Modeling Teaching Using Information and Communication Technologies in Early Childhood Education with Functional Diversity: The Case in Spain

Abstract

(1) Background: The objective of this study was to verify a model proposed adapted to the case of teachers working in the field of early childhood education of children with functional diversity. The model analyses the relationship between the perceptions, use, and necessity of Information and Communication Technologies (ICTs) and the level of ICT training. (2) Methods: Snowball sampling was used to acquire a random sample of 254 in-service teachers working in early childhood education centers including children with functional diversity in Andalusia (Spain). A partial least squares–structural equation model was developed. (3) Results: A positive relationship between teachers’ perceptions of ICTs and their use was found, corroborating the findings of previous studies on pre-service teachers working with students with functional diversity. However, unlike previous studies, no significant relationship was found between ICT use and needs and the level of ICT training. The novelty of this study is that the participants are in-service teachers, the education context is early childhood, and the model includes age as a control variable, which had a negative impact in this case in Spain. (4) Conclusions: The importance of improving teacher training to optimize the use of ICTs is emphasized. The professionals interviewed highlighted the need for early assessments and increased material and human resources in public educational institutions. The urgency of administrative restructuring to expedite the delivery of financial aid and the recruitment of more specialized professionals is underscored.

1. Introduction

Digital technology can improve opportunities for quality learning as well as equality and equity for all students (Fernández Batanero et al., 2017; Legorburu-Fernández et al., 2023; Pegalajar Palomino, 2015), regardless of their functional abilities (Bonilla-del-Río et al., 2018; Mañas-Viniegra et al., 2023). Similarly, digital technology provides benefits by increasing (among other things) self-esteem, autonomy, sensory stimulation, motivation, and attention, and can reduce frustration in response to errors (Mas García & Jordá, 2023). Additionally, it can improve the feasibility of teamwork development, increase the likelihood of modifying and adjusting evaluation methods, and enhance bidirectional communication between teachers and students (Morales García, 2023; Pegalajar Palomino, 2017).

Furthermore, digital technology enhances teachers’ competencies; however, prior training or education is necessary (Rodríguez-García et al., 2019; Shin et al., 2023). The initial training of teachers needs to be strengthened to improve their digital technology skills (Colomo-Magaña et al., 2023; Papavlasopoulou et al., 2024; Toledo Morales & Llorente Cejudo, 2016). Additionally, when working, it is important to ensure that student distraction is minimal and that learning is facilitated through educational applications (Pegalajar Palomino, 2015). Therefore, educational institutions must devise educational strategies, develop evaluation types, support the use of ICTs, and create a favorable learning environment (Iancu, 2024; Vega-Angulo et al., 2021).

Digital technology is a powerful tool to help teachers provide more inclusive education (Budnyk & Kotyk, 2020; Fernández Batanero, 2018; Mañas-Viniegra et al., 2023; Pegalajar Palomino, 2017; Polat et al., 2024; Sanahuja Ribés et al., 2020; Silva Sández & Rodríguez Miranda, 2018). The requirements and potential of ICTs are significant and require specific training to ensure effective and accessible implementation (Fernández Batanero & Rodríguez-Martín, 2017; López-Meneses & Fernández-Cerero, 2020). At the same time, it is important to change mindsets and attitudes towards developing critical reflective, collaborative, and creative teaching practices among teachers (Trujillo Torres et al., 2011; Eickelmann & Vennemann, 2017) and their subsequent adoption (Jordá Fabra et al., 2023). Teachers can motivate learning through personalization, access to information, interaction, communication and collaboration, flexibility, immediate feedback, memory support, and organization (Gallardo Montes et al., 2020; Spiteri & Chang Rundgren, 2017), which allow for the promotion and development of multiple literacies and constructive learning (Rodríguez Correa & Arroyo González, 2014).

The term “functional diversity” is often used instead of deficit, limitation, restriction, barrier, “disability”, or “handicap”, in order to emphasize that differences can be a natural part of human functioning. Instead of focusing on a person’s limitations, it emphasizes their abilities and capacities and the diversity in the way in which they can contribute to and participate in society in a valuable way, thereby integrating with conventional societal norms (Mañas-Viniegra et al., 2023). Teachers must provide opportunities for students to contribute and meet their needs and interests (Fernández Batanero & Rodríguez-Martín, 2017), raise awareness about the importance of their inclusion (Legorburu-Fernández et al., 2023), and ensure early interventions to reach their maximum potential (Benítez-Lugo et al., 2021). To achieve this, training in functional diversity and the development of creative activities are necessary (Bonilla-del-Río et al., 2018). Additionally, we can rely on the use of play and emotions to optimize learning, awareness, empathy, and inclusion for children with and without functional diversity (Benítez-Lugo et al., 2021).

Some individuals with functional diversity face barriers in accessing digital devices and platforms. To address this, it is necessary and crucial to consider digital accessibility and Universal Design for Learning (UDL) (Ok & Rao, 2019). This approach enables flexibility, adaptability, and personalized learning materials tailored to individual needs (Jordá Fabra et al., 2023). This includes designing user-friendly interfaces, using accessibility standards, providing options to adjust content presentation, alternative navigation options, and support for assistive devices and technologies (Silva Sández & Rodríguez Miranda, 2018). Specifically, this involves designing accessible websites, screen readers, video subtitles, audio transcriptions, adaptable user interfaces, augmentative and alternative communication (AAC) applications, assistive devices, augmented reality, virtual reality, social networks, and collaboration platforms (López-Meneses & Fernández-Cerero, 2020). It is necessary to provide skills and competencies for teachers’ professional learning and development (Fernández-Batanero et al., 2020). This underscores the need for training in web design and application development that meets accessibility standards (Fernández Batanero et al., 2018).

For students in general, digital technology enhances their academic training and removes social and architectural barriers to achieve inclusion (Vega-Gea et al., 2021). In the specific case of students with significant cognitive difficulties (SCDs), it is crucial for the teacher–student relationship to be functional, offering the student the opportunity to acquire skills and overcome limitations associated with their SCDs. This includes support for auditory comprehension (Nasir-Tucktuck et al., 2023, p. 209), visual connection comprehension (Malone et al., 2023), and cognitive or motor skills (Toledo Morales & Llorente Cejudo, 2016).

Literature Review and Hypotheses

In the literature, it is evident that the vast majority of early childhood education (preschool) teachers are at a satisfactory level of training in terms of information literacy and communication for early childhood education specialists. In Romania, a qualitative focus group method was used in a group of 15 female teachers from different kindergartens in Bucharest, aged between 32 and 58, to gather their opinions on the integration of information technologies in preschool learning (Iancu, 2024). This study revealed that improvements are needed both in the teaching content and in the organizational and pedagogical support to improve preschool teachers’ information literacy and communication competency. Preschool teachers at the kindergarten level are open to using new digital methods and tools (Iancu, 2024). In a systematic literature review, it was noted that there are a limited number of studies specifically focusing on the early childhood stage, necessitating the reliance on studies from the primary stage that analyze these dimensions (Iancu, 2024).

In their literature review, Eickelmann and Vennemann (2017) referred to a study comparing three European countries: Germany, Norway, and the Czech Republic. In this review, they stated that a 2013 study by the International Association for the Evaluation of Educational Achievement found that positive attitudes and beliefs are considered crucial determinants and predictors for teachers’ use of ICTs in teaching, although the strength of this association varied among countries.

Several recent studies in Spain focused on early childhood and ICTs. Pegalajar Palomino (2017) described the perceptions of 231 university students (pre-service teachers) studying early childhood education and primary education in Murcia, Spain, regarding the use of ICTs for the development of inclusive practices in regular classrooms. The results showed favorable perceptions among the future teachers on the use of ICTs. Fernández-Batanero et al. (2019) investigated the level of training and technological knowledge of 777 primary education teachers from Spain regarding the use of ICTs with individuals with disabilities (functional diversity) and found low levels of teacher proficiency. The analysis of variance (ANOVA) showed significant differences in the effects of different socio-demographic characteristics on teacher training; that is, the level of training was influenced by personal factors (gender, age), professional factors (teaching experience), and educational factors (qualifications).

In other countries, as the Republic of Serbia, Novković Cvetković et al. (2022) analyzed teachers’ attitudes towards the use of information and communication technologies in teaching. The study involved 269 primary school teachers in the Republic of Serbia and found that the teachers frequently used ICTs in their teaching. According to the results, the greatest perceived advantages of using ICTs was the improvement in teaching quality, the ability to conduct interesting lessons, and the quick and easy access to information.

In Agarbe (Portugal), Carrapiço et al. (2022) identified the socio-professional characteristics of 156 teachers in the first cycle of basic education and their relationship with the use of ICTs in their classrooms. They found that female teachers use ICTs more frequently (on a weekly basis), when compared to their male counterparts. Additionally, the study revealed that higher levels of education and training correlated with increased hours of ICT implementation in the classroom.

In Saudi Arabia, Shater et al. (2023) investigated the perceptions of 124 teachers in public schools regarding digital learning for students with disabilities and revealed that they had a positive attitude towards digital learning.

Gözüm et al. (2023) and Papadakis et al. (2024) adapted the Teacher Self-Efficacy Scale in the Use of ICT at Home for Preschool Distance Education (TSES-ICT-PDE) for preschool teachers in Turkey and Greece, respectively. They found that it was a reliable and valid scale for confirmatory analysis in both countries. Papadakis et al. compared the knowledge, communication, and technology factors of Greece and Turkey and found no significant differences. They pointed out that the key to enhancing ICT use in early childhood education is a training approach that emphasizes pedagogical over technological knowledge and suggested strategies to address the practical barriers to technology use in classrooms, such as hardware malfunctions, insufficient institutional and technical support (Kabadayı, 2006; Simsar & Kadim, 2017; Gözüm & Kaya, 2024), and a lack of systematic organization in teacher training programs (Gözüm & Kaya, 2024).

Imasuen and Iyamu (2024) studied the attitude towards and knowledge and application of ICTs in early childhood education among teachers in the Benin metropolis (Nigeria). The study included 560 teachers and found that the teachers’ attitudes towards ICTs in early childhood education were positive. The levels of knowledge and application of ICTs were moderate, and the teachers’ sex, age, and experience significantly predicted their attitude towards and knowledge and application of ICTs in teaching.

In the specific context of functional diversity in Spain, Colomo-Magaña et al. (2023) analyzed 284 future teachers’ perceptions regarding the role of ICTs in addressing diversity. They observed positive perceptions regarding the use of ICTs with students with functional diversity and including them in their training. The study revealed significant differences based on gender in the post-test phase.

We posed the following research questions regarding the use of ICTs by teachers working in early childhood education with children with functional diversity.

Regarding the influence of ICT perceptions, ICT use and the need for ICT training, we have established the following research questions (RQ);

• (RQ1) Which gap does our study fill?
• (RQ2) What studies support the proposed PLS-SEM model?
• (RQ3) Which relationships can be used for the structural equation model that works with the latent variables included in the model?
• (RQ4) Which latent variable has the greatest impact on the dependent variable (ICT training)?
• (RQ5) What is the consensus in the literature on this topic and what are the studied relationships with demography characteristics (age, years of experience, gender, and level of education)?

Now, we can respond to some of these questions:

(Response RQ1) Which gap does our study fill? Our study can address the gaps raised by Fernández-Batanero et al. (2019) by extending the analysis to different regions and to early childhood education. We analyzed the quality of preschool functional diversity education in a specific regional context: the Autonomous Community of Andalusia.

(Response RQ2) What studies support the PLS-SEM model? Gallardo Montes et al. (2020) and Colomo-Magaña et al. (2023) were the main references for our study. Gallardo Montes et al. (2020) validated scales using expert judgments and subsequent exploratory analysis. The validation process involved 122 educational and health professionals from the formal and non-formal education sectors, who were teachers of people with functional diversity, without specifying the educational level they were teaching at. Meanwhile, Colomo-Magaña et al. (2023) used these scales in the context of pre-service teachers in the education of young children with functional diversity. The latent variable’s validity was first verified through exploratory factor analysis (EFA) and later through confirmatory factor analysis (CFA), which were used to estimate a covariance-based structural equation model (CB-SEM).

(Response RQ3) Which relationships can be used for the structural equation model that works with their latent variables? The following hypotheses were proposed in the context of educating young children with diversity based on Colomo-Magaña et al. (2023):

H1 (H1+).

Professionals’ perceptions positively influence their uses and the necessity of ICTs in early childhood education for children with functional diversity (ICT_PERCEPT).

H2 (H2+).

Professionals’ perceptions on applying ICTs in early childhood education for children with functional diversity (ICT_PERCEPT) positively influence their level of ICT training for use with children with functional diversity (ICT_NECESS).

H3 (H3+).

The uses and necessity of using ICTs for teaching children with functional diversity (ICT_NECESS) positively influence professionals’ level of ICT training for early childhood education for children with functional diversity (ICT_TRAINING).

H4 (H4+).

Professionals’ perceptions of ICT training for early childhood education for children with functional diversity positively and indirectly influence their level of ICT training (ICT_TRAINING) for working with children with functional diversity based on uses and needs.

H5 (H5-).

Age impacts negatively the level of ICT training (control variable).

(Response RQ4) The response to RQ4 is given in the Results section.

(Response RQ5) What is the consensus in the literature on this topic and what are the studied relationships with demography characteristics (age, years of experience, gender, and level of education)? These characteristics are usually used as control variables, but no academic references are cited in most of the published articles. Although there may be studies that considered age as a control variable in educational research, we did not find any similar papers in our review of the relevant literature in relation to ICT training. To the best of our knowledge, the inclusion of age as a control variable in an SEM in this specific context, where training in ICT for early childhood education for children with functional diversity is the dependent variable, represents a novel contribution of our model. Nevertheless, we recommend that future research further investigate this aspect to confirm if this is an underexplored area.

Fernández-Batanero et al. (2019) conducted a survey of 341 physical education teachers from primary schools in the 17 Autonomous Communities of Spain concerning ICT training for teaching people with disabilities. Significant differences based on age and digital competence (DC) were found using ANOVA. The results indicated that the younger teachers (aged 20–30, 31–40, and 41–50) had higher DC compared to older teachers (51–60 years), which is in line with the findings from other studies (Alqurashi et al., 2017; Gudmundsdottir & Hatlevik, 2017). This higher level of digital knowledge may be due to the younger teachers being closer to the end of their studies, and the recent inclusion of training elements on the use of ICTs for teaching people with disabilities in teachers’ university curricula. However, Fernández-Batanero et al. (2019) argued for the need for a curricular re-structuring for teacher training in Spain, as specializations that provided students with more generalized knowledge have recently disappeared. This may have had negative repercussions on the practical aspects of the training activities for people with disabilities. The results also pointed out the individual efforts of the teachers to supplement the lack of initial training with experience (Fernández-Batanero et al., 2019).

Regarding the correlation between age and digital competence in general terms, in the Spanish context, there are several contradictory results. Some studies presented a negative correlation: there was an inverse relationship between age and digital competence (DC) level in 520 teachers from all over Spain teaching in early childhood, primary, and secondary education, and the authors concluded that older teachers tend to have lower levels of digital competence (Pozo Sánchez et al., 2020). A negative and average correlation (0.30 < r < 0.50) between age and DC in 75 teachers (23–65 years) for two educational groups (preschool–primary and secondary) in Catalonia (Spain) was also found in another study. This shows that older teachers have not received specific training in digital teaching in their initial training and that digital competencies were not considered in higher education qualifications before the adaptation to the European Education Area (Garcia i Grau et al., 2022). In our opinion, a lower level of digital competence may influence the level of ICT training.

Concerning the correlation between age and level of ICT training, there are contradictory results in Spain. One study pointed out that, in the case of teachers of people with functional diversity in Granada (Spain), age and training level had no relationship (Gallardo Montes et al., 2023). The preliminary results of our study found significant negative correlations between age and level of ICT training. Therefore, we proposed that there is an inverse relationship between age and the latent variables. Most of participants were woman and all of them had obtained university degrees. Thus, gender and level of education were not considered as control variables.

2. Materials and Methods

2.1. The Geographical Context and Participants in the Educational System for Children with Functional Diversity

It is advisable to contextualize the geographical context and participants in the early childhood education system for children with functional diversity in Spain. The focus of this study was on professionals from the region of Andalusia, Spain, and the analysis of early childhood education focused on the second cycle, which includes children aged 3 to 6 years. In Spain, the current protocol used by public institutions for children with functional diversity follows these stages: the children are referred to a pediatrician, then to a neuropediatrician (neurologist), and from there to mental health specialists, such as those specializing in functional diversity, autism or other fields. The majority of these children receive medication, at least for their sleep issues, which is the primary issue. Apart from health professionals, other professionals in the educational field (early childhood educators, speech therapists, teachers, therapists, etc.) conduct evaluations. Later, the children are referred to the early intervention care center (CAIT, “Centro de Atención Infantil Temprana”) to identify their needs and provide information. At this stage, CAIT sends a report to the school, establishing a connection between the family, school, and external services (CAIT) from this point onward. Additionally, in Andalusia, there are multiple associations that complement the work of CAIT (for example, in Granada, there is the MIRAME association). This protocol is very similar throughout the rest of Spain.

2.2. Data, Design, and Participants

The data were collected from November 2022 to September 2023 using Google Forms jointly with phone calls and snowball sampling. The research utilized quantitative and cross-sectional data. The free software G*Power version 3.1 (Faul et al., 2009; G*Power, 2007) was used to calculate the required sample size for the proposed structural equation model based on a small effect size of 0.10, a significance level of 5%, a statistical power of 95%, and a maximum of three predictors of the dependent variable, resulting in a minimum sample size of 132 participants. Age was included in determining the sample size. The final sample, after filtering out outliers and inconsistencies, consisted of 254 valid responses out of 271 responses.

The sample of 254 specialized teachers from Andalusia. Of these, 17.7% were men and 82.3% were women. The average age was 45.5 years, with a standard deviation of 8.416 years. Among the participants, 23.8% were special education teachers, and the rest were teachers with other specialties or educational degrees. All the participants (100%) had a university degree (

Table 1. Sociodemographic information.

Sociodemographic Variable	Frequency	Percentage
Sex
Men	45	17.7%
Women	209	82.3%
Age
24–33	24	9.4%
34–43	76	29.9%
44–53	110	43.3%
54–64	44	17.3%
Academic Qualification
Special Education (Speech and Language Therapy/Therapeutic Pedagogy)	60	23.6%
Early Childhood Education	160	63.0%
Education + Specialization	11	4.3%
Psychopedagogy/Psychology/Pedagogy	23	9.1%
Total	254	100%

).

All the educational centers use the Internet and 81.5% of the teachers have more than 10 years of experience. Regarding experience with teaching students with functional diversity, 3.9% had no experience in this field, 48.9% had more than 10 years of experience, and the rest had less than ten years. All the teachers work in early childhood education. The majority of the teachers, 91.6%, work in public schools (state-funded centers) and the rest in private schools. Over 65% work in urban educational centers, and the rest in rural educational centers. Additionally, more than 60% of the teachers work full time. The most relevant modalities that they work on (sensory, intellectual, and physical–motor, among others) accounted for approximately 47% of their focus, followed by behavioral (Attention Deficit Disorder (ADD) and Attention Deficit Hyperactivity Disorder (ADHD), among others) or emotional (anxiety, depression, etc.) modalities at 30%.

2.3. Indicators and Constructs

The questionnaire included 22 indicators (

Table 2. Construct and indicators.

Construct/Indicator	Description
ICT_PERCEPT	Professional’s perceptions of ICTs
fd101	Enhance teacher competencies.
fd102	Require guidance on searching, selecting, and evaluating ICT resources for the teaching–learning process.
fd103	Provide greater flexibility in the teaching–learning process.
fd104	Enable responses to educational needs.
fd105	Are easy to use in the context of addressing diversity.
fd106	Promote inclusion.
fd107	Offer multiple opportunities in the realm of diversity management.
fd108	Improve performance and effectiveness.
fd109	Increase motivation towards learning.
fd110	Enable access to information.
fd111	Allow for achieving objectives in a more flexible manner.
ICT_NECESS	ICT uses and necessities
fd201	Demand greater dedication and effort in my work.
fd202	Require specific training.
fd203	Require greater material resources and investment from the administration.
fd204	Help provide better support for diversity.
fd205	Selection of specific ICTs based on students’ needs.
ICT_TRAINING	Professional’s level of ICT training
fd301	I am aware of the main limitations that can affect their use.
fd302	I know different places on the Internet where specific resources can be found.
fd303	I can design activities using widely used educational software.
fd304	I feel prepared to assist them in using technical support and ICTs.
fd305	They facilitate the design and adaptation of activities.
fd306	They help with evaluation processes.

). The proposed model contains three constructs—two predictors and two endogenous constructs (one a priori dependent variable and one final dependent variable). These scales have been used with pre-service education teachers focused on education for children with functional diversity (Gallardo Montes et al., 2020). In our research, these scales include 11 indicators to measure professionals’ perceptions of applying ICTs in early childhood education of children with functional diversity (ICT_PERCEPT), five indicators to measure the ICT needs of children with functional diversity (ICT_NECESS), and six items to measure the ICT training level of professionals for working with people with functional diversity (ICT_TRAINING), considering the outer models as reflective. Each indicator was measured on a five-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree).

The descriptive measures of the indicators are shown in

Table 3. Descriptive analysis of indicators.

Item	No.	Type	Missing	Mean	Median	Minimum	Maximum	Standard Deviation	Excess Kurtosis	Skewness
Sex	1	0|1	0	0.823	1	0	1	0.382	0.901	−1.701
Age	2	MET	0	45.547	46	24	64	8.399	−0.546	−0.063
fd101	3	MET	0	4.358	4	3	5	0.616	−0.655	−0.411
fd102	4	MET	0	4.240	4	1	5	0.596	2.526	−0.590
fd103	5	MET	0	4.413	4	3	5	0.626	−0.587	−0.588
fd104	6	MET	0	4.358	4	3	5	0.647	−0.677	−0.512
fd105	7	MET	0	4.016	4	2	5	0.758	−0.507	−0.299
fd106	8	MET	0	4.177	4	2	5	0.679	−0.229	−0.388
fd107	9	MET	0	4.339	4	3	5	0.578	−0.665	−0.207
fd108	10	MET	0	4.323	4	3	5	0.600	−0.635	−0.278
fd109	11	MET	0	4.508	5	3	5	0.580	−0.486	−0.700
fd110	12	MET	0	4.390	4	3	5	0.597	−0.670	−0.406
fd111	13	MET	0	4.268	4	3	5	0.614	−0.600	−0.236
fd201	14	MET	0	4.134	4	2	5	0.662	0.211	−0.400
fd202	15	MET	0	4.276	4	3	5	0.623	−0.637	−0.277
fd203	16	MET	0	4.339	4	3	5	0.636	−0.679	−0.436
fd204	17	MET	0	4.350	4	3	5	0.614	−0.656	−0.387
fd205	18	MET	0	3.980	4	2	5	0.707	−0.067	−0.308
fd301	19	MET	0	3.929	4	2	5	0.642	0.396	−0.294
fd302	20	MET	0	4.016	4	2	5	0.687	0.907	−0.606
fd303	21	MET	0	3.425	4	1	5	1.02	−0.543	−0.412
fd304	22	MET	0	3.768	4	2	5	0.762	−0.251	−0.226
fd305	23	MET	0	4.165	4	2	5	0.673	0.439	−0.523
fd306	24	MET	0	4.150	4	2	5	0.711	0.464	−0.623
Source: Ringle et al. (2022).
Mardia’s multivariate Test                     b    z       p-value Skewness           1287.921    54,521.96877   0 Kurtosis          4023.459     28.03496      0

Source: https://webpower.psychstat.org/models/kurtosis/results.php?url=088214b17ed0123914fbd5e311e459b7.

. The values of skewness and excess kurtosis statistics are provided in Table 3. The indicators fall outside the acceptable range of normality (±1), indicating they do not follow a normal distribution. If the variables are not normally distributed (unidimensional), then the data will also not have a multivariate normal distribution. Mardia’s test of multivariate kurtosis was performed, with the null hypothesis being that the sample comes from a multivariate normal distribution and is rejected if the significance level is below 1% (Table 3).

2.4. Methodology and Data Analysis

This study employed Partial Least Squares Structural Equation Modeling (PLS-SEM) and the SmartPLS software (Ringle et al., 2022) for the analysis. The model evaluation included assessing the measurement models for internal consistency, convergent reliability, and discriminant validity using three approaches. The structural model evaluation examined the explanatory power or model fit through the coefficient of determination (R²), effect size (f²), multicollinearity analysis using Variance Inflation Factors (VIFs), and significance testing of path coefficients via bootstrapping. The global model fit indices lacked universal agreement. An approximate fit can be obtained using metrics such as the Standardized Root Mean Square Residual (SRMR) and Normed Fit Index (NFI). Common method bias (CMB) concerns were addressed in the model evaluation.

The proposed theoretical model is shown in Figure 1.

3. Research Results and Discussion

3.1. Test of Common Method Bias (CMB)

This study was conducted in accordance with the Declaration of Helsinki. The survey was anonymous and met ethical standards. Before answering the questionnaire (Table S1), the education professionals (participants) were informed by phone about the study’s purpose. All subjects gave their informed consent for inclusion before they participated in the study through e-mail. They completed the document and sent it back by ordinary post. Among other aspects, this minimizes the common method bias (CMB) that can occur in survey studies. Harman’s one-factor test is a commonly applied method for identifying CMB (F. Kock et al., 2021; Podsakoff et al., 2003); however, it is only a diagnostic test and yielded a factor that explains 0.35 of the variance, below the established threshold of 0.50. N. Kock (2015) proposed that a Variance Inflation Factor (VIF) greater than 3.3 not only indicates severe collinearity, but also suggests that common method bias may be affecting the model. The VIFs in the inner model ranged from 1.161 to 1.376, indicating that the model is free from common method bias.

3.2. Results of the Variance-Based Structural Equation Model

3.2.1. Measurement Models

In terms of the reliability of the indicators, not all the outer loadings reach the recommended minimum threshold of 0.7. However, Hair et al. (2017, pp. 113–114) established that factor loadings between 0.4 and 0.7 should be individually examined before considering removal from the measurement scale to meet specific requirements.

Various measures were used to assess the construct’s reliability: Cronbach’s alpha coefficient, Dijkstra–Henseler’s rho coefficient, composite reliability via RhoA and CR or RhoC- coefficients, and average variance extracted (AVE) for convergent validity, which all met the required thresholds (

Table 4. Item reliability: internal consistency, reliability, and convergent validity.

Construct (Composite)	Factor Loadings (Min.; Max.)	CA	Composite Reliability		AVE
			RhoA	RhoC
1. Professionals’ perceptions of applying ICTs in education...	(0.670; 0.840)	0.828	0.835	0.875	0.540
2. Necessity of applying ICTs in education...	(0.789; 0.919)	0.651	0.738	0.846	0.733
3. Professionals’ level of ICT training for educating...	(0.691; 0.855)	0.874	0.889	0.904	0.614

⁴ Note: Factor loadings > 0.707 (Carmines & Zeller, 1979); 0.4 ≤ loadings ≤ 0.7 (Hair et al., 2017, 2nd ed.). CA: Cronbach’s alpha > 0.7 (Cronbach, 1951); RhoA: Dijkstra–Henseler rho (ρA) > 0.7 (Dijkstra & Henseler, 2015); CR or RhoC: composite reliability > 0.7 (Werts et al., 1974); AVE: average variance extracted > 0.5 (Fornell & Larcker, 1981). Obtained from the * 95% confidence intervals. All items were significant (p < 0.001). Source: Prepared by authors based on Ringle et al. (2022).

).

Three criteria were used to analyze the discriminant validity: the Fornell–Larcker Criterion, the heterotrait–monotrait ratio of correlations (HTMT) (Henseler et al., 2015), and the cross-loadings, which all met the recommended thresholds (

Table 5. Discriminant validity approaches.

Fornell–Larcker Criterion and HTMT Criteria
Construct			1	2	3
1. Professionals’ perceptions of applying ICTs in education...			0.735	0.505	0.659
2. Necessity of applying ICTs in education...			0.384	0.856	0.298
3. Professionals’ level of ICT training for educating...			0.578	0.239	0.783
Note: The square roots of the average variance extracted (AVE) in bold are located on the diagonal. Over-diagonal and under-diagonal HTMT correlations indicate the construct’s inter-correlations.
Cross-Loading Approach
Constructs
Indicator	1. Professionals’ perceptions of applying ICTs in education...	2. Necessity of applying ICTs in education...	3. Professionals’ level of ICT training for educating...
fd101	deleted	deleted	deleted
fd102	deleted	deleted	deleted
fd103	deleted	deleted	deleted
fd104	0.688	0.242	0.427
fd105	0.670	0.174	0.456
fd106	0.730	0.316	0.309
fd107	deleted	deleted	deleted
fd108	0.840	0.305	0.511
fd109	0.725	0.322	0.401
fd110	deleted	deleted	deleted
fd111	0.745	0.329	0.423
fd201	0.382	0.919	0.249
fd202	0.257	0.789	0.142
fd203	deleted	deleted	deleted
fd204	deleted	deleted	deleted
fd205	deleted	deleted	deleted
fd301	0.325	0.172	0.701
fd302	0.361	0.157	0.691
fd303	0.449	0.265	0.810
fd304	0.436	0.112	0.850
fd305	0.546	0.180	0.855
fd306	0.534	0.224	0.775

Note: the highest values are marked in bold. Source: prepared by authors based on Ringle et al. (2022).

).

3.2.2. Structural Model

There is controversy regarding the use of global model fit indices (Benítez et al., 2020). In this study, two commonly used indices for this purpose, the standardized root mean square residual (SRMR = 0.08) and the normed fit index (NFI = 0.76), had values very close to the established threshold of <0.08 (Hu & Bentler, 1999). Based on these indices, the model could be considered satisfactory (Hair et al., 2019; Hu & Bentler, 1999).

The coefficient of determination, R², assesses the explanatory power of the model within the sample (in-sample). For the final dependent composite (R² of Professional ICT Training = 0.375), it was moderate, and for the other dependent variable (R² of ICT Needs = 0.147), it was weak (Hair et al., 2017, 2019). The collinearity analysis did not find any concerning issues. Additionally, the composite of professional perceptions showed the largest effect size (f² = 0.451) on the professionals’ level of ICT training (

Table 6. Measures of the inner model.

Dependent Construct	Explanatory Power		Inner VIF			Effect Size f2
	R2		Constructs			Constructs
			1	2	3	2	3
1. Professionals’ perceptions of applying ICTs in education...						0.164	0.451
2. Uses and necessity of applying ICTs in education...	0.147 *	Weak a		1.000	1.164		0.000
3. Professionals’ level of ICT training for educating...	0.375 *	Moderate a			1.170
Age							0.065

Note. Minimum cut-off: R² ≥ 0.1 (Falk & Miller, 1992); inner VIF (Variance Inflation Factor) ≤ 3.3 (F. Kock et al., 2021). * p < 0.05. Tested by bootstrapping 10,000 samples; percentile configuration, one-tailed. a: classification according to Hair et al. (2017). Source: prepared by authors based on Ringle et al. (2022).

).

Table 7. Analysis of direct and mediating effects and final decisions.

H	Path	Direct Effect	Sig.	Decision	H	Path	Total Indirect Effect	Sig.	Decision	Type of Mediation	Total Effect
H1+	1 −> 2	0.384 ***	Yes	Support	-	-	-	-	-	-	0.384
H2+	1 −> 3	0.572 ***	Yes	Support	H4+	1 −> 2 −> 3	0.002	no	Reject	No Mediation	0.572
H3+	2 −> 3	0.004 (ns)	No	Reject	-	-	-	-	-	-	-
H5-	4 −> 3	−0.202 ***	Yes	Support

Note: H, proposed hypothesis; Sig, significant; ns, not significant; *, p < 0.05; **, p < 0.01; ***, p < 0.001. Total indirect effects = total effects − direct effects. Source: prepared by authors base on Ringle et al. (2022). 1. Professionals’ Perceptions of ICTs; 2. Use or necessity of ICTs; 3. Professionals’ level of ICT training; 4. Age (control variable).

presents the analysis results of the effects (direct and indirect) and the final decisions on the hypotheses. H1, H2 and H4 are supported while H3 and H4 are rejected (2. Needs -> 3. Training), which shows that only the indirect effects were not significant. Figure 2 shows the estimated model. The response to the research question RQ4 is that the perception on ICT has the highest direct positive impact on ICT training following by the control variable age with negative direct effect.

4. Discussion

Three of the five hypotheses are supported by data in the model adapted to the case of teachers working in early childhood education of children with functional diversity for Andalusia (Spain).

4.1. Relationship Between Perceptions of ICTs and the Uses and Necessity of ICTs

Regarding the relationship between the perceptions of ICTs and the uses and necessity of ICTs, it was concluded that the data support a positive effect of ICT use on early childhood education of children with functional diversity, corroborating the findings of Colomo-Magaña et al. (2023). Additionally, another study on early childhood education in Benin City, Nigeria, conducted by Imasuen and Iyamu (2024), showed that preschool teachers’ attitudes towards and application of ICTs were also positive. Furthermore, our results align with those obtained from specialized teachers (early childhood, primary, and secondary) in the use of ICTs in inclusive education in Spain (Vega-Gea et al., 2021). Finally, Shin and Jang (2023), who focused on secondary school teachers in Korea, found a positive relationship between perceptions and the use of ICTs.

Pegalajar Palomino (2017) found favorable perceptions of future teachers towards the didactic possibilities of ICT resources in the teaching–learning process of students with educational needs, and their contribution to professional development processes, inclusion of students, and analysis of the teaching function.

4.2. Relationship Between Perceptions and ICT Training

There was a direct relationship between perceptions of ICTs and level of ICT training, which is consistent with the findings of Colomo-Magaña et al. (2023). In our model, we found that perceptions of ICTs influence the level of ICT training, whereas Colomo-Magaña et al. (2023) found positive covariances. The perception of the utility of ICTs and the rapid software updates could explain this positive relationship.

Several studies on primary education university students focusing on inclusive education (Fernández Batanero et al., 2017; Matheu Pérez et al., 2024; Toledo Morales & Llorente Cejudo, 2016; Vega-Gea et al., 2021) concluded that there was limited knowledge regarding the application of ICTs for people with disabilities, and that their knowledge varied depending on the type of disability. Therefore, measures need to be taken regarding initial teacher training to provide future teachers with the opportunity to acquire skills and competencies in using and integrating ICTs in inclusive classrooms. Thus, universities should improve teacher training in the use of ICTs for diversity support (Fernández-Batanero et al., 2020).

4.3. Relationship Between Uses and Necessity of ICTs and Level of ICT Training

Our study’s results do not support a positive relationship between the use or necessity of ICTs and ICT training, unlike Colomo-Magaña et al. (2023). One possible reason could be that the constructs of use and necessity are not related to the same indicators in both models. Furthermore, the different participant populations—namely, teachers working (in-service) with children with functional diversity vs. future teachers (i.e., pre-service teachers)—could also contribute to this disparity. Finally, the different level of education and age of children could be another reason.

4.4. Relationship Between Uses and Necessity of ICTs and ICT Training

The indirect relationship between perceptions of ICTs and ICT training is not supported. This could be interpreted that the perceptions of use and necessity do not necessarily motivate undergoing ICT training. Furthermore, the separation of use and necessity should be considered to see if it changes the model.

4.5. Age

Although there are studies that have considered age as a control variable in educational research, our review of the relevant literature did not find any similar papers in relation to ICT training. To the best of our knowledge, the inclusion of age as a control variable in this specific context represents a novel contribution of our model. Nonetheless, we recommend that future research further investigate this aspect to confirm if this is an underexplored area.

Fernández-Batanero et al. (2019) found that older teachers have lower levels of digital competence compared with younger teachers. The reason could be the modification of university curricula to remove specialized courses in functional diversity. ICTs undergo rapid and tremendous changes over time. In our opinion, the lower level of digital competence, though not included in our model, of older people may impact their confidence and produce anxiety when using ICTs. Their ICT training may also be affected due to a lack of time or effort or a lack of available resources to put into practice the concepts in training sessions (robotics, 3D modeling, artificial intelligence, among others). Additionally, as they near retirement age, this type of training that requires effort, concentration, and time is less interesting since they will not need to apply this specific educational knowledge in everyday life. Therefore, it is reasonable to expect an inverse relationship between age and level of ICT training, which is in line with the findings from other Spanish studies (Garcia i Grau et al., 2022; Pozo Sánchez et al., 2020).

Some studies concerning age and digital competence pointed out significant differences in digital ability between different age groups (Fernández-Batanero et al., 2019). Digital ability was related to the use of ICTs, training, and age. Therefore, we should tailor the type of training to specific age groups: generations X, Y, and Z, or dichotomous groups (e.g., younger versus older), among others. In general, older age correlates with a lower cognitive ability, less concentration, greater anxiety at work, and more stress, among other implications of old age. Therefore, time to retirement is another factor that should be taken into account because these people may not be motivated to undergo training in the use of ICTs for students with functional diversity or other types of disabilities.

Finally, for training in general, Shin et al. (2023) emphasized the need for continuous and long-term teacher training (professional development) in the use of educational technology and teaching methods based on their recent study targeting elementary school teachers in Korea. They also highlighted that the production and exchange of teaching materials had the least amount of support and recognized it as an area that should be prioritized (Kim, 2022 cited in Shin et al., 2023). This support should include budgetary support for content creation and a platform for content sharing.

5. Conclusions

5.1. Theoretical Implications

The considered type of analysis was applied for the first time in the context of early childhood education of children with functional diversity, with the aim of addressing the research gaps identified by various authors (Fernández-Batanero et al., 2019; Gallardo Montes et al., 2020; Colomo-Magaña et al., 2023). The rising incidence of these issues in children is raising concerns and prompting a search for solutions, not only in Spanish society but also in other countries. Therefore, this study contributes to the improvement of teacher training and, consequently, to that of the students.

5.2. Practical Implications

The interviewed professionals emphasized the importance of early assessments. In this regard, they perceive the personal and material resources to be scarce. The majority of centers are public and have limited human and financial resources to invest in this type of student. Therefore, the scarcity of resources is the main limitation faced by public institutions. The assistance received is often insufficient and delayed. In some cases, it takes two or more years to receive assistance when it is needed at a young age. The solutions involve restructuring and streamlining the bureaucracy of Spanish public administrations to achieve faster delivery of financial aid for purchasing ICT equipment. Moreover, this should be accompanied by technical support and maintenance support. Another solution is increasing the hiring of professionals. However, the primary focus should be on expanding teacher training and developing strategies for family involvement. Concerning age, different types of training could be offered for different age groups, in order to motivate the teachers.

6. Limitations and Future Research Lines

This study has several limitations. The study was cross-sectional and the results are specific to the region of Andalusia, Spain. Regarding future studies, several avenues can be explored:

(1)

Comparisons of outcomes of this type of education with technology integration and which professionals (such as therapeutic pedagogy for integration support, PTAI, and regular teaching staff) utilize these technologies in different countries.

(2)

At a theoretical level, we can extend the model by adding how decisions are made in public and/or private schools. Additionally, we can study the impact of new technologies on students with professionals involved in their education.

(3)

Future research could analyze the perceptions of parents or guardians of children with functional diversity to determine if they perceive improvements during this educational stage. It would be valuable to compare these perceptions with those of teachers or professionals.

(4)

Future studies should use larger sample sizes.

(5)

Additionally, it should consider performing multigroup analysis based on gender, age, years of experience, and type of school, among others.

(6)

Other control variables should be added such as gender, type of professional, years of experience, and level of education, among others.

Apart from advocating for increased training, other crucial questions arise: Should this enhanced training focus more on psychological aspects and train teachers from a perspective that fosters better social relationships with children with functional diversity? Or should it solely emphasize technological training? Or does it require a balance of both? Is technology replacing human interactions? Answering these questions will require further investigation and expansion of this study. Therefore, validating this idea would involve jointly analyzing both types of training within the same model.

Finally, we would like to acknowledge and express gratitude not only for the role of the regional and national public educational institutions in Spain, but also for the involvement of non-governmental organizations in promoting and assisting schools and parents with children with functional diversity at any educational stage.