1 Introduction

In today's world, it is necessary to go beyond the mere use of information technologies that have given shape to the future. Computer programming hereafter, programming, as a means of information technologies, is the current method of producing and seeking solutions to problems (Digital Promise, 2021, March). It's clear that today's younger generation must have programming skills, because their interaction with computers goes beyond routine tasks like posting blogs, writing documents, or archiving photos.In our increasingly computational world, students need to learn to create, modify, customize, and adapt tasks via computers. As one of the twenty-first century skills, the basic programming and sub-learning objectives related to these basic concepts and thinking skills inherent in programming have gained priority in today's world (Balanskat & Engelhardt, 2015). In primary school, students mainly need to learn computer basics, fundamental concepts to computing (Wing, 2006), and how computers process algorithms rather than memorizing the syntax of programming languages. As a concept that meets these needs, Wing (2014, para. 5) defined computational thinking (CT) as "the thought processes involved in formulating a problem and expressing its solution(s) in such a way that a computer–human or machine- can effectively carry out". CT must be thought of more broadly than programming or computer science (Israel-Fishelson, & Hershkovitz, 2022; OECD, 2020). It includes skills such as problem solving, decomposing, pattern abstraction and recognition, algorithm design, algorithmic thinking, debugging, and evaluation which skills are fundamental components of programming basics, and that are needed in programming (Eight Reasons Why Every Child, 2021; Yadav et al., 2016). In same way, the relevant literature reports that these skills are directly related to performance in programming (Buitrago Flórez et al., 2017; Tsarava et al., 2017; Wing, 2006). Academic achievement in programming, which measures the acquisition of these key concepts mostly captured by the CT skill, is associated with CT. Thus, learning basic programming concepts and then to code is associated with acquiring the skill CT.

CT and concepts are indispensable prerequisites in today's worldlike other basic skills (International Society for Technology in Education (ISTE), 2021). CT, which has been incorporated into most national curricula, is not only fundamental to computer science, but is also needed independently to solve real-life problems (Bitesize, 2021; Gülbahar & Kalelioğlu, 2018). In addition, Kite et al. (2021) points out that code-centered one-size-fits-all present instructions of CT is not appropriate for all students and the risk that CT is not accessible to all students could be eliminated by integrating it into the core academic curriculum. In order to meet the need to expand the teaching and learning of programming and computer science, new pedagogical approaches are being explored that promote deeper conceptual understanding (Grover et al., 2019). To equip students with basic programming skills, block-based programming applications, other computer-based applications, and unplugged activities are mostly preferred (Brackmann et al., 2017; Merino-Armero et al., 2022). Recently, more and more studies have shown that basic programming learning objectives can achieve through unplugged activities where the computer is not essential (Balanskat & Engelhardt, 2015; Bocconi et al., 2016; Brackmann et al., 2016; Kirçali, & Özdener, 2022). With the help of these positive developments, it is possible to teach the conceptual framework of programming through unplugged activities to students who may not have access to a computer under various circumstances, in order to prepare them for programming classes (Brackmann et al., 2017). Unplugged activities are activities that support the teaching and learning of computer science using cards, pieces of string, crayons, puzzles, games, and crosswords (CS Unplugged, 2021; Gülbahar & Kalelioğlu, 2018). Moreover, digital resources and tools used in education are costly (Merino-Armero et al., 2022), and they may not be accessible to all students (Kite, et al., 2021; Unnikrishnan et al., 2016), and also students may feel uncomfortable using computers (Bell et al., 1998). However, unplugged activities can be a way to avoid some other disadvantages (power outages or an Internet interruption, technical details of computers, the failure of hardware, radiation emissions, etc.) that can occur when using digital resources in the classroom. They prefer unplugged activities in countries where there are insufficient resources for computer education and if teachers’ training in programming is inadequate (Brackmann et al., 2016; Pérez-Marín et al., 2018; Tsortanidou et al., 2022). Accordingly, the digital gap is still there. Therefore, it is possible to teach students in the lower grades the basics of programming concepts without computers (Merino-Armero et al., 2022). One of the key approaches we have for this is unplugged pedagogy. But the question of whether unplugged activities are interchangeable with plugged-in activities requires answers for many different aspects. This study seeks to examine effective strategies for teaching foundational CT practices and basic programming concepts. To this end, this study addresses if plugged-in activities can be replaced with unplugged activities based on the learning objectives.

1.1 Theoretical Background

1.1.1 Computer Programming and Computational Thinking Skill

Computer programming used to be taught by teaching programming languages (e.g. C#, Python, Java, etc.) that are used to programme computers. However, since computers are used by everyone in all fields, beyond its importance for professionals who need to know programming languages to do programming work, but also to understand how computers work (Wing, 2006). Instead of teaching specific programming languages, teaching programming logic, computational thinking, algorithmic thinking and problem-solving skills have become increasingly important (Ramadhan, 2000). In a review study by Falkner and Vivian (2015) of programming education curricula in Australia, it found that the focus is more on course content than on pedagogy, while pedagogical support is neglected. Similarly, Lahtinen et al. (2005) reported that it limited novice programmers to superficial “line-by-line” programming rather than using meaningful programming structures that required computational thinking skills relevant to programming. Although computer education offers a fun and effective learning experience, there is no single pedagogical solution for all grade levels (Garneli et al., 2015). While learning programming, when a programmer transition from process learning to synthesize creative programming in an interdisciplinary way, it will create a difference in terms of the digital ability needs of today (Futschek, 2006; Resnick & Siegel, 2015; Romero & Barberà, 2014). Del Olmo-Muñoz, et al. (2020) claimed that a mixed approach that mixes unplugged and plugged-in activities is better for the early years of primary education than only through plugged-in activities. In this way, besides text-based programming languages, there are some computer tools that are widely used in both programming and programming instruction. One such classification is made by BuitragoFlórez et al. (2017) into three subgroups, as shown in Fig. 1.

Fig. 1
figure 1

A classification of the tools used for teaching programming

Full size image

Robotic kits and concrete media tools (Arduino, GogoBoards etc.), web-based simulation writing tools (Agentsheets and Agentcubes) and graphical (visual) programming environments (Scratch, Alice, Kodu, Greenfoot, LightBot, etc.) can be mentioned as examples. Newly developed tools can form additional groups as well as be integrated into existing ones. With the increasing knowledge of visual programming tools and the growing importance given to teaching programming related concepts to students at younger ages, these studies are even conducted with preschool children (Fessakis et al., 2013). In addition, a review study of information and computer technology (ICT) instructional pedagogy by Waite (2017) indicates that learning models and instructional techniques need to be identified at the initial stage and then the environment for instruction should be determined. The characteristics of the environment where the instruction will take place are listed under four headings in Fig. 2 (Waite, 2017).

Fig. 2
figure 2

The characteristics of the environment for teaching programming

Full size image

With Papert (1980) starting to use some tools that inspired him within the scope of physical programming pedagogy, robots that could be programmed started to be used in the teaching of programming. Physical programming is a process where software and hardware come together and students make "repairs" with wires, sensors, LEDs, and other elements (Meyers, 2019). On the other hand, researches on unplugged computer science have attracted as much attention as alternative tools (Bell et al., 2009; Brackmann et al., 2017; Grover et al., 2019). These are kinesthetic opportunities designed to help students relate complex concepts to their own lives and internalize them (CS Fundamentals Curriculum Guide, 2021). They increase students' motivation, thinking skills, imagination, and computational thinking (Nishida et al., 2008; Sun et al., 2021). In addition, unplugged activities in unplugged implementation pedagogy familiarize students with most basic concepts that are distinct from distracting features and technical details of computers such as game playing and the Internet surfing during instruction (Computer Science Unplugged, n.d.). Indeed, as recommended by the Computer Science Unplugged movement, direct computer programming is not essential for developing computational thinking concepts in students (CS Unplugged, 2021). In the research carried out with plugged-in and unplugged activities in two groups, the increase in CT skills level of the unplugged group in the first phase of instruction was more remarkable (Del Olmo-Muñoz, et al., 2020). There are also recommendations to go beyond text-based programming languages when teaching programming to secondary school students (Szlávi & Zsakó, 2006). As indicated by research studies, everyday practical materials for teaching basic programming should be developed that take into account the development of learners at each stage (Lister, 2016).

Even though programming is at the heart of data processing and computer science, they should teach it to develop CT skills, not in terms of passive application of syntax (Buitrago Flórez et al., 2017; Futschek, 2006). In a review study by Lye and Koh (2014) on teaching and learning the skill of CT, they noted students should not only practice this skill but also reflect on it. The statements about CT in the literature, a fundamental skill for everyone, anticipate new learning objectives that include CT for early stages in schools (Wing, 2008).CT refers to the idea of finding a solution to a problem by following an algorithmic solution after examining the problem and identifying its variables and invariants (Wing, 2006, 2010).It is not only essential for teaching programming skills, but also includes cognitive understanding, which is crucial for computer literacy (Buitrago Flórez et al., 2017; Tsaravaet al., 2017), as well as all basic concepts and applications (Korucu et al., 2017). Digital Promise (2021, March) stating that CT is a skill set that need to be integrated with the core academic subject matter. Due to these characteristics, CT, which is increasingly used in many fields, is attracting the attention of governments and educational institutions, which are working towards including it in primary school curricula (Bougot-Robin et al., 2016; Tsarava et al., 2017; Ministry of National Education (MoNE), 2018a, p.27; Wing, 2014). Contrarily, CT is still mystical to many educators (Digital Promise, 2021, March). Its practical applications are introduced even in early childhood (Lee et al, 2022). Therefore, it is important to know more about which core school subjects are relevant to CT for primary school students. For this reason, the correlation and regression between CT and the acquisition of basic programming skills, as well as the extent to which the components of CT are related to the acquisition of basic programming skills and their contribution to the formation of academic achievement scores, are considered significant data.

1.1.2 Related Studies on Unplugged and Plugged-in Activities in Teaching Programming

Learning computational concepts and developing process thinking will help students acquire more sustainable knowledge and skills rather than the rapidly changing technology (Strnad, 2018) because considering that students can carry out functions by using their own mental tools like a computer can enable them to acquire CT via unplugged activities (Kim et al., 2013). In addition, rather than aiming to teach programming to children, by using unplugged activities or challenging class activities toward CT, it is possible to obtain real positive effects at primary (del Olmo-Muñoz et al., 2020) and secondary school levels even with teachers with low level of teaching experience (Moreno-León & Robles, 2015; Ung et al. (2022);. through a preliminary investigation, they revealed that teachers do not seem to know what computational thinking means in general, similar the study done by Tsortanidou et al. (2022). In research by del Olmo-Muñoz et al. (2020) concluded that unplugged activities seem beneficial in terms of development of CT skills and motivation of 84 s-grade students. They used activities extracted from Code.org courses for instructional intervention.

A study comparing high school students on block-based and text-based platforms found that students on the former platform developed higher levels of learning objectives and more positive attitudes toward programming courses (Weintrop & Wilensky, 2019). In a study by Duncan et al. (2014) was found that 28 of the 47 introduction-to-programming courses used the block-based platforms. Similarly, Bers (2018) stated that ScratchJr is the most popular free introductory programme for kids. On the other hand, positive results regarding the development of the CT were obtained in studies related to project based Arduino educational robot applications (Karaahmetoğlu & Korkmaz, 2019), Minecraft-based coding activities (Kutay & Oner, 2022), programmable floor robot named Bee-Bot (Angeli &Valanides, 2020), the application of unplugged activities (del Olmo-Muñoz et al., 2020; Merino-Armeroet al., 2022), the Scratch application (Bers, 2018; Rodríguez-Martínez et al., 2020), and the learning game, Zoombinis (Asbell-Clarke et al., 2020). As an example, in a study by Yallihep, &Kutlu (2020) a significant development was revealed in the achievement levels regarding programming concepts of 5th grade students studying in the Lightbot platform when compared to those whose studies were strictly guided by the MoNE ITS (Information Technology and Software) curriculum (MoNe, 2018b). It seems that an attempt was made to teach the basic programming concepts with only one tool, and thus to evaluate only one tool.

On the other hand, in terms of both computer programming and basic ICT life skills, CT is a skill that needs to be developed at a young age (Conde et al., 2017; ISTE, 2021). Thus, the methods and approaches that can be effective in teaching this skill should be studied according to the principles of accessibility, cost-effectiveness and suitability for students, and lessons should be derived from the results. It can be observed that unplugged activities in lower grades are significantly beneficial at lower costs when access to computers is limited and when disadvantages related to technical failures and some computer-related problems (child screen time out, the distraction of attention, radiation emission, etc.) are to be avoided, and even in conditions where teachers do not have sufficient programming skills (Moreno-León & Robles, 2015; Nishida et al., 2008). In addition, some studies have found positive results on unplugged activities in relation to the studied variables such as motivation, CT, gender, academic achievement (del Olmo-Munoz et al., 2020; Delal, & Oner, 2020; Nishida et al., 2008; Kirçali & Özdener, 2022). By reviewing studies, it is observed that unplugged activities were developed, introduced (Bell et al., 2009; Sendurur, 2019) adapted to the grade level in the subject (Meyers, 2019; Nishida et al., 2008), or their effect was investigated alone (Delal & Oner, 2020) or compared to a computer application (del Olmo-Munoz et al., 2020; Kirçali & Özdener, 2022). It cannot be guaranteed that all learning objectives of a curriculum or syllabus will be achieved using block-based or visual programming tools with their unique interfaces, activity flows, and scopes, as the CT involves different thinking skills (CS Unplugged, 2021; Yadav et al., 2016). CT exploration is limited in Primary Education.

The aim of the present study was to compare the unplugged activities with their equivalent plugged-in activities used in teaching programming basics in 6th-grade information and software technologies (IST) courses to find their effect on students' academic achievement in basic programming, computational thinking, and opinions. To this end, the answers to the following research questions were sought:

  1. 1.

    Is there a significant difference between the pre and post achievement tests in basic programming within the plugged-in group and the unplugged group?

  2. 2.

    Is there a significant difference between plugged-in and unplugged groups in terms of academic achievement in basic programming?

  3. 3.

    Is there a significant difference between the plugged-in and unplugged groups in terms of computational thinking skills?

  4. 4.

    Is students' computational thinking ability a predictor of academic achievement in basic programming?

    4a. Is each sub-dimension of the computational thinking scale a meaningful predictor of academic achievement in programming?

  5. 5.

    What are the opinions of the participating students about the plugged-in or unplugged activities that were conducted in their group?

2 Method

2.1 Design

In this study, sequential-explanatory research design, one of the mixed-methods, was used sincethe quantitative data were collected and analyzed first, and then the results were examined and used to plan the qualitative phase. The purpose of this design is to use qualitative data to further explain quantitative results (Creswell, 2013). In this study, for the quantitative phase, a quasi-experimental design was designed with two groups. It was conducted to examine whether there were differences between the groups in computational thinking skills and academic achievement of basic programming. A pre-test academic achievement was administered to test the equivalence of the two groups before intervention. For qualitative phase, 12 students were interviewed, three students from the upper and lower subgroups of each group. Interviews were conducted to learn students' opinions of the implementations and to reveal their experiences. The research design processes of the study are shown in Fig. 3.

Fig. 3
figure 3

Details regarding the research design employed in the study

Full size image

2.2 Sample

The study was conducted with a total of 84 participants consisting of 42 girls and 42 boys 6th-grade students (between the ages of 10 and 11) in two different schools. The sample was obtained through non-random convenience sampling method while the classes were randomly assigned to the groups. Table 1 shows the homogeneity of the groups. According to Piaget (1977), children between the ages of seven and eleven can solve problems that relate to concrete objects, but not abstract concepts or phenomena. In fact, unplugged activities are more related to concrete materials and concrete phase of child development. Moreover, the "unplugged" activities of the national curriculum for 6th grade were designed for this stage of development and compared with the "plugged-in" activities. Accordingly, sixth grade students are the participants in this study.

Table 1 Demographic Characteristic of the Participants
Full size table

2.3 Instruments

Based on the aim of the study, three different instruments were used to collect data. While the academic achievement test and interview protocol were developed by the researchers of the present study, the Computational Thinking Skills Scale was developed by Korkmaz et al., (2015a, b).

2.3.1 Academic Achievement Test

For the development of the academic achievement test, the purpose of the test was initially identified, the learning objectives were specified, and subsequently the table of levels of objectives was prepared, accordingly the test items were written, revised and the pilot test form was prepared. The pilot test was conducted with 104 6th grade students who responded to the preliminary implementation form, which consisted of 33 questions. There are studies in the literature on the development of achievement tests with as many or fewer respondents as in this study, e.g. Djambong et al. (2018), Lin (2018), and Marie and Sreekala (2015). It was conducted on an easily accessible sample near the teacher, one of the authors. The number of questions was limited by the amount of testing time available for nearly 10 years old students, 40 min. Item difficulty and item discrimination analysis in the achievement test was calculated to make the test items discriminant for high and low scorers. Based on the preliminary item analysis results of the pilot study, 5 items were removed from the form; the mean score of the 28 items remaining in the form was 15.67. The coefficients of skewness and kurtosis of the data were within acceptable ranges, 0.31 and -0.52, respectively (Tabachnick & Fidell, 2013). The standard deviation was found to be 5.49, and a Cronbach alpha reliability coefficient of 0.82 indicated that the test was reliable.

2.3.2 The Computational Thinking Scale

The original version of the computational thinking scale, which was adapted to the secondary school level in this study, was developed by Korkmaz et al. (2017). The final version of the scale consists of 5 sub-dimensions with a total of 22 items and is based on a 5-point Likert scale, with “5” indicating the most positive and “1” the most negative. In the present study, the reliability coefficients of the post-test administered to the unplugged group and the plugged-in group were found to be 0.82 and 0.71, respectively.

2.3.3 Interview protocols

Two open-ended interview sheets were prepared for each group with 11 questions related to the activities in their group (See in Fig. 4). The aim of the interview questions was to conduct an in-depth analysis of the students' opinions regarding the challenges they encountered during class, their expectations, feelings, performance levels, anxiety and satisfaction, the methods they used in solving problems, the solution methods they applied to algorithms and problems, and their self-assessment of their own learning. In accordance with the purpose of the study, the design of the interview forms was made using the literature and taking into account that the questions must not be predictable, must be easy to understand without leading to short answers, must not provoke a negative reaction from the respondent, and must encourage a detailed and explanatory answer (Yıldırım & Şimşek, 2016). It is important that each question focuses on a single dimension so that in-depth responses can be obtained (Creswell, 2013). Since students have difficulty giving long answers, more questions were written. Then, it was ensured that two academics reviewed the form and two students who were not part of the sample responded.

Fig. 4
figure 4

Interview procedure of groups

Full size image

2.4 Procedure

The intervention process of the study was conducted in the 6th-grade ICT course in four different classes over 7 weeks by the same teacher. It extended over a total of 14 lessons—2 h per week. The steps followed in the intervention process are shown in Fig. 5.

Fig. 5
figure 5

The implementation process of study

Full size image

Here, the basic programming learning objectives of the intervention were part of the curriculum of the 6th grade MoNE curriculum (MoNE, 2018b). Its unplugged lesson plans were used in the unplugged group as teaching materials for the study. Besides, some of the basic programming teaching tools (code.org, scratch, compute.it; lightBot) and some other applications that enable computer interactions (kahoot.it; draw.io; learningapp.org) were used in the plugged-in group. During the 7-week intervention process, unplugged and plugged-in activities that differ from the common presentation and narrative are listed by objectives for both groups in Table 2. Based on these objectives, plugged-in activities have been developed or selected to correspond to the unplugged activities in the program, and presented weekly in Table 2.

Table 2 Plugged-in and Unplugged Teaching Activities and Practices Used for Instruction Intervention
Full size table

The activities, B1 worksheets, that were performed in week 5 and week 7 in the unplugged group is presented as an example (See Fig. 6). These activities are part of MoNE curriculum of information written by Gülbahar Güven et al. (2018).

Fig. 6
figure 6

Pattern Recognition (6.2.7.B1) and Algorithmic thinking (6.2.5.B1): Sample activity for unplugged group

Full size image

The other common and different implementation processes in the groups are shown in Fig. 7.

Fig. 7
figure 7

The characteristics of the teaching activities used in the groups

Full size image

The unplugged group followed both common and unique instructions in the traditional classroom setting, as listed in Fig. 7; activities were either done in groups or pairs or individually in class or completed as homework. In the plugged-in group, computer applications were used to address the targeted learning objectives (See in Table 2). These included various block-based programming applications as well as Draw.io (2021), a drawing platform, and a platform for preparing interactive educational materials, as well as other platforms such as LearningApp and Kahoot. In the plugged-in group, the computer applications were also used for summative assessment, with the goal of continuing the computer interactions. After the implementations, the scales were administered to both groups as a posttest. Following this, the interviews were conducted.

2.5 Data Analysis

In the present study, data were collected before and after the 7-week intervention process. The dependent variables were academic performance in basic programming and computational thinking, while the independent variables were the unplugged and plugged-in instructions. Instead of removing students with no answer (NA) from our analysis, we run the missing data analysis and assigned mean values to the missing data. This is because we do not know why the students did not answer them. It may be due to various circumstances, for example, that he/she skips the question because this happens to young students. Then the normality tests were conducted and the assumptions for the inferential statistics were supplied. The Shapiro–Wilk test for the pre-test achievement scores yielded values of p = 0.118 and p = 0.090 in the groups, respectively, (p > 0.05). In addition, the “Hedges’ g” value was also calculated to explain the effect size of the implementation. In the sub-dimensions of the CTs and overall, the skewness and kurtosis values were found to fall between ± 2, which indicate a normal distribution. The values that the regression analysis yielded met the required assumptions, and the test was conducted accordingly. For the analysis of quantitative data, students' responses were transcribed into written form. A descriptive analysis was conducted to identify student responses in both groups.

3 Results

3.1 Comparison of academic achievement scores before and after instruction

The present study examined whether there was a significant difference among the groups in terms of academic achievement and CT skills. First, the mean scores of the pre-academic achievement test before the intervention were compared between the groups using a t-test for independent samples; the results are presented in Table 3.

Table 3 Independent Sample t-Test of Pre-Achievement Test Scores
Full size table

As can be seen in Table 3, pre-test scores did not significantly differ among the groups (p > 0.05). Accordingly, the groups' pre-implementation level of basic programming was equivalent. Next, Table 4 summarizes the difference between students' within-group pre-test and post-test academic achievement scores using dependent sample t-Test.

Table 4 Dependent Sample t-Test of Pre-test and Post-test Scores for Academic Achievement
Full size table

Table 4 shows that there was a significant statistical difference between pre-test and post-test academic achievement scores in favor of the post-test of students in both, the unplugged group (t = 9.30, p < 0.05) and the plugged-in group (t = 5.65, p < 0.05). In this way, both sets of activities had a significant effect on students' academic performance in basic programming.

3.2 The comparison of the achievement scores of the groups after the intervention

Table 5 shows that the results of the comparison between the academic achievement scores of the students in the unplugged and plugged-in groups after the intervention were examined using the Independent Sample t-Test.

Table 5 Independent Sample t-Test of Post-Achievement Test Scores
Full size table

According to Table 5, the post-test mean score of the unplugged group (MUnplugged = 57.923, SD = 2.600) was higher than that of the plugged-in group (MPlugged-in = 48.510, SD = 3.110). Moreover, there was a significant difference among the groups’ post-achievement test scores (t(82)=2.337; p < 0.05). This evidences that the use of unplugged activities in the classroom supported greater improvement and had a more positive effect on student academic achievement than the use of plugged-in activities. Moreover, with a value of d = 0.51, the Hedges’ g effect size had a moderate level of practical significance.

3.3 The Across-Group Comparison of the Computational Thinking Scale Scores after the Intervention

The level of students CT was investigated by analyzing the descriptive data in the computational thinking scale, the results of which are presented in Table 6.

Table 6 Descriptive Statistics related to the Computational Thinking Scale after the intervention
Full size table

The total scale means of the unplugged group and that of the plugged-in group can be seen in Table 6. The visual descriptive CT levels of the students in both groups are shown in Fig. 8.

Fig. 8
figure 8

The CT levels of the students in the groups

Full size image

In both groups, the majority of the students had a high level of CT (Nunplugged = 25; Nplugged-in = 24). After controlling for pre-test achievement scores, there was not a significant effect of teaching condition on computational thinking skill scores (F(1,81) = 0.108, p = 0.743, ηp2 < 0.01). Estimated marginal means were similar in the plugged-in (M = 3.516, SE = 0.089) and unplugged (M = 3.557, SE = 0.085) conditions. Pre-test achievement was significantly related to computational thinking skill test scores (F(1,81) = 8.948, p = 0.004, ηp2 = 0.099 > 0.01).

3.4 The computational thinking skill as a predictor of academic achievement in the unplugged group

No significant correlation was identified between the examined variables in relation to the students in the plugged-in group (p > 0.05). In the unplugged group, there is a significant positive correlation of moderate level between students’ academic achievement and their CT (r = 0.536, p < 0.01). Therefore, a basic linear regression analysis was performed for the unplugged group to identify whether the level of CT predicted academic achievement, between which a moderate degree of positive linear correlation was found. The model developed was found to be significant (F(1, 42) = 16.956, p < 0.05, R2 = 0.271) (see Table 7).

Table 7 The Predictor Model of the Academic Achievement of Students in the Unplugged Group
Full size table

Based on the model, it was revealed that the levels of CT significantly predicted students’ post-test achievement scores. The model accounts for 27% of the academic achievement.

3.5 Examining the Correlation between Academic Achievement Scores and the Computational Thinking Scale Subdimensions for the Unplugged Group

When the correlation between the academic achievement scores of the students in the unplugged group and the sub-dimensions of the computational thinking scale is examined, a significant correlation is observed between some variables (p < 0.05). The obtained results are presented in Table 8.

Table 8 The correlation between the subdimensions of the computational thinking scale and academic achievement in the unplugged group
Full size table

Accordingly, it was found that there was a positive moderate correlation between academic achievements and the sub-dimensions creativity, algorithmic thinking and problem solving of the computational thinking scale, while there was a positive but weak correlation between critical thinking and academic achievement. Then, the results obtained by the multiple regression analysis are presented in Table 9.

Table 9 The Predictor Model of the Academic Achievement of Students in the Unplugged Group by the Computational Thinking Scale Subdimensions
Full size table

The hypotheses predicting academic achievement for subdimensions other than creativity was rejected. The hypothesis that only the creativity subdimension of the computational thinking scale is a predictor of academic achievement was confirmed by the resulting regression model (p < 0.05). Creativity explained 29% of academic achievement according to the resulting model.

3.6 Findings Regarding Students’ responses to interview

The responses received from six students from each of the groups – plugged-in and unplugged – were categorized. The groups were coded as S1_x (plugged-in) and S2_x (unplugged). In Table 10, the codes for the themes and the frequencies are presented for each group separately, and the responses that were commonly expressed are indicated with a different bullet point.

Table 10 The process-related opinions of the students in the plugged-in and unplugged groups
Full size table

In both groups, students liked the classroom activities. They rated the teacher's classroom management as positive. One student in plugged-in group said, "I was really excited when we were going to play games." I did not feel sad at all. The games were fun (S1_4)." In the plugged-in group, feelings of excitement and a sense of collaboration were emphasized. Students in the unplugged group, on the other hand, indicated that the lessons were educational and that they had more freedom during the lessons. One of them said, "I felt happy. We learned how to solve the questions." (S2_4). They highlighted that they learned the material and said, "I felt happy. We learned how to solve the questions." (S2_4).

As a barrier to learning, anxiety must be kept at a low level. In fact, all students who participated in both groups of activities indicated that they felt no anxiety throughout the learning process. They attributed the lack of anxiety to the fact that they were able to ask their questions easily and had no difficulty learning. One of the unplugged groups said, "I was not anxious because I could ask the questions that were on my mind. The teacher explained them right away" (S2_3). Furthermore, making lessons palatable to students makes learning conducive. In fact, students in both groups found the learning process enjoyable. Some students in unplugged group mentioned that it was fun to learn certain topics. Especially in the plug-in group, playing games was found to be entertaining. One in the plugged-in group said, "I had fun because I learned how to solve the questions; I learned how to use the computer and I had fun when we played games” (S2_3).

Another consideration is that if the instructions are challenging, this makes learning difficult. Fortunately, the plugged-in group expressed the opinion that learning technology made their lives easier. In the unplugged group, it was pointed out that the activities facilitated learning because they were practical, but three students indicated that they faced challenges when they did not understand how to perform the activities. Because block-based programming environments for children allow trial and error, they are not afraid to make mistakes. Therefore, plugged-in activities challenged them rather less. Students were also not bored because of the gaming effect of the plugged-activities. However, two students in the unplugged group reported boredom and one said, "I was bored. I got bored solving the algorithm."(S2_2). However, all students in the plugged-in group indicated that they were not bored. The reasons that they were not bored are that they were able to express their opinions freely and understood the topics covered in class well.

Topics mainly include understanding and creating algorithms and related concepts. For example, students were asked about the effect of solving problems using algorithms. Three students in the unplugged group indicated that using algorithms to solve problems makes things easy, saying, "Algorithms are a very simple method. More explanatory." (S2_3). Similarly, five students in the plugged-in group stated that using algorithms was useful as it enabled one to do the steps in the work in a sequential and organized way. They also stated that they planned and carried out the work step by step, using a problem-solving method that they improved by using an algorithm. Similarly, two students from the unplugged group used the word “nice”, and one student from each group used the word “good”, stating that they were able to understand how to solve the problems, how to develop problem solving methods like finding the solution by following a step-by-step approach. It seems that the unplugged group developed a better understanding of how to use algorithms to solve problems, saying, "I can find the answers to questions instantly. I learned how everything is or will be."(S2_3).

After that comes the question about their activities. The unplugged group rated activities as engaging in activities, learning more effectively, enjoying using pen and paper, and learning with visual aids. They said “They enabled me to learn more effectively because I like it when I do them on paper.” (S2_2)“They enabled me to learn more effectively because when you assign an activity, I remember it. I still study with activity worksheets.” (S2_6). Similarly, plugged-in group reported learning more effectively through trial and error, using the computer more proficiently and learning with visual aids. Furtermore, students then gave their opinions on the impact of the activities on their academic performance. The majority of the unplugged group stated that the activities contributed to their achievement and that they learned during the activities. One said, "The activities in the course allowed me to get a slightly higher score; I learned the algorithm better through these activities."(S2_1). One student said, “The activities in the course did not help me get a higher score because I did not feel like it was real lessons. We did not do a tight class” (S1_1). The conclusion is that the computer activities and subsequent paper and pencil exam might make it difficult for the plugged-in group to transfer their knowledge. But the unplugged group had already learned with paper and pencil and kinesthetic activities. This is also consistent with the quantitative results of academic achievement scores.

4 Discussion

4.1 Within-Group and among-groups Differences in Academic Achievement

The purpose of the present study was to compare the plugged-in and unplugged activities in terms of programming academic achievement, hereafter academic achievement, and CT skill in two groups. There was no significant difference in pre-test academic achievement scores among unplugged and plugged groups. This shows the equivalence of the groups' preliminary knowledge prior to the intervention. It was found that the difference within-group pre and post-academic achievement test scores, was in favor of the post-test, suggesting that both unplugged and plugged-in activities led to learning. The intervention resulted in improved academic achievement and CT skill. Importantly, a significant difference was found in academic achievement in favor of unplugged activities. Similarly, there are studies in the related literature that report a positive impact of unplugged activities on achievement in basic programming (Tsortanidou et al., 2022; del Olmo-Munoz et al., 2020; Gülbahar & Kalelioğlu, 2018; Kalelioğlu, 2018; Moreno-León & Robles, 2015; Rodriguez, 2015). Unplugged activities facilitate the understanding of basic concepts related to computer science and the concept of programming (Kalelioğlu, 2018; Rodriguez, 2015). In the present study, the students in the unplugged group stated that what they learned became clear in their minds, that they arrived at solutions more easily, learned algorithm more effectively and found the topics easy, which verifies its significant difference among the two groups. This shows that unplugged activities help students focus better on the subject matter by eliminating the destructive effects of computers or other digital tools. Besides that, the plugged-in platforms are standardized and this limits the adaptability of them to students' local experiences in the classroom. However, the ease of adapting and redesigning unplugged activities could possibly support the learning of the unplugged group. In a study by Pérez-Marín et al. (2018), the results showed that it was possible to teach basic programming concepts in a short time through the use of unplugged activities, and they observed a development of CT skills. Moreover, in the present study, students in the unplugged group commented on the reasons that positively affected their learning, such as the fact that they liked the pen and paper activities and the effectiveness of learning through visual posters. Showing relationships between concepts and how concepts interact with each other on visual aids or posters seems to be very useful for teaching basic concepts.

In fact, students who participated in the plug-in activities cited as reasons for this difference that some of them felt they were not participating in a typical course, that the course was not taught effectively, and that they could achieve higher scores if there were no computer games in the classroom. However, this may not be a barrier to the fact that, as Rodríguez-Martínez, et al. (2020) found, integrating CT with Scratch applications resulted in higher student academic achievement compared to students exposed to the traditional method. In the present study, each platform used was designed or selected by considering various characteristics of the various plugged-in activities that could address the corresponding learning objective, in order to diversify and adapt the activities as much as possible to the level of knowledge of the students. If the teaching practice of computer programming is adapted to the age level of the younger students, they will internalize it more easily (Pérez-Marín et al., 2018). In conclusion, sixth grade students can learn easily when activities and metaphors are adapted to the level and experience of them and topics that are not related to the learning objectives are eliminated. Brown et al. (1989) stated that when teaching simple and concrete computer science concepts, students could work collaboratively, solve problems, and think creatively using unplugged activities. Zhan et al. (2022) supports this situation for unplugged activities, also states that it offers more interaction. During the interviews of the plugged-in group in the present study, the students indicated that they felt enthusiastic while playing, that they were happy, and that they were not bored or anxious. Even though students viewed the process as positive, they did not feel they were in a typical course. In a study by Yallihep and Kutlu (2020) which investigated the effect of games on students' understanding of programming concepts, it was reported that teaching through games did not have a positive effect on students' attitude towards the course. Meerbaum-Salant et al. (2013), who evaluated the teaching materials they developed via Scratch with secondary school students, reported that they experienced difficulties in teaching certain topics such as compilation, variables, and flow. However, to avoid bias, for which platform the academic achievement test was developed should be taken into consideration when evaluating these results.

On the other hand, there are certain negative effects related to unplugged activities. For example, Tsarava et al. (2022) and Jun (2018) stated that unplugged activities are gaining popularity, but standards were lacking in the assessment of these activities. Similarly, students in the present study attributed their inability to achieve higher scores to the fact that several assessment questions were different from classroom activities. It may be that students were unable to relate the hands-on activities to those in the pen-and-paper exams. Related to this point, Falkner and Vivian (2015) reported that there is a further need for unplugged materials to be prepared in a standardized manner and for assessment and critical thinking materials to be developed for primary school students as part of the programming process. Rodríguez-Martínez et al. (2020) made it clear that students had little prior experience with CT and they implied that the development of computational concepts does not happen by chance at this stage of students' education. The long time it took the students to complete the activities also showed us that. When they get into it, it goes faster. In the present study, the finding that the impact of the teacher's classroom management approach was viewed positively by the students shows a consistency with the findings reported in other studies that unplugged activities contribute positively to students' long-term memory span and motivation level due to the interactions they have with each other and with their teachers (Zhan et al., 2022), and that close and effective supervision by teachers can facilitate more effective learning than leaving students alone with the devices (Meerbaum-Salant et al., 2013).

4.2 The Difference among the Plugged-in and Unplugged Groups in terms of the Computational Thinking Skill

No significant difference was found between the two groups of activities. Five sub-dimensions of the computational thinking scale were addressed in our study namely creativity, algorithmic thinking, cooperativity, critical thinking and problem solving. Both groups scored approximately the same mean scores in these sub-dimensions.

In both groups, the students' CT skills were at an intermediate or high level. The unplugged group achieves slightly higher mean scores only in the sub-dimensions algorithmic thinking, cooperativity and problem solving of the computational thinking scale. Similar results have also been reported in related literature (Atman Uslu et al., 2018; Brackmann et al., 2017; Chen et al., 2017; Çelik&Özdener, 2019; Davies, 2008; Harris, 2018; Pérez-Marín et al., 2018; Zhan et al., 2022). Similarly, Merino-Armero et al. (2022) said that the unplugged approach contributed to students CT skills. In a study by Brackmann et al. (2017), it was reported that students who engaged in unplugged activities developed more skills than those who engaged in plugged-in activities. In an experimental study, it was found that the game-based unplugged group showed better academic performance (Kuo & Hsu, 2019). In a study of teachers by Harris (2018), it was highlighted that the pre- and post-intervention groups showed similar levels of skill and that teachers learned in a more holistic, enjoyable, and confident manner through the use of unplugged activities.

The finding that students gain depth of interaction in the unplugged activities (Kuo & Hsu, 2019; Zhan et al., 2022) supports the high means score in favor of the unplugged group in the cooperativity sub-dimension in our study. One of the reasons could be that unplugged activities provide more opportunities for collaboration. On the other hand, Yıldız et al. (2017), in their study investigating whether there was a difference between students' computational skills in relation to the type of digital games they played, found that the level of critical thinking skills was low among secondary school students. This result is consistent with the critical thinking sub-dimension being rated low in both groups compared to the other sub-dimensions and the overall scale.

The fact that there was no significant difference among the groups in terms of intervention activities should not be viewed negatively. Thus, in the posttest, the level of CT was high in both groups. Similarly, there are studies that report that unplugged activities develop these skills at least as much as plugged-in applications and that they motivate students (Atman Uslu et al., 2018; Brackmann et al., 2017; Çelik & Özdener, 2019; Davies, 2008; Pérez-Marín et al., 2018). While block-based programming activities have a positive effect on the development of CT (Karaahmetoğlu & Korkmaz, 2019; Turan, 2019), there are also studies that find no significant difference between the compared applications (Atman Uslu et al., 2018; Gülbahar & Kalelioğlu, 2014). So, based on these results, it can be claimed that when students have no prior familiarity with computers, their interest in how computers operate can be developed based on knowledge of unplugged activities. Because they are focused on the subject and the task and are not afraid of breaking the computer or don't want to play games on the other windows of the computer. In parallel with these results, students with unplugged activities are able to explore computer science concepts and fundamentals more deeply without the presence of a computer, which they usually consider as a tool or toy (Bell et al., 2009).

As in the present study, Hsu, et al. (2018), who conducted a literature review study on how CT should be taught and researched and the reasons why no significant difference was found in the CT, recommended that the method and content of instruction should vary as students grow older and progress to higher grade levels. The group activities and active learning activities that Zeybek (2019) recommended to develop students' cooperative characteristics seem to indicate unplugged activities. For example, the students in the unplugged group expressed in their opinions that the lessons and games were integrated and that they felt happy, relaxed and had a sense of freedom. Zhan et al. (2022) also found that the unplugged group was more engaged and this improved their classroom interaction and communication. Indeed, having students engage in unplugged activities as a group or in pairs seems to enhance communication, interaction, and learning. Merino-Armero et al. (2022) discussed how CT can be introduced in schools. Tsortanidou et al. (2022) highlight the explicit need to integrate CT and related concepts into the primary curriculum. We believe that the preliminary readiness of teachers is a precondition to integrate it into the primary curriculum explicitly. Sendurur (2019) analyzed the unplugged activities designed by prospective computer teachers. He pointed out that the design process showed that the teacher candidates' designs were incompetent. In fact, we have also seen how important it is for students in the classroom that teachers place value on unplugged activities. When teachers were familiar with unplugged activities and knew how to use and adopt them, students felt freer and perceived the lessons as more fun and educational, they said. Students in the unplugged group may have felt freer because the way activities were conducted was based on creativity and the possibility of different solutions. However, for rural students who mostly have little experience with computers, unplugged activities can be an effective way to improve their CT. In fact, both pre-service and in-service teachers need to know how to design an unplugged activity and how to develop possible solutions to potential implementation challenges. In our implementation experience, the students' reading level and the time required for the activity were very critical to their performance. Similarly, Tsarava et al. (2022) indicated that CT has a positive association with students' linguistic thinking ability.

4.3 The Correlation between the Variables in Groups

In the present study, a moderate positive correlation was found between academic achievement and CT in the unplugged group. Academic achievement posttest scores were significantly predicted by CT. This correlation indicates that development in the CT increases academic achievement in basic programming classes. There are also other studies that found this correlation (Chen et al., 2017; Conde et al., 2017; Donley, 2018; Korkmaz et al., 2018; Korkmaz et al., 2015a, b; Román-González et al., 2018; Taylor et al., 2010).

A study by Korkmaz et al. (2018) found a correlation between achievement in programming and CT. A study by Román-González et al. (2018) has empirically supported the statement that CT and coding achievement are cognitively correlated in middle school. In addition, Donley (2018) found a moderate significant correlation between academic achievement and creativity. This finding shows consistency with the finding in our study regarding the correlation between academic achievement and its predictor, the CT, and the creativity subdimension. Korkmaz et al. (2018) also showed a moderate correlation between achievement and problem solving-another subdimension of CT. Similarly, Chen et al. (2017) reported that students who did not follow the codes in their observations found a more effective and creative solution method and showed effective CT. Moreover, Korkmaz et al., (2015a, b) found that programming performance can be predicted by the CT. Thus, as creativity increases, academic achievement is also expected to increase, since the only sub-dimension of CT, which is accounting for academic achievement, is creativity.

4.3.1 The correlation between academic achievement scores and subdimensions of the computational thinking scale for the unplugged group

Since the CT is a predictor of academic achievement, the correlation between academic achievement and the subdimensions of the CT was examined for the unplugged group. Accordingly, it was found that there was a positive moderate correlation between academic achievement and the sub-dimensions creativity, algorithmic thinking and problem solving of the computational thinking scale, while there was a positive but weak correlation between critical thinking and academic achievement. With the exception of the cooperativity sub-dimension of CT, all sub-skills can be considered effective in enhancing performance in the basic concepts of programming related to unplugged activities. Based on the developed model, 29% of academic achievement is predicted by creativity sub-dimension alone. This could mean that CT is a skill closely related to creativity. We can emphasize that creativity-enhancing teaching practices are needed to improve students' basic programming skills. Also, in a study by Atiker (2019), a significant difference in the creativity sub-dimension was found between groups in favor of the experimental group. Likewise, Berry (2014) argues that creativity has informed ICT teaching for many years and that it would be wrong to neglect this important factor within CT. By focusing on creative thinking within CT, students are helped to engage with new ideas (Brooks, 2019). Similarly, students in the present study felt that developing new methods of solving problems through unplugged activities helped them understand how to solve problems in a relatively easier way than before. In this way, children learn how to analyze problems and improve solutions. Learning problem-solving skills can also decrease their anxiety about the subject and teaching.In addition, they believed that algorithms helped in getting the job done faster and saved them from making mistakes. Atun and Usta (2019) stated that programming instruction developed students' CT as well as problem solving skills. Finally, Nouri et al. (2020) supported with evidence that there is a strong correlation between programming and creativity.It must be noted because computer programming is the way of utilizing of utilizing of computers to form solutions of crucial problems of human being, it gives emphasis on the creativity and problem-solving skills.

5 Conclusions

The aim of the present study was to compare the unplugged activities and their equivalent plugged-in activities used in teaching basic programming in 6th grade ICT course and to reveal student opinions. Programming basics and CT are rapidly being integrated into primary education curricula, and tools and methods are being developed to teach them. As the findings of this study show, the literature recommends unplugged activities as an effective option for making these skills accessible to all basic education students. This study makes a contribution to the impact of unplugged activities for teaching basic computer skills, especially in countries and regions with limited access to computers and Internet technology. In doing so, it demonstrates that digital tools such as computers and tablets are not essential for the acquisition of these skills by comparing unplugged activities with plugged-in activities based on learning objectives. The platforms that could address the related learning objective was selected for plugged-in group by taking into consideration different features of various plugged-in activities; attempts were made to add variety to the activities and to adapt to the students’ level of knowledge as much as possible. As a recommendation, the type of teaching activities must be selected according to the conditions, interests, level of knowledge of the students, etc. This study has some limitations. Because of physical circumstances of classrooms (the limited numbers of computers or other materials), some activities were conducted in groups or pairs. In addition, analysis of item difficulty and item discrimination in the achievement test was conducted for 104 respondents.

A significant difference in academic performance in basic programming was found in favor of the unplugged activities and that the unplugged approach-maintained students’ CT skill compared to the plugged-in group. Also, the creativity sub-dimension of CT explains 29% of academic achievement in the unplugged group. This suggests the importance of CT to academic achievement in basic programming. Creativity subdimension of the computational thinking scale is a predictor of academic achievement was confirmed by the resulting regression model (p < 0.05). Also, the students in the unplugged group expressed their positive perspective that lesson and games were integrated, and that they felt happy, relaxed, and a sense of freedom. In the present study, students in the unplugged group indicated that what they learned became clear in their minds, they arrived at solutions more easily, learned algorithms more effectively, and found the topics easily, confirming the significant difference between the two teaching practices. However, students who participated in plugged-in activities indicated that they did not feel they were taking a regular course, that the course was not taught effectively, and that they would score higher if there were no games in the course. Indeed, the use of unplugged activities for the basic ICT skills development needs to be extended especially to learning environments in rural areas where there is limited access to computers so as not to widen the digital gap.

6 Recommendations

Taking into account the findings of the present study and the knowledge and experience gained during the research, planning, implementation and follow-up phases, some recommendations were made. The goal of extending algorithmic thinking skills to lower grades requires studies and research on more effective and comprehensive content designs. Therefore, it can be suggested that more unplugged activities can be developed or adapted to the cultures.

  • Researchers can develop standardized assessment instruments that measure basic thinking skills independent of instructional content, thereby eliminating some uncertainties and biases. Standardized tests can be developed to assess academic achievement that incorporates unplugged methodology.

  • Depending on the nature of the learning objectives, both researchers and teachers should consider the different pedagogical characteristics, target audiences, and teaching methods of tools to select.

  • It is recommended that studies should be conducted on the method and content of different unplugged activities for students of different age groups in different grade levels.

  • A study can be conducted to investigate how much learning is achieved when the unplugged activities are presented individually in one group and in groups or pairs in another group. In conclusion, may be recommended to use a larger sample size to examine the CT and other variables.

  • Students with computer experience and students with limited experience can be compared on study variables with the same unplugged activities. So, in which group CT and programming basics would be developed more through unplugged activities.