The physical components of embedded systems that can be touched or seen, such as the CPU, motors, actuators, and other physical pieces, are referred to as hardware. Software, on the other hand, refers to the capacity to create and write programs in a variety of languages to solve technical challenges.

In fact, thinking about difficult talents as skills that can be shown is the simplest way to remember them. Furthermore, embedded systems are typically designed to interact with the physical world via sensors and actuators (Stern, 2001; Adunola, 2011).

Embedded system education should be multidisciplinary in nature and includes several aspects of control and signal processing, computing theory, real-time processing, optimization, and evaluation as well as systems architecture and engineering, according to the ARTIST Guidelines for a Graduate Curriculum on Embedded Software and Systems (Abele et al., 2015). Many universities, such as Ghent University (Schmidt-Atzert, 2004) and Universität Stuttgart (Schuler and Schuler, 2006), provide embedded system courses with the goal of addressing the practical difficulties that arise in the design and development of embedded systems, which include both software and hardware interfaces.

Meanwhile, Oakland University provides a Master of Science in Embedded Systems course (Marwedel, 2011), which takes 32 graduate credits, with 16 core requirements, and at least 12 depth requirements. Others (Ilakkiya et al., 2016) created an embedded system engineering curriculum for a 4-year undergraduate degree (ESE). A body of education knowledge for embedded system engineering was defined in the suggested curriculum, and it comprises 15 knowledge categories, as shown in Table 1.

On the other hand, Dick (2016) created an integrated series of courses spanning early undergraduate to graduate course levels that were designed to link lower-order and higher-order cognition levels of Bloom’s taxonomy (Lima et al., 2014), which grouped learning into six cognitive levels: 1) knowledge; 2) comprehension; 3) application; 4) analysis; 5) synthesis; and 6) evaluation. Dick (2016) created a first-year course, a senior course, and a graduate course in embedded systems that covered the embedded system knowledge area (CE-ESY), which was one of the 18 accrediting areas and covered 11 knowledge units. In Table 2, the embedded system’s knowledge units are displayed.

Soft skills are nontechnical abilities that enable us to maintain self-control and have positive interactions with others. To be more specific, soft talents can be characterized as follows: “Skills that go beyond academic success, such as positive values, leadership abilities, teamwork, communication skills, and lifelong learning”—Roselina appears in Lima et al., (2013).

Employees must be able to use suitable information, skills, and attitudes in every element of their lives in order to improve the competitiveness of both economy and competition in today’s knowledge-based economy (ARTIST network of excellence, 2005; John, 2016; Bertels et al., 2009; Haetzer et al., 2011). Furthermore, businesses and employers respond to economic needs by hiring workers who can communicate, think critically, and solve problems.

In summary, higher education must be redesigned to improve the employability of graduates (Catalog.oakland.edu, 2014). Employers indicated that fresh graduates lacked effective communication and critical thinking abilities, had insufficient knowledge in their industry, and had weak problem-solving skills, according to the National Graduate Employability Blueprint (Bloom, 1956; Seviora, 2005; Rover et al., 2008; Shakir, 2009; Alliance for Excellent Education, 2011).

Furthermore, the issue of unemployment and work preparedness is a global phenomenon that is primarily attributable to poor graduate quality (Sahlberg, 2006; Johanson, 2010; Saavedra and Opfer, 2012; Catalog.oakland.edu, 2014), rather than a lack of employment options. As a result, several researchers have proposed that higher education institutions reform engineering education in order to keep up with industry expectations and bridge the gap between academic perceptions and industry expectations or employability skills for entry-level engineers (NGEM and Ministry of Higher Education, 2012; Hanapi and Nordin, 2014; Seetha, 2014; Alias, 2016; Rahmat et al., 2017; Ting et al., 2017; Fernandez-Chung and Yin Ching, 2018; Karzunina et al., 2018). As a result, the Ministry of Higher Education has mandated that institutions of higher learning should include soft skills in their curricula.

Critical thinking entails combining skills such as assessing arguments, inferring inductive or deductive conclusions, judging or evaluating, and making decisions or solving issues (Howieson et al., 2012). Indeed, these critical abilities are referred to by a variety of terms, including cognitive thinking, reflective judgment, and epistemological development (Itani and Srour, 2016). For example, intelligence, scientific problem-solving, meta-cognition, drive to learn, and learning styles are all constructs and approaches that make up cognitive talents.

Despite the fact that each term has a different goal and application, cognitive abilities have been recognized as an important college student outcome because of their “applicability and value across a wide range of academic areas” (Itani and Srour, 2016). Many educators are interested in teaching critical thinking to their pupils (Howieson et al., 2012). It is widely agreed that critical thinking is an important and vital issue in modern education.

When employing engineering graduates, communication skills tend to be one of the most significant qualifications (Srour et al., 2013). Despite its importance, some studies have focused on employers’ growing concern about engineering graduates’ deteriorating communication skills (Yoder, 2011; Baytiyeh, 2012; AUB (American University of Beirut), 2013; Selinger et al., 2013; Crystal, 2016).

Teamwork has evolved as one of the most important skills for job readiness in the 21st century (Md Zabit et al., 2018), making it a necessary practice in higher education, with lecturers frequently giving group assignments that need student collaboration (Pascarella and Terenzini, 2005).

The development of small-scale business management abilities through entrepreneurship education has been linked to improved living conditions. Furthermore, entrepreneurial education teaches students to work not only for a living but also for themselves. In this context, it has been reported that fresh university graduates enter the work market on a yearly basis (King, Goodson, and Rohani).

The term “lifelong learning” comes from an article by Edger Faure for UNESCO titled “Learning to be” (Markes, 2006). From infancy through old age, one’s actions are reflected in lifelong learning. As a result, everyone should be taught how to learn. In addition, lifelong learning refers to a learning process that can be continued in a variety of settings without returning to school.

The cognitive domain focuses on how a student acquires and applies knowledge, which can range from simple material memory to more complicated and abstract activities. Furthermore, this domain refers to a hierarchy of learning skills that allows instructors to measure students’ learning in a systematic manner. Furthermore, the cognitive domain can be split into six key categories: 1) knowledge, 2) comprehension, 3) application, 4) analysis, 5) synthesis, and 6) evaluation (Meier et al., 2000; Scott and Yates, 2002a; Scott and Yates, 2002b), which range from simple to complicate and tangible to abstract.

Lorin and Krathwohl (Martin et al., 2005) identified various flaws in the cognitive domain. They asserted that, compared to the other levels, the knowledge level has a dual character. As a result, the Revised Bloom’s taxonomy cognitive domain was introduced, with some alterations to its six levels. The cognitive domain modernizes the taxonomy for use in the twenty-first century, emphasizing the active character of teaching and learning in the learning process of students rather than their behaviors. It also provides students with clear and well-defined objectives that allow them to connect their current learning to old knowledge (Barton, 2007; Ahles and Bosworth, 2004). Figure 1 illustrates the variances between the original cognitive domain and the 2001 revised one.

This topic is concerned with issues involving the emotional aspect of learning, and it encompasses everything from basic willingness to information intake as well as beliefs, ideas, and attitudes. It also takes into account student’s knowledge of ethics, global awareness, and leadership. As illustrated in Table 3, the affective domain is divided into five categories: 1) receiving, 2) responding, 3) valuing, 4) organizing, and 5) characterizing.

In the context of engineering, Timothy argued in Nguyen, (2017) that “the affective domain is involved with how one confronts interpersonal issues in their professional work or the realm of professional ethics, where a clear relationship is vivid with the value system to which one must follow.” This idea is similar to ABET criterion (f), which states that students must “understand professional and ethical duties” (Faure, 1972).

Learners’ attitudes of awareness, interest, attention, concern and responsibility, ability to listen and respond to interactions with others, and ability to demonstrate those attitudinal characteristics or values that are appropriate to the test situation and field of study are all examples of affective learning. Krathwohl’s affective domain taxonomy (Qamar et al., 2016), which gives a set of criteria for classifying educational results related to the complexity of thinking in the emotive domain, is possibly the most well-known of all of the affective taxonomies.

Again, the taxonomy is organized from the simplest to the most complicated feelings. Although it is related to the emotional, this taxonomy can be used in conjunction with Bloom’s cognitive taxonomy (Yildirim and Baur, 2016). All learning areas and levels can benefit from Krathwohl’s taxonomy. It gives a framework for instructors to create a series of activities that will help students develop their personal relationships and value systems.

Future educators can instill accountability in their deserving students and recognize that each learner is unique based on individual differences, allowing them to pinpoint his or her strengths and weaknesses in order to further improve the students. As with all taxonomies, there must be a very clear instructional intention for growth in this area specified in the learning objectives when labeling objectives use this domain (Syed Ahmad and Hussin, 2017).

This section explains how to apply the SLR used in this study (Lorin et al., 2000; Boyle et al., 2016; Hamari et al., 2016), including how to define research questions and how to define a search strategy.

The research questions emerged as a critical feature that must be well defined because they were used to gather data and determine the study’s final outcomes. The following research questions influenced the expected outcomes in this study:

RQ1. What are the key body of knowledge (BOK) courses for undergraduate students interested in learning the embedded system design?

RQ2. What cognitive, emotional, and psychomotor skills are necessary for success in the embedded system design course?

RQ3. How can the instrument’s validity and reliability be assessed?

RQ4. Are students prepared to learn about the embedded system design?

The authors determine the number of articles published in this specific region during the last two decades to answer RQ1 through RQ4. The SLR allows you to explicitly explain your search approach, which aids in the discovery of relevant and useful articles for your research. As a result, the outcomes of the search strategy are heavily influenced by keywords and databases. Articles on embedded readiness, as well as instrument development and validation, were searched for in this study.

ScienceDirect, Web of Science, Scopus, Emerald, and Google Scholar were used to search for similar article titles in this study. The relevance of these five databases to electrical and electronics engineering, as well as engineering education disciplines, was chosen to assure complete and inclusive data collection.

The Web of Science, established and maintained by Thomson Reuters, is the world’s largest online database, providing access to six databases (Krathwohl, 2002).

Scopus, on the other hand, is an Elsevier-owned online database that contains over 22,000 abstracts and articles from 5,000 publishers in the scientific, technological, medical, and social sciences fields (Ferri, 2015). Aside from that, Google Scholar is a free web search engine that indexes full texts or metadata of academic literature across a number of publishing formats and disciplines, with over 160 million pages indexed as of May 2014 (Chugh and Madhravani, 2016).

The study keywords were chosen based on the research questions (RQs). The chosen keywords were cross-checked against the study RQs to confirm that they matched the research objectives and expectations. As a result, the search keywords listed in Table 4 were used to find the relevant articles.

To identify the primary articles, the search phrase shown in Table 5 was used to implement a search on the four (4) selected web databases using Boolean operators. Meanwhile, the OR operator was utilized to link the synonyms, and the AND operator was employed to chain the important key phrases. During the search process, we also used stemming and wildcards searching tips from another digital library (Norasiken et al., 2016; Sönmez, 2017; Smt and Philip, 2017).

This section explains how the review was actually carried out. As a result, the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) was used to conduct the review, which followed the stated methodology shown in Figure 2. The procedure included identifying a research pilot search, selecting a study, and assessing the quality of the results.

Based on the specified search string, a pilot search was conducted to find as many results as possible relevant to embedded systems and preparedness as well as instrument development and validation. As a consequence, 390 articles were initially identified from the selected databases, including 151 articles from ScienceDirect, 121 articles from Web of Science, 40 articles from Scopus, 52 entries from Emerald, and 38 articles from Google Scholar, as shown in Table 6. Finally, after using the reference management tool, a total of 92 studies were found and eliminated.

While the primary articles for this study were chosen based on a two-level inclusion and exclusion criterion used on the 298 articles that were chosen, 56 studies were discarded for failing to meet the criteria. Following that, after reviewing the titles and abstracts of the publications, 86 were rejected. After reviewing the full texts, 114 articles were eliminated from the 156 articles that were chosen.

The remaining 42 articles were then divided into three categories based on the RQs, with 20 articles linked to embedded systems (ES), 12 entries linked to readiness measurements (RMs), and 10 articles associated to instrument development and validation (IDV). The SLR was used to incorporate the remaining articles as primary studies.

The quality assessment criterion was used in this study to examine the quality of the primary articles in order to eliminate biases and risks to validity in empirical investigations as well as serve as more accurate inclusion and exclusion criteria, as proposed by Kitchenham (2004). As a result, a checklist of questions for embedded system categories, readiness measurement, and instrument creation and validation was created, as illustrated in Tables 7–9, respectively. Meanwhile, the answer scale illustrated in Table 10 was utilized to analyze and validate the primary selected articles, as indicated by Maiani de Mello et al., (2014).

This section contains the report of the performed review, which displays the SLR’s final phase. The study’s findings are presented according to a predetermined review methodology. Using the data gathered, the primary studies were extracted throughout the review based on the established inclusion and exclusion criteria. In addition, descriptive statistical tools were used to present the results.

Around 42 main articles were chosen for SLR based on the established selection criteria. Nonetheless, none of the source publications contained all three search terms of study-embedded systems, instrument development, and readiness. Instrument development and readiness, on the other hand, were mentioned frequently in the articles chosen. In reality, as shown in Figure 3, ten of the research articles chosen were journal publications, and ten were conference articles. The most recent research studies [ES1], [ES2], [ES3], [ES4], [ES5], [ES6], and [ES7] were published in 2016, while the oldest one was published in 2005 [ES20]. The primary studies and their publication year are displayed in Figure 4, which indicated an increasing trend in the number of publications.

The relevant articles from the data extracted in the data extraction form had been used to answer each RQ, as shown in Table 11.

RQ1: What are the essential body of knowledge (BOK) courses for undergraduate students interested in learning the embedded system design? The courses included in the body of knowledge for the embedded systems design course in electrical and electronics engineering, computer engineering, and computer science degrees were the topic of this research question. Nevertheless, most of the reviewed articles did not mention or have misconception about a concept or a weak understanding of a concept regarding BOK courses.

RQ2: What cognitive, emotional, and psychomotor skills are necessary for success in the embedded system design course? The focus of this research topic was on both hard and soft abilities that students need to succeed in an embedded systems design course. In total, 3 of the 20 articles in the SLR dealt on the development of the embedded systems design courses curriculum. [ES15], [ES18], and [ES20] were the studies in question. Furthermore, [ES2], [ES3], [ES4], [ES5], [ES8], [ES10], [ES16], [ES17], and [ES19] were discovered in 17 research articles, with nine of them focusing on providing hands-on teaching materials for embedded system courses. The remaining eight articles, [ES1], [ES6], [ES7], [ES9], [ES11], [ES12], [ES13], and [ES14], measured the effect of implementing project-based learning (PBL) on the embedded systems design course, while [ES15] evaluated master’s and degree embedded systems programs at three universities, namely the University of Eindhoven’s Master in Embedded Systems program, the University of Freiburg’s Master of Science in Embedded Systems Engineering program, the University of Pennsylvania’s Master of Science in Engineering in Embedded Systems program, and the University of Freiburg’s embedded systems engineering degree program. These articles presented an embedded system curriculum and teaching module that could be included in any degree or master’s program in electrical and electronics engineering, as well as computer science studies, at Berlin’s Humboldt University.

Following that, [ES20] describes a 98-credit hour, 4-year undergraduate embedded systems engineering program at the University of Waterloo in Canada. It offered a curriculum that included solid mathematical foundations and in-depth understanding of the programming and engineering design, as well as hardware and software integration, which appears to be the primary application of embedded systems. Furthermore, the first 3 years of the curriculum were covered by the core, while the final year was an elective. A variety of laboratories and class projects were incorporated into the curriculum to give students a wide range of hands-on experience as well as the opportunity to apply fundamental design tools. Meanwhile, [ES18] is another study that presented the embedded systems curriculum.

The difficulties that software-oriented students encountered while studying hardware courses, specifically microprocessors and embedded controllers, at the Institute of Informatics, Faculty of Natural Sciences and Mathematics at the Saints Cyril and Methodius University of Skopje in the Republic of Macedonia, were presented in this study.

The study advocated changing the hardware course structure to mimic the software course structure that students are familiar with. It also proposed a teaching module on microprocessors and embedded controllers for the institution. The following articles were primarily descriptive and quantitative in nature: in fact, [ES1], [ES2], [ES3], [ES4], [ES5], [ES6], [ES7], [ES8], [ES9], [ES10], [ES11], [ES12], [ES13], [ES14], [ES16], [ES17], and [ES19] stressed the impact of PBL on the embedded systems design in comparison to traditional methods.

RQ3: How can the instrument’s validity and reliability be assessed? This study looked into instrument validation through a four-part psychometric assessment: 1) content validity, 2) pilot test, 3) reliability, and 4) construct validity.

The degree to which a designed instrument fully examines the construct of interest is referred to as content validity. As a result, the content validity index (CVI) was utilized to assess the instrument’s content validity in [IDV1].

A pilot test is an initial study designed to assess the produced instrument’s reliability, comprehensibility, relevance, and acceptability. According to the information gathered, none of the articles had conducted a pilot test study.

The reliability analysis was carried out as part of the primary investigations in [IDV1], [IDV2], [IDV3], [IDV4], [IDV5], [IDV6], [IDV7], [IDV8], [IDV9], and [IDV10] by using Cronbach’s alpha to analyze the internal consistency of the instruments.

Construct validity is the degree to which a designed instrument measures what it is supposed to assess.

Construct validity was assessed using dimensionality analysis, item analysis, and the p Wright map in [IDV6] and [IDV10]. Confirmatory factor analysis (CFA), principal component analysis (PCA), and item analysis were employed by [IDV2], [IDV8], and [IDV9], respectively. To perform construct validity, [IDV1] used item analysis and CFA, whereas [IDV3] used dimensionality analysis and differential item functioning (DIF).

In addition, [IDV4] used dimensionality and ANOVA analysis, whereas [IDV5] used dimensionality, CFA, and the Wright map. In [IDV7], item analysis, Wright maps, and Pearson’s correlations were also utilized.

RQ4: Are students prepared to take a course on the embedded system design? This topic centered on the description of a systematic method for creating an instrument that would be used to assess students’ readiness to take an embedded systems design course. According to the SLR, no studies found a link between preparedness and embedded systems design courses.

In reality, 12 of the 42 primary articles in SLR provided data on students’ readiness, with three (3) studies [RM1], [RM2], and [RM3] measuring the aspect of students’ readiness to learn science courses at the tertiary level. Meanwhile, [RM4] and [RM5] supplied information on students’ readiness for mathematics in higher education, and [RM6], [RM7], and [RM8] provided information on higher school students’ readiness for college.

Students who gained in-depth prior knowledge from earlier courses relating to the areas they sought to learn had higher grades in their target course, according to these research studies. [RM9], [RM10], [RM11], and [RM12] also included information on students’ readiness to accept mobile learning.