Education has been throughout history as one of the main pillars that sustains the development of our society. Education allows people to extend their knowledge and understanding of the environment that surrounds them. Being such an important area, it is logical that many organizations deepen educational issues in different social environments, as this contributes to increasing production and the scope of knowledge. In this era where cultural, social and technological changes are advancing in shorter periods, schools in all educative levels need to provide the right tools, so that students can develop the curiosity, creativity, self-regulation, collaborative learning and many other skills that are necessary to cope with the changes in their environment (Khan, 2012) .
In international context, some organizations like the United Nations Organization (UN), the Organization for Economic Cooperation and Development (OECD), the United Nations Educational, Scientific and Cultural Organization (UNESCO) and the United Nations Economic Commission for Latin America and the Caribbean (ECLAC) put some statements about what is to be desirable in education for the future: goals, programs and the importance to achieve better opportunities. In general terms they provide work guidelines from elementary to higher education about what need to be the focus in education. The programs are based on the particular needs of each society, from infrastructure to the characteristics of the students; with this, they can be prepared for the problems they might face in the future (ECLAC, 2018; OECD, 2017; UN, 2015; UNESCO, 2016) .
Is it worth mentioning that in Mexico some of the most important universities are aware of this recommendation from international organizations? They acknowledge the importance and the impact this could have in the outcomes of education because some of them cover high school education to higher education; with this they can shape a better opportunities for the students in their academic goals (BUAP, 2007; IPN, 2004; ITESM, 2018; UABC, 2013; UANL, 2008) . From the vast majority of factors that involve the education process in education, we believe the abilities within the student are the crucial factor of their academic success, like curiosity, creativity, self-regulation as a few of the skills that could help the students with their learning process; in this document we focus on the particular aspect of collaborative learning and how teachers can evaluate this particular skill in their students. For this reason, the main objective of this study is to design and evaluate the psychometric properties of a scale that examines the frequency of use of teaching strategies that lead to collaborative learning in students.
2. Basic Concepts
Teaching strategies in high school education have changed drastically in recent years, although there are still some teachers who use traditional strategies that could lead to less effective learning in the students. Participation and interaction among educational actors, nowadays more and more are struggling with the application of strategies that provoke reflection and the joint construction of knowledge and teachers need to provide guidance in student so they can develop the required skills to ensure their academic goals (Khan, 2012) .
In 2007 Schroeder, Scott, Toison, Huang and Lee, cited by Ramírez Montoya (2013) , presented research data regarding the most effective teaching strategies in science, found that the strategies with which students performed better were: question-answer, research, evaluation, significant contexts, use of technology and collaborative learning.
In various studies, both theoretical and empirical, it is beneficial to bring together a group of students who are assigned activities that involve mutual assistance and interaction between peers, which through dialogue and consensus, transmit, discuss, modify and transform the knowledge (Noh & Yusuf, 2018) , highlighting the fact that peers may have expertise or authority on some issues. The learning that is reproduced in the classrooms from the above is what is known as collaborative learning. Collaborative learning between peers let them choose their topic and engage their commitment to achieve their academic goals (Maldonado Pérez, 2008) . Also, the use of collaborative learning is now day’s extended in virtual environment, proven a useful tool in learning process through social networks (Marín-Juarros, Negre, & Pérez, 2014) .
2.1. Circle of Ideas
It is a variation of the brainstorming technique in which students generate them, but do not elaborate, explain, evaluate or question ideas. The members of the groups respond in turn to a question with a word, expression or short statement. The order of the answers is organized by passing from one student to another until everyone has had the opportunity to speak.
2.2. Conversations Groups
They are teams of between four and six students that are formed quickly to answer questions related to the topic. The purpose of the dialogue is to exchange ideas and these are effective to generate information in a short time (De Oca Recio & Manchado Ramirez, 2011) .
2.3. Critical Debates
In this strategy the students select an opposite view of their own in one specific topic. Then they form teams and dialogue, present and defend the issue in front of the opposing team (Martinez Lirola, Catala Cobos, & Diaz Soria, 2013) . Preparing, participating and listening to the debates have many benefits, can increase motivation, strengthen research skills, promote critical thinking and deepen from many points of view.
2.4. Role Playing Game
A simulated situation is created in which students deliberately represent or assume personalities or identities that they would not normally admit. It facilitates an environment of action for students to experience the emotional and intellectual responses of an assumed identity or an imagined circumstance; this facilitates students learning process (Chan, 2012) . In addition, role play can be fundamental for the acquisition of different skills and attitudes in the students (Martínez-Riera, Sanjuán-Quiles, Cibanal-Juan, & Pérez-Mora, 2011) . For some students, role playing game helps in their leaning process but some may experience difficulties based on their academics goals, if they just want to reproduce or memorize rather than understand a concept (Spiers et al., 2014) .
2.5. Test Teams
Students work in teams in order to prepare for the examinations called by the teacher and perform them, first answer the exam individually and then in a group. By working together to prepare for the exam, students will help each other to deepen the understanding of content.
Students work in small groups in order to develop their knowledge on a particular topic. Students must master a certain subject and pass it on to different teams. First they are part of a “puzzle” or team and transmit the content. Once understood by his teammates they move on to the next team and form a new “puzzle”. It is helpful in motivating students to take on the responsibility of learning something well enough to teach it to their peers. It also offers the opportunity for each student to become the center of attention (Reeve & Jang, 2006) .
2.7. Case Studies
The student teams review a written study of a real-life scenario that contains a problematic situation related to the area of knowledge (De Oca Recio & Manchado Ramirez, 2011) . These cases include a brief history of how the situation has developed and present a problem faced by a key character on the stage (Díaz De Salas, Mendoza Martínez, & Porras Morales, 2011) . Team members apply concepts from the subject to identify, evaluate alternative approaches to solving the problem and provides a better understanding through experience (Álvarez Álvarez & San Fabián Maroto, 2012) .
2.8. Structured Problem Solving
This strategy provides students with a procedure for solving a complex, content-related problem over a specific period of time. All members of the group have to agree on the solution and must be able to explain both the response and the strategy used to solve the problem. This technique helps students by dividing the process into concrete steps. Students learn to identify, analyze and solve problems in an organized way.
2.9. Collaborative Writing
Students work in groups of two or three to write an article. Each student participates in each sentence of the writing process: brainstorming, obtaining and organizing information and writing, reviewing and correcting the writing. Working together can help you learn and more effectively complete the writing phases of writing and produce better text with more grammatical accuracy and complexity (Storch, 2005) .
In this study we choose a quantitative approach aimed to design a set of subscales with evidence of validity and reliability, which measure the frequency of use of teaching strategies that lead to collaborative learning in the students, with the specific objectives of:
1) Design a set of subscales with Likert-type items where each one evaluates the frequency of use of a teaching strategy in the classroom.
2) Analyze the factor structure of each subscale.
3) Determine the inter-item reliability of each subscale.
The sample (Table 1) selected for this research was composed of 200 teachers from different public high schools in Mexico (de que materias en específico, de que estados de la república y si se tienen las cantidades por estado).
The only criterion of inclusion was that teachers had at least one-year teaching experience in any field.
The Teaching Strategies Scale for Collaborative Learning or EEDAC (See Appendix) (his acronym in Spanish) was designed using the theoretical frame of reference Barkley, Cross and Major (2005). The scale consists of 80 items integrated in 9 subscales; each of which contains indicators of the appropriate application of a teaching strategy that promotes collaborative learning. The subscales with their respective number of items are the following: Circle of Ideas Wheel (10), Conversation Groups (8), Critical Debates (11), Role Playing (12), Exam Teams (7), Puzzles (6), Case Studies (11), Structured Problem Solving (6) and Collaborative Writing (9). The items have a Likert format, whose response options are Always, Many Times, Sometimes, Few Times and Never, with a path of 5 to 1, all of which are positive items. The scale is located in the annex.
For the design of the scale, an extensive bibliographical review was carried out
Table 1. Characteristics of the sample.
on the collaborative learning construct and its classical and contemporary exponents, being chosen the theoretical reference model to Barkley, Cross and Major (2005). The construct teaching strategies for the development of the scale was chosen, which includes in itself the constructs teaching strategies and learning strategies. Teaching and learning strategies were delineated to those that promoted collaborative learning, selecting 9; each of which was represented by a subscale of those that make up the EEDAC. They were conceptually and operationally defined and indicators were derived to point the presence of each strategy. According to the indicators, the items were written. Because of the characteristics the construct, the EEDAC consists only of positive items that refer only to the presence of indicators of the use of strategies.
Likewise, the items were submitted to the judgment of three experts to assess the content validity before the application. The reliability of the evaluations made by the judges was assessed using the index proposed by Hernández Sampieri, Fernández-Collado, & Baptista Lucio (2010): reliability = total number of agreements/total number of coded analysis units, which is acceptable when exceeds the score of .85. The reliability index was calculated by strategy and 3 pairs of judges were formed: JAyB, JAyC and JByC. In the same way the total reliability was obtained, product of the sum of the reliabilities of the three pairs divided by 3.
A total of 200 scales were then applied to subjects with the characteristics already indicated in the participants section. The application was self-administered and the specific instructions given to the participants can be read in the Annex, in which the EEDAC is located. Once the scales were applied, the subjects’ responses were coded, puncturing 5 for the response category Always, 4 for Many times, 3 for Sometimes, 2 for Few times and 1 for Never. In the case of items that were not answered, a score of 3 was assigned, which is the average score to be obtained.
Regarding the evaluation of the construct validity and identification of the grouping of the items of each subscale of the EEDAC, an exploratory factorial analysis (EFA) was made. The method of principal component analysis (PCA) was applied as extraction method. To evaluate the reliability of the scale, an inter-item reliability analysis was performed with the Cronbach’s alpha (Cronbach, 1951). Both construct validity and reliability analysis was performed for each subscale, since in the case of different strategies, it was decided to consider each subscale as an independent scale and to verify its psychometric properties separately.
The first objective of the study was to design a set of subscales with Likert-type items, each one evaluating the frequency of use of a teaching strategy for collaborative learning in the classroom. In relation to this and the evaluations of the validity of content made by the three experts in the subject (judges), three items were detected that where two judges disagreed with their content validity, so it was decided to eliminate them and not integrate them on the scale.
The results obtained in the total reliability index of the subscales were greater than .85, that is, acceptable (Hernández Sampieri, Fernández-Collado, & Baptista Lucio, 2006), in each of the subscales. Objectives 2 and 3 aimed to assess the psychometric properties of the scale, so that the EFA was carried out in the first place to generate evidence of construct validity. As part of the EFA, the Barlett sphericity test (1950, 1951) was applied to ensure that the correlation matrix of each subscale was significant (p < .05) and to be able to reject the hypothesis of independence of the variables. The Kaiser-Meyer-Olkin (KMO) sample adequacy measures were also obtained by subscale. The values obtained in the KMO index and in the Bartlett test are shown in Table 2.
In order to be acceptable, the KMO index should be higher than .5 (Kaiser, cited by Hoffmann, 2008) and with respect to the data obtained in the KMO sample adequacy measures of the 9 subscales, the viability of the factor analysis is observed. Likewise, the Barlett sphericity test in each subscale is statistically significant (p < .05), which leads to the rejection of the hypothesis of independence of the variables and it deduces that it is appropriate to continue with the factorial analysis. In sum, both values, KMO and Barlett, fully justify the application of exploratory factor analysis. When selecting the number of factors, the K1 Rule (Guttman, 1954; Kaiser, 1960) and the Cattell Sedimentation Chart (1969) were used as the basis for the exploratory factor analysis.
This study seeks to verify the unifactorial structure of each subscale, as shown in Table 3, the suggested number of factors by the eigenvalue criterion greater than 1 (rule K1) is one factor for all subscales except for Circle of Ideas and Critical Debates. The EFA of the subscales Wheel of Ideas and Critical Debates suggested the presence of three and two factors, respectively. So we proceeded to re-apply the EFA using PCA extraction method, but now fixing a fixed number
Table 2. KMO and Barlett sphericity test.
Source: own elaboration.
Table 3. Percentage variance explained and factorial loads by subscale.
Source: own elaboration.
of factors to extract. As expected, each subscale measures a factor (that is, the strategy it measures), the number 1 was assigned. The percentage of the variance that is explained by a factor, and the factorial loads of the items in each subscale are shown in Table 3. For the analysis of each factor, only it considered those items whose saturation was greater or equal to .40, according to the inclusion criteria proposed by Hair, Anderson, Tatham, & Black (1999). In Table 3, the factor loads are indicated with an asterisk (*) which shows that it is impossible to include his item in the factor.
In the Circle of ideas subscale, the two items with factor loads less than .40 were eliminated (I ask my students to form a circle to start the activity and I indicate to my students that they must answer with a word, expression or a short statement), and the EFA was again made with the aforementioned extraction method. In the new EFA evidence of viability was obtained (KMO = .71, X2 = 253.15, df = 28, p = .000). The variance explained by the factor was 35.05, with an eigenvalue of 2.56 and, as expected, the ranges of factorial loads increased (.449 - .746).
Now when analyzing the reliability of the subscales using the Cronbach’s alpha, it observed the correlation of each item with the total of the subscale to which it belongs. This represents the discrimination index to determine whether or not to delete the item. That is, this value indicates what would be the value of Cronbach’s alpha if the item is deleted. Thus, the items whose scores obtained in the column “alpha if the item is deleted are eliminated” is greater than the Cronbach’s alpha value of the subscale in its entirety. In the case of the Circle of Ideas subscale, Cronbach’s alpha was obtained once the items with loads less than .40 were eliminated. Table 4 shows the Cronbach’s alpha values by subscale.
This research emerged from the interest of deepening the study of the application
Table 4. Cronbach’s alpha by subscale.
Source: own elaboration.
of strategies of collaborative learning in the classroom. Regarding the objective of developing subscales that measured the frequency of use of teaching strategies, it was found that, as a result of the content evaluation of the items by expert judges, the reliability of the content evaluations was acceptable; it followed that the analysis of content by the judges was adequate. After the content evaluation, the scale was applied to a sample for the evaluation of its psychometric properties and it was affirmed that there was evidence of construct validity, through the EFA. Rule K1 indicated that one factor was the one recommended for each, which was congruent, since each measured the frequency of use of a single collaborative learning teaching strategy. As shown in Table 3, the factor loadings of the items are adequate for each factor of the EEDAC.
Following in the same line, reliability was also evaluated. According to De Villes (2003), the internal consistency indexes obtained in all the subscales were interpreted as very good, except for Circle of Ideas, which obtained a minimally acceptable reliability. The results obtained show the achievement of the objectives and provide evidence that the Teaching Strategies Scale for Collaborative Learning (EEDAC) is a scale that has acceptable psychometric properties that can contribute to future research related to the subject when performed in environments and similar conditions in Mexico. Thus, the main contribution of this study is to provide researchers in educational psychology a reliable and valid scale, especially those interested in collaborative work.
The theoretical framework in relation to the adequate application of strategies of collaborative learning by the teachers in the classroom is scarce. Therefore, it is considered that the present study will serve as a push for the knowledge of the same and its application in the classroom. However, it is recommended to have special care in the Circle of Ideas strategy, since its reliability is not very good and the factor analysis reports the presence of three factors in a first solution. This points out that the characteristics of such a strategy could be interpreted as different strategies. This could be due to the fact that the Circle of Ideas is particularly similar to the Storm of Ideas and this strategy is not related to collaborative learning since it can be done individually. Because the EEDAC contains only collaborative learning strategies, the activities carried out by the teacher in order for the student to learn are left out but are more related to meaningful learning (Méndez & González, 2012).
Escala de Estrategias Docentes de Aprendizaje Colaborativo (EEDAC)
Scale of Teaching Strategies for Collaborative Learning (EEDAC)
Instrucciones: A continuación se te presentan 9 series de ítems sobre estrategias docentes que pudieras utilizar para promover el aprendizaje colaborativo en tus estudiantes, así como las siguientes opciones de respuesta:
Instructions: Below are 9 series of items on teaching strategies that you could use to promote collaborative learning in your students, as well as the following response options.
En las columnas de la derecha, marca con una ☑ ó ☒ la frecuencia con la cual utilizas la estrategia docente que se describe. Intenta usar poco la categoría de respuesta A veces, a menos que realmente así sea tu caso. No hay respuestas correctas o incorrectas.
In the columns on the right, mark with a ☑ or ☒ the frequency with which you use the teaching strategy described. Try using the response category Sometimes, unless it is really your case. There are not correct or incorrect answers.