Exploring the Effect of Training in Visual Block Programming for Preservice Teachers

This study evaluates the effectiveness of visual block programming-based instruction and its possibilities in the training of future teachers. In particular, the application Scratch, a visual programming environment, was employed to introduce pre-service teachers to programming. The study followed a mixed-method design with a sample of 79 pre-service teachers. A quantitative approach was used to evaluate the gains in the participants’ knowledge of computational concepts and attitudes towards Scratch as a pedagogic tool. A qualitative analysis aimed at evaluating the participants’ knowledge concerning programming applications, and their perception about possible difficulties in the implementation of programming in educational contexts. Positive results were obtained for programming in the classroom, with significant improvements in innovation, collaboration, active learning, motivation, and fun for the students. After the experiment, the subjects highlighted Scratch as a fundamental block programming tool and the need for teacher training in this field. The need to improve the implementation of visual block programming in Education Degree curricula is supported.


Introduction
Society and the labor market demand mastery of the essential tools to function effectively in the information age. One of these technology-related tools, commonly known as digital competence, is programming [1] which is a means of creating something new in the digital world, solving problems, implementing ideas and using creativity. Programming provides benefits and advantages in different areas of elementary education, regarding motivation, digital competence and skills to create. It is introduced through block-based visual languages [2], and Scratch is one of the most widely used products worldwide [3]. However, despite growing interest in introducing computational thinking and programming into K-12 contexts [4], most teachers find it difficult to use Scratch in the classroom [3].
Technologies already have a significant impact on learning, teaching and creative thinking in university education [5]. However, in pre-service Teacher Education Programs as well as in primary and secondary education, new approaches are needed to integrate technology into education and train digitally competent teachers [6,7]. The European Higher Education Area aims to clarify and resolve this issue by examining how European Higher Education Institutions generate new teaching formulas, aimed at promoting active and dynamic learning, technology use, and student-centered methodologies in response to changing demands [8].
Scratch application was really valuable and accessible; something that has also been argued by other authors [30].
In a study with elementary students, Sáez-López et al. [2] reported statistically significant improvements in the understanding of computational concepts and practices due to the use of Scratch in the classroom, which suggests recommending educational authorities to implement programming in elementary educational environments. Students can approach computational practices and create their own content related to curricular areas, especially in the social sciences and arts, given the characteristics of the visual content these areas present, allowing the creation of colorful, dynamic, and motivating projects from an active perspective. These conclusions are reaffirmed in the study by [31] which demonstrated the acquisition of basic concepts of computer programming in the elementary education setting.
These investigations highlight the advantages related to student motivation, fun, commitment and enthusiasm through this pedagogical approach. Students are generally in favor of this pedagogical design, highlighting its usefulness and the active learning provided. The importance of an educational design that includes the approach of a visual programming language to understand the elements of logic, mathematics, and the creation of content in art and history, brings about significant improvements both in elementary [2] and higher education [32] through the understanding of computational concepts and active approaches.
From our position as trainers of trainers, and from the application of approaches focused on visual block programming, we consider the impact generated by these emerging innovations in the initial training of future teachers, as well as the real presence of these key trends in the university training context. To address this, we specify the following aims: • O01-To assess the effect of a visual block programming-based intervention on pre-service teachers' mastery of coding. • O02-To assess the effect of a visual block programming-based intervention on pre-service teachers' opinions about the potential of programming as a pedagogical tool. • O03-To assess the pre-service teachers' satisfaction and enjoyment based on the visual block programming-based intervention.

Materials and Methods
The study followed a pre-experimental design as educational ethics prevented researchers to configure a control group, which is a clear limitation of the experiment and confers an exploratory nature to this piece of research. The intervention was carried out in two classroom-based instruction sessions of 5 h each, in two different weeks. Additionally, 10 h of autonomous student work were added, which was carried out outside the classroom. Figure 1 summarizes the different phases of the intervention. Scratch application was really valuable and accessible; something that has also been argued by other authors [30]. In a study with elementary students, Sáez-López et al. [2] reported statistically significant improvements in the understanding of computational concepts and practices due to the use of Scratch in the classroom, which suggests recommending educational authorities to implement programming in elementary educational environments. Students can approach computational practices and create their own content related to curricular areas, especially in the social sciences and arts, given the characteristics of the visual content these areas present, allowing the creation of colorful, dynamic, and motivating projects from an active perspective. These conclusions are reaffirmed in the study by [31] which demonstrated the acquisition of basic concepts of computer programming in the elementary education setting.
These investigations highlight the advantages related to student motivation, fun, commitment and enthusiasm through this pedagogical approach. Students are generally in favor of this pedagogical design, highlighting its usefulness and the active learning provided. The importance of an educational design that includes the approach of a visual programming language to understand the elements of logic, mathematics, and the creation of content in art and history, brings about significant improvements both in elementary [2] and higher education [32] through the understanding of computational concepts and active approaches.
From our position as trainers of trainers, and from the application of approaches focused on visual block programming, we consider the impact generated by these emerging innovations in the initial training of future teachers, as well as the real presence of these key trends in the university training context. To address this, we specify the following aims: • O01-To assess the effect of a visual block programming-based intervention on pre-service teachers' mastery of coding. • O02-To assess the effect of a visual block programming-based intervention on pre-service teachers' opinions about the potential of programming as a pedagogical tool. • O03-To assess the pre-service teachers' satisfaction and enjoyment based on the visual block programming-based intervention.

Materials and Methods
The study followed a pre-experimental design as educational ethics prevented researchers to configure a control group, which is a clear limitation of the experiment and confers an exploratory nature to this piece of research. The intervention was carried out in two classroom-based instruction sessions of 5 h each, in two different weeks. Additionally, 10 h of autonomous student work were added, which was carried out outside the classroom. Figure 1 summarizes the different phases of the intervention. Before the first session, the students took a pre-test in which their prior knowledge of computational concepts, their opinion on coding in educational contexts and their knowledge of coding applications and current needs were measured. The instruction was structured in two modules. In the first one, the students were introduced to basic computational concepts (sequences, loops, conditionals) through the development of simple projects using Scratch. In the second module, Before the first session, the students took a pre-test in which their prior knowledge of computational concepts, their opinion on coding in educational contexts and their knowledge of coding applications and current needs were measured. The instruction was structured in two modules. In the first one, the students were introduced to basic computational concepts (sequences, loops, conditionals) through the development of simple projects using Scratch. In the second module, more advanced computational concepts (parallelism, events, operators and data) were explored, as well as project-design practices through the creation of games where the pedagogical advantages of coding in the teaching of Social Sciences could be highlighted.
The following computational concepts [21] were used in this study: • Sequence: To create a program it is necessary to think about the order of the steps. • Iteration (loop): "always" and "repeat" can be used for iteration (repetition of a series of instructions).

•
Conditional statements: "if" can be used to establish a condition. • Threads (parallel execution): Execution of two independent orders that are executed in parallel. • Event Management: For example, when a key is pressed or when objects are clicked.
Subsequently, the students had to develop, as homework, an original project with Scratch in which the concepts and programming learned during the instruction had to be used in order to enhance their creativity. This work would be evaluated by the teacher and would be posted in Scratch Studio, where all the projects developed could be reviewed. Once the instruction was finished, the participants were again invited to carry out the post-test following the same procedure used in the pre-test.

Participants
The study population are university students in the second year of the Degree in Primary Education in the Faculty of Education, Albacete (University of Castilla-La Mancha), enrolled in the subject "Social Sciences II: History and its didactics". The sample, which was non-probabilistic and intentional, consisted of 79 individuals. The tests from the design are applied as related samples linked to a pre-test/post-test design. The experimental group consisted of 72.2% female students and 27.8% male students, which turns out to be a representative sample of the gender disparity that exists in teacher studies, in which there is always a greater number of women. The average age was almost 21 (20.85). The group was quite homogeneous, because it consisted of students of almost the same age and educational level.

Instruments
This study uses the survey technique and a questionnaire as instruments, consistent with the objectives of the research. In dimension 1: Results and computational concepts, the Visual Blocks Creative Computing Test (VBCCT) [2] was used. The test was made up of 20 multiple-choice items with 4 options and one correct answer, in which students must answer questions related to computational concepts and computational practices. Figure 2 shows an example of the questions used.
In dimension 2: Programming in educational contexts, a questionnaire with four dimensions adapted from Cózar-Gutiérrez and Sáez-López [33] was used. The questionnaire includes different sub-dimensions: (1) Knowledge and mastery of the programming, (2) Visual block programming in educational contexts, (3) Active learning, and (4) Fun. Items in the first two sub-dimensions were adapted from [2], while items for Active Learning and Fun were adapted from [34,35], respectively. The descriptive data of the four sub-dimensions were assessed from the pre-experimental approach, applying, again, one sample t-tests between the pre-and post-test. The results were analyzed through an inferential analysis with one sample t-tests, except for sub-dimension 4, Fun, which was exclusively measured in the post-test. Multimodal Technol. Interact. 2020, 4, x FOR PEER REVIEW 5 of 11 This information allows us to determine if there are significant differences before and after the intervention carried out with the students. The level of significance (α) was 0.05 in all the cases. Reliability calculated through Cronbach's alpha coefficient is 0.83 in the first dimension, and an average of 0.71 for the different scales in the second dimension; as it is always greater than 0.6, it is considered acceptable [36]. In dimension 3: Programming applications and current needs, the qualitative information of open questions is analyzed.
The analysis followed a data triangulation approach [37] using quantitative and qualitative information, which makes it possible to determine whether there is evidence to support the validity of the results and minimize error variance [38]. The data triangulation was implemented using quantitative information collected in the aforementioned test in the first dimension, and the questionnaire in the second dimension. We obtained qualitative data from the open questions in the third dimension.

Dimension 1: Results and Computational Concepts
For dimension 1, a pre-experimental design was applied, analyzing data with the one-sample t-test [39]. The Creative Visual Block Computing Test (TVBCC) was administered with a pre-test/post-test design, which makes it possible to check if there were significant improvements after the intervention. The values in the 20-item test emphasize the importance of an educational design that includes a visual programming language to understand the elements of logic, mathematics, and content creation, providing improvements as highlighted in this analysis.
The mean post-test scores for the 79 participants was 11.43 (SD = 2.52), which was compared to the mean pre-test score of 6.59 (SD = 2.18). The results revealed a significant difference between pre-and post-test scores on the TVBCC test (t = 17.04, p < 0.001, d = 1.92). It can thus be affirmed that This information allows us to determine if there are significant differences before and after the intervention carried out with the students. The level of significance (α) was 0.05 in all the cases. Reliability calculated through Cronbach's alpha coefficient is 0.83 in the first dimension, and an average of 0.71 for the different scales in the second dimension; as it is always greater than 0.6, it is considered acceptable [36]. In dimension 3: Programming applications and current needs, the qualitative information of open questions is analyzed.
The analysis followed a data triangulation approach [37] using quantitative and qualitative information, which makes it possible to determine whether there is evidence to support the validity of the results and minimize error variance [38]. The data triangulation was implemented using quantitative information collected in the aforementioned test in the first dimension, and the questionnaire in the second dimension. We obtained qualitative data from the open questions in the third dimension.

Dimension 1: Results and Computational Concepts
For dimension 1, a pre-experimental design was applied, analyzing data with the one-sample t-test [39]. The Creative Visual Block Computing Test (TVBCC) was administered with a pre-test/ post-test design, which makes it possible to check if there were significant improvements after the intervention. The values in the 20-item test emphasize the importance of an educational design that includes a visual programming language to understand the elements of logic, mathematics, and content creation, providing improvements as highlighted in this analysis.
The mean post-test scores for the 79 participants was 11.43 (SD = 2.52), which was compared to the mean pre-test score of 6.59 (SD = 2.18). The results revealed a significant difference between pre-and post-test scores on the TVBCC test (t = 17.04, p < 0.001, d = 1.92). It can thus be affirmed that there were significant gains after the interventions. Indeed, the effect size d is 1.92, which can be considered Multimodal Technol. Interact. 2020, 4, 65 6 of 11 as large [40]. Consequently, the implemented program improved the students' ability to understand the concepts and logic of programming to create multimedia products related to the curriculum.

Dimension 2: Coding in Educational Contexts
To evaluate dimension 2, one-sample t-tests were performed for each item (Table 1) and for each sub-dimension. Concerning the first sub-dimension (Knowledge and mastery of the programming), the participants' scores were significantly higher in the post-test (M post = 3.00, SD post = 0.47) compared to the pre-test (M pre = 2.38, SD pre = 0.50, t = 11.64, p < 0.001). In addition, the difference can be seen as large (d = 1.32). An in-depth analysis of each item in this dimension indicates that pre-service teachers have positive opinions concerning the importance of programming knowledge for their future career, and also in their training at the university (items 1.1, 1.2, and 1.3). The gains in these items are moderate-medium sized (see Table 1). The intervention was particularly effective at increasing the participants' experience in programming (item 1.4), and qualifying them to use block programming languages to create content in educational contexts and use them as classroom tools (items 1.5 and 1.6), with large effect sizes. Concerning the second sub-dimension (Visual block programming in educational contexts), the scores revealed that the intervention improves pre-service teachers' opinions in favor of the pedagogic use of these programming languages. Thus, the comparison between pre-test (M pre = 2.78, SD pre = 0.72) and post-test scores (M post = 3.46, SD post = 0.46) showed the huge benefits derived from the intervention in this sub-dimension (t = 13.07, p < 0.001, d = 1.48). After the experiment, the students stressed that visual block programming enhances creativity, collaborative advantages, communication, skills development, innovation and motivation in educational contexts (items 2.1, 2.2, 2.3, 2.4, 2.5, and 2.6). The effect can be seen as large for all these items.
Regarding the third sub-dimension (Active learning), the experience led participants to consider that programming activities promote active involvement and learning (items 3.1 and 3.4) and make the subjects more interesting (3.3). Globally, the participants' scores were again significantly higher in the post-test (M post = 3.32, SD post = 0.49) compared to the pre-test (M pre = 2.71, SD pre = 0.73, t = 11.18, p < 0.001, d = 1. 24) Finally, the last sub-dimension (Fun), which was only measured in the post-test as it exclusively refers to motivational variables related to the activities developed during the experience. A descriptive analysis indicated that the participants felt extremely happy, motivated, relaxed, and comfortable during the activities (see Table 1).
In short, all the inferential tests, together with the descriptive analysis, point to a very positive impact of the intervention in all the sub-dimensions, promoting the students' mastery of coding, their attitude towards visual block programming as an educational tool, and their enjoyment.

Dimension 3: Programming Applications and Current Needs
A qualitative analysis was conducted based on the participants' answers to a survey. In the design of the instrument, we opted for a mixed questionnaire consisted of two open questions as an instrument because it enables students to respond freely. Considering the purpose of this paper, only the answers of one question were included in the analysis. Indeed, the frequencies provided by the subjects to the question provided elements and factors of interest for the study, which were numbered and accounted for to reinforce the values obtained in the quasi-experimental design and descriptive analysis.
In particular, the question asked students "What problems or obstacles do you perceive to implement visual block programming in educational contexts?" The sample subjects highlighted the need for training, time, and effort in the design and development of these activities, as well as the need for resources and means (Figure 3).

3.
improved through visual block programming. In short, all the inferential tests, together with the descriptive analysis, point to a very positive impact of the intervention in all the sub-dimensions, promoting the students' mastery of coding, their attitude towards visual block programming as an educational tool, and their enjoyment.

Dimension 3: Programming Applications and Current Needs
A qualitative analysis was conducted based on the participants' answers to a survey. In the design of the instrument, we opted for a mixed questionnaire consisted of two open questions as an instrument because it enables students to respond freely. Considering the purpose of this paper, only the answers of one question were included in the analysis. Indeed, the frequencies provided by the subjects to the question provided elements and factors of interest for the study, which were numbered and accounted for to reinforce the values obtained in the quasi-experimental design and descriptive analysis.
In particular, the question asked students "What problems or obstacles do you perceive to implement visual block programming in educational contexts?" The sample subjects highlighted the need for training, time, and effort in the design and development of these activities, as well as the need for resources and means ( Figure 3).

Discussion
Many studies in the last few years have analyzed visual block programming in educational contexts and have yielded positive results similar to those found in our study [2,3,29,31]. Based on triangulation of the results, it can be concluded, in a structured way, that: 1. The subjects positively valued receiving training in programming during their initial teacher training. Based on the TVBCC values, statistically significant improvements were obtained after the program was implemented, improving the ability of university students to understand the concepts and logic of programming to create multimedia products as part of their training to be future teachers (dimension 1).
2. The formative action significantly favored knowledge about, attitudes towards, and the application of programming in initial stages of teacher training. The subjects considered programming to be essential in the training future teachers, and very important during initial university education (items 1.1, 1.2, and 1.3).
3. Although many of the participants had experience in programming before the intervention, pre-service teachers improved their skills in the implementation of visual block programming after the intervention, which points in the same direction as the results obtained in [3].
4. Block programming in educational contexts (sub-dimension 2) showed positive and significant values for the importance of creativity (item 2.1), interactive, collaborative advantages, and the development of competencies (items 2.2, 2.3, and 2.4). Educational innovation processes were also possible (item 2.5), and, as in [2], the motivation of the students was increased and greatly enhanced (item 2.6).
5. With high values and significant improvements, it appears that integrating programming into the classroom encourages active learning (sub-dimension 3), while emphasizing that coding allows the topic being worked on to be more interesting (item 3.3), and programming encourages active participation and learning of the essential lesson content (items 3.1, 3.2, and 3.4).
6. These practices provided the students with satisfaction and fun (sub-dimension 4), and the students stated that they were happy, motivated and liked the programming activities (items 4.1, 4.2, 4.3, 4.4, and 4.5).
7. The main obstacle to implementing block programming appeared to be the need for teacher training, due to the difficulty of these pedagogical approaches and computational practices, something that may be related to the pedagogical beliefs of teachers about the barriers in the use of technology in the classroom [41] and their perceived competence [42]. To a lesser extent, the need for material resources and the difficulty in curricular design for these types of activities were also mentioned, which underpins what was previously denoted in [29].
Once the availability of resources, class planning and initial and continuous teacher training have been overcome, the implementation of visual block programming in educational contexts creates pedagogical advantages that allow greater student activity and prominence, taking advantage of the strength of interest and motivation that these approaches elicit, through collaboration, communication, creativity and student satisfaction.
International studies and reports have proposed that these key trends be adopted in the short term, leading to a change in educational practices [1]. University education and initial teacher training must adapt to these challenges and the demands of today's society, taking into account the emerging trends that prospective professionals will encounter in their immediate future [33]. This and other studies [43,44] provide evidence concerning the attitudes, assessments, and perspectives of the application of programming in university contexts among future teachers. Due to the difficulty of ensuring adequate application, and the essential need for training to avoid the barriers and difficulties that arise due to erroneous pedagogical application, now is a key moment for the introduction of educational technology during initial teacher training in university contexts, since if such programming training is delayed, the opportunity to incorporate these technologies into the classroom will be lost. How to teach Scratch at elementary classrooms is beyond the scope of this study as the training was focused on developing a good command in Scratch, but it was not aimed at teaching the future teachers how to introduce this visual language to elementary students. However, this issue is essential to computer sciences education and should be addressed in future studies. This study highlights the positive assessment of programming in the classroom by future teachers during their training period, with special focus on the advantages in terms of motivation, innovation, collaboration, active learning, and student satisfaction. Highlighting Scratch as a fundamental tool, and teacher training in this field as a challenge to overcome, visual block programming in higher education should be a reality that fosters the integral training of future teachers.