Energy Literacy of Prospective Primary School Teachers: A Comparative Study in a Portuguese University

Abstract

The word “energy” is one of the most used in our daily language. People are constantly reminded to save energy, use it efficiently, and increase the use of renewable energy, which is much more environmentally friendly than non-renewable energy. Citizens need to develop a good level of energy literacy, that is, a general predisposition, and, at the same time, a competence for engaging in desirable behaviours in the area of energy consumption. Teachers are a key element in promoting energy literacy. This article aims to compare the energy literacy of first- and third-year students enrolled in the 3 year long Basic Education undergraduate programme of a Portuguese university, who will become primary school teachers. To attain the research objective, 203 prospective teachers (first year: 98; third year: 105) were invited to answer a questionnaire on knowledge about energy (cognitive dimension), behaviour (behavioural dimension) concerning energy, and attitudes (affective dimension) toward energy. Concerning the cognitive dimension, statistically significant differences were obtained between first and third year groups and between third year rural and urban participants; differences between participants with parents with diverse levels of education are not statistically significant. The same applies to comparisons within the affective and the behavioural dimensions. Even though participants seem to believe that they know more than they do, the undergraduate programme in Basic Education seems to cause a small improvement, which is not enough for prospective teachers to reach a good level of energy literacy. Therefore, more attention should be paid to the development, in the Basic Education programme, of energy literacy, considering its diverse dimensions.

1. Introduction

The word “energy” is one of the most used in our daily language. This is because we need energy to live, enjoy activities, and work; we eat and drink because we require energy; different foods have diverse energy potential; home utilities consume energy; and workplaces are energy-dependent. In addition, we are constantly reminded to save energy, use it efficiently, and increase the use of renewable energy. Renewable energy is energy from resources that are naturally replenished and may be used over and over again (Çelikler, 2013), that is, environmentally friendly energy. In contrast, non-renewable energy causes global warming, ocean acidification, and adverse human health effects. In sum, energy is everywhere and is needed for everything, but renewable energy is key to a sustainable future.

The concept of energy is abstract (Avcı, 2021; Feynman et al., 1963), and there is no conceptual definition of energy that states what it is; there is only an operational definition of energy that says how energy manifests itself and how it can be measured. From a scientific viewpoint, energy is understood as the ability to do work, and can be transferred from one body to another, and transformed from one form of energy (kinetic and potential) to the other, and from one type of energy (e.g., nuclear, electric, gravitational, thermal, etc.) to another. Despite these transformations and transfers, the total amount of energy of an isolated system remains constant, as stated by the principle of the conservation of energy (Feynman et al., 1963).

From a societal perspective, energy use is a measure of the level of progress of countries and societies (Şen & Temel, 2023) and has been increasing steadily. The point is that some energy sources currently in use are finite and pose environmental risks. The use of fuel from such sources emits gases that contribute to global warming and may ultimately lead to climate change. To protect the environment, citizens need to make appropriate energy choices not only in terms of the energy sources selected (renewable or non-renewable), but also in terms of the amount of energy used. Although the production and use of renewable energy have been increasing (EEA, 2025), as innovative solutions for harnessing energy from renewable sources have gradually become economically competitive (Scharnigg & Sareen, 2023), renewable (or green) energy also entails environmental and economic costs. Therefore, reducing energy use and increasing energy efficiency should align with the choice of environmentally friendly (or green) energy sources (EEA, 2025).

For humankind to attain those goals, citizens need to develop a good level of energy literacy as “citizenship understanding” of energy (DeWaters & Powers, 2008); that is, they should hold a general predisposition and, at the same time, a competence for engaging in desirable behaviours in the area of energy consumption (Białynicki-Birula et al., 2022). Energy literacy is more than a multidimensional concept; it is an interdisciplinary concept that encompasses both scientific and social science literacies (US Department of Energy, 2017). It encompasses energy-related concepts and processes that belong to the science domain, as well as issues related to energy use and economy, and energy reduction and savings, which pertain to the social sciences domain. Energy literacy encompasses understanding how energy functions in the environment and in our social and personal lives (Ramachandran et al., 2024). This understanding is necessary for individuals and communities to make informed decisions and consumer choices, both at home and at work, and to support appropriate national and international energy policies (US Department of Energy, 2017), namely by voting for parliament or government.

An energy-literate person can understand the nature and role of energy in the universe and in our lives, and apply this understanding to answer questions and solve problems (US Department of Energy, 2017). They understand how energy is used in everyday life; the impact that energy production and consumption have on all spheres of our environment and society; the awareness of the need for energy saving efficiency and the need to develop alternatives to fossil fuel-based energy sources; and the impact of cognition on personal energy-related decisions and actions on the global community. They also strive to make choices, take action, and reflect understandings and attitudes with respect to energy resource development and energy consumption and are equipped with the necessary skills to do so (DeWaters et al., 2007; DeWaters & Powers, 2013). More recently, the US Department of Energy (2017) described an energy-related person as someone who:

• “can trace energy flows and think in terms of energy systems
• knows how much energy he or she uses, for what, and where that energy comes from
• can assess the credibility of information about energy
• can communicate about energy and energy use in meaningful ways
• is able to make informed energy use and energy decisions based on an understanding of impacts and consequences
• continues to learn about energy throughout his or her life.” (p. 1).

Whatever the list of features, energy literacy encompasses conceptual and procedural knowledge of energy and related concepts and processes, as well as attitudinal knowledge that may foster responsible attitudes toward energy. Together, conceptual and procedural knowledge (cognitive dimension) and attitudinal knowledge (affective dimension) about energy may foster responsible and appropriate behaviours towards energy (DeWaters et al., 2007). Even though low energy literacy mean scores in the knowledge dimension may coexist with moderate mean scores in the behavioural and affective dimensions (Ayata et al., 2024), concerns about energy problems need to be grounded in knowledge and skills for people to be able to effectively contribute toward energy-related solutions (DeWaters & Powers, 2011).

As expected, energy knowledge appears to be fundamental for achieving good energy literacy (Martins et al., 2022). However, energy literacy levels are influenced by a variety of personal, contextual, and educational factors (Naim et al., 2025), some of which affect the dimensions of energy literacy in different ways and, consequently, require specific (Akitsu & Ishihara, 2019) and contextualized interventions (Cerda et al., 2025). For instance, students in diverse national and institutional contexts exhibit different attitudes and report engaging in different types of energy-saving behaviours, and they also hold differing perceptions of their own energy use (Cotton et al., 2016). The experience of energy poverty (EPAH, 2022) may lead to higher levels of energy literacy (Białynicki-Birula et al., 2022), while people living in rural areas may hold more positive attitudes and values toward environmentally friendly energy than urban students (Bahrami & Mohammadi, 2021).

Awareness of the consequences of energy choices and use, social expectations regarding energy sources, and social pressure to use energy more effectively play a critical role in linking basic energy knowledge to energy-saving behaviour (Akitsu & Ishihara, 2019). Energy price awareness seems to have a positive and significant impact on knowledge and behaviour (Martins et al., 2020b), and financial expertise has a positive and significant impact on energy knowledge (Akitsu & Ishihara, 2019; Martins et al., 2020b). However, living in one’s own home and being responsible for electricity expenses do not affect energy literacy (Martins et al., 2020b).

Individual’s level of education seems to have a positive and significant impact on energy knowledge (Martins et al., 2020b). Parents’ level of education is a factor of students’ knowledge of energy (Białynicki-Birula et al., 2022) and energy literacy (Bahrami & Mohammadi, 2021). Energy literacy appears to increase with school grade (Akitsu & Ishihara, 2019) and with age beyond school. Martins et al. (2020a) concluded that age positively affects energy literacy, with senior people showing better overall and dimension-specific performance, except for behaviour.

Regarding gender differences, some studies report no differences in energy literacy between genders (Akitsu & Ishihara, 2019; Martins et al., 2021), whereas others find gender differences, with males scoring higher than females (Bahrami & Mohammadi, 2021). However, while women may show less knowledge on energy-related issues than men (Martins et al., 2020b), they may demonstrate a more positive attitude and more correct behaviour towards energy choices and use (Martins et al., 2020a, 2020b, 2021) and show greater concern and more predisposition to act in favour of energy conservation than men (Castañeda-Garza et al., 2025). This is consistent with Akitsu and Ishihara’s (2019) conclusion that basic energy knowledge does not necessarily increase overall energy literacy. As DeWaters et al. (2013) pointed out, it may be that some people “are aware of and sympathetic to the energy problems we face, yet may lack knowledge and skills to effectively contribute toward remediating these problems through individual efforts.” (p. 71).

The aforementioned information highlights the need not only for schools to provide energy education but also for it to begin early in children’s school lives (Kandpal & Bromanand, 2014). This requires curricula that integrate renewable energy (Ramongalo, 2024) and support energy literacy as a multidimensional concept, rather than curricula that focus solely on content knowledge, as they often do (Mosscrop, 2025). For a school to promote energy literacy, it is not enough to have the topic in the curriculum. Teachers, including primary school teachers, also need to possess a good level of energy literacy (in its components) and understand how it impacts the Sustainable Development Goals across not only educational but also environmental, social, economic, and leadership dimensions (Daoudi, 2024).

Because energy literacy is a relatively new area, except for physics teachers, most in-service teachers did not learn about energy, energy sources, and energy economy in school or university, and may be underprepared to teach these topics, which are relevant components of the cognitive dimension. Şen and Temel (2023) found that pre-service science teachers have a low level of comprehension about renewable energy, and Zyadin et al. (2014) concluded that secondary school teachers have limited knowledge of renewable energy (slightly higher for male teachers) and neutral perceptions regarding its use (slightly higher for female teachers). Ismail et al. (2025) found that science teachers had a weak sense of their ability to teach green energy content through technology, and Antink-Meyer and Aldeman (2021) concluded that teachers’ science content knowledge is necessary but not sufficient for effective teaching about energy technologies. Research also suggests that prospective science teachers hold misconceptions and incomplete understandings of energy (Avcı, 2021) and renewable energy (Şen & Temel, 2023), and that their lack of knowledge negatively influences their awareness of renewable energy (Guven & Sulun, 2017). Similar conceptual difficulties are expected among prospective primary school teachers who receive weaker training in science and are expected to prepare future generations of schoolchildren about renewable energy and resources (Çelikler, 2013). Therefore, they “should have more knowledge and more awareness than the rest of the society so that renewable energy could become widespread and gain social acceptance.” (Şen & Temel, 2023, p. 412).

Stylos et al. (2023) found that, although pre-service primary school teachers have low-to-moderate knowledge of energy concepts and challenges, they exhibit a satisfactory level of affect and behaviour regarding energy. They also concluded that women performed better than men on the affective and behavioural dimensions of energy literacy, and that pre-service teachers who had opted for a science or technology major in high school performed better on the cognitive and affective dimensions than those who had pursued the humanities. While these gender-related results are consistent with those obtained by Doğru and Çelik (2019), they contradict those obtained by Çelikler (2013), who found no gender differences in the awareness of pre-service primary school teachers of renewable energy. Stylos et al.’s (2023) results regarding high school major areas are consistent with Çelikler’s (2013) findings concerning differences by years of study, which indicate that more advanced prospective primary school teachers (PPSTs) show greater awareness than their less experienced counterparts. Kara (2020) concluded that prospective Turkish teachers were generally of the opinion that developed countries are the ones that use renewable energy sources most, and Campa-Bousoño et al. (2025) found that Spanish prospective primary school teachers have limited knowledge about the different methods of energy production and rely on the media as the primary source of information. In addition, they concluded that when prospective primary school teachers accept the adoption of renewable and clean energy sources, they prefer that power plants not be located in their neighbourhood (Campa-Bousoño et al., 2025).

So far, research suggests that the different dimensions of energy literacy are not equally developed among prospective teachers and that the strongest dimension depends on the research context. Thus, this article aims to compare the energy literacy of first- and third-year PPSTs enrolled in the Basic Education undergraduate programme of a Portuguese university. Basic Education (a three-year-long programme) is the first step in primary school teacher education; a Master’s programme (a two-year-long programme) is the second and final step. Primary school teachers can teach up to sixth grade, that is, 12-year-olds. Basic Education comprises courses in science, science education, and environmental education, which together constitute the main primary science training component of PPSTs. Therefore, the comparison allows the exploration of the impact of undergraduate courses on PPSTs’ energy literacy and provides insight into the level of energy literacy of these future primary school teachers.

2. Materials and Methods

2.1. Research Questions and Design

In line with the research objective presented in the previous section, this study aims to answer the following research questions: How does the energy literacy of 1st-year PPSTs enrolled in the Basic Education undergraduate programme at a Portuguese university compare with that of their 3rd-year counterparts? Is there any relationship between PPSTs’ energy literacy and the features of the place where they live? Is there any relationship between PPSTs’ energy literacy and their parents’ education levels?

To answer these questions, a cross-sectional comparative study of 1st- and 3rd-year PPSTs enrolled in a basic education undergraduate programme was conducted. Third-year students had already completed their undergraduate training in science and environmental education. Thus, based on the research reviewed, they were expected to hold a higher level of energy literacy, namely at the cognitive level, than their first-year counterparts. To address the research questions, data from both groups were compared and statistically analysed, with attention paid to the components of energy literacy, their parents’ education level, and the place where PPSTs live. Thus, following McMillan and Schumacher (2014), this study adopts a quantitative research paradigm and a comparative research design.

2.2. Participants

Participants in the study are undergraduate students attending a Basic Education programme at a Portuguese university. Successful completion of this programme is required for enrolment in one of the primary school teacher training Master’s programmes, which enable graduates to teach 6- to 12-year-old children. Before enrolling in the undergraduate programme, students complete secondary school (12th grade) in a STEM or humanities area. During the undergraduate programme, students take a variety of course units, namely in science and environmental education, in which they address energy in social, environmental, and scientific contexts. In contrast to 1st-year students, 3rd-year PPSTs who were invited to participate in the study had already completed the course units in science and environmental education. Therefore, the latter are expected to have stronger and better-balanced energy literacy, considering its different dimensions.

Table 1 shows the main features of the 203 participants in the study. It should be noted that, regardless of the year of study, the majority of participants are female, as teaching at the primary school level typically attracts more women than men. As expected, while most 1st-year students are under 20 years old, most 3rd-year students are 20–21 years old, because most students are 18 when they enter university. However, a few students may enter university under the “over-23 program”, which may explain the presence of a few over-22-year-olds across the years of study. Regardless of the year of their studies, most participants were used to living in rural areas.

Table 1. Main features of the participants (%).

PPSTs’ Year of Study	Gender			Age (Years)					Living Area
	Female	Male	Other	17	18–19	20–21	22–23	≥24	Rural	Urban
1st year (n1 = 98)	94.9	5.1	0.0	12.2	76.5	6.1	0.0	5.1	53.1	46.9
3rd year (n3 = 105)	95.2	3.8	1.0	0.0	9.5	79.0	4.8	6.7	61.0	39.0

(N = 203).

Focusing on parents’ education, Table 2 shows that, across all groups, more than half of fathers completed only 9th grade. This percentage is lower among mothers, who have a higher proportion of higher education (first-degree and postgraduate).

Table 2. Participants’ parents’ education level (%).

PPSTs’ Year of Study	Mother’s Education Level			Father’s Education Level
	≤9th Grade (n1 = 42; n3 = 45)	12th Grade (n1 = 32; n3 = 30)	Higher Ed. (n1 = 24; n3 = 30)	≤9th Grade (n1 = 52; n3 = 65)	12th Grade (n1 = 33; n3 = 25)	Higher Ed. (n1 = 13; n3 = 15)
1st year (n1 = 98)	42.9	32.7	24.5	53.0	33.7	13.3
3rd year (n3 = 105)	42.9	28.6	28.6	62.0	23.8	14.3

(N = 203).

2.3. Data Collection Technique and Instrument

Inquiry through a questionnaire (McMillan & Schumacher, 2014) is the technique most used to assess energy literacy in the literature reviewed. Most studies, including that by Ayata et al. (2024), used questionnaires derived from the instrument developed by DeWaters and colleagues (DeWaters et al., 2007; DeWaters & Powers, 2013). Typically, those questionnaires include the three dimensions of energy literacy mentioned above: cognitive (that is, about knowledge on energy, energy sources, energy conservation, environmental implications of energy choices, etc.), affective (that is, about attitudes toward energy, energy usage, energy saving, etc.), and behavioural (that is, about the ways of dealing with energy and energy-related issues) dimensions.

The questionnaire used by Ayata et al. (2024) to analyse contributions of epistemological beliefs on energy literacy in lower-secondary school students in Turkey served as the basis for the questionnaire used in the present research. The original questionnaire includes a knowledge dimension in the form of a multiple-choice test with a total of 30 questions, and affective and behavioural dimensions in the form of a 5-point Likert scale (1: Strongly disagree; 5: Strongly agree) with 17 and 10 items. It also includes questions on personal data to characterise the participant in the study, and four questions on self-perceptions of energy-related issues (knowledge, behaviour, sources of knowledge, and energy saving). The cognitive dimension included some questions that did not fit the Portuguese context well, because, for instance, the terminology was not consistent with that used in Portugal, which acknowledges two forms of energy (kinetic and potential), and several types of energy (mechanical energy, electric energy, thermal energy, chemical energy, etc.). Only 20 cognitive items were included in the questionnaire for this study, which was considered a good number within the context of the whole questionnaire, which is quite long. Questionnaires that are too long increase the probability of random answers, which in turn, decreases the reliability of the data (McMillan & Schumacher, 2014). The 20 questions address the key issues related to the cognitive dimension of energy. The items/questions were translated by two authors; their translations were compared, and a consensus version was reached. The latter was pilot-tested, and a few words were replaced to increase the readability of the questionnaire.

The original questionnaire was validated in the Turkish context and showed a Cronbach’s alpha of over 0.7 (as expected), regardless of the energy literacy dimension considered. Given the cultural differences between Turkey and Portugal, as well as differences in the target population and the changes introduced in the cognitive section of the questionnaire, following Tavakol and Dennickhe’s (2011) recommendation, a construct validity analysis was conducted. Then, Cronbach’s alpha was computed for each dimension (or construct) and per year of study For the behavioural dimension, Cronbach’s alpha values were above 0.7 for both groups (1st-year: 0.73; 3rd-year: 0.75). For the affective dimension, the 3rd-year alpha was above 0.7, but for the 1st-year groups, it was significantly below 0.7 (0.56). Subsequently, based on an item–total correlation analysis, four items with very low correlations in the 1st-year groups were removed. With 13 items, the Cronbach’s alpha values improved, and the 1st-year approached 0.7 (1st-year: 0.68; 3rd-year: 0.75), which was considered acceptable. For the cognitive dimension, Cronbach’s alpha values are lower than for the other dimensions (1st-year: 0.63; 3rd-year: 0.52). Given the nature of the items in this dimension, the Kuder Richardson (KR-20) was computed from two splits of the test (even versus odd items), and a small increase was obtained for the 3rd-year group (1st-year: 0.63; 3rd-year: 0.54). According to Tavakol and Dennickhe (2011), the test’s broad scope and the range of concepts it encompasses may contribute to lower scores on the cognitive dimension. It can be expected that participants master some concepts better than others, which may lower the alpha value. However, as Taber (2018) argues, high alpha values may be undesirable when developing a test of scientific knowledge or understanding, as the items would likely be redundant.

2.4. Data Collection Procedures

After approval by the Ethical University Board (protocol code CEICSH 071/2025, 30 March 2025), one of the authors informed the PPSTs of the study’s objectives and invited them to participate, and assured them that the questionnaire was anonymous, that data would be analysed confidentially, and that they could withdraw at any time, with no penalty. All the PPSTs that were present accepted to participate and were asked to sign the informed consent form, which they did, and no one gave up. Afterward, participants were asked to answer a digital version, prepared in google forms (by Google, Mountain View, CA, USA), of the questionnaire under examination conditions. This was necessary because the cognitive section of the questionnaire should be completed without assistance. Data collection took place during four class sessions (two per year group), in the presence of one of the authors, who was not the class tutor.

2.5. Data Analysis Procedures

Data were analysed (through Microsoft Excel, Redmond, WA, USA). per study year, and results for the two years were compared, For the self-perception questions, relative frequencies per possible answer were computed. A Chi-square test was used to analyse statistical differences between the two groups. For the behavioural and affective dimensions, the qualitative Likert scale was converted into a quantitative scale, ranging from 1 (totally disagree; most undesirable) to 5 (totally agree; most desirable). The mean scores for each dimension and student were computed. Differences between the two groups were analysed using an unpaired t-test. For the cognitive dimensions, correct answers were awarded 1 point, and incorrect answers 0 points. This resulted in a minimum of 0 and a maximum of 20 points per participant on this dimension. An unpaired t-test was used to analyse the differences between the means of the two groups. It was not possible to analyse differences between male and female students due to the small number of men in the sample. Differences between urban and rural living participants, per year, were analysed through an unpaired t-test. Differences among participants with mothers and fathers with varying levels of schooling, by year, were analysed using ANOVA. The Chi-square test was used to analyse data (percentages) related to participants’ self-perception of energy-related issues. As usual in education (McMillan & Schumacher, 2014), the value of α taken as reference for considering that differences are statistically significant is 0.05.

3. Findings

3.1. Participants’ Self-Perception of Energy-Related Issues

The data presented in Table 3 show that most participants report knowing little or nothing about energy. As could be expected, third-year PPSTs show a higher perception of their own knowledge of energy than their first-year counterparts. The Chi-square test shows that the differences are not statistically significant (p = 0.148, that is, p > 0.05).

Table 3. Participants’ perception of their own knowledge of energy.

Own knowledge About Energy Literacy	1st Year (n1 = 98)		3rd Year (n3 = 105)
	f	%	f	%
No/little knowledge	65	66.3	56	53.3
Moderate knowledge	29	29.6	41	39.0
Quite a lot/Very much knowledge	4	4.1	8	7.6

(N = 203).

First-year students’ perception of saving energy is lower than that of their third-year counterparts (Table 4). Nearly 40% of the first-year participants, against about 32% of the third-year, save energy sometimes or less than that. The Chi-square test shows that the differences are not statistically significant (p = 0.348; that is, p > 0.05).

Table 4. Participants’ perception of their energy-saving behaviours.

Behaviour Toward Energy Saving	1st Year (n1 = 98)		3rd Year (n3 = 105)
	f	%	f	%
Never/Rarely save	2	2.0	4	3.8
Sometimes save	36	36.7	30	28.6
Often save	53	54.1	65	61.9
Always save	7	7.1	6	5.7

(N = 203).

Table 5 shows that more than one-fifth (21.9%) of the third-year PPSTs stated that they talk to their relatives about energy very often. The first-year PPSTs seem to have conversations on energy saving less frequently than their third-year counterparts do. The Chi-square test shows that there are no statistically significant differences between the two groups (p = 0.157, that is, p > 0.05).

Table 5. Participants’ conversations with their relatives on energy saving.

Conversations with Relatives on Energy Saving	1st Year (n1 = 98)		3rd Year (n3 = 105)
	f	%	f	%
Never/Occasional	16	16.3	14	13.3
Rarely	14	14.3	12	11.4
Often	58	59.2	56	53.3
Vey often	10	10.2	23	21.9

(N = 203).

The internet and the school, in that order, are the two sources of information that most participants recognise as contributing to their knowledge about energy (Table 6). In this case, the percentages are similar for both groups. The Chi-square test shows that there are no statistically significant differences between the two groups (p = 0.976; that is, p > 0.05).

Table 6. Participants’ perceptions of the contribution of diverse sources of information to their knowledge on energy.

Contribution to Their Knowledge on Energy	1st Year (n1 = 98)		3rd Year (n3 = 105)
	f	%	f	%
School	34	34.7	36	34.3
Books, newspapers, or magazines	2	2.0	1	1.0
Friends or family	13	13.3	15	14.3
Internet	43	43.9	46	43.8
Television programmes	6	6.1	7	6.7

(N = 203).

3.2. Affective Dimension of Energy Literacy

Third-year PPSTs exhibit a slightly higher performance (3.98 versus 4.10) and less dispersion on the affective dimension than their first-year counterparts, as their standard deviation is lower (Table 7). The differences between the two groups are statistically significant (p = 0.053).

Table 7. Participants’ performance on the affective dimension of energy literacy.

Year	Group	Mean	SD
1st	Global (n1 = 98)	3.98	0.401
	Rural (n = 52)	4.05	0.437
	Urban (n = 46)	3.91	0.346
3rd	Global (n3 = 105)	4.10	0.454
	Rural (n = 64)	4.15	0.495
	Urban (n = 41)	4.03	0.395

(N = 203).

When comparing rural and urban living PPSTs, per year of studies, the results indicate a slightly better performance of the PPSTs living in rural areas, regardless of the year, even though they have more variability of answers, as shown by the standard deviation values. However, comparisons of the rural/urban results for each year of study, through an unpaired t-test, suggest that the differences are not statistically significant in the first-year group (p = 0.087) and in the third-year group (p = 0.184).

An analysis of the results based on the mothers’ level of education (Table 8) suggests that even though there seem to be some small differences between the performance of children of mothers with different education levels, ANOVA showed that the differences among the three subgroups are not statistically significant in any of the study groups (p > 0.05). The same applies to the subgroups of children of fathers with different education levels. However, it should be kept in mind that the reduced dimension of some subgroups may introduce some disturbance in the results of this analysis.

Table 8. Participants’ performance on the affective dimension of energy literacy according to their parents’ education level.

Parents	Levels	1st Year (n1 = 98)			3rd Year (n3 = 105)
		Mean	SD	p	Mean	SD	p
Mother	Up to 9th grade (n1 = 42; n3 = 45)	3.99	0.436	0.647	4.08	0.447	0.694
	Up to 12th grade (n1 = 32; n3 = 30)	4.02	0.368		4.07	0.475
	Higher education (n1 = 24; n3 = 30)	3.92	0.362		4.16	0.431
Father	Up to 9th grade (n1 = 52; n3 = 65)	3.98	0.388	0.996	4.10	0.443	0.986
	Up to 12th grade (n1 = 33; n3 = 25)	3.98	0.440		4.10	0.528
	Higher education (n1 = 13; n3 = 15)	3.99	0.325		4.12	0.341

(N = 203).

3.3. Behavioural Dimension of Energy Literacy

Concerning the behaviour dimension, Table 9 shows that the overall mean performance is slightly above the midpoint, regardless of the group (3.65 versus 3.64), with the third-year group showing more variability than the first-year one. The differences between the two year groups are not statistically significant (p = 0.863). When comparing rural and urban living PPSTs, per year of studies, the results indicate similar performances of the PPSTs living in rural and in urban areas, regardless of the year, even though the urban group has more variability of answers, as shown by the standard deviation values. However, comparisons of the rural/urban results for each year of study, through an unpaired t-test, suggest that the differences are not statistically significant in the first-year group (p = 0.883) and in the third-year group (p = 0.818).

Table 9. Participants’ performance on the behavioural dimension of energy literacy.

Year		Mean	SD
1st	Global (n1 = 98)	3.65	0.609
	Rural (n = 52)	3.66	0.596
	Urban (n = 46)	3.64	0.630
3rd	Global (n3 = 105)	3.64	0.616
	Rural (n = 64)	3.63	0.616
	Urban (n = 41)	3.65	0.624

(N = 203).

The analysis of the results of the behavioural dimension based on the mothers’ level of education (Table 10) suggests that even though there seem to be some small differences between the performance of children of mothers with different education levels (in the first-year group), ANOVA showed that the differences among the three subgroups are not statistically significant in any of the study groups (p > 0.05). The same applies to the subgroups of children of fathers with different education levels, even though small differences were observed among the three subgroups, for both study groups.

Table 10. Participants’ performance on the behavioural dimension of energy literacy according to their parents’ education level.

Parent	Levels	1st Year (n1 = 98)			3rd Year (n3 = 105)
		Mean	SD	p	Mean	SD	p
Mother	Up to 9th grade (n1 = 42; n3 = 45)	3.71	0.573	0.666	3.50	0.607	0.109
	Up to 12th grade (n1 = 32; n3 = 30)	3.59	0.632		3.69	0.597
	Higher education (n1 = 24; n3 = 30)	3.62	0.616		3.79	0.593
Father	Up to 9th grade (n1 = 52; n3 = 65)	3.67	0.589	0.735	3.62	0.645	0.438
	Up to 12th grade (n1 = 33; n3 = 25)	3.59	0.541		3.76	0.579
	Higher education (n1 = 13; n3 = 15)	3.74	0.788		3.51	0.477

(N = 203).

3.4. Cognitive Dimension of Energy Literacy

Table 11 shows the overall mean performance for the cognitive dimension. Neither year group reached the 10 mid-point. However, the third-year group approached it (its performance was 9.83 against 8.31 of the first-year group) and showed less variability of answers than the first-year group. Comparison through unpaired t-test indicates that the differences between groups are statistically significant (p = 0.0005; that is, p < 0.05). When comparing rural and urban living PPSTs, per year of studies, the results indicate that the rural PPSTs’ performance was below the urban’s, in the first-year group, but the opposite occurred in the third-year group. Comparisons of the rural/urban results for each year of study, through an unpaired t-test, suggest that whereas the differences are not statistically significant in the first-year group (p = 0.263), the opposite happens in the third-year group (p = 0.004).

Table 11. Participants’ performance on the cognitive dimension of energy literacy.

Year		Mean	SD
1st	Global (n1 = 98)	8.31	3.209
	Rural (n = 52)	7.96	3.093
	Urban (n = 46)	8.70	3.326
3rd	Global (n3 = 105)	9.83	2.927
	Rural (n = 64)	10.48	2.766
	Urban (n = 41)	8.81	2.909

(N = 203).

Table 12 shows that the analysis of the results of the cognitive dimension based on the mothers’ level of education suggests that even though there seem to be some small differences between the performance of children of mothers with different education levels (especially in the first year), ANOVA showed that the differences among the three subgroups are not statistically significant in any of the study groups (p > 0.05). The same applies to the subgroups of children of fathers with different education levels, even though small differences were observed among the three subgroups, for both study groups.

Table 12. Participants’ performance on the cognitive dimension of energy literacy according to their parents’ education level.

Parent	Levels	1st Year (n1 = 98)			3rd Year (n3 = 105)
		Mean	SD	p	Mean	SD	p
Mother	Up to 9th grade (n1 = 42; n3 = 45)	7.95	2.609	0.643	9.87	2.963	0.935
	Up to 12th grade (n1 = 32; n3 = 30)	8.59	3.673		9.67	2.891
	Higher education (n1 = 24; n3 = 30)	8.54	3.366		9.93	2.851
Father	Up to 9th grade (n1 = 52; n3 = 65)	8.46	2.990	0.878	9.82	2.806	0.723
	Up to 12th grade (n1 = 33; n3 = 25)	8.15	3.267		9.56	3.125
	Higher education (n1 = 13; n3 = 15)	8.08	3.710		10.33	2.936

(N = 203).

4. Discussion and Conclusions

Energy is a commodity that, more or less consciously, everybody uses every day, and whose use may have important environmental implications. This is why children should learn, both at home and at school, to manage energy from a very young age. For primary school teachers to give a positive contribution to that, they need to develop a good level of energy literacy themselves, namely during their initial teacher education programmes.

The research involved two comparison groups; therefore, some caution is warranted when interpreting the results. First-year PPSTs may have developed their energy literacy in school or through online sources, as they reported when asked about their information sources. Thus, a comparison of first- and third-year PPST literacy indicates that, during their undergraduate programme, prospective primary school teachers may have developed the diverse dimensions of energy literacy differently.

The cognitive dimension is further developed in the third-year group. This means that the science and environmental education courses they took may have positively impacted their knowledge, but the level attained is low for a teacher who must teach about energy. These results are consistent with those reported by Akitsu and Ishihara (2019), who concluded that holder students have greater knowledge of energy literacy. Participants’ performance on the cognitive dimension aligns with their self-perceived knowledge of energy, which is slightly higher for the third-year group. An interesting result was that, although first-year rural students were behind urban students, third-year rural PPSTs were ahead of their urban counterparts. This result may be explained in light of the study by Bahrami and Mohammadi (2021), which reported that students in rural areas tend to hold more positive attitudes and values toward environmentally friendly energy than their urban counterparts. Consequently, these more positive attitudes may contribute to a greater motivation to learn about environmentally friendly energy and energy in general. Concerning parents’ education, no statistically significant difference was obtained among students who have parents with different levels of education. These results contradict those obtained by Białynicki-Birula et al. (2022) and Bahrami and Mohammadi (2021), who concluded that parents’ education level impacts their children’s knowledge of energy. The results obtained in the present study are hard to interpret because information on the specific education area of parents holding higher education was not collected. It would be expected that parents holding education in science and technology-related areas would be more influential on their children’s energy literacy than their human and social sciences counterparts. However, this hypothesis would need to be studied.

Concerning the affective dimension, third-year students are ahead of their first-year counterparts, but the difference is not statistically significant. This suggests that undergraduate learning may have had only a small effect on PPSTs, likely because attitudes are difficult to change and do not command as much teacher educators’ attention as the knowledge dimension. These results are inconsistent with Stylos et al. (2023), who found that pre-service primary school teachers show a rather satisfactory behaviour regarding energy. Although differences are not statistically significant, third-year PPSTs living in rural areas may perform slightly better than PPSTs in urban areas. This may be because in rural areas, families have lower income than families in urban areas (Rodrigues, 2019), and therefore may be more careful about energy consumption and also more familiar with private renewable (namely solar) energy production. Similarly, no statistically significant differences were found among the PPSTs with parents who have different education levels. This result is consistent with that obtained by Bahrami and Mohammadi (2021), who conclude that parents’ level of energy literacy does not have a significant impact on their children’s affective dimension of energy literacy. This may mean that parents are not sufficiently aware of how much people’s relationship with energy affects life on earth at global, local, and even individual levels. If this is true, it may also mean that parents do not promote their children’s attitudes towards energy. However, these conjectures would require empirical support.

Performance on the behavioural dimension was lower than on the affective dimension for both year groups. This may be because people tend to adopt convenient behaviours that require less effort and attention. In addition, according to Akitsu and Ishihara (2019), this result may be explained by the absence of peer social pressure regarding energy-saving behaviours, as all students showed similar performance on the behaviour dimension. Moreover, the fact that they do not directly pay the electricity bill may reduce the impact on their actual behaviour, which is broadly consistent with the findings of Martins et al. (2020b). This may also explain why the first- and third-year PPSTs yielded similar results across groups and when divided into rural/urban subgroups. These results are inconsistent with PPSTs’ perception of their energy behaviour regarding energy saving, as PPSTs showed a considerably high level, which was higher for the third-year group. In addition, this result may indicate that PPSTs are not aware of their actual behaviour and require support to develop such awareness. The behaviour dimension results contradict those reported by Stylos et al. (2023), who found that pre-service primary school teachers exhibit a relatively satisfactory level of behaviour regarding energy. Concerning parents’ education, no statistically significant differences were found among the PPSTs. Again, this result is consistent with Bahrami and Mohammadi’s (2021) results on the behavioural dimension. This result may mean that, regardless of education level, parents exhibit similarly appropriate or inappropriate behaviours, or that they do not promote, or succeed in promoting, their children’s behaviour regarding energy.

Most first- and third-year PPSTs are used to discussing energy-related issues with their relatives. Most PPST fathers’ highest level of education is ninth grade. This means that during their conversations, PPSTs and, in particular, third-year PPSTs, should be expected to contribute to their parents’ and other relatives’ energy literacy development. To do so and to develop their future students’ energy literacy, they need to strengthen the cognitive and behavioural dimensions of energy literacy. The improvement attributed to the undergraduate programme in Basic Education is not enough. Therefore, greater attention should be paid to the development of energy literacy, given its diverse dimensions in this undergraduate programme. Echoing Cerda et al. (2025), a way forward could be to adopt an integrated, contextualised approach across curriculum units, namely, environmental, science, and science education areas, fostering coherent learning experiences about energy. Some place-based project work would probably enable PPSTs to use and further develop their knowledge of energy, how the place is dealing with energy transition, and how inhabitants should behave towards energy use, reduction, and choices. In addition to improving their energy literacy, this would better equip PPSTs to teach their future students energy literacy.

This research has some limitations. Consistent with the research questions, the study focused on a single university. Consequently, its results cannot be generalised. Although there is a mandatory number of science credits in Portuguese undergraduate programmes in Basic Education, other universities may have different science course syllabi, leading to varied outcomes. The use of a cross-sectional sample imposes limitations on conclusions regarding the development of energy literacy. To overcome this limitation, it would be worthwhile to conduct a longitudinal study following a cohort of PPSTs from the first to the third year. This would enable the evaluation of the impact of initial teacher education on PPST energy literacy, namely of urban and rural students, and the collection of information to improve the Basic Education programme and/or to organise training courses on energy literacy for newly formed teachers. Such courses should pay attention to the strategies used to engage and encourage students to be more energy-efficient (Cotton et al., 2015), which should be supported by institutional practices relative to energy usage. An important consequence of this would be the enhancement of prospective teachers’ energy literacy, with implications beyond the university community, namely for their future students’ personal and professional lives.