Teaching Factors Influencing ICT Use in Writing Instruction

Abstract

The increasing integration of Information and Communication Technologies (ICT) in education has highlighted their potential to support writing instruction. However, research examining how ICT is used to teach writing, as well as the factors influencing this use across educational stages, remains limited. Accordingly, this study investigates differences in ICT-related training and the use of ICT for writing instruction among language and literature teachers in primary and secondary education, together with the internal and external factors shaping such use. A survey based on an ad hoc questionnaire was completed by 360 Spanish teachers. The results revealed significant differences between educational stages in both ICT-related training and the use of ICT for teaching writing. Primary school teachers tended to focus on low-level writing processes (e.g., spelling), whereas secondary school teachers placed greater emphasis on higher-level writing skills and more advanced ICT-based activities, despite the overall limited use of ICT for instructional purposes. Internal and external factors varied across educational levels and were positively associated with ICT integration; however, self-efficacy emerged as the main significant predictor of ICT use for writing instruction. Overall, the study identifies key barriers to ICT integration and suggests practical strategies to enhance its use in writing instruction, including targeted teacher training in ICT-supported writing practices and professional development initiatives aimed at strengthening teachers’ self-efficacy.

1. Introduction

The widespread use of Information and Communication Technologies (ICT) in daily life has made them an essential component of all areas of society, and particularly in education. ICT tools are now fundamental for teaching across various domains, including literacy and specifically, writing competence (Gómez-Fernández & Mediavilla, 2022). Incorporating ICT into classrooms has been linked to increased autonomous learning, motivation, and academic performance among students (Zhang et al., 2024). As a result, national (LOMLOE, 2020) and international educational policies (European Commission, 2023) emphasize the development of digital competence in both students and teachers, as well as the creation of innovative learning environments and instructional resources. In this context, the Eurydice report (Eurydice, 2019) highlights two complementary perspectives in digital education: the development of digital competences for students and teachers, and the pedagogical use of technologies to support, enhance, and transform teaching and learning processes. Driven by this growing emphasis on ICT integration, research has focused on how to incorporate digital technologies into instructional practices (Blink Learning, 2021) and on developing guidelines to support teachers in making pedagogically meaningful use of ICT (Vuorikari et al., 2022). Teachers play a fundamental role in this process, as they are responsible for translating technological possibilities into educational practice (Almerich et al., 2023; Ertmer, 2005; Kirschner & De Bruyckere, 2017; Suárez-Rodríguez et al., 2018). Consequently, the success of ICT integration largely depends on teachers’ training, beliefs, and willingness to incorporate digital tools into their teaching (UNESCO, 2023). However, empirical evidence consistently shows that teachers’ instructional use of ICT remains below expectations (e.g., González & de Pablos, 2015; Mutisya, 2021; Scherer & Hatlevik, 2017), suggesting that access to technology alone is insufficient. This limited integration highlights the need to identify the factors that influence how and why teachers use ICT in their instructional practices (Area-Moreira et al., 2020; Olszewski & Crompton, 2020). This will be the focus of the present study.

From a theoretical perspective, the present study is explicitly grounded in Ertmer’s model of barriers to technology integration in education (Ertmer, 1999; Ertmer et al., 2015). This model originally distinguishes between first-order barriers, referring to external and contextual conditions (e.g., access to resources, technical and institutional support), and second-order barriers, referring to internal and personal factors (e.g., teachers’ attitudes, self-efficacy, and theoretical orientations).

Although Ertmer’s original formulation refers to first- and second-order barriers, subsequent research has widely operationalized this framework using the distinction between external and internal factors, which has become the most common conceptualization in the literature. Recent empirical models explicitly adopt this internal-external distinction to examine teachers’ ICT integration (e.g., Almerich et al., 2024).

This framework is particularly relevant for writing instruction, as teaching writing involves complex instructional processes that require deliberate pedagogical decisions regarding content, task design, feedback, and the selection of instructional tools. Consequently, understanding teachers’ use of ICT for writing instruction requires examining not only contextual conditions, but also teachers’ internal dispositions, such as attitudes, self-efficacy, and training related to both writing pedagogy and ICT use.

In the present study, the internal-external conceptualization of Ertmer’s model provides the theoretical basis for organizing the study variables and guiding the analytical approach. Accordingly, the factors examined are systematically structured into internal and external dimensions, which are described in the following section (Ertmer et al., 2015; Nunes et al., 2022; Suárez-Rodríguez et al., 2018).

1.1. Internal Factors Influencing Teachers’ Use of ICT

These factors refer to individual aspects and beliefs about how learning occurs (Nunes et al., 2022), including background variables (e.g., gender, age, teaching experience, and educational level), competence-related factors (i.e., training in ICT and level of digital competence), and internal psychological factors, such as teachers’ self-efficacy, attitudes toward ICT, and theoretical orientations.

Looking at the first of these aspects, Goswami and Shukla (2020) showed that men generally have higher levels of ICT training than women. ICT training tends to be higher among younger teachers and those with fewer years of experience (Goswami & Shukla, 2020; Martínez-Rico et al., 2022). Similarly, López-Belmonte et al. (2020) reported that teachers with bachelor’s degrees have higher levels of ICT training than those with diplomas, university degrees, and/or doctorates.

Furthermore, a positive relationship has been demonstrated between the use of ICT and teachers’ training and level of digital competence. Studies with teachers across different educational stages and subject areas indicate that those with low levels of training and/or digital competence rarely use ICT in their teaching practice (Portillo-Berasaluce et al., 2022).

In addition, research on the level of self-efficacy in the use of ICT—understood as teachers’ perceptions of their ability to effectively apply ICT in the teaching-learning process (Nunes et al., 2022)—indicates that teachers with high self-efficacy tend to use ICT more frequently in their classrooms (Aivazidi & Michalakelis, 2023).

Previous studies have also analyzed attitudes toward ICT, understood as teachers’ perceived evaluations of the usefulness of ICT in teaching. These studies indicate that teachers who believe ICT has positive effects on learning tend to use it more frequently (Nunes et al., 2022). However, teachers with negative attitudes toward ICT either use it infrequently or reject it altogether.

Finally, teachers’ theoretical orientations play a key role in shaping instructional practices and pedagogical decision-making within the teaching–learning process. Research has distinguished between traditional orientations, typically associated with teacher-centered approaches (e.g., lectures), and constructivist orientations, in which teachers adopt a guiding role and students take a more active part as knowledge creators. Evidence suggests that constructivist orientations are associated with a greater use of ICT in classroom practice, whereas more traditional orientations tend to be linked to more limited or instrumental uses of digital technologies (Nunes et al., 2022; Almerich et al., 2024).

1.2. External Factors Influencing Teachers’ Use of ICT

The availability of ICT resources and support from key groups (e.g., school leadership or families) is particularly important according to previous studies. The availability of ICT resources, as highlighted by Mateus et al. (2022), is essential for their effective use, and a lack of such resources results in infrequent use (Liu et al., 2025). Osorio-Saez et al. (2021) demonstrated that support from school leadership and families influences integration of ICT in the classroom. When these groups see ICT as a challenge beyond their capabilities, they tend to reduce their commitment to incorporating it into the teaching-learning process.

Other external factors linked to ICT use include the type of school ownership and the educational stage. According to authors such as Almerich et al. (2023), Goswami and Shukla (2020), and Portillo-Berasaluce et al. (2022), teachers in private schools have higher levels of ICT training than those in public schools. Likewise, Portillo-Berasaluce et al. (2022) suggest that primary school teachers have less digital training than secondary school teachers.

1.3. The Present Study

Despite the substantial body of research on teachers’ use of ICT in education, existing evidence has largely been generated from general or cross-curricular perspectives, without systematically accounting for the specific instructional demands associated with different subject areas. While this research has provided valuable insights into overall patterns of ICT adoption, it has paid limited attention to how disciplinary characteristics shape teachers’ pedagogical uses of digital technologies. As a result, our current understanding of ICT integration remains largely detached from the distinctive pedagogical and instructional requirements of specific domains such as writing.

Only a small number of studies have examined teachers’ ICT use in language-related domains, and research in this area therefore remains particularly scarce. Specifically, few studies have addressed ICT use in the teaching of literature (Rodríguez-Muñoz & Ruiz-Domínguez, 2021) and literacy more broadly (García-Tati & Sánchez-Moreno, 2023; Nunes et al., 2022). However, these studies have generally adopted broad or generic analytical approaches, without focusing explicitly on writing instruction as a distinct instructional domain. Consequently, despite the relevance of writing as a key academic competence, empirical evidence on how teachers use ICT specifically for the teaching of writing remains limited.

Writing is a particularly complex competence that involves the coordination of multiple processes, including task design, content development, feedback provision, and revision. These processes place specific pedagogical demands on teachers and may condition not only whether ICT is used, but also how, when and for what purposes it is integrated into instruction. In this context, empirical research aimed at understanding how teachers use ICT to teach writing is especially relevant, given the central role of writing in students’ academic, personal, and professional development (Blink Learning, 2021). Moreover, teachers are increasingly expected to possess the digital competences required to guide students in educational contexts characterised by digital writing practices (Herrero, 2019).

Beyond describing patterns of ICT use, there is also a need to better understand the factors that facilitate or hinder its integration into writing instruction. Previous research has identified a range of internal factors (e.g., attitudes, self-efficacy) and external factors (e.g., availability of resources, institutional and professional support) associated with teachers’ ICT use. However, to our knowledge, no studies have examined relative contribution and interplay of these factors within a single, theoretically grounded framework applied specifically to writing instruction. Such an approach is necessary to move beyond fragmented and descriptive evidence and to gain a more comprehensive understanding of the conditions under which teachers integrate ICT into the teaching and learning of writing.

Guided by the internal-external framework of ICT integration (Ertmer, 1999; Ertmer et al., 2015), the present study seeks to address these gaps by examining teachers’ use of ICT for writing instruction from a comprehensive and integrative perspective that brings together instructional practices, teacher-related factors, and contextual conditions.

Specifically, this study pursues three main aims. First, it examines differences between primary and secondary education teachers in their knowledge and training related to ICT and ICT-supported writing instruction, as well as in their use of ICT for teaching writing, considering the writing-related content addressed, the types of instructional and learning activities implemented through ICT, and the ICT tools and resources used. Second, it analyses differences between primary and secondary education teachers in the internal and external factors associated with the use of ICT for writing instruction. Third, it identifies which internal and external factors predict teachers’ use of ICT for writing instruction in primary and secondary education.

To address these aims, the following research questions (RQ) were formulated:

RQ1. To what extent do primary and secondary education teachers differ in (a) their knowledge and training related to ICT and ICT-supported writing instruction, and (b) in their use of ICT for writing instruction, considering the writing-related content addressed, the types of instructional and learning activities implemented through ICT, and the ICT tools and resources used?

RQ2. Are there differences between primary and secondary education teachers in the internal and external factors associated with the use of ICT for writing instruction?

RQ3. Which internal and external factors predict the use of ICT for writing instruction among primary and secondary education teachers?

2. Materials and Methods

2.1. Study Design and Variables

The present study follows a cross-sectional, non-experimental design based on self-report data. Its aim is to examine Spanish language and literature teachers’ use of ICT for writing instruction across educational levels, as well as the internal and external factors associated with such use.

The main outcome variable is teachers’ use of ICT for writing instruction, operationalised as the frequency of ICT use to support writing-related explanations and conduct writing activities with students. Responses were provided on a 7-point scale ranging from never to daily, and an average score was computed to represent overall ICT use for teaching writing.

To characterise ICT-supported writing instruction and examine differences across educational levels, additional descriptive indicators were considered, including the writing-related content addressed through ICT, the types of instructional and learning activities implemented, and the digital tools and resources employed.

A set of explanatory variables was included to account for factors related to ICT integration in writing instruction according to Ertmer’s framework (Ertmer, 1999; Ertmer et al., 2015). Internal factors comprised teachers’ theoretical orientations toward writing instruction, self-efficacy for teaching writing with ICT, and attitudes toward ICT. External factors referred to contextual conditions, including time availability, institutional and social support, and access to ICT resources. Teachers’ ICT-related knowledge and training, as well as sociodemographic and professional variables (e.g., gender, years of teaching experience, educational level) were included as background variables.

This design allows for comparisons across educational levels (1st–3rd primary, 4th–6th primary, and 1st–4th secondary education) and for the examination of associations and predictors (regression-based) of teachers’ frequency of ICT use for writing instruction.

2.2. Participants

The sample consisted of 360 Spanish language and literature teachers from primary and secondary education in northern Spain, obtained through purposive sampling by contacting all primary and secondary schools by phone.

Table 1. Sample distribution by educational level.

	1st–3rd Year Primary (n = 110; 30.56%)		4th–6th Year Primary (n = 121; 33.61%)		1st–4th Year Secondary (n = 129; 35.83%)		Total (n = 360)
Sex	n	%	n	%	n	%	n	%
Female	96	87.27	95	78.51	95	73.64	286	79.44
Male	14	12.73	25	20.66	30	23.26	69	19.17
I’d rather not say	0	0.00	1	0.83	4	3.10	5	1.39
Qualification
Diploma	48	43.64	60	49.59	0	0.00	108	30.00
Bachelor’s Degree	20	18.18	25	20.66	71	55.04	116	32.22
University Degree	28	25.45	19	15.70	7	5.43	54	15.00
Master’s Degree	13	11.82	16	13.22	42	32.56	71	19.72
Doctorate (PhD)	1	0.91	1	0.83	9	6.98	11	3.06
Ownership of the school
Public	88	80.00	93	76.86	96	74.42	277	76.94
Private/semi-private	22	20.00	28	23.14	33	25.58	83	23.06
Years of Teaching Experience
M	17.1		20.5		16.6		18.1
SD	8.07		9.81		10.09		9.55

presents the sample distribution based on educational level, sex, qualifications, school ownership, and years of teaching experience. Based on previous studies (Sánchez-Rivero et al., 2021), primary education was divided into two levels to allow for a more balanced comparison in terms of educational stages, as well as to ensure a more homogeneous number of participants in each group.

2.3. Instruments

After reviewing the existing literature, only one previous study was identified that developed a questionnaire addressing the use of ICT tools in the context of writing instruction (Avidov-Ungar & Amir, 2018). However, that instrument focused on general aspects of ICT integration in first language (L1) instruction and assessment, examining teachers’ technological–pedagogical mastery, frequency of use, and attitudes toward ICT tools. It did not specifically address instructional content, teaching practices, or the types of writing activities carried out with ICT in the classroom.

Consequently, no comprehensive instrument was found that specifically captured teachers’ use of ICT for teaching writing together with the internal and external factors associated with such use. Therefore, an ad hoc questionnaire was developed for the present study using Google Forms. The questionnaire was designed by integrating adapted scales from previous validated instruments for certain constructs (e.g., teachers’ theoretical orientations, self-efficacy, and attitudes) together with newly developed or substantially recontextualized items addressing external conditions and ICT-based writing practices. The instrument development was guided by the phases proposed by Muñiz and Fonseca-Pedrero (2019).

Content validity was assessed through expert judgment, yielding an overall content validity coefficient (CVCtc) of 0.91, which indicates excellent validity. This process also included the evaluation of linguistic clarity, cultural appropriateness, and conceptual alignment of both adapted and newly developed items. The final questionnaire consisted of 60 items distributed across five sections (See Supplementary Material).

The first section includes 11 items assessing sociodemographic variables (e.g., sex and age), the frequency of ICT use (e.g., Approximately how many minutes per week do you dedicate to the use of ICT in your Spanish language and literature classes?), and time devoted to teaching writing, both with (e.g., Approximately how many minutes per week do you dedicate to teaching writing using ICT?) and without ICT (e.g., 10. Approximately how many minutes per week do you spend teaching different aspects of writing in your classes?).

The second section aimed to evaluate teachers’ training for the use of ICT in the classroom, distinguishing between general digital competence (e.g., Do you have any type of certification attesting to your level of digital competence?) and specific preparation for using ICT to teach writing, focusing on their perceived training in this area (e.g., What preparation for teaching writing through ICT do you feel you have received after completing college?). This section comprised a total of 13 items developed based on international digital competence frameworks (e.g., DigComp 2.2; Vuorikari et al., 2022) and aligned with the objectives outlined in Component 19 of the Recovery, Transformation, and Resilience Plan (European Commission, 2023), which emphasizes training and certification in digital skills. The internal consistency of this section was acceptable (Cronbach’s α = 0.77).

The third section addresses external factors influencing ICT use for teaching writing and consisted of nine items assessing time availability to use ICT (e.g., indicate your degree of agreement on whether the time available to use ICTs during classes can have an influence on the use of ICTs to teach writing), the availability of ICT resources (e.g., indicate your degree of agreement on whether the availability of ICT resources at the school can influence the use of ICT to teach writing), and support from key groups for using ICT in the classroom (e.g., indicate your degree of agreement as to whether the support for using ICT by colleagues at the school (management team, teachers, etc. can influence the use of ICTs to teach writing). Items were adapted from Nunes et al. (2022) and were rated on a five-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree). This section showed excellent internal consistency (α = 0.94).

The fourth section, consisting of 25 items, assessed internal factors. First, 13 items on teachers’ theoretical orientations were included (α = 0.70), based on the Writing Orientation Scale by Graham et al. (2002), which differentiates between (a) correct writing, focusing on correct spelling, copying good models, and producing good texts in a single draft (e.g., before students begin a writing assignment, the teacher should remind them to write with correct spelling); (b) explicit instruction, related to the formal teaching of writing skills and strategies (e.g., it is important to teach students planning and/or revision strategies); and (c) natural learning, focused on both formal and incidental learning methods (e.g., students acquire spelling and grammatical rules progressively as they practice and respond to written messages). Importantly, the Writing Orientation Scale does not explicitly address the use of ICT in writing instruction, as its purpose is to capture teachers’ general pedagogical orientations toward writing instruction, independently of the instructional tools employed. These orientations have been previously identified as relevant internal factors associated with teachers’ use of ICT in literacy instruction (e.g., Nunes et al., 2022). Second, six additional items from the Teacher Efficacy Scale for Writing by Graham et al. (2001) were adapted to analyze teachers’ self-efficacy (e.g., I feel able to appropriately use ICT to teach writing) (α = 0.81). Third, six items on attitudes toward ICT (α = 0.88) were included (e.g., ICT can be useful in improving students’ writing proficiency), based on Nunes et al. (2022) and Tondeur et al. (2015). All three scales used a five-point response scale, where 1 = strongly disagree and 5 = strongly agree.

The final section focuses on teachers’ use of ICT for teaching writing and includes ten questions on the frequency of use of different ICT tools and resources (e.g., indicate how often you use ICT to support your writing explanations, for example, through videos or slide presentations), the types of writing content covered using ICT (e.g., select the writing content that you work on in your classes through the use of ICT), and the types of writing activities carried out by students and teachers using ICT (e.g., How often do you use ICT to perform the following writing activities?). These questions were designed based on Sánchez-Rivero et al. (2021), Suárez-Rodríguez et al. (2018), and previous studies (González-Laguna et al., 2024) in which the writing contents and learning and instructional activities were identified. One of the ten questions was open-ended, four used a dichotomous scale (Yes/No), and the remaining items used a 7-point scale, where 1 = never and 7 = daily. The overall reliability was α = 0.93.

Psychometric Properties of the Questionnaire

Psychometric analyses were conducted separately for each section of the questionnaire, considering the nature, origin, and measurement purpose of the items. Exploratory and confirmatory factor analyses were applied only to Likert-type scales intended to measure reflective latent constructs. Descriptive sections (e.g., socio-demographic variables, training background, and frequency of ICT use) were not subjected to factor analysis, as they do not represent latent variables.

For the scale assessing external factors influencing ICT use for teaching writing, an exploratory–confirmatory approach was adopted due to the adaptation and integration of items from previous research and the absence of a clearly established latent structure in this specific context. To provide stronger evidence of construct validity and to avoid capitalization on chance, the total sample (N = 360) was randomly divided into two independent subsamples of equal size.

An Exploratory Factor Analysis (EFA) was conducted on the first subsample (n = 180) using minimum residual extraction with oblimin rotation. Sampling adequacy was excellent (KMO = 0.88), and Bartlett’s test of sphericity was significant (χ²(36) = 1604.83, p < 0.001), confirming the suitability of the data for factor analysis. Parallel analysis and inspection of the scree plot supported a two-factor solution, accounting for 74.8% of the total variance. The first factor, labelled Facilitating Conditions, captured items related to time availability, training, and institutional and social support for using ICT to teach writing. The second factor, labelled ICT Resources, represented the availability of technological resources, including hardware, software, and access to ICT at school. All items showed strong loadings on their intended factors (λ = 0.65–0.94), with no salient cross-loadings and satisfactory communalities.

The two-factor structure was subsequently tested through Confirmatory Factor Analysis (CFA) in the second subsample (n = 180) using the diagonally weighted least squares (DWLS) estimator, given the ordinal nature of the data. The model demonstrated excellent fit according to incremental and residual-based indices, including CFI = 0.976, TLI = 0.967, and SRMR = 0.061. The RMSEA value was elevated (RMSEA = 0.177, 90% CI [0.152, 0.203]), a result that has been widely documented in models with few degrees of freedom, categorical indicators, and DWLS estimation (Kenny et al., 2015; Shi et al., 2019; Xia & Yang, 2019). Under such conditions, RMSEA tends to overestimate model misfit and should be interpreted with caution. In this context, the combination of very high CFI and TLI values together with an acceptable SRMR provides strong support for the adequacy of the proposed measurement model. All standardized factor loadings were statistically significant (p < 0.001) and of high magnitude (λ = 0.65–0.94), indicating that the items were robust indicators of their respective latent factors. Overall, these results provide solid evidence for the factorial validity of the external factors scale and support its use for subsequent analyses in the present study.

The internal factors section comprised three theoretically grounded constructs: teachers’ conceptions about writing instruction, self-efficacy for teaching writing with ICT, and attitudes toward ICT. As these constructs were assessed using scales derived from well-established instruments with a clearly defined a priori factorial structure (Graham et al., 2001, 2002), Confirmatory Factor Analyses (CFAs) were conducted to examine their construct validity in the present sample.

Teachers’ conceptions about writing instruction were assessed through three correlated latent factors representing correct writing (CW), explicit instruction (EI), and natural learning (NL) orientations. A three-factor CFA model was tested and showed a moderate but acceptable overall fit, χ²(51) = 222.18, p < 0.001, with RMSEA = 0.096 and SRMR = 0.085, while incremental fit indices were modest (CFI = 0.801; TLI = 0.742). Despite the non-optimal fit indices, all items loaded significantly on their intended factors (p < 0.001), with standardized loadings ranging from 0.23 to 0.84, and inter-factor correlations being moderate to high (r = 0.32–0.76), supporting a theoretically coherent and interpretable structure. Internal consistency estimates were acceptable for research purposes (ω = 0.62–0.75), in line with previous applications and adaptations of the Writing Orientation Scale (Graham et al., 2002).

Teachers’ self-efficacy for teaching writing with ICT was assessed using a unidimensional CFA model adapted from the Teacher Efficacy Scale for Writing. The model showed good fit, with excellent incremental fit indices (CFI and TLI > 0.95) and low residuals (SRMR < 0.06). All items loaded strongly and significantly on the latent factor, and reliability estimates indicated good internal consistency (α and ω > 0.80).

Teachers’ attitudes toward ICT were also examined through a one-factor CFA model. Results supported a clear and well-defined latent structure, with excellent model fit (CFI and TLI > 0.95; SRMR < 0.06) and high internal consistency (α and ω > 0.85).

Overall, these CFA results provide robust evidence for the construct validity and reliability of both factors, supporting their use in subsequent inferential analyses.

2.4. Statistical Analysis

To analyze differences between primary and secondary education teachers in their knowledge, training, and use of ICT for writing instruction (RQ1), a set of complementary analyses was conducted. First, teachers’ knowledge and training related to ICT and ICT-supported writing instruction were examined. Specifically, two repeated-measures analyses of variance (RM-ANOVAs) were conducted to analyze within-teacher differences. In the first RM-ANOVA, the within-subject factor was the type of digital knowledge (technical, methodological, and pedagogical). In the second RM-ANOVA, the within-subject factor was training for teaching different writing-related content through ICT, including high-level processes, low-level processes, and knowledge of text genres. These analyses also included teachers’ accreditation status and perceived digital training as grouping variables.

Additionally, differences in the use of ICT for teaching writing were examined. Contingency tables were used to explore differences between educational levels in the use of ICT for teaching writing content and learning activities, allowing the identification of the presence or absence of ICT use across instructional domains. In addition, one-way analyses of variance (ANOVAs) were conducted to examine differences in the instructional activities performed through ICT, as well as in the types of ICT tools and resources used.

Second, to analyze differences in internal and external factors related to ICT use for teaching writing (RQ2), one-way ANOVAs were conducted.

Third, to analyze the influence of internal and external factors on teachers’ use of ICT for teaching writing (RQ3), a linear regression analysis was conducted. Prior to this, a correlation matrix was computed to examine the associations between all study variables and the average use of ICT for teaching writing, to identify relevant predictors to be included in the regression model. In line with previous research, sociodemographic variables that have been shown to influence teachers’ use of ICT (e.g., gender, years of teaching experience, and educational level; Nunes et al., 2022) were also considered in the analyses.

Normality was assessed through visual inspection of Q–Q plots. Homogeneity of variances was tested using Levene’s test; when violated, Welch’s ANOVA was applied, followed by Games–Howell post hoc comparisons. For regression analyses, assumptions were evaluated based on residual diagnostics, including normality, homoscedasticity, independence of errors, and multicollinearity (VIF).

All analyses were performed using jamovi software version 2.3.28.

3. Results

3.1. Training and Use of ICT for Teaching Writing Competence

3.1.1. Training in ICT and ICT-Supported Writing Instruction

Significant differences were found in the level of certified digital competence between educational levels (F(2, 301) = 3.89; p = 0.021; η²p = 0.025). Post hoc analysis showed that these differences were observed between secondary teachers (M = 3.23) and those teaching at both primary levels (M = 3.50; p ≤ 0.048). While secondary teachers held certifications ranging from lower-intermediate level (B1) (n = 38) to upper-intermediate level (B2) (n = 36), most primary teachers had a B2 level certification. However, there were no significant differences in teachers perceived levels of digital competence, regardless of their certified level (F(2, 357) = 0.697, p = 0.499, η²p = 0.004). Additionally, more than two thirds of teachers across all educational levels (68.18–70.25%) reported not having received offers of ICT-related training courses.

The RM-ANOVA examining the types of digital content addressed in digital training courses indicated no differences based on educational level, nor any interaction effects between educational level and the types of digital content covered (Fs < 1.17; ps > 0.303). However, significant differences were found among the types of digital content delivered (F(2, 690) = 278.99; p < 0.001; η²p = 0.447). Regardless of educational level, teachers reported that technical knowledge was less frequently addressed in digital training courses (M = 2.71), whereas methodological knowledge (M = 3.52) and pedagogical knowledge (M = 3.51) were covered more frequently.

Finally, with regard to training for teaching writing content through ICT, there were significant differences between types of writing content (F(2, 714) = 34.05, p < 0.001, η²p = 0.087) and between educational levels (F(2, 357) = 3.10; p = 0.046; η²p = 0.017). Differences were found across all writing content types (p < 0.001), with high-level writing processes being the area in which perceived training was lowest. Differences between educational levels were found between secondary teachers (M = 2.12) and teachers in the earlier stages of primary education (M = 1.95; p = 0.064), indicating that secondary teachers perceived themselves as better prepared to teach various writing content through ICT.

3.1.2. Writing Content and Learning Activities Through ICT

Table 2. ICT-based learning content and activities by educational level.

	1st–3rd Primary (n = 110)		4th–6th Primary (n = 121)		1st–4th Secondary (n = 129)
Contents	n	%	n	%	n	%
Grammar	78	70.9	93	76.9	90	69.8
Spelling	90	81.8	98	80.9	93	72.1
Typing/Keyboarding	22	20.0	21	17.4	12	9.3
Handwriting	34	30.9	23	19.0	16	12.4
Planning	27	24.5	48	39.7	66	51.2
Revision	26	23.6	36	29.8	64	49.6
Features of textual genres	35	31.8	55	45.5	82	63.6
Learning activities
Lesson visualization	55	50.0	68	56.2	55	42.6
Planning	21	19.1	33	27.3	52	40.3
Writing	27	24.5	67	55.4	82	63.6
Revision	15	13.6	36	29.8	53	41.1
Analysis of textual genres	24	21.8	43	35.5	76	58.9
Content practice	56	50.9	77	63.3	68	52.7

shows a progressive increase in teaching high-level cognitive writing processes through ICT as the educational level progresses. Secondary-school teachers focused more on content related to planning, revising, and the characteristics of different text genres through ICT than primary-school teachers, who mainly used ICT to teach lower-level cognitive aspects of writing (e.g., spelling). Teachers also indicated that secondary students engaged in a greater number of ICT-based activities involving planning, drafting, revising, and analyzing text genres than primary students. However, upper-primary education, teachers made greater use of lesson visualization and content practice activities.

3.1.3. Instructional Activities

Regardless of educational level, the use of ICT in instructional writing activities was relatively infrequent (

Table 3. Descriptive analysis of ICT usage frequency in instructional activities.

	1st–3rd Primary (n = 110)		4th–6th Primary (n = 121)		1st–4th Secondary (n = 129)		p
Instructional Activities	M	SD	M	SD	M	SD
Previous knowledge activation	4.53	1.81	4.53	1.91	3.67	1.90	<0.001
Support for oral explanations	5.25	1.74	5.22	1.65	4.42	1.87	<0.001
Modelling	3.15	2.04	3.21	1.90	3.05	1.81	0.784
High-quality textual examples	3.41	1.94	3.68	1.84	3.93	1.74	0.095
Self-assessment criteria	2.75	1.87	3.11	1.77	3.17	1.69	0.176
Text publishing	2.65	1.97	3.55	1.83	3.45	1.78	<0.001
Digital writing spaces	2.18	1.64	2.80	1.87	2.95	1.79	0.002
Feedback	2.82	2.09	3.45	1.98	3.49	1.95	0.022
Scaffolding	2.57	2.04	3.29	1.99	3.26	1.86	0.011

). The most common activities were activating prior knowledge and using ICT to support oral explanations, while other activities used ICT less frequently. There were differences between educational levels in the use of ICT for activating prior knowledge (F(2, 237) = 8.48; p < 0.001; η²p = 0.046), supporting oral explanations (F(2, 236) = 8.47; p < 0.001; η²p = 0.048), providing digital writing spaces (F(2, 233) = 7.39; p < 0.001; η²p = 0.043), publishing (F(2, 237) = 6.62; p = 0.002; η²p = 0.033), and providing feedback (F(2, 234) = 3.86; p = 0.022; η²p = 0.022) and scaffolding (F(2, 234) = 4.63; p = 0.011; η²p = 0.027).

The post hoc analysis indicated that secondary teachers used ICT less frequently for activities related to activating prior knowledge and supporting oral explanations than primary teachers did (p ≤ 0.001). Furthermore, teachers from 1st to 3rd grade of primary engaged in activities related to text publishing, digital writing spaces, feedback, and scaffolding less frequently than teachers from 4th to 6th grade of primary and secondary teachers (ps ≤ 0.049).

3.1.4. ICT Tools and Resources

Regardless of educational level, most ICT tools and resources were used infrequently (

Table 4. Descriptive analysis of the frequency of use of ICT tools and resources.

	1st–3rd Primary (n = 110)		4th–6th Primary (n = 121)		1st–4th Secondary (n = 129)		p
	M	SD	M	SD	M	SD
ICT tools
Computer	4.45	2.19	4.75	1.86	3.94	2.11	0.006
Tablet	2.70	2.04	3.15	2.30	2.22	1.97	0.003
Projector	3.86	2.61	4.02	2.50	4.14	2.25	0.686
Mobile phone	2.14	2.12	2.06	2.05	2.57	1.92	0.094
Digital whiteboard	5.57	2.05	5.48	1.93	4.15	2.35	<0.001
TV	1.23	0.84	1.43	1.41	1.36	1.03	0.337
Camera	1.95	1.67	1.72	1.51	1.36	1.00	0.003
Video cameras	1.77	1.51	1.64	1.38	1.39	0.92	0.037
ICT Resources
Word processor	3.10	1.97	4.19	1.74	4.10	1.99	<0.001
Slide presentations	3.15	2.01	3.33	2.05	3.69	2.07	0.113
Grammar/spell checkers	2.67	2.04	3.08	2.02	3.26	2.02	0.082
Intelligent Tutoring Systems (ITS)	1.53	1.28	1.71	1.51	1.53	1.37	0.541
Virtual classes	1.79	1.49	2.05	1.68	1.86	1.69	0.448
Chats and forums	1.69	1.51	1.99	1.80	2.52	2.05	0.002
Blogs	2.12	1.75	2.10	1.79	1.96	1.43	0.692
Social networks	1.82	1.69	1.67	1.54	1.91	1.52	0.470
Digital games	3.81	1.89	3.60	1.81	2.40	1.59	<0.001
Educational videos	4.86	1.79	4.49	1.78	3.47	1.80	<0.001
Email	2.76	2.15	3.93	2.12	3.76	2.07	<0.001

). The most commonly used tools were the digital whiteboard, computer, and projector, with a frequency ranging from once a week to several times per month (M ≥ 4.01). The most frequently used ICT resources, with usage ranging from once to several times per month, included educational videos, word processors, email, and slide presentations (M ≥ 3.40). The rest of the tools and resources were either not used or used only minimally, ranging from once to several times per year.

Based on educational levels, there were significant differences in use of the following tools: computers (F(2, 234) = 5.29; p = 0.006; η²p = 0.027), tablet (F(2, 234) = 5.85; p = 0.003; η²p = 0.033), digital whiteboard (F(2, 236) = 15.79; p < 0.001; η²p = 0.089), and photo (F(2, 215) = 6.05; p = 0.003; η²p = 0.029) and video cameras (F(2, 216) = 3.36; p = 0.037; η²p = 0.016). Secondary-school teachers used computers, tablets, and digital whiteboards less frequently than teachers from 4th to 6th grade primary (ps ≤ 0.004). Secondary teachers also used digital whiteboards, photo cameras, and video cameras less frequently than teachers from 1st to 3rd grade of primary (ps ≤ 0.054).

There were also significant differences between educational levels in the use of five ICT resources: word processors (F(2, 235) = 11.25; p < 0.001; η²p = 0.061); chats and forums (F(2, 238) = 6.43; p = 0.002; η²p = 0.035), digital games (F(2, 231) = 24.85; p < 0.001; η²p = 0.115), educational videos (F(2, 236) = 19.49; p < 0.001; η²p = 0.099), and email (F(2, 235) = 9.99; p < 0.001; η²p = 0.054). The differences were primarily observed between secondary and primary teachers. Primary teachers used digital games and educational videos more frequently than secondary teachers (ps ≤ 0.001). However, secondary teachers used chats and forums more often than 1st to 3rd grade of primary teachers (ps < 0.001). In addition, secondary teachers and teachers in 4th to 6th grade primary used email and word processors more frequently than teachers in 1st to 3rd grade primary (ps ≤ 0.001).

3.2. Internal and External Factors in Teaching Writing Through ICT

In the analyses of variance, Welch’s ANOVA was applied to compare differences in internal and external factors according to the educational level at which teachers work. This analysis identified significant differences only in attitudes about the use of ICT across educational levels (F(2, 237) = 3.49, p = 0.024). Specifically, these differences were observed between teachers in secondary education (Grades 1–4; M = 3.36) and those in early primary education (Grades 1–3; M = 3.59, p = 0.036), as well as between secondary education and upper primary education (Grades 4–6; M = 3.59, p = 0.038). For the remaining external and internal factors that may influence ICT use, no significant differences were found across teachers’ educational levels (Fs < 2.07, ps > 0.128) (

Table 5. Descriptive analysis of internal and external factors related to ICT use.

	1st–3rd Primary (n = 110)		4th–6th Primary (n = 121)		1st–4th Secondary (n = 129)		p
	M	SD	M	SD	M	SD
Internal Factors
Orientation: Correct writing	3.44	0.61	3.35	0.63	3.52	0.66	0.128
Orientation: Explicit instruction	4.15	0.63	4.21	0.66	4.28	0.58	0.279
Orientation: Natural learning	3.45	0.59	3.46	0.68	3.34	0.62	0.233
Self-efficacy	3.11	0.59	3.09	0.74	3.06	0.78	0.838
Attitudes	3.59	0.60	3.59	0.65	3.36	0.84	0.024
External factors
Facilitating Conditions	377	0.99	3.72	1.01	3.82	0.80	0.715
ICT Resources	3.78	1.26	3.83	1.18	3.90	1.01	0.972

).

3.3. Influence of Internal and External Factors on ICT Use for Teaching Writing

To examine the association between internal and external factors and the use of ICT for teaching writing (supporting explanations and conducting writing activities), Pearson correlation analyses were conducted. The results revealed significant positive correlations between the average use of ICT for teaching writing and external factors, specifically facilitating conditions (r = 0.138, p = 0.009) and resources (r = 0.107, p = 0.042). Significant positive associations were also observed between ICT use and internal factors, including self-efficacy (r = 0.403, p < 0.001) and attitudes (r = 0.300, p < 0.001).

Regarding sociodemographic variables, only years of teaching experience and general digital competence showed significant associations with ICT use. Specifically, ICT use was positively associated with years of teaching experience (r = 0.116, p = 0.028), whereas a negative association was found with digital competence (r = −0.128, p = 0.015). Sex and educational stage were not significantly associated with ICT use.

For the linear regression analysis, an initial model was first estimated including the factors that showed significant associations in the correlation analysis to examine their influence on the use of ICT for teaching writing (

Table 6. Regression model.

Model 1	Estimate	SE	t	p
Educational Stage	−0.154	0.112	−1.368	0.172
Teaching Experience	0.014	0.009	1.520	0.129
Digital Competence	0.230	0.117	1.966	0.050
Self-Efficacy	0.771	0.150	5.136	<0.001
Attitudes	0.213	0.152	1.400	0.162
Facilitating Conditions	0.074	0.169	0.441	0.659
ITC Resources	−0.001	0.136	−0.008	0.993
Model 2	Estimate	SE	t	p
Self-efficacy	0.899	0.129	6.967	<0.001
Digital Competence	0.264	0.116	2.286	0.023

). Assumption testing indicated that the normality assumption was met (Kolmogorov–Smirnov, p = 0.293), no evidence of heteroscedasticity was observed (p > 0.594), the independence of errors assumption was satisfied (Durbin–Watson, p = 0.422), and no multicollinearity issues were detected among predictors (VIF < 3.44). This model explained 19.9% of the variance in the use of ICT for teaching writing (p < 0.001). However, none of the external factors showed a significant effect (ps > 0.659), nor did attitudes (p = 0.162), years of teaching experience (p = 0.129), or educational stage (p = 0.172). Digital competence showed a marginal effect (p = 0.050), whereas self-efficacy showed a positive and statistically significant effect (p < 0.001).

Based on these results, a second regression model was estimated including only self-efficacy and digital competence (Table 6). Assumption testing for this second model indicated that the normality assumption was met (Kolmogorov–Smirnov, p = 0.619), no evidence of heteroscedasticity was observed (p > 0.130), the independence of errors assumption was satisfied (Durbin–Watson, p = 0.244), and no multicollinearity issues were detected among predictors (VIF < 1.06). Regarding the results, the linear regression model was statistically significant and explained 17.8% of the variance in the use of ICT for teaching writing (R² = 0.178). Self-efficacy positively predicted the use of ICT in writing instruction (B = 0.90, p < 0.001), as did digital competence (B = 0.26, p = 0.023). In this second model, self-efficacy emerged as the key predictor, showing a positive and highly significant effect, whereas digital competence showed a positive but smaller effect compared to self-efficacy.

4. Discussion

This study aimed to examine differences between primary and secondary education teachers in their knowledge and training related to ICT and ICT-supported writing instruction, as well as in their use of ICT for teaching writing, considering the writing-related content addressed, the instructional and learning activities implemented through ICT, and the ICT tools and resources used. In addition, the study sought to analyze differences in the internal and external factors associated with teachers’ use of ICT for writing instruction and to identify which of these factors predict ICT use in writing instruction across educational levels. To address these aims, three research questions were formulated. The first research question examined the extent to which primary and secondary education teachers differ in their knowledge and training related to ICT and ICT-supported writing instruction, as well as in their use of ICT for teaching writing, considering the content, activities, and technological resources involved. The second research question focused on differences between educational levels in internal and external factors associated with ICT use for writing instruction. Finally, the third research question aimed to identify the key internal and external predictors of teachers’ use of ICT for writing instruction in primary and secondary education. Based on the results, the following sections discuss the main findings and contributions of the study for each research question.

In relation to the first research question, the results indicate that teachers across educational stages generally reported an intermediate level of digital competence, with slightly higher levels among primary teachers. This finding contrasts with previous studies reporting lower levels of digital training in primary education teachers (e.g., Portillo-Berasaluce et al., 2022) and may reflect differences in teacher education models across countries, where curricular content and pedagogical approaches to ICT integration vary by educational level. In line with previous research (Yang et al., 2022), secondary teachers appear to receive more specialized, domain-specific ICT training, whereas primary teachers tend to receive more generalized, cross-curricular training. In this context, the limited availability of ICT-related training courses reported by teachers is particularly noteworthy, as such opportunities could help address existing training needs and reduce disparities in the effective pedagogical use of ICT across educational stages (Gutiérrez-Martín et al., 2022).

Considering teachers’ specific training in the use of ICT for writing instruction, it is important to note that perceived training in this area was low and unevenly distributed across writing processes. Training related to low-level cognitive processes (e.g., transcription, spelling, and surface-level aspects) was consistently higher than training focused on higher-level processes such as planning, revision, and work on textual genres, across all educational stages. In line with this interpretation, previous research has shown that most ICT-related training opportunities and commonly used digital tools tend to focus on supporting low-level writing processes, whereas fewer resources and professional development initiatives explicitly address higher-level processes and self-regulation in writing (González-Laguna & Fidalgo, 2021; Strobl et al., 2019). Importantly, although secondary teachers reported feeling better prepared to teach writing through ICT than primary teachers, perceived training in higher-level writing processes remained low even at higher educational stages.

From an instructional perspective, this limited preparation in higher-level writing processes is reflected in teachers’ pedagogical use of ICT as shown in the present study. Overall, ICT-supported writing instruction was primarily oriented toward low-level cognitive processes, particularly in the early grades of primary education. In contrast, higher-level writing processes were addressed more frequently through ICT as the educational level increased, although their overall presence remained limited. This progression is consistent with previous findings indicating that instructional focus in writing tends to shift toward more complex processes as students advance through compulsory education (Aznárez-Mauleón, 2022).

Importantly, from an educational perspective, this pattern reveals a misalignment between current instructional practices and evidence-based recommendations for writing instruction. Research has consistently emphasized the need to foster higher-level writing processes from the earliest stages of writing development (e.g., Arrimada et al., 2018), as effective writing requires the coordinated development of both low- and high-level cognitive processes (Sánchez-Rivero et al., 2021). The limited use of ICT to support these processes in compulsory education and specifically in the early educational stages may therefore restrict students’ opportunities to develop robust writing competence.

A similar relationship between teachers’ preparation and instructional practice was observed in the implementation of specific teaching activities through ICT. While teachers frequently reported using ICT to support explanations and content practice, practices with a stronger instructional impact, such as providing feedback, scaffolding, offering digital writing spaces, and publishing students’ texts, were less commonly implemented, particularly in the early grades of primary education. In line with Ertmer’s framework (Ertmer et al., 2015; Tondeur et al., 2017), these findings suggest that insufficient pedagogical knowledge and training may partly explain the reduced implementation of these high-impact practices through ICT.

Taken together, these results indicate that the integration of ICT in writing instruction remains largely instrumental rather than pedagogically transformative. Beyond confirming previous trends, the present study contributes to the literature by offering a comprehensive, writing-specific analysis that integrates teachers’ training, writing-related content, instructional practices, and digital tools across compulsory educational stages. By explicitly linking teachers’ preparation to their pedagogical uses of ICT, this fine-grained perspective advances existing research grounded in Ertmer’s model (Ertmer, 1999) and identifies concrete targets for teacher education and professional development aimed at enhancing the instructional use of ICT for writing.

In relation to the second research question, the results indicate that differences between primary and secondary education teachers in the internal and external factors associated with ICT use for writing instruction were limited. Statistically significant differences emerged only in teachers’ attitudes toward the use of ICT, with secondary teachers reporting slightly less positive attitudes than teachers in both early and upper primary education. In contrast, no significant differences were found across educational stages in the remaining internal and external factors examined.

Focusing first on internal factors, the fact that attitudes were the only dimension differentiating educational stages is particularly relevant. Teachers’ self-efficacy for using ICT in writing instruction and their general teaching orientations did not vary significantly across primary and secondary education, suggesting that teachers at different stages perceive themselves as similarly capable of using ICT and share comparable pedagogical orientations. However, secondary teachers’ slightly less positive attitudes toward ICT may reflect the specific instructional and curricular demands of this educational stage. This result may partly reflect the fact that writing instruction in secondary education is frequently situated within content-heavy curricula and high-stakes assessment contexts, potentially limiting teachers’ openness to experimenting with ICT-based instructional practices and shaping their perceptions of the pedagogical value of digital tools for writing (Tondeur et al., 2017). This result is consistent with previous research showing that teachers’ attitudes toward ICT are sensitive to contextual and curricular pressures and may vary even when levels of self-efficacy and access to resources remain comparable (e.g., Nunes et al., 2022). In contrast, primary education contexts may offer greater curricular flexibility, leading teachers to perceive ICT as a more supportive resource for learning, engagement, and instructional innovation in writing.

Turning to external factors, the absence of significant differences across educational stages in facilitating conditions and access to ICT resources suggests that the contextual conditions for using ICT in writing instruction are relatively homogeneous. This finding indicates that disparities in infrastructure, availability of digital tools, or institutional support are unlikely to explain differences in teachers’ use of ICT for writing across primary and secondary education. This interpretation is consistent with evidence reported both at the national and international levels, which indicates that access to ICT resources in schools is generally widespread and relatively stable across educational stages in Spain and across educational systems worldwide (OECD, 2019; Ministerio de Educación y Formación Profesional, 2019).

In relation to the third research question, the results show that teachers’ self-efficacy for teaching writing with ICT emerged as the strongest and most consistent predictor of ICT use in writing instruction. Although several internal and external factors were associated with ICT use at the correlational level, the regression analyses indicated that, when these factors were considered simultaneously, self-efficacy was the variable that uniquely and robustly explained teachers’ instructional use of ICT for writing.

This finding can be meaningfully interpreted in light of Ertmer’s (1999, 2005) model of technology integration. In line with this framework, once basic external conditions are in place, teachers’ internal factors, particularly those closely related to instructional decision-making, play a decisive role in shaping classroom practices. In the present study, self-efficacy, understood as teachers perceived capability to teach writing effectively using ICT, functioned as a proximal determinant of practice, outweighing both contextual variables and more general dispositions toward technology such as attitudes.

The prominent role of self-efficacy is theoretically coherent given the specific demands of writing instruction. Teaching writing involves coordinating complex instructional processes, such as planning, drafting, revision, and feedback, which require deliberate pedagogical decision-making and may increase teachers’ perceptions of risk when integrating digital tools. In cognitively demanding domains such as writing, teachers’ confidence in their ability to apply ICT pedagogically appears to be a more decisive driver of instructional use than general attitudes or access to resources, a pattern also reported in previous research across educational contexts (Tondeur et al., 2017).

Although attitudes toward ICT and contextual variables showed significant associations with ICT use at the correlational level, their predictive value did not remain significant once self-efficacy was included in the regression models. This suggests that favourable dispositions toward ICT or facilitating conditions may operate as more distal influences, which are insufficient to translate into instructional use unless accompanied by a strong sense of pedagogical competence. From this perspective, self-efficacy appears to act as a key mechanism through which teachers’ general beliefs and attitudes toward ICT are enacted—or not enacted—in instructional practice (Demircan et al., 2025; Scherer et al., 2019).

In addition to self-efficacy, general digital competence showed a positive but modest contribution to the prediction of ICT use for writing instruction. This result is consistent with research highlighting the importance of teachers’ digital competence as a foundational condition for ICT integration (European Commission, 2023; Portillo-Berasaluce et al., 2022). However, the comparatively smaller effect of digital competence in the present study suggests that technical skills alone are insufficient to explain instructional use in a complex domain such as writing. Rather, teachers’ ability to translate digital skills into pedagogically meaningful practices appears to depend largely on their perceived instructional competence.

Importantly, external factors such as facilitating conditions and access to ICT resources did not contribute to explaining ICT use in the regression models. Rather than contradicting previous findings, this result complements the discussion of the second research question by indicating that, within the studied context, external conditions for ICT use are relatively stabilized and do not account for meaningful variation in instructional practices. Once minimum access is ensured, differences in teachers’ use of ICT for writing are better explained by internal, pedagogically oriented factors (Tondeur et al., 2017).

Taken together with the findings of the second research question, these results extend Ertmer’s model to the specific and cognitively demanding domain of writing instruction, an area that has received limited attention in previous research. By jointly examining internal and external factors and linking them to actual instructional use, the present study provides domain-specific empirical evidence showing that self-efficacy constitutes the central leverage point for promoting meaningful ICT integration in writing instruction. These findings underscore the need for professional development initiatives that move beyond access-oriented or purely technical approaches and explicitly aim to strengthen teachers’ confidence in teaching writing with ICT.

4.1. Limitations and Future Research Directions

The main limitations of the present study are related to the sampling procedure and the characteristics of the sample. In this type of research, participants tend to be teachers who show a higher affinity with the construct being measured. Consequently, it is possible that the sample overrepresents teachers who are more motivated to integrate ICT into their classroom practices. Future studies should therefore consider alternative sampling strategies, such as stratified random sampling, to reduce potential self-selection bias.

Additionally, the sample would need to be expanded to include teachers from other autonomous communities in order to enhance the generalizability of the findings, as the present study was conducted in a single Spanish region. Nevertheless, the existence of common digital competence frameworks at the European level and a shared national curriculum that establishes minimum requirements for ICT training suggests that the results may be cautiously generalizable to similar educational contexts.

Moreover, although the questionnaire was grounded in previously validated instruments and its psychometric properties were examined using a large sample and independent exploratory and confirmatory analyses, some measurement-related limitations should be acknowledged. First, the absence of a formal pilot study constitutes a methodological constraint. A pilot phase could have provided additional evidence regarding item clarity and preliminary dimensionality prior to large-scale administration, particularly for adapted or context-specific items. Second, although the exploratory and confirmatory models for most latent constructs showed good to excellent fit, some scales yielded acceptable but not optimal fit indices. While the factor structures were theoretically coherent and items loaded significantly on their intended dimensions, these results suggest that the questionnaire would benefit from further refinement. Accordingly, future research should consider combining pilot testing with mixed-method validation approaches (e.g., cognitive interviews or item-level qualitative feedback) and revisiting the factorial structure of certain sections to further strengthen construct measurement.

Another limitation concerns the exclusive reliance on self-report measures to assess teachers’ use of ICT for writing instruction and the associated internal and external factors. Although self-report questionnaires provide valuable insights into teachers’ perceptions, beliefs, and reported practices, they may not fully capture how ICT is actually implemented in classroom settings. Future studies could therefore complement self-report data with qualitative methods, such as interviews or classroom observations, to triangulate findings and gain deeper insight into teachers’ instructional decision-making and the ways in which ICT is used to support writing instruction in practice.

Finally, the present study focused exclusively on teachers and classroom-based uses of ICT for writing instruction. While teachers play a central role in shaping instructional practices, writing development is also influenced by learning experiences beyond the school context, particularly within the family and home environment. Previous research has highlighted the importance of family practices and home literacy environments in supporting students’ writing development, including the growing role of digital tools in these contexts (e.g., Neumann, 2018). Future research would therefore benefit from adopting a more ecological perspective by examining how ICT-supported writing practices are also promoted at home, how families engage with digital tools to support writing, and how school- and home-based practices may interact. Such an approach would provide a more comprehensive understanding of the conditions that support the development of writing competence in digitally mediated learning environments.

4.2. Theoretical Contribution

This study contributes to research on ICT integration in education by extending existing theoretical frameworks to the specific and cognitively demanding domain of writing instruction. While prior research has largely examined teachers’ ICT use from general or cross-curricular perspectives (e.g., Almerich et al., 2024), the present study applies an internal–external framework to analyze ICT use specifically for teaching writing, an instructional domain that has received comparatively limited attention in educational technology research.

By organizing teachers’ training, instructional practices, and ICT use within Ertmer’s internal–external framework (Ertmer, 1999), this study provides a domain-sensitive account of how different factors interact to shape ICT-supported writing instruction. In line with subsequent work building on this framework (Tondeur et al., 2017), the findings show that when external conditions are relatively homogeneous, meaningful variation in teachers’ instructional use of ICT can be explained primarily by internal factors closely linked to pedagogical decision-making.

More specifically, the results identify teachers’ self-efficacy for teaching writing with ICT as a key explanatory mechanism linking preparation, beliefs, and instructional practice. While attitudes and contextual factors were associated with ICT use at a correlational level, self-efficacy emerged as the most proximal predictor when multiple factors were considered simultaneously. This finding refines existing models of technology integration by highlighting the importance of domain-specific instructional confidence, particularly in complex learning domains such as writing that require the coordination of multiple cognitive and pedagogical processes (Graham, 2018).

4.3. Educational Implications

The findings of this study have important implications for educational policy and teacher education, both in initial teacher training and throughout teachers’ professional careers. Overall, the results indicate that integrating ICT into writing instruction requires more than access to technology or the development of general digital competence. Instead, effective ICT-supported writing instruction demands targeted pedagogical training that explicitly addresses how digital tools can be used to support key writing processes and instructional practices. These findings may inform the design of specific courses, workshops, or professional development programs aimed at equipping teachers with the knowledge and skills needed to integrate ICT meaningfully into writing instruction (Grgić & Bolliger, 2025).

In this regard, the results highlight the central role of teachers’ self-efficacy in shaping the instructional use of ICT for writing. Teachers who feel confident in their ability to use digital tools to teach writing are more likely to incorporate ICT in pedagogically meaningful ways, whereas those with lower self-efficacy may avoid or limit their use, even when technical skills and access to resources are available. Consequently, professional development programs should not only focus on enhancing teachers’ technical knowledge, but also explicitly aim to strengthen their instructional confidence and self-efficacy for teaching writing with ICT.

To achieve this, professional development initiatives should adopt practice-based models that provide teachers with opportunities to engage in hands-on experiences with ICT-supported writing activities, receive guided support, and reflect on their instructional decisions in authentic teaching contexts. Such approaches can simultaneously foster digital competence, pedagogical knowledge, and self-efficacy, thereby promoting sustained changes in instructional practice (e.g., Harris et al., 2023). From a policy perspective, these findings suggest that investments in teacher development should prioritize pedagogically oriented professional development over purely access- or infrastructure-focused measures, creating more favorable conditions for enhancing students’ writing competence in digitally mediated learning environments.