaps hold. Moreover, we attempt to differentiate between general and specic com-
ponents of perceived usefulness. In this regard, we argue that measurement models that assume a perfect item-factor link (i.e., each item is
assigned to only one factor) may not fully describe the structure (Muthe n & Asparouhov, 2012a, b; Song & Lee, 2012). The construct overlap
is likely to occur in scales that measure perceived usefulness of ICT related to specic learning goals. Second, we study how perceived
usefulness relates to further constructs as a step of construct validation (Kane, 2013), expecting positive relations of teachers' perceived
usefulness to their self-efcacy and use of ICT. This expectation is in line with existing research in the context of technology acceptance
(Donnelly et al., 2011; Govender & Govender, 2009; Teo, 2009). In addition, we expect a negative relation to teachers' age (O'Bannon &
Thomas, 2014).
Addressing these two aspects provides evidence on construct validity on the one hand and a synergism between substantive theory and
methodology on the other hand: We propose a conceptualization of perceived usefulness of ICT that is specic for teaching and learning and,
furthermore, present a modeling approach that is exible enough to represent this conceptualization. Taken together, we approach the
following research questions:
To what extent do the assumptions of four correlated factors and construct overlaps of teachers' perceived usefulness of ICT for teaching
and learning hold? (Research Question 1)
To what extent is teachers' perceived usefulness of ICT related to self-efcacy, the use of ICT for teaching and learning, and teachers'
age? (Research Question 2)
4. Method
4.1. Sample and procedure
We used a nationally representative data set of 1653 Norwegian teachers who participated in the International Computer and Infor-
mation Literacy Study (ICILS) in 2013. The study administered a questionnaire on teachers' background, ICT use, ICT self-efcacy, perceived
usefulness of ICT, and further constructs (Fraillon et al., 2014; Ottestad, Throndsen, Hatlevik, & Rohatgi, 2014). Since a number of teachers did
not take the entire questionnaire, the sample used in the present study consisted of N 1190 teachers in 132 schools (63.4% female).
Teachers' mean age was 45.4 years (SD 11.4) and ranged between 23 and 72 years. In sum, 24.9% reported mathematics and natural
sciences as their main subjects, whereas 48.9% had their major subject in language, arts, and humanities. Finally, 26.2% reported other
subjects as their majors (e.g., physical education). The administration and coding procedures of the questionnaire were performed according
to predened quality standards, ensuring the objectivity and reliability of the coding. Schools and teachers were randomly sampled and
voluntarily participated in the study.
4.2. Measures
In order to test our assumptions on the structure of perceived usefulness (Research Question 1) and its relation to further constructs
(Research Question 2), we used teachers' reports on the perceived usefulness of ICT for teaching and learning and external variables such as
ICT self-efcacy, the frequency of using ICT in classroom instruction, and age (Fraillon et al., 2013; Fraillon et al., 2014; Ottestad et al., 2014).
To obtain measures of scale reliability, we estimated McDonald's u1 in addition to Cronbach's a (Yang & Green, 2011).
4.2.1. Perceived usefulness of ICT for teaching and learning

Teachers' perceived usefulness of ICT for teaching and learning referred to the degree to which teachers believe that using ICT would
increase their teaching performance with respect to specic goals and purposes (Davis, 1989; Teo, 2011a). In ICILS 2013, the construct was
assessed by 15 items that represented four facets of perceived usefulness: fostering interest and learning (3 items; u .83, a .83), fostering
communication and collaboration (3 items; u .72, a .72), fostering information retrieval (2 items; u .60, a .61), ICT as causing
problems and creating obstacles (7 items with a negative formulation; u .79, a .84). Table 1 contains the item wordings and labels.
1
u (Sjlij)2/([Sjlij]2 Sdii) with factor loadings li and residual variances dii for an item i.
Teachers had to rate the items on a 4-point agreement scale ranging from 0 (strongly disagree) to 3 (strongly agree). The entire scale showed
an acceptable reliability of u .68 and a .89.
4.2.2. ICT self-efcacy

Teachers' self-efcacy in using ICT can be regarded as an indicator of their beliefs in their ICT skills on the one hand, and in their skills of
using ICT for teaching and learning on the other hand (Lee & Lee, 2014; Niederhauser & Perkmen, 2010). In the present study, the construct
was assessed by a scale comprising 14 items. Teachers had to rate to what extend they believe how well they could perform ICT tasks on a 3-
point scale (0 I do not think I could do this, 1 I could work out how to do this, 2 I know how to do this). The tasks presented in this scale
reected basic ICT skills (e.g., Storing your digital photos on a computer, 6 items; u .87, a .93), more advanced ICT skills (e.g., Collaborating
with others using shared resources such as [Google Docs], 4 items; u .80, a .83), and skills of using ICT for teaching and learning purposes
(e.g., Monitoring students' progress, 4 items; u .80, a .88). Scale reliability was acceptable (u .72, a .94).
4.2.3. Use of ICT in classroom instruction

Since the use of ICT is multifaceted (e.g., Schroeders & Wilhelm, 2011), ICILS 2013 used a measure that covered multiple aspects such as
the use of ICT for assessment and feedback (2 items; u .65, a .69), collaboration among students (4 items; u .75, a .79), and fostering
the development of students' skills (4 items; u .77, a .80). Teachers had to rate how often they use ICT for these purposes (0 never,
1 sometimes, 2 often). The entire scale showed an acceptable reliability of u .71 and a .88.
4.3. Statistical analyses
To approach our rst research question, we applied different procedures of modeling the factor structure of teachers' perceived use-
fulness: First, we specied a Bayesian conrmatory factor-analytic model, assuming four correlated factors (multidimensionality) and a
perfect item-factor link (i.e., no cross-loadings; Fig. 1a, Model 1). We compared this model with a unidimensional model in order to test our
hypothesis on the multidimensionality. Second, we compared Model 1 to a model, which also assumed multidimensionality but allowed for
cross-loadings (Model 2, Fig. 1b). Following this procedure, we specied nested factor models with and without cross-loadings to sort out
the general and specic components of the perceived usefulness construct and to test the existence of construct overlaps after controlling
for general perceived usefulness (Models 3 and 4, Fig. 1 c, d; Gustafsson & berg-Bengtsson, 2010; Muthe n & Asparouhov, 2012b). These
steps were undertaken to obtain evidence on factorial validity (Kane, 2013). In a last step, we tested the relations of the perceived usefulness
factors to constructs such as ICT self-efcacy, ICT use, and teachers' age by means of correlation and regression analyses (Fig. 2a, b).
4.3.1. Bayesian analysis

In our modeling approach, we chose Bayesian instead of maximum-likelihood estimation for a number of reasons: Bayesian analysis
considers parameters such as factor loadings, correlations, and variances as variables rather than point estimates (Kruschke, 2010). Hence,
these parameters have a specic distribution, which may contain so-called priors that reect researchers' assumptions or knowledge
derived from previous studies (van de Schoot et al., 2014). Priors can be non-informative or informative according to the degree of un-
certainty in their distribution. For instance, a large variance in a uniform distribution of a parameter is less informative than smaller var-
iances. In general, researchers using Bayesian analysis state their knowledge about the parameters (i.e., by setting up a set of priors) and
combine this knowledge with the information provided by the data (Baldwin & Fellingham, 2013; Kruschke, 2011; Muthe n & Asparouhov,
2012a, b). Furthermore, in Bayesian analysis, conrmatory factor-analytic and structural equation models can be specied more exibly than
Fig. 1. Bayesian Models of Perceived Usefulness for Teaching and Learning: (a) Conrmatory Factor-Analytic Model without Cross-Loadings, (b) Structural Equation Model with
Cross-Loadings, (c) Nested Factor Model without Cross-Loadings, (d) Nested Factor Model with Cross-Loadings.
Fig. 2. Bayesian (a) Structural Equation Model and (b) Nested Factor Model with Cross-Loadings and a Covariate.
in maximum-likelihood estimation, since model parameters are described by a distribution with a mean and standard deviation (e.g.,
MacCallum, Edwards, & Cai, 2012; Song & Lee, 2012). This advantage is of particular interest in describing substantial elements of the factor
structure such as item cross-loadings or residual (co-)variances (Muthe n & Asparouhov, 2012a). In particular, if we assume a perfect item-
factor link for teachers' perceived usefulness of ICT, a model without any cross-loadings is supposed to t the data reasonably well. In such a
model, items assigned to a factor have a distribution with a mean different from zero; but cross-loadings are strictly zero (Fig. 1a). By
contrast, if our assumption is that the different factors of perceived usefulness are overlapping for substantive reasons, such a model may no
longer hold. Hence, cross-loadings that deviate from zero can be introduced (Fig. 1b). In particular, Muthe n and Asparouhov (2012a) argued
that using informative priors for cross-loadings with mean zero and a small variance, for instance l ~ N (0, 0.01), may be more appropriate
than restricting cross-loadings to zero or any other point estimate (Kaplan & Depaoli, 2012).
Given our assumption on an overlap between the factors of teachers' perceived usefulness of ICT for teaching and learning, we apply
Bayesian analysis with non-zero cross-loadings and compare this model to a model, which assumes exact-zero cross-loadings (Research
Question 1; Fig. 1a, b). We expect the model with cross-loadings to better represent the structure of perceived usefulness than the model
without cross-loadings. On the basis of the results on the factor structure, we add further constructs to the model and obtain relations
among latent variables from the resulting model (Research Question 2; Fig. 2a, b).
4.3.2. Model estimation, t, and convergence

In the present study, the Monte Carlo Markov Chain (MCMC) algorithm based on Gibbs sampling with at least 50,000 iterations and eight
chains was employed. This algorithm represents an iterative procedure, which randomly draws samples from the posterior distribution to
approximate the distribution of parameters (for details on this algorithm, see Song & Lee, 2012). We evaluated the convergence of the
algorithm by investigating the graphical plots of the Markov chains, autocorrelations, and the potential scale reduction factor (PSR). The
latter compares the variances between and within the chains and indicates convergence if close to 1.0 (Kaplan & Depaoli, 2012; Song & Lee,
2012). We chose 0.01 as convergence criterion (Muthe n & Asparouhov, 2012a).
To evaluate model t, we used posterior predictive checking; looking at the discrepancies between the data generated by the model Drep
and the observed data Dobs. In particular, we relied on the Bayesian posterior predictive p value (PP p), which quanties the discrepancies for
a given test statistic T: PP p value Pr [T(Drep) T(Dobs) j Dobs] (Kruschke, 2013). In our case, the Bayesian PP p value of the c2 statistic served
as an indicator for the t of the Bayesian models (van de Schoot et al., 2014). A PP p value above 0.05 indicates a good t, whereas values
closer to zero indicate a poor t of the model (Muthe n & Asparouhov, 2012a). Perfect model t is indicated by PP p 0.5. All analyses were
employed in the statistical package Mplus 7.3 (Muthe n & Muthe n, 1998e2014). Teachers' ratings were treated as ordered-categorical data.
5. Results
5.1. Factor structure of teachers' perceived usefulness of ICT for teaching and learning
On the basis of our theoretical assumptions on the structure of the construct, we hypothesized three positive factors of perceived
usefulness of ICT for (1) fostering interest and learning, (2) fostering collaboration and communication, (3) fostering information retrieval,
along with (4) a negative factor of perceiving ICT as causing problems and creating obstacles in teaching and learning. In order to test our
assumptions of four correlated factors and construct overlaps, we compared a number of models: First, we specied a unidimensional factor
model, which showed a Bayesian PP p value of .000, indicating that the observed and replicated c2 values for this model differed, 95% CI
Dc2 [417.41, 566.69]. Hence, we did not have support on accepting this model.
Second, we specied a four-factor model with correlated but non-overlapping factors (Fig. 1a; Model 1). This model showed reasonable
convergence, as indicated by a PSR value close to 1.0 (Table 2). The factor correlations ranged between .47 and .87 with the highest cor-
relation between fostering interest and learning' and fostering collaboration and communication. Regarding the factor loadings, suf-
ciently high values above .55 were obtained. Although the model's PP p value of .006 was higher than that of the unidimensional model,
indicating that a differentiation between the four factors ameliorated the model t, comparing the c2 statistic of the data replicated under
the model with the actual data indicated a poor overall t (Bayesian posterior predictive p value close to zero; Table 2). Hence, we had
evidence on model misspecication and thus rejected this model.
Third, we incorporated cross-loadings with informative priors (i.e., l ~ N (0, 0.01)) to the four-factor model (Fig. 1b; Model 2) and found
that the Bayesian PP p value of .165 indicated a reasonable model t (Table 2). Moreover, convergence was indicated by a PSR value close to
1.0 and factor correlations ranged between .47 and .80. Interestingly, the factor of ICT as causing problems and creating obstacles in teaching
and learning showed the lowest correlations to the other factors. It is noteworthy that small but signicant cross-loadings occurred for
items Interest3, Collab2, Collab3, and Negative5 (ranging between .20 and .24; Table 2). For instance, the item ICT improves academic
performance of students (Interest3) was originally assigned to Factor 1. Model 2 showed that this item also loaded on fostering information
retrieval and ICT as causing problems and obstacles, two factors that may determine students' academic performance. In contrast, a low
overlap existed between Factors 2 to 4. Consequently, the correlations among the latent variables that involved Factor 1 were overestimated
in Model 1 but not in Model 2. To test the robustness of these results, we varied the prior variances of cross-loadings and studied how this
variation affects model t, the range of cross-loadings, and factor correlations. Our results suggested that increasing the prior variance, that
is, allowing for an even larger construct overlap, improved the model t and increased some of the factor correlations (Appendix A).
In light of the high correlations between the factors in Models 1 and 2 and in order to disentangle the general and specic factors of
teachers' perceived usefulness of ICT, we specied a nested factor model in a fourth step (Gustafsson & berg-Bengtsson, 2010). The model
assumed a general perceived usefulness factor and three specic but uncorrelated factors (Fig. 1c; Model 3). For reasons of model identi-
cation, the factor of Fostering information retrieval was set as a reference. This model did not contain cross-loadings and showed a better
t than the unidimensional model but the overall t was not acceptable (Table 3, Nested Factor Model without Cross-Loadings). All loadings
on the general factor and the specic factors were positive and signicant expect for that of item Interest3, indicating that both the general
and specic components of perceived usefulness capture substantial variance in item responses.
Finally, following the same strategy as for the models without any nested factor, we added cross-loadings to the model in order to test
whether the construct overlap is still apparent after controlling for general perceived usefulness (Fig. 1d; Model 4). The resulting model
Table 2
Bayesian Conrmatory Factor Analysis (BCFA) and Bayesian Structural Equation Modeling (BSEM) of teachers' perceived usefulness of ICT for teaching and learning.
BCFA without CrosseLoadings (Model 1) BSEM with CrosseLoadings (Model 2)
Factor 1 Factor 2 Factor 3 Factor 4 Factor 1 Factor 2 Factor 3 Factor 4

Factor Loadings
Factor 1: Usefulness of ICT to foster interest and learning
Interest1 .81 (.02)* e e e .83 (.07)* .00 (.05) .03 (.06) e.01 (.03)
Interest2 .78 (.02)* e e e .87 (.07)* e.01 (.06) e.04 (.06) e.03 (.03)
Interest3 .78 (.02)* e e e .50 (.08)* .10 (.06) .15 (.06)* .09 (.03)*
Factor 2: Usefulness of ICT to foster collaboration and communication
Collab1 e .68 (.03)* e e e.01 (.09) .75 (.09)* e.04 (.07) e.02 (.04)
Collab2 e .56 (.03)* e e e.19 (.09)* .69 (.10)* .10 (.07) .01 (.04)
Collab3 e .79 (.02)* e e .24 (.10)* .59 (.11)* e.06 (.06) .00 (.04)
Factor 3: Usefulness of ICT to foster information retrieval
Info1 e e .77 (.02)* e e.01 (.08) e.07 (.06) .81 (.07)* e.01 (.04)
Info2 e e .77 (.02)* e .06 (.07) .05 (.06) .73 (.06)* .00 (.04)
Factor 4: Using ICT in teaching and learning causes problems and creates obstacles
Negative1 e e e .56 (.03)* .07 (.07) e.02 (.06) e.07 (.06) .58 (.05)*
Negative2 e e e .63 (.03)* .04 (.07) .01 (.06) e.01 (.06) .60 (.05)*
Negative3 e e e .63 (.03)* .00 (.07) e.03 (.06) .01 (.06) .64 (.05)*
Negative4 e e e .70 (.02)* .11 (.07) e.01 (.06) e.08 (.06) .69 (.05)*
Negative5 e e e .55 (.03)* e.20 (.08)* .10 (.07) .01 (.06) .63 (.05)*
Negative6 e e e .71 (.02)* e.04 (.07) e.04 (.06) .03 (.06) .74 (.05)*
Negative7 e e e .79 (.02)* .01 (.07) .00 (.05) .04 (.06) .76 (.05)*
Factor Correlations
Factor 2 .87 (.02)* e e e .80 (.05)* e e e
Factor 3 .80 (.03)* .68 (.04)* e e .71 (.07)* .68 (.08)* e e
Factor 4 .60 (.03)* .47 (.04)* .54 (.04)* e .55 (.07)* .47 (.08)* .54 (.07)* e
Model Fit
95% CI Dc2 [12.08, 111.79] [e24.25, 71.85]
Bayesian PP p .006 .165
Model Convergence
PSR Max 1.004 1.011
Note. The table shows the Bayesian estimates and the posterior standard deviations (in brackets). Statistically signicant estimates have a 95% credibility interval that does not
cover zero (indicated by an asterisk *). 95% CI Dc2 95% condence interval for the difference between the observed and the replicated c2 values, PP p Posterior predictive p
value, PSR Max Maximum of the potential scale reduction factor after 50,000 iterations. Number of parameters: 109; N 1190. The model with cross-loadings is based on an
informative prior with a normal distribution, N (0, 0.01).
Table 3
Nested factor models of teachers' perceived usefulness of ICT for teaching and learning with and without cross-loadings.
Nested Factor Model without CrosseLoadings (Model 3) Nested Factor Model with CrosseLoadings (Model 4)
General Factor Factor 1 Factor 2 Factor 4 General Factor Factor 1 Factor 2 Factor 4
Factor Loadings
Factor 1: Usefulness of ICT to foster interest and learning
Interest1 .77 (.03)* .41 (.16)* e e .72 (.03)* .49 (.06)* .02 (.04) .01 (.02)
Interest2 .73 (.03)* .31 (.14)* e e .66 (.04)* .49 (.07)* .05 (.05) .00 (.03)
Interest3 .81 (.02)* e.06 (.08) e e .74 (.03)* .17 (.06)* .07 (.04) .06 (.03)*
Factor 2: Usefulness of ICT to foster collaboration and communication
Collab1 .56 (.03)* e .81 (.08)* e .54 (.04)* .08 (.06) .55 (.07)* e.02 (.03)
Collab2 .49 (.03)* e .25 (.05)* e .51 (.03)* e.05 (.05) .36 (.08)* e.01 (.03)
Collab3 .69 (.02)* e .18 (.04)* e .62 (.03)* .22 (.06)* .29 (.05)* e.01 (.03)
Factor 3: Usefulness of ICT to foster information retrievala
Info1 .60 (.03)* e e e .68 (.03)* e.07 (.05) e.09 (.05)* e.02 (.03)
Info2 .72 (.03)* e e e .81 (.03)* e.07 (.05) e.04 (.04) e.03 (.03)
Factor 4: Using ICT in teaching and learning causes problems and creates obstacles
Negative1 .34 (.03)* e e .46 (.03)* .32 (.04)* .08 (.05) .00 (.04) .47 (.04)*
Negative2 .42 (.03)* e e .47 (.03)* .41 (.04)* .04 (.05) .00 (.04) .47 (.04)*
Negative3 .37 (.03)* e e .52 (.03)* .38 (.04)* .02 (.05) e.03 (.04) .51 (.04)*
Negative4 .45 (.03)* e e .54 (.03)* .42 (.04)* .10 (.05)* e.01 (.04) .56 (.04)*
Negative5 .27 (.03)* e e .51 (.03)* .29 (.04)* e.09 (.05)* .08 (.04) .51 (.04)*
Negative6 .41 (.03)* e e .60 (.03)* .43 (.04)* e.02 (.04) e.03 (.04) .58 (.04)*
Negative7 .51 (.03)* e e .60 (.03)* .52 (.04)* .01 (.04) e.03 (.04) .59 (.04)*
Factor Correlations
Factor 1 .00 e e e .00 e e e
Factor 2 .00 .00 e e .00 .00 e e
Factor 4 .00 .00 .00 e .00 .00 .00 e
Model Fit
95% CI Dc2 [9.44, 110.00] [e24.19, 71.41]
Bayesian PP p .010 .163
Model Convergence
PSR Max 1.097 1.052
Note. The table shows the Bayesian estimates and the posterior standard deviations (in brackets). Statistically signicant estimates have a 95% credibility interval that does not
cover zero (indicated by an asterisk*). 95% CI Dc2 95% condence interval for the difference between the observed and the replicated c2 values, PP p Posterior predictive
p value, PSR Max Maximum of the potential scale reduction factor after 50,000 iterations. Number of parameters: 109; N 1190. The model with cross-loadings is based on
an informative prior with a normal distribution, N (0, 0.01).
a
This factor was used as a reference in order to identify the nested factor models.
outperformed the nested factor model in terms of model t and revealed ve signicant cross-loadings, showing that the overlap also
existed in this model (Table 3, Nested Factor Model with Cross-Loadings). Factor loadings on the general factor and the factors to which the
items were originally assigned to were positive and signicant without any exceptions.
Against this background, we have evidence that (a) the four correlated factors of perceived usefulness of ICT for teaching and learning can
be distinguished, and (b) there is an overlap between these factors even after controlling for general perceived usefulness. The correlated-
factors and the nested factor models with cross-loadings (Models 2 and 4) can be accepted as representations of the construct. We would
like to point out that these models show similar t but provide different information on the construct structure. Whereas Model 2 assumes
four correlated factors, Model 4 allows us to disentangle the general and specic components of perceived usefulness.
5.2. Relations among the factors of perceived usefulness, self-efcacy, ICT use, and age
In order to address our second research question on the relation between the perceived usefulness factors and external constructs such
as ICT self-efcacy, ICT use, and teachers' age, we analyzed the correlations between these constructs on the basis of Model 2 rst2 (Fig. 2a).
In a second step, we specied a regression model with the external constructs as outcomes and the factors obtained from the nested Model 4
as predictors3 (Fig. 2b). Our expectations were that perceived usefulness is positively related to self-efcacy and ICT use, whereas teachers'
age shows a negative effect.
5.2.1. Correlation analysis

For all correlation models, the t was acceptable. Regarding teachers' self-efcacy, we found support for our expectations: We obtained
positive correlations between the three positively formulated factors of perceived usefulness and self-efcacy in different kinds of skills
(Table 4). Accordingly, the negatively formulated factor (i.e., ICT as causing problems and creating obstacles) showed negative correlations,
indicating that lower levels of self-efcacy are linked to higher levels of believing that the use of ICT creates obstacles and causes problems in
classroom instruction. For self-efcacy in using ICT for teaching, there were almost no differences in the correlations to the perceived
usefulness factors. However, the factors fostering interest and learning and fostering collaboration and communication had lower cor-
relations with self-efcacy in basic ICT skills than fostering information retrieval and ICT as causing problems and creating obstacles in
teaching and learning.
2
Given the high factor correlations in the correlated-factors model (Model 2), multicollinearity is present, thus compromising the assumptions of regression models.
Hence, we only report correlations for this model.
3
In this nested factor model (Model 4), the Bayesian approach only allowed for specifying the relations among the factors in a regression.
Table 4
Correlations between teachers' perceived usefulness of ICT for teaching and learning and covariates based on Model 2.
Covariates Correlations r (posterior SD) 95% CI Dc2,

Bayesian PP p
Perceived usefulness
Factor 1: Interest & Factor 2: Collaboration & Factor 3: Information Factor 4: Problems &
learning communication retrieval obstacles
Teachers' SelfeEfcacy
Selfeefcacy in teaching with ICT .31 (.04)* .27 (.05)* .28 (.05)* e.30 (.04)* [e30.86, 87.05] .172
Selfeefcacy in basic ICT skills .13 (.05)* .10 (.06) .27 (.05)* e.31 (.05)* [e32.55, 97.30] .163
Selfeefcacy in advanced ICT skills .23 (.04)* .24 (.04)* .21 (.05)* e.23 (.04)* [e23.96, 93.36] .123
Teachers' ICT Use in School
ICT use to foster collaborative skills .35 (.04)* .38 (.05)* .26 (.05)* e.24 (.05)* [e24.16, 94.91] .122
ICT use for assessment .33 (.04)* .35 (.05)* .27 (.05)* e.21 (.05)* [e29.07, 77.21] .187
ICT use for skill development .35 (.04)* .38 (.04)* .30 (.05)* e.20 (.05)* [e20.72, 97.22] .106
Teachers' Age e.12 (.04)* e.04 (.04) e.10 (.04)* .08 (.03)* [e20.64, 80.94] .120
Note. 95% CI Dc2 95% condence interval for the difference between the observed and the replicated c2 values, PP p Posterior predictive p value. Statistically signicant
estimates have a 95% credibility interval that does not cover zero (indicated by an asterisk *).
Regarding teachers' use of ICT in schools, moderate and positive correlations to the four factors of perceived usefulness were obtained
(Table 4). The negatively formulated factor was again negatively associated with ICT use. The correlations of Factors 1 and 2 of perceived
usefulness to the three aspects of ICT use did not differ substantially. By contrast, the relations to Factors 3 (fostering information retrieval)
and 4 (ICT as causing problems and creating obstacles in teaching and learning) were lower than for Factors 1 and 2.
Regarding teachers' age, we found negative relations to the positively formulated factors and a positive relation to the negatively
formulated factors of perceived usefulness. These results indicate that older teachers tended to regard the use of ICT for different purposes as
less useful, and potential problems and obstacles for teaching and learning as more present than their younger colleagues. However, the
relations were rather small (Table 4).
5.2.2. Regression analysis

For all regression models, the t was acceptable and variance explanation ranged between 9% and 16% (Table 5). The results indicated
that the general perceived usefulness factor had positive and the highest regression coefcients for all constructs. Hence, higher levels of
general perceived usefulness are associated with higher levels of self-efcacy and ICT use. Furthermore, Factor 4 (causing problems and
creating obstacles) was signicantly related to all self-efcacy aspects. As expected, Factor 2 (fostering collaboration and communication)
was signicantly related to ICT use that involved collaboration and communication skills. In addition to these regression analyses, we
regressed the perceived usefulness factors of Model 4 on teachers' age and found signicant and negative relation to the general factor
(b .09, posterior SD .04), Factor 1 (b .11, posterior SD .05) but not for Factors 2 (b .07, posterior SD .05) and 4 (b .03,
posterior SD .04).
In sum, high levels of self-efcacy and ICT use were associated with high levels of perceived usefulness in fostering different instructional
activities, but with low levels of perceiving the use of ICT as problematic. Furthermore, teachers' age had a negative effect. We had evidence
on differential relations for some of the perceived usefulness factors.
6. Discussion
6.1. Summary of results
The main goal of the present study was to test hypotheses on the structure of teachers' perceived usefulness of ICT for teaching and
learning (Research Question 1) and study its relations to constructs such as self-efcacy, ICT use, and age (Research Question 2). Using
Bayesian analysis, we found that (a) four factors of perceived usefulness can be distinguished, representing different teaching goals in using
ICT and the problems caused (i.e., fostering interest and learning, collaboration, information retrieval; ICT as causing problems and creating
Table 5
Regression of Covariates on Teachers' Perceived Usefulness of ICT for Teaching and Learning and Covariates based on Model 4.
Covariates Regression coefcients b (posterior SD) R2 95% CI Dc2,

Bayesian PP p
Perceived usefulness
General Factor Factor 1: Interest & Factor 2: Collaboration & Factor 4: Problems &
learning communication obstacles
Teachers' SelfeEfcacy
Selfeefcacy in teaching with ICT .31 (.04)* .07 (.06) .02 (.06) .14 (.05)* .13 [e32.14, 86.41] .183
Selfeefcacy in basic ICT skills .24 (.05)* e.13 (.07) e.08 (.08) .21 (.06)* .14 [e32.79, 96.27] .166
Selfeefcacy in advanced ICT skills .22 (.04)* .07 (.06) .09 (.06) .13 (.04)* .09 [e25.37, 93.20] .127
Teachers' ICT Use in School
ICT use to foster collaborative skills .31 (.04)* .12 (.07) .19 (.06)* .06 (.05) .16 [e22.84, 94.89] .116
ICT use for assessment .31 (.04)* .11 (.06) .12 (.06) .02 (.05) .13 [e28.85, 75.33] .192
ICT use for skill development .34 (.04)* .08 (.07) .15 (.06)* e.01 (.05) .15 [e21.98, 96.64] .110
Note. 95% CI Dc2 95% condence interval for the difference between the observed and the replicated c2 values, PP p Posterior predictive p value. Statistically signicant
estimates have a 95% credibility interval that does not cover zero (indicated by an asterisk *).
obstacles); (b) general and specic components of perceived usefulness can be distinguished; (c) construct overlaps as manifested by
signicant cross-loadings exist; (d) teachers' perceived usefulness of ICT is positively related to self-efcacy and the use of ICT, whereas the
relation to age is negative.
6.2. Structure of teachers' perceived usefulness of ICT (research question 1)
Our rst research question was concerned with the factor structure of teachers' perceived usefulness of ICT for teaching and learning. We
hypothesized four different factors of the construct which referred to the use of ICT to foster students' interest and learning (Factor 1),
collaboration and communication skills (Factor 2), and skills in information retrieval (Factor 3), along with the problems and obstacles in
using ICT for teaching and learning (Factor 4). On the one hand, these factors describe different aspects of developing students' digital
competences as learning goals (Antonietti & Giorgetti, 2006; Wikan & Molster, 2011); and the perceived problems associated with the
integration of ICT on the other hand. Using Bayesian analysis, which assumed four correlated factors without any construct overlap, we
found correlations up to r .87 among the latent variables, compromising our expectations of the distinction between these factors.
However, this model did not represent the data well and could not be used to describe the factor structure. Our further analyses revealed the
existence of cross-loadings, conrming our expectations on construct overlaps (Muthe n & Asparouhov, 2012a). In particular, the four factors
of perceived usefulness interfered with each other, that is, for instance, perceiving ICT as being useful for fostering students' interest and
learning, and perceiving ICT as being useful for fostering specic competences such as information retrieval or communication go together,
since fostering interest and the acquisition of specic competences are parts of students' learning processes (Aesaert, Vanderlinde, Tondeur,
& van Braak, 2013; Ferrari, 2013). Hence, the link between items and latent variables is not perfect. We point out that the existence and
substantive meaning of these imperfect links may vary across different teacher samples, since teachers may understand the items on
perceived usefulness differently, depending on their conceptions of the learning goals proposed (Antonietti & Grigorietti, 2006). It is
therefore essential for researchers to interpret the construct overlaps and the degree to which they exist for their specic samples. In our
sample, the items Interest3 (ICT improves academic performance) and Collab3 (ICT helps students develop skills in planning) were sen-
sitive towards an overlap. This nding is not surprising because the item formulations allow different interpretations: For instance, teachers
may relate academic performance to a number of skills including information retrieval, communication or collaboration. In addition, skills
of planning may also be considered elements of learning. In our study, we were also able to separate the general and specic aspects of
perceived usefulness by showing the preference of nested factor models. Even in these models, the cross-loadings were apparent, sup-
porting the existence and hypothesis of construct overlaps. Moreover, the nested models allow researchers to study the components that are
specic to the four factors of perceived usefulness after controlling for general perceived usefulness (Gustafsson & berg-Bengtsson, 2010).
This approach appears to be useful, particularly in strengthening the robustness of dimensionality results. We notice that our approach was
conrmatory rather than exploratory, as we tested hypotheses on the elements of the factor structure.
Taken together, these overlaps reect the degree to which teachers can distinguish between their perceptions of the usefulness of ICT for
specic learning goals. However, we encourage future research to disentangle the conceptual meaning of these overlaps in detail. But more
importantly, we point out that a unidimensional view of the construct may overgeneralize the complex nature of teachers' perceptions of
using ICT in classroom instruction (Antonietti & Giorgetti, 2006; Teo, 2011a). A multidimensional view of perceived usefulness provides
more differentiated information on how teachers perceive ICT with respect to different aspects, as represented by the four factors
(Herna ndez-Ramos et al., 2014).
6.3. Relations to self-efcacy, ICT use, and age (research question 2)
Regarding our second research question, we found that teachers who perceive themselves as competent in ICT perceive the use of ICT as
useful for teaching and learning, and vice versa. Accordingly, teachers who see problems and obstacles in ICT use tend to have a low self-
efcacy in ICT and vice versa. These ndings support our expectations, since previous research has shown that ICT self-efcacy can be
regarded as an important predictor of teachers' ICT use with moderate effects (e.g., b .32e.46; Kreijns et al., 2013; Lee & Lee, 2014; Teo,
2009). However, it is currently unclear how self-efcacy, perceived usefulness, and ICT use are structurally related. From a perspective of
planned behavior, perceiving an educational tool as useful may interact with teachers' beliefs about their ICT skills when considering the
prediction of the use of ICT (e.g., Niederhauser & Perkmen, 2010; Terzis & Economides, 2011). Nevertheless, our results advocate the
empirical relation and distinction between self-efcacy and perceived usefulness beliefs. This distinction can be interpreted in light of
theories on teachers' beliefs: The two refer to different aspects of the belief system; self-efcacy refers to teachers' personal beliefs whereas
perceived usefulness refers to behavioral and normative beliefs (Antonietti & Giorgetti, 2006; Aubusson et al., 2014; Chien et al., 2014;
Klassen & Tze, 2014).
The results on the positive relations between teachers' perceived usefulness as measured by positive perceptions (Factors 1 to 3) and the
use of ICT are in line with existing research in the context of technology acceptance. More precisely, previous studies have identied a
positive and direct effect of perceived usefulness on teachers' intentions to use ICT (e.g., b .22; Teo, 2009) and the actual integration of ICT
in classrooms (e.g., b .36; Sang et al., 2010). Notably, negative perceptions which reect the problems and obstacles teachers refer to when
using ICT (Factor 4) were negatively correlated with the actual ICT use. This nding appears reasonable, since the more problems teachers
see in using ICT for classroom instruction, the less likely they would integrate ICT in classrooms. Interestingly, the obstacles and problems
mainly referred to doubts with respect to students' learning and the development of skills rather than mere organizational problems, which
could have affected teachers' perceptions of the usefulness of ICT (Tondeur, Valcke, & van Braak, 2008).
Regarding the relations between perceived usefulness and teachers' age, higher age was associated with higher levels of perceiving the
problems and obstacles of ICT use. Interestingly, the largest effects were present for the factors of Fostering students' interest and learning
(Factor 1) and Fostering students' skills of information retrieval (Factor 3), whereas the lowest effect was found for Fostering collaboration and
communication (Factor 2). This may indicate a tendency of mistrust among older teachers with respect to fostering motivational and skill-
related aspects of learning. In contrast, ICT-mediated collaboration and communication was regarded as useful, independent from age. Our
ndings on the age dependency of perceived usefulness support existing research which shows that age matters for the use and perceptions
of ICT (O'Bannon & Thomas, 2014; Vanderlinde, Aesaert, & van Braak, 2014).
Interestingly, the nested factor model revealed the same tendencies of relations among perceived usefulness and the other constructs.
Since the general factor was signicantly related to self-efcacy, ICT use, and age, we interpret this factor as construct-relevant. In contrast to
models which capture methodological and construct-irrelevant variance, this factor may represent general perceived usefulness (Gustafsson
& berg-Bengtsson, 2010). The specic factors thus reect the specic components of perceived usefulness in the hypothesized dimensions.
Taken together, the relations between the constructs under investigation support our expectations on the role of teachers' perceived
usefulness of ICT for teaching and learning. Furthermore, we argue that teachers' choices of using ICT are related to different beliefs, of which
perceived usefulness can be regarded as an important determinant (Aubusson et al., 2014; Davis, 1989; Ertmer et al., 2012).
6.4. Limitations and future directions
The present study has a number of limitations, which point out to future research in the context of technology acceptance and ICT
integration in classrooms: First, since we were aimed at studying the factor structure of teachers' perceived usefulness of ICT (Research
Question 1) and the relations to constructs such as ICT self-efcacy, ICT use, and age (Research Question 2), we did not take on the structural
relations between the constructs, which are, for instance, proposed in models of technology acceptance (e.g., Davis, 1989; Teo, 2014). We
therefore regard our study as a rst step to examine teachers' perceived usefulness of ICT in more detail. Hence, we encourage researchers to
take a closer look at these relations on the basis of the proposed modeling approach. Second, we did not study the effects of teachers'
perceived usefulness on students' digital competences. Since the structure of perceived usefulness differentiated between different factors,
it would be interesting to study how these factors relate to the adjacent skills. The relations obtained from these analyses may inform
researchers about the impact of perceived usefulness on students' learning outcomes in ICT (e.g., Tondeur et al., 2008). In order to strengthen
the results of such analyses, a larger number of items, particularly for measuring the information retrieval factor would be desirable.
Moreover, future research may be concerned with the obstacles teachers perceive at different levels of the educational system, particularly
those created by the school environment (Aubusson et al., 2014; Ertmer et al., 2012; Tondeur et al., 2008).
6.5. Conclusion and implications
The main idea the present study wants to convey is that teachers' perceived usefulness of ICT for teaching and learning can be
conceptualized as a multidimensional construct that comprises more than one factor according to the goals of teaching and learning. Hence,
becoming more specic about teachers' job performance in assessing perceived usefulness provides measures that reveal detailed infor-
mation about the specic views of teachers on the usefulness of ICT for instructional purposes (Niederhauser & Perkmen, 2010). This in-
formation is crucial for identifying the specic needs along with potential reasons for reservations towards integrating and using ICT in
classroom instruction.
Moreover, we were able to provide evidence that the multidimensional conceptualization of perceived usefulness can be aligned with
specic learning goals, which reect dimensions of students' digital competence. Finding support for multidimensionality and the differ-
entiation between four factors conrmed our expectations on the structure of the construct. In addition, given that relations to ICT self-
efcacy, ICT use, and age were in line with existing research, our ndings lend evidence on construct validity. Hence, we argue that the
measure used in ICILS 2013, indeed, provides a valid and reliable assessment of perceived usefulness. Besides this substantive perspective,
we suggest how to describe the construct by Bayesian structural equation models. For instance, if researchers are interested in disentangling
the general and specic components of perceived usefulness, nested factor models are considered appropriate tools. Additionally, our
research identies challenges such as construct overlaps, which should be accounted for in specic teacher samples.
In light of the substantive-methodological perspective taken in the paper, we encourage researchers working on theoretical models in
the context of technology acceptance to become more specic in the conceptualization and structure of perceived usefulness rather than
relying on general conceptualizations and restrictive assumptions such as unidimensionality that may not reect its structure.
Appendix A
Sensitivity Analyses of Using Different Prior Variances for Normally Distributed Cross-Loadings.
Prior Variance 95% cross-loading limit 95% CI Dc2 Bayesian PP p Range of cross-loadings Range of Factor correlations
0.01 .20 [e24.25, 71.85] .165 e.20 to .24 .47e.80
0.02 .28 [e29.21, 65.41] .228 e.25 to .32 .47e.77
0.03 .34 [e31.16, 63.32] .248 e.26 to .36 .46e.75
0.04 .39 [e31.19, 62.68] .256 e.26 to .38 .46e.74
0.05 .44 [e31.87, 61.86] .258 e.27 to .41 .45e.72
0.10 .62 [e31.63, 61.59] .261 e.28 to .47 .42e.69
Note. 95% CI Dc2 95% condence interval for the difference between the observed and the replicated c2 values, PP p Posterior predictive p value. Number of parameters:
109, N 1190. The 95% cross-loading limits refer to a normal distribution with mean zero (Muthen & Asparouhov, 2012a).
Acknowledgments
Special thanks to the Norwegian ICILS group for their support. We would like to thank the two anonymous reviewers for their thorough
review and constructive comments on previous versions of the paper.
References
Aesaert, K., van Nijlen, D., Vanderlinde, R., & van Braak, J. (2014). Direct measures of digital information processing and communication skills in primary education: using item
response theory for the development and validation of an ICT competence scale. Computers & Education, 76, 168e181. http://dx.doi.org/10.1016/j.compedu.2014.03.013.
Aesaert, K., Vanderlinde, R., Tondeur, J., & van Braak, J. (2013). The content of educational technology curricula: a cross-curricular state of the art. Educational Technology
Research and Development, 61, 131e151. http://dx.doi.org/10.1007/s11423-012-9279-9.
Anjewierden, A., Gijlers, H., Kolloffel, B., Saab, N., & de Hoog, R. (2011). Examining the relation between domain-related communication and collaborative inquiry learning.
Computers & Education, 57, 1741e1748. http://dx.doi.org/10.1016/j.compedu.2011.03.010.
Antonietti, A., & Giorgetti, M. (2006). Teachers' beliefs about learning from multimedia. Computers in Human Behavior, 22, 267e282. http://dx.doi.org/10.1016/
j.chb.2004.06.002.
Aubusson, P., Burke, P., Schuck, S., Kearney, M., & Frischknecht, B. (2014). Teachers choosing rich tasks: the moderating impact of technology on student learning, enjoyment,
and preparation. Educational Researcher, 43, 219e229. http://dx.doi.org/10.3102/0013189X14537115.
Baldwin, S. A., & Fellingham, G. W. (2013). Bayesian methods for the analysis of small sample multilevel data with a complex variance structure. Psychological Methods, 18,
151e164. http://dx.doi.org/10.1037/a0030642.
Bong, M., & Skaalvik, E. M. (2003). Academic self-concept and self-efcacy: how different are they really? Educational Psychology Review, 15, 1e40. http://dx.doi.org/10.1023/
A:1021302408382.
Carstens, R., & Pelgrum, W. J. (2009). Second information and technology in education study e SITES 2006 Technical report. Amsterdam: IEA.
Cheung, R., & Vogel, D. (2013). Predicting user acceptance of collaborative technologies: an extension of the technology acceptance model for e-learning. Computers &
Education, 63, 160e175. http://dx.doi.org/10.1016/j.compedu.2012.12.003.
Chien, S.-P., Wu, H.-K., & Hsu, Y.-S. (2014). An investigation of teachers' beliefs and their use of technology-based assessments. Computers in Human Behavior, 31, 198e210.
http://dx.doi.org/10.1016/j.chb.2013.10.037.
Claro, M., Preiss, D. D., San Martin, E., Jara, I., Hinostroza, J. E., Valenzuela, S., et al. (2012). Assessment of 21st century ICT skills in Chile: test design and results from high
school students. Computers & Education, 59, 1042e1063. http://dx.doi.org/10.1016/j.compedu.2012.04.004.
Davis, F. D. (1989). Perceived usefulness, perceived ease of use, and user acceptance of information technology. MIS Quartely, 13, 319e340. http://dx.doi.org/10.2307/249008.
Donnelly, D., McGarr, O., & O'Reilly, J. (2011). A framework for teachers' integration of ICT into their classroom practice. Computers & Education, 57, 1469e1483. http://
dx.doi.org/10.1016/j.compedu.2011.02.014.
Ertmer, P. A., Ottenbreit-Leftwich, A. T., Sadik, O., Sendurur, E., & Sendurur, P. (2012). Teacher beliefs and technology integration practices: a critical relationship. Computers &
Ferrari, A. (2013). DIGCOMP: A framework for developing and understanding digital competence in Europe. Luxembourg: Publications Ofce of the European Union,. http://
dx.doi.org/10.2788/52966.
Fraillon, J., Ainley, J., Schulz, W., Friedman, T., & Gebhardt, E. (2014). Preparing for life in a digital age e The IEA International Computer and Information Literacy Study Inter-
national report. Amsterdam: IEA.
Fraillon, J., Schulz, W., & Ainley, J. (2013). International computer and information literacy Study: Assessment framework. Amsterdam: IEA.
Govender, D., & Govender, I. (2009). The relationship between Information and Communications Technology (ICT) integration and teachers' self-efcacy about ICT. Education
as Change, 13, 153e165. http://dx.doi.org/10.1080/16823200902943346.
Grifn, P., Care, E., & McGaw, B. (2012). The changing role of education and schools. In P. Grifn, B. McGaw, & E. Care (Eds.), Assessment and teaching of 21st Century skills (pp.
1e15). Dordrecht: Springer. http://dx.doi.org/10.1007/978-94-007-2324-5_1.
Gustafsson, J.-E., & berg-Bengtsson, L. (2010). Unidimensionality and interpretability of psychological instruments. In S. E. Embretson (Ed.), Measuring psychological con-
structs: Advances in model-based approaches (pp. 97e121). Washington, DC: American Psychological Association.
Hermans, T., Tondeur, J., van Braak, J., & Valcke, M. (2008). The impact of primary school teachers' educational beliefs on the classroom use of computers. Computers &
Herna ndez-Ramos, J. P., Martnez-Abad, F., Garca Pen ~ alvo, F. J., Herrera Garca, M. E., & Rodrguez-Conde, M. J. (2014). Teachers' attitude regarding the use of ICT. A factor
reliability and validity study. Computers in Human Behavior, 31, 509e516. http://dx.doi.org/10.1016/j.chb.2013.04.039.
Ibrahim, A. M. (2001). Differential responding to positive and negative items: the case of a negative item in a questionnaire for course and faculty evaluation. Psychological
Reports, 88, 497e500. http://dx.doi.org/10.2466/pr0.2001.88.2.497.
Ifenthaler, D., & Schweinbenz, V. (2013). The acceptance of Tablet-PCs in classroom instruction: the teachers' perspectives. Computers in Human Behavior, 29, 525e534. http://
dx.doi.org/10.1016/j.chb.2012.11.004.
Kane, M. T. (2013). Validating the interpretations and uses of test scores. Journal of Educational Measurement, 50, 1e73. http://dx.doi.org/10.1111/jedm.12000.
Kaplan, D., & Depaoli, S. (2012). Bayesian structural equation modeling. In R. H. Hoyle (Ed.), Handbook of structural equation modeling (pp. 650e673). New York, NY: Guilford
Press.
Kikis, K., Scheuermann, F., & Villalba, E. (2009). A framework for understanding and evaluating the impact of information and communication technologies in education. In
F. Scheuermann, & F. Pedro (Eds.), Assessing the effects of ICT in education (pp. 69e82). Luxembourg: Publication Ofce of the European Union. http://dx.doi.org/10.2788/
27419.
Klassen, R., & Tze, V. (2014). Teachers' self-efcacy, personality, and teacher effectiveness: a meta-analysis. Educational Research Review, 12, 59e76. http://dx.doi.org/10.1016/
j.edurev.2014.06.001.
Kreijns, K., Van Acker, F., Vermeulen, M., & van Buuren, H. (2013). What stimulates teachers to integrate ICT in their pedagogical practices? The use of digital learning
materials in education. Computers in Human Behavior, 29, 217e225. http://dx.doi.org/10.1016/j.chb.2012.08.008.
Kruschke, J. K. (2010). Bayesian data analysis. Wiley Interdisciplinary Reviews: Cognitive Science, 1, 658e676. http://dx.doi.org/10.1002/wcs.72.
Kruschke, J. K. (2011). Doing Bayesian data analysis. Burlington, MA: Elsevier.
Kruschke, J. K. (2013). Posterior predictive checks can and should be Bayesian: comment on Gelman and Shalizi, Philoshophy and the practice of Bayesian statistics. British
Journal of Mathematical and Statistical Psychology, 66, 45e56. http://dx.doi.org/10.1111/j.2044-8317.2012.02063.x.
Lee, Y., & Lee, J. (2014). Enhancing pre-service teachers' self-efcacy beliefs for technology integration through lesson planning practice. Computers & Education, 73, 121e128.
http://dx.doi.org/10.1016/j.compedu.2014.01.001.
Liaw, S.-S., & Huang, H.-M. (2013). Perceived satisfaction, perceived usefulness and interactive learning environments as predictors to self-regulation in e-learning envi-
ronments. Computers & Education, 60, 14e24. http://dx.doi.org/10.1016/j.compedu.2012.07.015.
Lloyd, M. (2005). Toward a denition of the integration of ICT in the classroom. In Australian Association for Research in Education (Ed.), Proceedings AARE '05 education
research e Creative Dissent: Constructive Solutions. Parramatta: AARE. Retrieved from eprints.qut.edu.au/3553/1/3553.pdf [12/2/2015].
Mac Callum, K., Jeffrey, L., & Kinshuk. (2014). Comparing the role of ICT literacy and anxiety in the adoption of mobile learning. Computers in Human Behavior, 39, 8e19. http://
dx.doi.org/10.1016/j.chb.2014.05.024.
MacCallum, R. C., Edwards, M. C., & Cai, L. (2012). Hopes and cautions in implementing Bayesian structural equation modeling. Psychological Methods, 17, 340e345. http://
dx.doi.org/10.1037/a0027131.
Marsh, H. W., Morin, A. J. S., Parker, P. D., & Kaur, G. (2014). Exploratory structural equation modeling: an integration of the best features of exploratory and conrmatory
factor analysis. Annual Review of Clinical Psychology, 10, 85e110. http://dx.doi.org/10.1146/annurev-clinpsy-032813-153700.
Muthe n, B., & Asparouhov, T. (2012a). Bayesian structural equation modeling: a more exible representation of substantive theory. Psychological Methods, 17, 313e335. http://
dx.doi.org/10.1037/a0026802.
Muthe n, B., & Asparouhov, T. (2012b). Rejoinder to MacCallum, Edwards, and Cai (2012) and Rindskopf (2012): Mastering a new method. Psychological Methods, 17, 346e353.
http://dx.doi.org/10.1037/a0029214.
Muthe n, B., & Muthe n, L. (1998e2014). Mplus 7.3 [Computer software]. Los Angeles, CA: Muthe n & Muthen.
Niederhauser, D. S., & Perkmen, S. (2010). Beyond self-efcacy: measuring pre-service teachers' instructional technology outcome expectations. Computers in Human Behavior,
26, 436e442. http://dx.doi.org/10.1016/j.chb.2009.12.002.
OECD. (2014). PISA 2012 Results: Creative problem solving: Students' skills in tackling real-life problems (Volume V). Paris: OECD Publishing. http://dx.doi.org/10.1787/
9789264208070-en.
Ottestad, G., Throndsen, I., Hatlevik, O., & Rohatgi, A. (2014). Digitale Ferdigheter for Alle? Norske Resultater Fra ICILS 2013. Oslo: Senter for IKT i utdanningen.
Oye, N. D., Iahad, N. A., & Rahim, N. A. (2014). The history of UTAUT model and its impact on ICT acceptance and usage by academicians. Education and Information Tech-
nologies, 19, 251e270. http://dx.doi.org/10.1007/s10639-012-9189-9.
O'Bannon, B. W., & Thomas, K. (2014). Teacher perceptions of using mobile phones in the classroom: age matters! Computers & Education, 74, 15e25. http://dx.doi.org/10.1016/
j.compedu.2014.01.006.
Player-Koro, C. (2012). Factors inuencing teachers' use of ICT in education. Education Inquiry, 3, 93e108.
Preckel, F. (2014). Assessing need for cognition in early adolescence. European Journal of Psychological Assessment, 30, 65e72. http://dx.doi.org/10.1027/1015-5759/a000170.
Proctor, M. D., & Marks, Y. (2013). A survey of exemplar teachers' perceptions, use, and access of computer-based games and technology for classroom instruction. Computers
& Education, 62, 171e180. http://dx.doi.org/10.1016/j.compedu.2012.10.022.
Pynoo, B., Devolder, P., Tondeur, J., van Braak, J., Duyck, W., & Duyck, P. (2011). Predicting secondary school teachers' acceptance and use of digital learning environment: a
cross-sectional study. Computers in Human Behavior, 27, 568e575. http://dx.doi.org/10.1016/j.chb.2010.10.005.
Sang, G., Valcke, M., van Braak, J., & Tondeur, J. (2010). Student teachers' thinking processes and ICT integration: predictors of prospective teaching behaviors with educational
technology. Computers & Education, 54, 103e112. http://dx.doi.org/10.1016/j.compedu.2009.07.010.
van de Schoot, R., Kaplan, D., Denissen, J., Asendorpf, J., Neyer, F., & van Aken, M. (2014). A gentle introduction to Bayesian analysis: applications to developmental research.
Child Development, 85, 842e860. http://dx.doi.org/10.1111/cdev.12169.
Schroeders, U., & Wilhelm, O. (2011). Computer usage questionnaire: structure, correlates, and gender differences. Computers in Human Behavior, 27, 899e904. http://
dx.doi.org/10.1016/j.chb.2010.11.015.
Smarkola, C. (2008). Efcacy of a planned behavior model: beliefs that contribute to computer usage intentions of student teachers and experienced teachers. Computers in
Human Behavior, 24, 1196e1215. http://dx.doi.org/10.1016/j.chb.2007.04.005.
So, H.-J., Choi, H., Lim, W. Y., & Xiong, Y. (2012). Little experience with ICT: are they really the Net Generation student-teachers? Computers & Education, 59, 1234e1245. http://
Song, X.-Y., & Lee, S.-Y. (2012). A tutorial on the Bayesian approach for analyzing structural equation models. Journal of Mathematical Psychology, 56, 135e148. http://dx.doi.org/
10.1016/j.jmp.2012.02.001.
Teo, T. (2009). Modelling technology acceptance in education: a study of pre-service teachers. Computers & Education, 52, 302e312. http://dx.doi.org/10.1016/
j.compedu.2008.08.006.
Teo, T. (2011a). Factors inuencing teachers' intention to use technology: model development and test. Computers & Education, 57, 2432e2440. http://dx.doi.org/10.1016/
j.comped.2011.06.008.
Teo, T. (2011b). Modeling technology acceptance among pre-service teachers. In T. Teo (Ed.), Technology acceptance in education: research and issues (pp. 79e99). Rotterdam:
Sense Publishers.
Teo, T. (2014). Unpacking teachers' acceptance of technology: tests of measurement invariance and latent mean differences. Computers & Education, 75, 127e135. http://
Teo, T., Lee, C. B., Chai, C. S., & Wong, S. L. (2009). Assessing the intention to use technology among pre-service teachers in Singapore and Malaysia: a multigroup invariance
analysis of the Technology Acceptance Model (TAM). Computers & Education, 53, 1000e1009. http://dx.doi.org/10.1016/j.compedu.2009.05.017.
Teo, T., & Noyes, J. (2011). An assessment of the inuence of perceived enjoyment and attitude on the intention to use technology among pre-service teachers: a structural
equation modeling approach. Computers & Education, 57, 1645e1653. http://dx.doi.org/10.1016/j.compedu.2011.03.002.
Terzis, V., & Economides, A. A. (2011). The acceptance and use of computer based assessment. Computers & Education, 56, 1032e1044. http://dx.doi.org/10.1016/
j.compedu.2010.11.017.
Tondeur, J., Valcke, M., & van Braak, J. (2008). A multidimensional approach to determinants of computer use in primary education: teacher and school characteristics. Journal
of Computer Assisted Learning, 24, 494e506. http://dx.doi.org/10.1111/j.1365-2729.2008.00285.x.
Toomey, R. (2001). Information and communication technology for teaching and learning. Schooling Issues Digest No. 2.
n-Lo
Trista pez, A., & Ylizaliturri-Salcedo, M. A. (2014). Evaluation of ICT competencies. In J. M. Spector, M. D. Merrill, J. Elen, & M. J. Bishop (Eds.), Handbook of research on
educational communications and technology (4th ed.). (pp. 323e336). New York, NY: Springer. http://dx.doi.org/10.1007/978-1-4614-3185-5_26.
Vanderlinde, R., Aesaert, K., & van Braak, J. (2014). Institutionalised ICT use in primary education: a multilevel analysis. Computers & Education, 72, 1e10. http://dx.doi.org/
10.1016/j.compedu.2013.10.007.
Venkatesh, V., Morris, M. G., Davis, M. G., & Davis, F. D. (2003). User acceptance of information technology: toward a unied theory. MIS Quartely, 27, 425e478.
Wikan, G., & Molster, T. (2011). Norwegian secondary school teachers and ICT. European Journal of Teacher Education, 34, 209e218. http://dx.doi.org/10.1080/
02619768.2010.543671.
Yang, Y., & Green, S. (2011). Coefcient Alpha: a reliability coefcient for the 21st Century? Journal of Psychoeducational Assessment, 29, 377e392. http://dx.doi.org/10.1177/
0734282911406668.

Research Gap

Încărcat de

Informații document

Titlu original

Drepturi de autor

Formate disponibile

Partajați acest document

Partajați sau inserați document

Opțiuni de partajare

Vi se pare util acest document?

Este necorespunzător acest conținut?

Drepturi de autor:

Formate disponibile

Research Gap

Încărcat de

Drepturi de autor:

Formate disponibile

Computers & Education 88 (2015) 202e214

Contents lists available at ScienceDirect

Computers & Education

Becoming more specic: Measuring and modeling teachers' perceived

Item wordings Label

3. The present study

4.1. Sample and procedure

4.2.1. Perceived usefulness of ICT for teaching and learning

4.2.2. ICT self-efcacy

4.2.3. Use of ICT in classroom instruction

4.3. Statistical analyses

4.3.1. Bayesian analysis

4.3.2. Model estimation, t, and convergence

BCFA without CrosseLoadings (Model 1) BSEM with CrosseLoadings (Model 2)

Factor 1 Factor 2 Factor 3 Factor 4 Factor 1 Factor 2 Factor 3 Factor 4

5.2.1. Correlation analysis

Covariates Correlations r (posterior SD) 95% CI Dc2,

5.2.2. Regression analysis

6.1. Summary of results

Covariates Regression coefcients b (posterior SD) R2 95% CI Dc2,

6.2. Structure of teachers' perceived usefulness of ICT (research question 1)

6.3. Relations to self-efcacy, ICT use, and age (research question 2)

6.4. Limitations and future directions

6.5. Conclusion and implications

S-ar putea să vă placă și