A knowledge management approach to promote an energy culture in higher education

ABSTRACT

This document presents a Knowledge Management (KM) model aimed at promoting a strong energy culture regarding Saving and Efficient Use of Energy (SEUE) in higher education students. A survey was carried out and the analysis revealed a significant association between the study subjects’ prior tacit knowledge about SEUE and their KM situation and attitudes about it. We have also used factor analysis as a statistic technique to extract latent variables from the ones already existing. The findings suggest the design and use of didactic materials and educational strategies that promote training and development practices in the context of an effective and transparent knowledge transfer process, as well as individual practices in the change of energy use behaviour. The educational strategies must prevent previously inherited beliefs from distorting the new necessary knowledge, i.e., the tacit knowledge higher education students must acquire, assimilate and disseminate to strengthen a culture of SEUE.

KEYWORDS:

• Knowledge management
• energy saving
• higher education
• factor analysis
• educational strategies

1. Introduction

Sometimes, even in small groups with similar interests, there is a wide diversity of opinions, judgements or perceptions (tacit knowledge) (Nonaka & Takeuchi, 1995) about energy sources and their effects on the environment. When energy-related problems are discussed and analysed in forums or different media, they are usually perceived as unrelated to everyday life. The population often gets confused due to the abundance of information about energy reaching them through the mass media, since the messages transmitted are not always reliable or valid, sometimes being outright myths (Brown, Gumerman, Sun, Sercy, & Kim, 2012; Georgescu-Roegen, 1975; Maccracken, 2009). For this reason, it is necessary to generate a critical information and knowledge base to establish consensus and sound judgements about such problems.

The scientific knowledge generated and disseminated by higher education institutions is stripped of as much subjectivity as possible (Trivella & Dimitrios, 2015). The education sector, in its role of providing the population with valid explicit knowledge, must be attentive to this situation and create opportunities for dissemination and promotion of key issues such as energy saving and efficiency, both for comfort and for the survival of human beings (Annan-Diab & Molinari, 2017; Bizerril, Rosa, Carvalho, & Pedrosa, 2018; as cited in Jones, Trier, & Richards, 2008; Ntona, Arabatzis, & Kyriakopoulos, 2015).

Teachers, researchers and experts must promote and build reference frameworks in students that allow them to construct their tacit knowledge by contrasting facts and information in search of consensus and common goals, with all of them participating together in energy-related projects and programmes. This scenario must apply in the same way to the promotion of a strong culture of Saving and Efficient Use of Energy (SEUE). SEUE consists in optimising the consumption of energy resources but keeping the expected results unaltered (Pérez-Tello, Campbell-Ramírez, Suástegui-Macías, & Reinhardt, 2018).

When using Knowledge Management (KM) models, it is helpful to analyse the tacit knowledge of the subjects involved in such models, in order to establish KM strategies (Shabahat Husain, 2015). Such strategies can make use of guiding questions designed to turn knowledge into advantages that benefit the organisation. One of those questions has to do with the so-called KM Situation, i.e., the factors that promote the creation and use of knowledge as well as the barriers that prevent it. One of the methods to collect this information is the application of questionnaires to the study subjects, thus obtaining knowledge in a structured manner. The strategic knowledge (know-what) of the organisation not only allows to convert existing knowledge but also helps to identify the knowledge that is required to offer new strategic options (North & Kumta, 2014, pp. 155–162).

The attitudes of subjects towards the environment are formed by their beliefs, the influence of certain circumstances in their life and their intentions. Analysing and understanding these attitudes allows us to build a framework based on their tacit knowledge and thus promote a rational use of energy. Educational strategies that take students’ attitudes and cognitive level into account contribute to raise awareness of the influence their actions have regarding SEUE (as cited in Ntona et al., 2015). The necessary update to the second layer of the three-layer KM model presented in this paper demands that we know how higher education students perceive their own KM situation and attitudes about SEUE as well as those of the community that they make up.

The present research was carried out in the city of Mexicali, capital of Baja California, Mexico, described as an area with a warm-dry climate and low precipitation (Government of Baja California, 2018). Baja California has no connection to the national electricity grid; this implies that it must generate its own electricity to satisfy the demand of a population in constant growth, whose economic activity depends on a considerable number of multinational companies that export their products to other countries (Quintero-Núñez, Muñoz-Meléndez, Campbell-Ramirez, & Díaz-Gonzalez, 2013). On top of this, Mexicali has a particularly extreme climate that requires the use of air conditioning to maintain the comfort of its inhabitants, but the electricity consumed this way depends on factors such as building materials, internal heat loads, thermal insulation and energy management by the user, among others (Pérez-Tello et al., 2018). Since 1990, Mexicali has been a precursor in the design and application of programmes and actions aimed at saving energy and using it efficiently, such as thermal insulation of homes and replacement of low efficiency air conditioners and refrigerators for their high efficiency counterparts (Integral Programme for Systematic Saving, 2016; Suástegui Macías, Pérez Tello, Campbell Ramírez, & Magaña Almaguer, 2013). Regional development and its value creation in the community are driven through the implementation of these actions and by giving due relevance to knowledge as an innovation dynamics (Schiuma & Lerro, 2010). The actors with a strategic role that try to identify and illustrate the precise knowledge assets of their region are solidly oriented towards the achievement of regional objectives (Lönnqvist, Käpylä, Salonius, & Yigitcanlar, 2014). Baja California’s Energy Profile for the years 2011–2025 states that the criterion of energy saving through the efficient use of energy arises in response to the continuous reduction of primary resources and environmental considerations. One of the conclusions of this profile’s prospect is to implement energy saving and efficient use actions in the education sub-sector (Quintero-Núñez et al., 2013). Saving and efficient use of energy, in addition to being crucial in regions of extreme climates, concerns the whole world due to its direct impact on the future energy requirements of modern cities (Pérez-Tello et al., 2018).

The purpose of this work is to implement the updating of the second layer in a three-layer KM model that promotes a strong energy culture in higher education students. An extensive survey application was conducted in order to probe the KM situation and attitudes regarding SEUE. Mexicali has an economic activity based on industry and construction; for this reason the selected subjects are students with an engineering or architecture professional profile (Industrial Development Commission of Mexicali, 2018; International Energy Agency [IEA] & World Bank, 2015).

Following, the reference framework section introduces a general description of KM – main axis of this work – and goes on to establish the existing interrelations between education, KM models and energy saving, concluding with a presentation of the adaptation of a SEUE-specific KM model. Next, the methodology we used is introduced, describing the main characteristics of the study subjects along with the applied questionnaire and the statistical techniques we selected for the survey analysis. Then the research findings are discussed, based on general aspects of descriptive statistics, a hypothesis testing and factor analysis. After this, we present the discussion and some recommendations for future work. Finally, the conclusions and main findings, as well as the scope and limitations, are pointed out.

2. Theoretical framework

2.1. Knowledge management

Knowledge is inherent to the human being and, although it is closely linked with information, the latter needs to acquire a significant value in a given context in order to transform into the former. The interpretations, reflections and experiences in which information underlies are also needed for the construction of knowledge (as cited in Shabahat Husain, 2015; Trivella & Dimitrios, 2015). Based on the definitions given by Hernández Rivera (2014) & Rodríguez Elías (2007), we define KM as the discipline that, through a systematic approach, interrelates technology, processes and people by planning, coordinating and controlling efficient knowledge flows that affect the maturity of an organisation.

KM allows making the distinction between knowledge value and information sources, and it can adapt to any other discipline in its particular context. The multidisciplinarity that KM grants enables the joint collaboration of the team with other institutions sharing similar objectives (as cited in Drodge, 2003; Shabahat Husain, 2015; Trivella & Dimitrios, 2015). Hence, the merging of education and engineering expertise is necessary for developing strategies aimed at knowledge acquisition and dissemination.

We have searched in the existing literature for KM models that promote SEUE directed to the education sector but we have not found any. That is why we decided to continue our search based on two types of interrelations that are detailed in the following subsections; one refers to the relationship between the education sector and energy saving, and the other one considers KM Models and energy saving. After addressing these issues in particular, we introduced one of the main contributions of this research, which is a KM model that promotes SEUE in higher education students.

2.2. Education sector and energy saving

In the review of literature, we have found papers whose results prove the significant association of interest and responsibility concerning energy saving with the behaviour and attitudes about it in students of secondary and higher education. These studies also highlight the benefit of promoting such attitudes towards energy saving so that students have an accurate knowledge of the subject, since this directly impacts the individual responsibility on SEUE (Chalfoun, 2014; Choeisuwan, 2015; Cotton, Miller, Winter, Bailey, & Sterling, 2016; Ntona et al., 2015). Chalfoun (2014) describes a method for greening university campus buildings based on energy audits in order to identify energy efficiency opportunities. Once these opportunities have been identified, this author proposes certain future work strategies, calculating the economic benefits, water saving and the reduction of CO₂ emissions that would be generated in the university if these strategies were applied. Choeisuwan (2015) studies the association between receiving information, the support of society and the responsibility for energy saving with energy saving behaviour (which includes the notions of knowledge, attitudes and action). This author has found high and moderate level associations after making a statistical analysis of the mentioned variables. Cotton et al. (2016) states that energy literacy is irregular in the USA and UK. Through an online survey in a higher education institution, he concludes that generalised misconceptions about energy can reduce the effectiveness of energy saving behaviours. This author claims that the results of his work pose challenges to understand the agency and effectiveness in energy relations. Ntona et al. (2015) investigates the complex views or attitudes about energy and its use in relation to the environment of high school students. The results point to the need for a radical change in terms of human behaviour patterns. We have not found in the reference literature an approach that encompasses the education sector and the SEUE from KM and that takes the previous knowledge of the students as a reference of association.

Problems such as environmental pollution, economic expense and survival risk of future generations due to inadequate consumption habits and excessive growth of population require projects that promote education in society for the benefit of its own preservation (McGrath & Powell, 2016; Quaglione, Cassetta, Crociata, & Sarra, 2017). This work arises as a follow-up to the same line of research that inspired actions in SEUE already implemented in Mexicali, to continue promoting a strong energy culture, not only in a general sense to the entire population, but specifically in the higher education sector (Aini, Chan, & Syuhaily, 2013; Belaïd & Garcia, 2016; Faham, Rezvanfar, Movahed Mohammadi, & Rajabi Nohooji, 2017; Integral Programme for Systematic Saving, 2016; Ouyang & Hokao, 2009; Suástegui Macías et al., 2013; Urban & Ščasný, 2012).

2.3. KM models and energy saving

Although there are multiple ways to model the knowledge of organisations, each of them allowing the structuring of ideas, plans and particular actions, there is no specific model that can be implemented in any type of organisation indistinctly (Bimba et al., 2016; Trivella & Dimitrios, 2015). KM is properly established when the modes of knowledge conversion rotate as a spiral in which knowledge increases as the cycle goes by and repeats itself (Nonaka & Takeuchi, 1995). Without merging the formulated strategies into the spiral knowledge process, the increase and quality of knowledge will be compromised (North & Kumta, 2014).

With regard to KM models that promote energy saving, a search of related works has been carried out. The models available in the literature are those that focus on measuring electricity consumption in universities and then applying specific energy saving measures (Chung & Rhee, 2014; Sipan, 2016; Zhou, Yan, Zhu, & Cai, 2013) or models that focus on changing the population’s energy saving behaviour (Cotton, Miller, Winter, Bailey, & Sterling, 2015; Quaglione et al., 2017). A specific KM model to promote SEUE where knowledge processes are fundamental parts in the execution of it (Zaim, Muhammed, & Tarim, 2018) has not been found in the literature.

2.4. Adaptation of KM model regarding SEUE

The convergence of multiple disciplines such as pedagogy, didactics and engineering in a single project whose main element is the valid knowledge about SEUE, as well as the lack of a pre-existing KM model, has resulted in the adaptation of a new one (Figure 1) based on the SECI Model by Nonaka and Takeuchi (1995) and in the layered model by Sandoval Yáñez (2013). It is a model for the establishment of a KM system that allows the continuity of potential future work on the subject.

Figure 1. Three-layer KM model to promote SEUE

Left side of Figure 1 shows the name and number of the knowledge layer corresponding to the graphic representations shown at the same height on the right side of the image. Starting with the layer 1 (bottom), called Information Sources, it is possible to identify and distinguish sources of tacit knowledge (teachers, researchers and experts) and sources of explicit knowledge (articles, books and data).

After the corresponding analysis, the information sources of layer 1 (Figure 1) are recorded, in a first process called acquire knowledge located within the layer 2 called Knowledge Management. For this process, several knowledge acquisition techniques are used such as related lists, structured and unstructured interviews, document analysis, classification of concepts, among others (Fouché, 2006; Gavrilova & Andreeva, 2012; Rao, 2005; Wiig, De Hoog, & Van Der Spek, 1997).

The next process in Figure 1 is combine knowledge, which means the generation of explicit knowledge suitable for the higher education level from the explicit knowledge existing at the expert level. Then comes the structuring knowledge process, which makes it possible to store knowledge in a knowledge base and thus to present that knowledge to users (higher education students), determining in that way layer 3 of the model, called Knowledge Presentation. Guided by the aforementioned model process, we have developed several didactic resources, some of which were presented in a web page (Reinhardt, 2019). The final process that is still part of layer 2 and that has not been executed so far is update, within which the main contribution of this paper is located.

3. Research methodology

The methodology we have chosen for this project was carried out following by and large the procedure shown by Ntona et al. in 2015, which consisted of study subject selection, design of a questionnaire based on previously existing ones, hypothesis testing and factor analysis. The KM situation is not analysed in the quoted research, but it is considered in this research. We chose the methodology of Ntona et al. (2015) because we needed to know aspects of the students’ tacit knowledge that we did not have, which we could extract not only through explicit variables but also from implicit variables (analysis factor, hypothesis testing and descriptive statistics). The use of a questionnaire facilitates this type of data collection and processing, since the samples are large enough for us to rule out personalised interviews.

We have combined the use of quantitative methods (objective in nature, whose observations in numbers verify the relationships between concepts through deduction) and qualitative methods (subjective in nature, aimed at descriptive and inductive models). The qualitative method was used to design the questionnaire and required to explore, analyse and investigate in various sources of information on what would the most appropriate questions be for the instrument with which we would later collect the data. The quantitative method was used for sample design, descriptive statistical analysis, hypothesis testing and factor analysis. In the next subsections we will explain each method used in detail. The complementarity of both methods is due to the fact that they invigorate each other in studies with demanding initial conditions (Abalde & Muñoz, 1992; Johnson, Onwuegbuzie, & Tunrer, 2007). We consider that the purpose of this research makes both kinds of methods necessary.

3.1. Selection of study subjects

Considering that the place where the research was conducted is a border city whose main economic activities are manufacturing and construction (Industrial Development Commission of Mexicali, 2018), undergraduate students with a professional profile related to engineering and architecture were selected as study subjects (see Table 1). The criterion of considering the main economic sectors to select the study subjects is due to these students having an academic profile whose probability of working within these sectors in the future is high (IEA & World Bank, 2015). The border region of Mexicali experiences rapid industrial growth and, without saving and efficient use of energy, air pollution can become very severe; in addition, the demand for energy in this sector is increasing (Quintero-Núñez et al., 2013). Regarding the construction sector, in order to make proper use of energy, many factors must be taken into account in the building: thermal insulation, internal loads, building orientation, materials used, among others (Pérez-Tello et al., 2018). Knowing in first instance the KM Situation and Attitudes can give us very relevant information to update the second layer of the KM Model for the promotion of SEUE, allowing teachers, experts and researchers to continue training, from that perspective, the people who will become agents of change in the city in the medium term.

Table 1. Professional profile of the surveyed subjects

Professional Profile	Number of students	Percentage (%)
Renewable Energy Engineering	50	17.1
Energy Engineering	51	17.4
Mechanical Engineering	70	23.9
Electric Engineering	23	7.8
Electromechanical Engineering	15	5.1
Industrial Design	11	3.8
Architecture	73	24.9
Total	293	100

Three higher education institutions, those with the most enrolments in Mexicali, were selected to carry out the field work. Having considered the selected profiles as a population, a probabilistic method (simple random sampling) was chosen and carried out; its definition and calculation method required that at least 278 surveys be carried out to attain a 95% confidence level and 5% margin of error (Wright & Tsao, 1985). A total of 293 surveys from seven different professional profiles were taken into account, of which 71.3% belong to engineering and 28.7% belong to architecture. The 293 surveys represent 7.38% of the total number of students who attend any of the three educational institutions selected and who belong to one of the professional profiles mentioned in Table 1 (State Educational System of Baja California, 2018).

The age of the selected study subjects is above 17 years old. This is considered adequate because at this age the students already have well-established logical-abstract thinking which belongs to the formal stage (Piaget & Inhelder, 1959), which is required to act with critical thinking in the role of agents of change with respect to the promotion of SEUE. Of the 293 surveys, 212 are men (72.4%) and 81 are women (27.6%). This distribution has been completely randomised, without the authors having chosen a particular criterion.

3.2. Methods

In Social Sciences, internal validity and external validity are regarded as the main conceptual tools for analysing the validity of a research study. Internal validity is related to the reliability of the causal inferences that arise from the results obtained in an investigation, while external validity refers to how generalisable those results are outside the experimental conditions that the investigation had (Jiménez-Buedo, 2011). This is why, although validity refers to inferences made from the final results, the coherence of the designed instruments and the reliability of the collected data are essential for such analysis to be possible.

To establish the information collection instrument, a wide literature review was carried out (D. Cotton et al., 2016; Gao, Wang, Li, & Li, 2017; Irma & Guaderrama, 2016; Leygue, Ferguson, & Spence, 2017; North & Kumta, 2014; Ntona et al., 2015; Quaglione et al., 2017; Thøgersen, 2017) and, after consulting academic peers with expertise in SEUE, KM and Higher Education, the decision was made to select questions from two different questionnaires, one referring to the KM situation (North & Kumta, 2014, p. 176) and another referred to attitudes about SEUE (Ntona et al., 2015, p. 10). We considered that both sections should have the same design and the same scale to be able to make an integral assessment of the aforementioned aspects. Some modifications were also made to the wording of the questions for greater clarity. Regarding the KM Situation – whose authors North and Kumta (2014) define as the factors that promote the creation and use of knowledge, as well as the barriers that prevent it – we selected for our adapted questionnaire those questions that could be applied to educational communities and discarded the questions containing concepts specific to the business sector, such as benchmarking. In the second questionnaire (Ntona et al., 2015), only the questions related to attitudes about SEUE applicable to higher education students were selected, since we had already collected the other types of questions–related to habits and sources of information–when we established the KM Model regarding SEUE in 2015 (Reinhardt MS. 2017). It was fundamental in the design of a single instrument to consider that the response time of the students is not excessive. The scale used was of the Likert type from 1 to 5, corresponding to 1 the lowest score and 5 to the highest score. The number 5 in each question is an indicator that there is greater sensitivity to knowledge about SEUE.

The Cronbach’s Alpha was computed to estimate the reliability of the instrument. The Cronbach’s Alpha coefficient we have obtained for the survey applied is 0.704, which is considered an acceptable value (Tavakol & Dennick, 2011). This result allows this tool to be replicated also in other geographical areas, adjusting the interpretation to the particular context in which the research takes place. When replicating the study in other geographical areas, it is also recommended to make a pilot test and calculate the Cronbach’s Alpha to check if the instrument needs adjustments for that particular environment.

After requesting some data such as gender, field of study and age, all respondents were asked to tell if they had prior knowledge about SEUE. To ensure that we obtain accurate information about prior knowledge, we explained to students that those who had taken a subject on SEUE or related topics (e.g., Energy audit, Thermal and energy adaptation, among others) should mark “yes” on this option, while those who had not taken one should mark “no,” unless they had completed a technical course in SEUE outside the university. Following that, the two sections mentioned above are located (see Tables 2–3). Note that at the top of each section the student is told the topic and the role from which he or she must interpret the statements. The adapted tool has a total of 15 items.

Table 2. Adapted questionnaire: Students’ KM situation

Knowledge Management Situation Answer the following questionnaire from the Saving and Efficient Use of Energy perspective, and place yourself as part of the higher education student community of Mexicali, Baja California.
Knowledge and learning
Information equals knowledge.	1	2	3	4	5	Knowledge is generated from information.
We slowly learn from other communities in a situation similar to ours.	1	2	3	4	5	We quickly learn from other communities in a situation similar to ours.
Knowledge is transferred inefficiently in our community.	1	2	3	4	5	There is an effective knowledge transference inside our community.
The communication in our community is inefficient.	1	2	3	4	5	The communication in our community is efficient.
There is no transparency in knowledge.	1	2	3	4	5	The transparency in the knowledge is established through knowledge maps, etc.
The training and development are directed to individual learning processes.	1	2	3	4	5	There are common training and development practices, teamwork and knowledge transference.
There are inefficient knowledge transference groups.	1	2	3	4	5	There are groups that gather and transfer knowledge.
We don’t have any collaborative project (altogether).	1	2	3	4	5	The collaborative projects (altogether) that we do promote teamwork.
In our community, the work area and social environments are delineated.	1	2	3	4	5	In our community, the work area and social environments promote teamwork.

Table 3. Adapted questionnaire: Students’ attitudes

Attitudes Answer the following questionnaire from the Saving and Efficient Use of Energy perspective, and place yourself as part of the higher education student community of Mexicali, Baja California.
Attitudes
The harmonious coexistence of human beings and the environment is not a precondition for our survival.	1	2	3	4	5	The harmonious coexistence of human beings with the environment is a precondition for our survival.
There are enough energy reserves and there’s no reason for concern over its depletion in the future.	1	2	3	4	5	There are reasonable grounds for concern over the depletion of energy reserves.
The efforts to save energy do not require immediate concern.	1	2	3	4	5	Our main concern should be to encourage the efforts to save energy.
Saving energy is a matter of concern only for the government.	1	2	3	4	5	We have an important role in saving energy in our house and school.
Everything we do has no effect. We can’t generally bring change because other persons take the decisions.	1	2	3	4	5	We try to find other partners with the same interest in saving energy.
We apply practices to save energy when we are in company of other person(s).	1	2	3	4	5	We apply practices to save energy even if we are alone.

The survey was administered face-to-face. The students answered the survey in less than 15 minutes. Once the field work was completed, the database was prepared and the data was loaded and analysed in the statistics software SPSS for Windows, version 19 (IBM Corp, 2010).

3.3. Information analysis procedure

As a first step, a descriptive analysis of the surveyed groups, i.e., degrees, gender and age, was made. Next, a working hypothesis to be proven through χ² test with a significance level of 2% was proposed as follows:

H₀: Previous knowledge about SEUE in higher education students lacks a significant association with their KM situation and attitudes about it.

H₁: Previous knowledge about SEUE in higher education students has a significant association with their KM situation and attitudes about it.

Furthermore, a factor analysis of the 15 item questions was carried out in order to explore the possibility of discovering new pieces of knowledge that had been implicit within the explicit variables. Factor analysis, a statistical technique used extensively in Social Sciences to explore the psychometric properties of a scale or an instrument, does not consider variable dependency as an a priori condition (Finch, 2013; Osborne, 2015).

The score obtained from all the study subjects was added up, after which the total score of the 15 test items was recoded. This recoding was done with the “Transform Variable → Recode into Different Variable” tool of SPSS. Three levels were established for the obtained score, using as a range the difference between the highest score possible (75 points) and the lowest score possible (15 points). Such difference was divided by three, giving place to the low, medium and high levels. The subjects’ age was also recoded into ranks in order to make a clear presentation in the next section.

4. Research findings

The results obtained in the surveys to update a KM model for the promotion of SEUE are analysed below.

4.1. Descriptive statistics

In this section a synthesis about descriptive statistics is presented through numerical descriptions in the form of tables and text. As mentioned in section 3 of the research methodology, the sample consisted of 293 undergraduate students who answered the survey during the second semester of 2018. Table 4 shows an age profile of students under analysis.

Table 4. Age ranges of the study subjects

Age Ranges	Number of students	Percentage (%)
17 to 20 years old	91	31.1
21 to 24 years old	175	59.7
25 years old or more	27	9.2
Total	293	100

4.1.1. KM situation

Table 5 shows the statements which in Table 2 are on the right side with the frequency of scoring obtained according to the total number of respondents. It is recalled again that the survey instructions asked students to respond from the perspective of SEUE, as a member of a higher education community.

Table 5. Students’ KM situation

	Frequency of obtained score in ascending order
	1		2		3		4		5
	N	%	N	%	N	%	N	%	N	%
I1. Knowledge is generated from information.	2	0.7	5	1.7	22	7.5	47	16	217	74.1
I2. We quickly learn from other communities in a situation similar to ours.	30	10.2	34	11.6	65	22.2	83	28.3	81	27.7
I3. There is an effective knowledge transference inside our community.	35	11.9	52	17.8	82	28	83	28.3	41	14
I4. The communication in our community is efficient.	24	8.2	58	19.8	77	26.3	90	30.7	44	15
I5. The transparency in the knowledge is established through knowledge maps, etc.	10	3.4	24	8.2	88	30.1	95	32.4	76	25.9
I6. There are common training and development practices, teamwork and knowledge transference.	17	5.8	28	9.5	68	23.2	96	32.8	84	28.7
I7. There are groups that gather and transfer knowledge.	17	5.8	26	8.9	78	26.6	98	33.4	74	25.3
I.8 The collaborative projects (altogether) that we do promote teamwork.	14	4.8	19	6.5	57	19.4	94	32.1	109	37.2
I.9 In our community, the work area and social environments promote teamwork.	17	5.8	45	15.4	78	26.6	92	31.4	61	20.8

It is observed that most students (74.1%) have a clear position that information and knowledge about SEUE are not synonymous. The situation with the following three items (I2-I4) and the last item (I.9) is different since more than 20% have decided to give them the lowest scores. That is to say, 20% of students consider that their community learns slowly from other communities, that the knowledge transfer is not effective, that communication is not efficient within their community and that social and work spaces are delimited in a way that does not promote teamwork.

It is quite clear that, from the perspective of the students, the projects they develop together promote teamwork (I.8) because the scores 4 and 5 reached almost 70%. On the items referred to knowledge transfer groups (I.7), its transparency (I.5) and training and development practices (I.6) between 30% and 40% have given it a score of 3 or less, a situation that merits attention.

4.1.2. Attitudes

Table 6 shows the statements from the right side of Table 3 with the frequency of scores obtained according to the total number of respondents.

Table 6. Students’ attitudes

	Frequency of obtained score in ascending order
	1		2		3		4		5
	N	%	N	%	N	%	N	%	N	%
I10. The harmonious coexistence of human beings with the environment is a precondition for our survival.	7	2.4	11	3.7	19	6.5	43	14.7	213	72.7
I11. There are reasonable grounds for concern over the depletion of energy reserves.	6	2.1	8	2.7	23	7.9	42	14.3	214	73
I12. Our main concern should be to encourage the efforts to save energy.	4	1.4	7	2.4	7	2.4	50	17	225	76.8
I13. We have an important role in saving energy in our house and school.	2	0.7	2	0.7	27	9.2	32	10.9	230	78.5
I14. We try to find other partners with the same interest in saving energy.	15	5.1	12	4.1	59	20.2	90	30.7	117	39.9
I15. We apply practices to save energy even if we are alone.	9	3	19	6.5	62	21.2	62	21.2	141	48.1

Table 6 shows that more than 80% of the students have given the highest scores to items I10, I11, I12 and I13, recognising that human beings must live harmoniously with the environment and that there are sufficient reasons for concern about the depletion of energy reserves. The majority of them also consider that the main concern of their community should be to encourage efforts to save energy and they assume an important role in energy saving, both in their home and in the educational institution they attend.

Two aspects that have not been as positive as those mentioned are those reflected in the lowest scores (1, 2 and 3) of items 14 and 15. Around 30% tend to think that what they do as a measure of SEUE has no effect because the decisions that bring change depend on other people, and they also apply energy saving practices when they are in someone’s company but not necessarily when they are alone.

4.2. Hypothesis testing

Hypothesis testing was conducted between the variable corresponding to prior knowledge and the recoded variable of the total score of the 15 test items that comprehensively reflect both the KM situation and the students’ attitudes concerning SEUE. This recoded variable was called KM Situation and Attitudes. Initially, three levels of variable recoding were proposed: low, medium and high, but when computing relative frequencies, it was noted that the students are between the middle and high level, that is, none of them had a low level.

Table 7 shows the relationship between both variables. While students with previous knowledge about SEUE have obtained higher relative frequencies than expected at the high level, students without prior knowledge have obtained higher relative frequencies than expected at the middle level. Out of the total number of students, 132 (45.1%) possess knowledge about SEUE (N Real = Yes), while 161 (54.9%) have not received education about it in courses or workshops (N Real = No).

Table 7. Prior knowledge vs. KM situation and attitudes

			KM Situation and Attitudes
			Middle Level		High Level
		Frequency	N	Row %	N	Row %
Prior Knowledge	Yes	Real	24	18.2	108	81.8
		Expected	33.3	25.2	98.7	74.8
	No	Real	50	31.1	111	68.9
		Expected	40.7	25.3	120.3	74.7

Hypothesis testing yielded a χ² of 6.368 and a bilateral asymptotic significance of 0.012. We can thus reject H₀ and accept H₁ since, judging from the results of the hypothesis testing, both variables are significantly associated with a level of 2%. The same significant association is observed in some subgroups of students if we apply a cross-analysis regarding age, gender and academic profile (engineering vs. architecture). Significant differences were detected in the age range of 21 to 24 years (χ² of 4.283 and a bilateral asymptotic significance of 0.038); gender (χ² of 2.879 and a bilateral asymptotic significance of 0.090 for males, χ² of 4.046 and a bilateral asymptotic significance of 0.044 for women) and the engineering academic profile (χ² of 3.086 and a bilateral asymptotic significance of 0.079).

4.3. Factor analysis

Attempting to find variable groups with a common meaning among the 15 variables that make up the questionnaire, factor analysis was carried out using the Principal Component Analysis (PCA) as a method. To perform factor analysis under the basic assumptions, we proceeded to calculate the Kaiser-Meyer-Olkin index (KMO), which was found to be 0.744, with a Bartlett’s Test of Sphericity significance below 0.05 (0.000). These values indicate that there are significant correlations between the variables. The components that were extracted are those whose eigenvalues were above one. To detect factor loads, we used the Varimax rotation method, which is frequently used in this type of analysis (Birasnav, Albufalasa, & Bader, 2013; Kaiser, 1958; Takano et al., 2010).

Figure 2 shows the scree plot that allows one to visualise the choice of four components that explain 48.8% of the variance. The remaining components are discarded because they are already below the elbow of the curve. A reference line was added in order to appreciate the representativeness of the first four components.

Figure 2. Scree plot

On Table 8, the four extracted components are shown with their factorial loads.

Table 8. Component matrix^a

	Component
	1	2	3	4
I4	.744	.141	.068	−.216
I3	.692	.053	.136	−.224
I5	.581	.009	.131	.231
I2	.538	−.014	.097	.148
I13	.034	.715	−.020	.259
I14	.073	.642	.147	−.142
I12	−.095	.622	−.026	.305
I15	.154	.494	.060	.026
I11	−.129	.460	.241	.363
I7	.051	−.001	.800	.164
I8	.241	.190	.687	−.038
I6	.379	−.056	.465	.406
I9	.260	.229	.450	−.355
I1	.134	.176	−.074	.697
I10	−.083	.215	.174	.598

Extraction Method: principal component analysis.

Rotation Method: Varimax Kaiser Normalisation

^a4 components extracted.

A graph of components in rotated space showing the grouping of the different items is also shown in Figure 3. The items that make up Component 1 (I4, I3, I5, I2) are related to the intra and inter-group knowledge flow of the students. Component 2 (I13, I14, I12, I15, I11) confirms that said items refer to the attitudes of the students concerning saving and efficient use of energy. Component 3, composed of items I7, I8, I6 and I9, is linked to the group cohesion perceived by the students. Regarding Component 4, it is considered that two variables are too few to conceptually abstract an implicit variable, therefore no name was given to it. We believe it has to do with essential underlying concepts that should be known before the subject is aware of what SEUE implies, such as the fact that information and knowledge are not synonymous (I1) and that harmonious coexistence of humans with the environment is a requirement for survival (I10).

Figure 3. Component plot in rotated space

5. Discussions and future work

5.1. Three-layer KM model about SEUE

The situation presented in the introduction deals with the need to implement an updating on Knowledge Management (layer 2) of a three-layer KM model to promote SEUE. This allows the layer 2 cycle to start again, having valid information about the perspective of subjects for whom the model was adapted. In a new beginning of the layer 2 cycle, the didactic material and educational strategies to be produced must take advantage of and continue to promote the position of the students reflected in the surveys, which refers to their civil responsibility to save energy and use it efficiently. Increasing the sample size and including other professional profiles could be very useful to further perfect the model.

The database that resulted from the survey field work presented in this research complements Information Sources (layer 1), since we now have a new explicit source of information about the KM situation and attitudes towards SEUE complementing the sources that were already established in the original model, such as role definition templates, books and data from surveys of higher education students about their technical knowledge. The knowledge base of Knowledge Presentation (layer 3), which already has a knowledge base in an online information system about SEUE, will have to be complemented when the strategies proposed in the next section as future work are carried out in the higher education sector.

5.2. Educational strategies

There are weak points in the study subjects that were exposed in the analysis of the surveys, which are important to reinforce when designing new materials and strategies concerning SEUE. Teachers, researchers and experts in technical-scientific knowledge must promote more individual energy saving practices, encouraging students to apply them even when they are not in the company of other people, in addition to strengthening the notion that individual actions do affect the energy problems society faces today.

5.2.1. General educational strategies

In order to strengthen an energy culture, the saving and efficient use of energy must begin to be addressed as a subject of study not only in those areas of knowledge that are evidently related, like Physics or Ecology, but also in Geography (relationship between size of energy systems with demand patterns) or the Sociology of Technology (socioeconomic or political drivers of energy use), among others. This demands for modifications in the curricula which must take into account, in addition to what has been mentioned, the influence of the different cultures that can enrich said curricula with different points of view (Sovacool, 2014).

It is important to establish a historical trend in the consumption patterns of higher education students, promote in them ecological initiatives of transport and distance learning, as well as forums to share information where discussion takes place about positive and negative implications in the use of energy, climate change, energy efficiency of renewable and non-renewable energies, among other relevant issues. It is also important to encourage students to form student campaigns that focus on real energy reduction plans (Altan, 2010).

5.2.2. Particular educational strategies

Particularly in the city of Mexicali, the formation of a student team called Energy Squad is proposed. Such a group may arise from meetings and inter-university agreements. Once this squad has been established, contact with the media and exchanges and systematisation of experiences among the members of the squad would be facilitated. The squad would be encouraged to reflect on the geographical location in which the students are living and their high dependence on resources such as electricity and water. In this context, transdisciplinarity could take place in the influence that students may have in the community they live in and vice versa (as cited in Montedonico, Herrera-Neira, Marconi, Urquiza, & Palma-Behnke, 2018). The integration of this student squad with teachers, researchers and experts working as guides is fundamental for the soundness of a strong energy culture.

On the other hand, it is advisable to organise meetings between students and the developers of the recent Virtual Energy Management System (Bonilla, Samaniego, Ramos, & Campbell, 2018) so that the former can have a clearer idea of the measurement and monitoring of electrical energy in a university building. Just as it is possible to monitor electrical parameters, carbon footprint and electricity cost, it is also advisable to monitor the three-layer KM Model presented in this research to contribute to the maturity of the higher education institutions that use it (Moller, Berkes, Lyver, & Kislalioglu, 2004). The monitoring proposed as future work would not only be about the situation of KM and attitudes about SEUE but also about longitudinal studies of specific notions about the use of energy, energy efficiency, thermal energy and air conditioning in higher education students.

5.3. Strategies from other perspectives and sources of knowledge

This work favours the knowledge of certain features that provide framework and reference of a particular section of the population who, in the not-so-distant future, will be part of the society that forges the region’s socio-technical and economic reality. As stated in Glück (2018), the strengthening of an energy culture requires the contribution of multiple and heterogeneous actors who are sometimes silenced or made invisible by diverse circumstances. A future strategy with a situational analysis approach that looks at higher education students from multiple perspectives and sources of information would allow the in-depth understanding of the complexity of the processes where an energy culture arises, without the reductionism of not contemplating certain social actors or non-human elements such as new technologies and infrastructure.

The formal and informal interactions with the government, policies with well-defined benchmarks and evaluation frameworks, programmes, political or scientific conferences, speeches, official documents and standards, material resources, digital services, as well as the mapping of relationship, roles and responsibilities are sources of information that should simultaneously allow the necessary transformations that lead to new understandings in the context where cultures emerge, change and persist. It is the object of these transformations to seek out more elements, processes and approaches, to establish best practices that promote learning and collaborative relationships by facilitating the means to visualise aspects that have remained invisible, fragmented or ambiguous, and give continuity to all of them (Altan, 2010; Glück, 2018; Quaglione et al., 2017; Trivella & Dimitrios, 2015).

The aforementioned sources of information and strategies from other perspectives that go beyond the strictly educational, in addition to influencing the KM model promoting SEUE in higher education students, must be known, used and implemented within said model based on the collaboration of multiple actors united in favour of the welfare of future generations. There must be a mutual correspondence between the model and the energy culture beyond the fact that, in terms of scope, the latter encompasses the former. There is still much to do in strengthening an energy culture, but the programmes applied, their evaluation, technical reports and this KM approach certainly contribute to a greater strength in the saving and efficient use of energy in Mexicali (Magaña-Almaguer, Pérez-Tello, & López-Badilla, 2016; Pérez-Tello et al., 2018; Quintero-Núñez et al., 2013; Suástegui Macías et al., 2018).

5.4. Reflecting on the results

5.4.1. Comparison with a related work

The attitudes probed in this research were already studied by Ntona et al. in 2015. The differences were that our investigation also took into account the KM situation of the students and that the study subjects were higher education students, and not high school students, as in aforementioned research. Neither have we evaluated the socio-demographic situation of the study subjects as Ntona et al. (2015) did, since it is not part of our current objectives. We have confirmed that the variables that form the component of attitudes proposed by the research of Ntona et al. (2015) are replicated in a similar way in our own analysis, even with better results in terms of score, a situation that could be due to the age difference of the respondents between both studies.

5.4.2. Contributions

In the factor analysis, it was noted that the KM situation has two implicit components, one is intra- and inter- community knowledge flow and another is group cohesion. Although group cohesion is a sine qua non condition of the effective knowledge flows of a community, having them within separate components is very useful for analysis. Knowledge flows bring dynamism to KM models (Flores-Rios, Pino, Ibarra-Esquer, González-Navarro, & Rodríguez-Elías, 2015). Having detected a component that contains implicit knowledge flows in students (mainly referred to transference and transparency of knowledge) greatly benefits our model, which has already taken into account the knowledge flows between the teachers, researchers and experts that compose it.

Furthermore, a component of two variables emerged (I1 and I10). We have not made any generalisation about it because it is an insufficient number of variables, although we note that it could depend on the foundations (basic notions) required to have a good level of knowledge about SEUE. It is recommended to investigate this in future work, adding items to the fourth extracted component that could be related to the two that are already part of it.

6. Conclusions

Globally, the world is facing problems related to energy, mainly those that arise from the depletion of non-renewable resources and the emission of polluting gases that cause climate change. Mexicali is not the exception, both for its main economic activity – the industry – and for its extreme weather, which requires an intensive use of air conditioning during the summer. This situation demands constant education for its population, especially young generations who already have critical thinking and who in turn can act as agents of change in the community they inhabit, thus collaborating with the work that experts in the field are already making and achieving a greater scope in the dissemination of valid knowledge.

This research has established the updating of layer 2 in a three-layer KM model to promote an energy culture in the higher education sector by collecting and analysing valid information on the perception and attitudes of the study subjects. Their perspective provides the basis for a new beginning of the layer 2 cycle that will continue to promote their active role in disseminating the topic of study and making decisions that benefit the community. The questionnaire used to collect data and information showed acceptable reliability, so it can be replicated and extended to other contexts taking into account their particular features. Additionally, we propose general and particular educational strategies that could favour the knowledge spiral of layer 2 and complement the knowledge base of layer 3 in the three-layer KM model. In its previous state, this model considered the technical dimension of higher education students (knowledge about energy efficiency of combined-cycle power plants, solar panels and appliances, among others), and in its current mode, the model also begins to contemplate the students’ cognitive dimension, having probed into their KM Situation and attitudes about SEUE (Choo, 2006).

This research has also shown that the study subjects’ prior knowledge about SEUE is associated with a high level of KM situation and attitudes in this regard, while the lack of prior knowledge about SEUE is associated to a medium level in the KM situation and attitudes in this regard (see Table 7). An in-depth analysis revealed that this same trend is repeated in the age range of 21 to 24, in the engineering profile and, although each gender remains significant in its particular set, greater significance was observed in the female gender. Another important finding has been the discovery of two components within the KM situation section in the applied survey that were implicit. These two components can now be studied in depth both at their own individual level and in the relationship between them.

6.1. Scope and limitations

The sample of study subjects is representative for students of higher education in the city of Mexicali with the profiles mentioned in Table 1. To replicate the factor analysis and hypothesis tests, as well as the basic descriptive analysis, the necessary internal and external validity tests should be carried out. A pilot test is recommended before proceeding with the full study.

There are some concepts in the questionnaire applied in this research that are complex in nature, such as information, knowledge and transparency, among others, and they may have affected the intelligibility of certain items by students. When replicating this study, it is recommended to provide respondents with shared definitions before they answer the questionnaire.

This has been a cross-sectional study that has given us information of which there was no history, but it is also recommended to do longitudinal studies. If this last type of study is carried out, it should be taken into account that the association of two phenomena may be conditioned by the temporal pattern of cultural consumption in a given context (Quaglione et al., 2017). Expanding the sample size to other academic profiles could give more representative results of the higher education sector.

Developing an energy culture in societies demands the consideration of very deep transformation processes, which in turn encompass the modes of reflection, diversity and complexity. Energy culture is built from the most diverse sectors of society and none of them, within their heterogeneity, is excluded; this promotes fairer and more democratic societies in energy terms (Glück, 2018). This work contributes with its KM approach to the strengthening of an energy culture.

Disclosure statement

No potential conflict of interest was reported by the authors.

Correction Statement

This article has been republished with minor changes. These changes do not impact the academic content of the article.

Additional information

Funding

This work was supported by the Consejo Nacional de Ciencia y Tecnología [686216/581498] and Universidad Autónoma de Baja California.