Analysis of Factors Influencing Households’ Preference Level for Solar Energy in Urban Areas of Southwest Nigeria

If sustainable development goals (SDGs) seven of affordable and clean energy will be achieved, access to clean energy such as solar energy needs dedicated attention. The study assessed respondents’ perception of solar energy and identified factors influencing their preference level in urban areas of Southwest Nigeria. Data were obtained from four hundred and eighty solar energy users and four hundred and eighty non-solar energy users leading to total of nine hundred and sixty respondents’ through a multi-stage sampling technique. Descriptive statistics, 5-likert scale type and tobit regression model were used to analyze the data collected. The respondents agreed that solar energy is the most appropriate source of energy compared to fossil fuel with mean score of 3.83. Price stability (0.83) attribute of solar energy is one of the reasons that makes it preferred by the respondent. A unit increase in household income will lead to 0.0879592 preference for solar energy in the study area. The study recommends that government at all levels should sensitize the households on the usage of solar energy in Nigeria to achieve the goal seven of SDGs.


INTRODUCTION
A large proportion of the world's population mostly the developing countries like Nigeria generate the energy they use from solid fuels and most of these fuels are known to pose a real threat to human health (Perera, 2018). According to World Health Organization (WHO), more than 40% of the world's population relies on polluting fuels (WHO, 2012). Particularly in the developing countries, approximately three billion people are using polluting crude oil, natural gas, solid fuels such as biomass, charcoal and animal waste for their daily purposes (Langbein et al., 2017). These fuels are combusted in poorly designed and inefficiently resulting in emission of noxious gases and products of incomplete combustion especially in the urban areas (Makonese et al., 2017), which are the major human source of Greenhouse Gases (GHG) and spurs climate change by releasing carbon dioxide into the atmosphere (Perera, 2018).
Also, the global share of the population with access to clean energy and technologies reached 61% in 2017, up from 57% in 2010. Despite this progress, close to 3 billion people still rely primarily on inefficient and polluting energy systems. If Sustainable Development Goals (SDGs) seven of affordable and clean energy will be achieved, access to clean energy such as solar energy needs dedicated attention. To reach universal access to clean energy by 2030, the annual rate of clean energy access needs to accelerate to 3%. If the current trajectory continues, 2.3 billion of the global population would remain without access to clean energy in 2030 (Perera, 2018).
In turn, due to the recent concerns about the negative externalities of the use of traditional fuels on the environment and health, household fuel transition from dirty fuels towards clean fuels in the developing countries has received growing research attention (Muller and Yan, 2018). Access to clean, affordable and efficient energy has become a challenge for the majority of the low to medium income households in developing countries (Kimemia and Annegarn, 2012;Makonese et al., 2017). These challenges coupled with the degradation of global environment caused by the Greenhouse Gas (GHG) emissions and the prediction of the depletion of the fossil fuel resources by scientists have all encouraged the global community to search for alternative sustainable and environmentally friendly energy resources (Kurnia et al., 2016). Amidst other alternative energy sources, solar energy has emerged and earned more attention across the world because solar energy is considered the most sustainable clean and renewable environment friendly energy source (Demirbas, 2010), and have been adopted by a large population of the developed world.
However, depending on the price of fossil fuels, solar energy is relatively more expensive than the fossil fuels in Nigeria, due to the high costs of installation (Dioha et al., 2012;Ajala et al., 2015), and they will have to be marketed at higher prices (Sivashankar et al., 2016). The need for rapid deployment of solar energy technologies has not received the political attention it deserves in Nigeria. High entry costs of solar energy solutions, a lack of consumer awareness of their benefits, financing gaps for producers seeking to enter the market, slow progress in the innovation of solar energy, and lack of infrastructure for solar energy production and distribution have together kept widespread solutions to this challenge out of reach.
Access to clean and renewable energy technologies has the potential to save millions of lives each year. Household air pollution resulting from the use of crude oil and natural gas to generate electricity, and the use of generator that emit GHG alone is responsible for some hundreds of thousand death annually in Nigeria (Federal Ministry of Health, 2018). Hence, the need to know the perception and the preference level of solar energy in urban area of Southwest Nigeria. This is with the aim of guiding the policy maker to know what attracts the respondents to use solar energy and the preference level of solar energy in Nigeria. This will reduce the degradation of global environment caused by the Greenhouse Gas (GHG) caused by use of crude oil and natural gas to generate electricity and reduce the use of generator that emits GHG using fossil fuels. ). An important aspect of vegetation of the four States is the prevalence of the tree crops. The major tree crops in the four States include: Cocoa, kola, oil palm and citrus. While the arable crops are yam, cassava and maize. The map of the study area is shown in Figure 1.

Data Collection and Sampling Procedure
A multi-stage sampling procedure was used in the selection of respondents. The first stage involved the selection of Ekiti, Ondo, Osun and Oyo State for the study. The second stage involved the purposive selection of capital cities of each State for the study because most of the emission of GHG takes place in the cities where there is urbanization influence on both consumption and production (Yonghong et al., 2017). These capital cities are: Ekiti State is Ado-Ekiti, Ondo State is Akure, Osun State is Osogbo and Oyo State is Ibadan. The third stage involved the purposive selection of 120 households using solar energy by snowball technique and random selection of 120 non-user households from each of the capital city. This led to a total of 960 respondents divided into 480 solar energy users and 480 non-users.

Nature and Source of Data
Primary data were used for this study. Primary data were obtained through the administration of well-structured questionnaire. The questionnaire was used to obtain relevant information on numbers of variables such as: socio-economic profile of the selected solar energy users and non-users such as age of consumers, household size, education, gender and occupation, willingness to pay criteria, their knowledge on solar energy and estimated annual income.

Data Analytical Procedure
Descriptive statistics comprising of frequency, mean and percentage were used to describe the socio-economic characteristics of the respondents. 5-likert scale type was used to determine the perception of solar energy users' and non-users' in the study area.
Respondents were asked to respond to statements relating to the economic attraction of solar energy usage, using Strongly Agreed (SA), Agreed (A), Indifferent (I), Disagreed (D) and Strongly Disagreed (SD). The responses were scored as 5,4,3,2 and 1 for SA, A, I, D, and SD respectively. The mean from each statement were obtained and used to classify the responses on each statement into SA (>4.5), A (3.5-4.4), I (2.5-3.4), D (1.5-2.4), SD (<1.5). The grand means for all the statement were calculated to be able to place the responses on a scale that enable a conclusion to be drawn on what the perception of the respondents is on solar energy.
Tobit Regression model was used to estimate the preference level of respondents for solar energy in the study area. The level of preference was measured using a scale of 1-5 (not preferred = 1, least preferred = 2, indifferent = 3, preferred = 4, most preferred = 5) based on if the consumers prefer the products or not. The following attributes were measured using the above scale: price, heat intensity, renewability, accessibility, easy to light up, price stability. The tobit regression model is state thus: Where Y i = the dependent variable that was used in examining the preference level of consumers for solar energy. Y i * = y i if 0 <y i < 1 (for those that prefer based on solar energy attributes) Y i * = y i if y i = 0 (for those that do not prefer) X i = vector of consumers' characteristics relevant in explaining the levels of preference for solar energy. β = vector of parameters to be estimated U i = normally distributed error term The independent (explanatory) variables that were used in examining the preference level of consumers for solar energy in the study area were as specified below: X 1 = Age of Consumer (years) X 2 = Consumption of clean (Solar) energy (years) X 3 = Marital Status (married = 1, 0 otherwise) X 4 = Gender of the consumer (male = 1, female = 0) X 5 = Year of Formal Education (years) X 6 = Household size (number of persons) X 7 = Estimated Annual income (₦) X 8 = Main Occupation of consumer (Public Sector = 0, Private Sector = 1) X 9 = Price of products (₦/Square m 2 ) X 10 = Familiarity with GHG (1 = Yes, No idea = 0) X 11 = Familiarity with climate change (1 = Yes, No idea = 0) X 12 = Knowledge of solar energy (1 = Yes, No idea = 0) X 13 = Membership of Environmental Association U i = Error term Table 1 presents the socioeconomic characteristics of the respondents. The table revealed that 55.0% (264) of the solar energy users were within the age bracket of 31 and 40 years while 61.7% (296) of the non-users were within 51 and 60 years. The mean age of solar energy users was 44.3 years and that of the non-users was 58.7 years. This implies that most of the users of solar energy were younger, hence they are expected to be agile and the need for renewable energy to ensure their productivity. This is in line with the findings of (John, 2016) that in developing Countries, active economy working class appreciates innovation of renewable energy. Sex distribution of the respondents shows that 72.9% (350) of the solar energy users were male and 27.1% (130) of them were female, while 65.0% (312) of the non-users were male and 35.0% (168) of them were female. This reflects that men are conscious of the need for alternative energy with respect to electricity in Nigeria. Since they fend for the family, therefore, they need to be productive which can only be possible with availability of innovations and technologies that will enhance energy sustainability like solar Most of the respondents of both solar energy user 92.1% (442) and non-users 58.8% (282) have spent above 10 years studying in a formal educational institution. The mean years of schooling of solar energy user was 17.4 years and non-ser was 12.2 years. This implies that most of the respondents had formal education. This may be attributed to the fact that the study was carried out among the urban populace where majority of the populace were highly educated and these set of people were also willing to respond to the research questionnaires. Education has always played a huge role in determine the preference for technology and consequently the welfare of the people (Obayelu et al., 2014). It is expected that education would provide people with opportunities to access information and understand the benefits of solar energy. This corroborates the fact that high level of education will raise the preference of the benefit that solar energy inherits (Yuan, 2011;Hast et al., 2015). This could account for the reason why the mean year of schooling of solar energy users was more than that of the non-users in the study area.

Socioeconomic Characteristics of the Respondents
Again, most (97.1%) (466) solar energy users and 50.4% (242) of non-users had household size <5. The mean household size for users of solar energy was four persons per household while that of the non-user was six persons per household suggesting an urban moderate family size (Obayelu et al., 2014). Again, as shown in Table 1, all the respondents both the users and non-users were aware of solar energy in the study area. Since the study area are the urban cities and State capital, therefore, solar energy innovation and technology is not new in the study area. According to Olayinka et al. (2014), in Nigeria, most citizens leaving in the urban cities are aware of solar energy revolution. Table 1 further revealed that 58.8% of the solar energy users are working in the private sector and 63.1% of non-users works in the public sector. This buttress the findings of Salman, (2019) that renewable energy is mostly appreciated by those who works in the private sector in Africa. Since they must source for alternative energy that is cheaper and that will enhance their productivity.
From Figure 2, 30.8% of the solar energy users earned between ₦201,000 ($558.33) and ₦300,000 ($833.33) monthly while 12.9% earned above ₦501,000 ($1,391.67). The mean estimated income per month for users of solar energy was ₦221,256.12 ($614.60). Figure 3 revealed that 40.1% of the non-users earned between ₦101,000 ($280.56) and ₦200,000 ($555.56) monthly and 10.6% earned above ₦501,000 ($1,391.67). The mean estimated income per month for non-users of solar energy was ₦141,098.75 ($391.94). This shows that the solar energy users earn more than non-users, which implies that they have the financial endowments to afford solar energy. This is in line with the finding of Andrea, (2019) that most users of renewable energy in the developing Countries are middle and upper-class citizens who have the financial capacity to afford it.

Distribution of Respondents by Perception of Solar Energy
The respondents' perception on the benefits, values and the environmental advantages of using solar energy were generally positive. Drawing inference from the class boundaries of means as shown in Table 2, the respondents agreed that solar energy is the most appropriate source of energy when compared to fossil fuel with mean score of 3.83. This could be because solar energy is renewable and very easy to maintain compare to fossil fuel and natural gas which is responsible for more than 75% electricity in Nigeria. Although, the respondents were indifferent that solar energy was easily accessible compared to fossil fuel with mean score of 2.55. This could be because solar energy technology is relatively new in the study area. Hence, is expensive and the technology is not yet perfected, therefore, many households are skeptical about the energy technology.
Again, with the mean score of 2.44, the respondents were indifferent in their perception about solar energy been faster than fossil fuel when compared. This could because most households make use of solar energy for electrical appliances with minimal load, because the technology is just emerging in Nigeria, therefore, the level of trust is highly minimal (Akinboro et al., 2012). The   respondents also agreed that solar energy is an important source of energy and supply of solar energy is reliable with mean scores of 4.18 and 3.51 respectively. Since the emergence of solar energy in Nigeria, it has served as good source of energy for many households in the urban cities. Many have abandoned their generators and making use of solar energy as an alternative means of generating electricity. The respondents agreed that solar energy production can be done in a sustainable manner with the mean score of 3.97. This could because solar energy is a renewable energy unlike fossil fuel and natural gas.
Equally, the respondents strongly agreed (4.53) that solar energy is environmentally friendly (Not emitting GHG). The implication of these is that consumers are very much concerned about their health, their choice of energy, place high value on clean energy and are likely to prefer the products. This in line with the findings of Gracia et al. (2011) who reported that preference level for clean and renewable energy is influenced by consumers perception on the benefits from using the clean and renewable energy. However, the respondents were indifferent (2.80) that solar energy is cheaper than fossil fuel when compared. The cost of installing solar energy in Nigeria is relatively high, therefore, many who are interested can not afford it. This could likely account for the indifferent perception of the respondents with respect to the affordability of solar energy.

Respondents' Preference for Solar Energy to Electricity and Generator
The result in Table 3 revealed a greater percentage of the respondents preferred solar energy (81.5%) to fossil fuel in the study area. Hence, this buttress the findings of Oyedepo, (2012) that Nigerians prefer solar energy to fossil fuel, but the government at all levels needs to support them to shift from fossil fuel to solar energy. Table 4 shows the preference level for solar energy based on selected attributes (affordability, heat intensity, renewability, emission, price stability, ease to light up and accessibility). It was revealed that the product is mostly preferred by the respondents based on its renewability with preference index of 0.91. This is  one of the inherent attributes of solar energy. It makes it different from the fossil fuel that can be used up and not renewable. This is one of the reasons that makes it desirable as energy. Followed by its emission attributes with preference index of 0.87, this is an attribute that shows that solar energy does not emit GHG like the fossil fuel that is commonly use in Nigeria. So, with the use of solar energy, environmental pollution is averted. Price stability (0.83) attribute of solar energy is one of the reasons that makes it preferred by the respondent. This refers to the stability in the cost of installation of solar energy appliances in Nigeria. The cost of installation of solar energy per square meter is stable over time.

Respondents Preference Level for Solar Energy Based on Selected Attributes
It does not fluctuate like the price of fossil fuel which is subject to fluctuation in foreign exchange.
Ease to light up attributes with the preference index of 0.78 is the quality of easy to operate attributes of solar energy appliances. For example, lighting solar energy streetlight is very easy. Therefore, operating solar energy appliances is very easy. The heat intensity attributes (0.73) is one of the reasons why the respondents prefer solar energy to fossil fuel. This implies that using solar energy does not lead to emission of heat like the fossil fuel. Findings indicate that solar energy attributes have a positive effect on consumers' utility. This is in line with the findings of Gracia et al. (2011) that clean energy acceptance by the consumers is influenced by its positive attributes. However, solar energy product is least preferred because of its affordability and for being not easily accessible with an index of 0.45 and 0.48. This means that respondents would consider changing to solar energy if it will be available in a wider range and affordable (Lloyd. 2015).

Determinants of Respondents' Preference for Solar Energy
As shown in Table 5, the determinants of preference for solar energy was analyzed using tobit regression model. Income and price of the product were significant at 1%, year of education and  familiarity with climate change were significant at 5%. Household size and gender were significant at 10%. With respect to household income, implies that a unit increase in household income will lead to 0.0879592 preference for solar energy in the study area. This means that income is a major determinant of solar energy consumption among the households in the study area. It shows that as households' disposable income increases, more money may be devoted to the purchase and consumption of solar energy. Numerous studies point to income as the major driver behind the uptake of clean energy. This is in line with the study by Ariyo et al. (2018) carried out in Nigeria that increase in disposable income induces households to choose electricity and solar energy over wood and kerosene. Again, Baiyegunhi and Hassan, (2014) revealed in their study that in rural Nigeria, the transition from dirty fuel to clean fuel is critically influenced by increasing disposable income. All these studies seem to corroborate the energy ladder concept, which emphases income in explaining the transition from 'inferior' traditional energy to "normal" clean modern energy (Muller and Yan, 2018).
The price of the product was statistically significant from zero at 1% level of significance and had a negative influence on preference level for solar energy. This indicates that a unit increase in the price of the product will lead to −0.0929461 decrease in the preference level for solar energy by the respondent. This is in line with studies by Gracia et al. (2011);Zhang and Koji, (2012); Kim et al. (2018) that an increase in price of clean energy significantly reduces the probability of using such energy.
At 5% level of significance, the coefficient of year of education had a positive and significant influence on the respondents' preference level for solar energy in the study area. The positive coefficient of 0.1909668 for the year of education implies that a unit increase in the year of education of the respondents will lead to 0.1909668 increase in preference level for solar energy in the study area. The study revealed that the literate tends to appreciate the use of clean energy than the illiterate. Also, the familiarity with climate change had a positive and significant influence on respondent's preference level. The positive coefficient of 0.1031613 for the respondents in the study implies that a unit increase in the knowledge of climate change probably contribute to the increase in preference level for solar energy in the study area. This result also agrees with the findings of Sivashankar et al. (2016) who established that familiarity with climate change has a positive influence on respondents' preference level for clean energy.
Household size also plays an important role in energy choices (Israel, 2002;Abebaw, 2007;Zhang and Koji, 2012) findings revealed that there is negative association of household size with per capita energy consumption. Household size of respondents had a negative coefficient of 0.0995086 and it is statistically significant at 10%. This means a unit increase in the household size will reduce their preference for solar energy which is regarded as a clean energy. This is in line with the finding of Baiyegunhi and Hassan, (2014) indicate that larger households prefer non-clean energy over clean modern solar energy. One possible reason is that household size is often larger in poorer households that cannot afford modern clean energy.
Gender was statistically significant at 10% significant level with a negative coefficient. The negative coefficient of −0.093917which was in favour of the female gender means female headed household are more likely to prefer clean energy as their source of energy. This corresponds with findings of Brew-Hammond, (2010); Kowsari and Zerriffi, (2011) who stated that femaleheaded households prefer modern clean energy to fossil fuel. This may be attributed to the fact that women are often responsible for household cooking and thus they are directly affected by the air pollution emitted from the burning of the fossil fuel.

CONCLUSION
This study assessed households' perception and factors that influence the preference level of solar energy in the Southwest Nigeria. Sex distribution of the respondents shows that 72.9% (350) of the solar energy users were male and 27.1% (130) of them were female, while 65.0% (312) of the non-users were male and 35.0% (168) of them were female. The mean years of schooling of solar energy user was 17.4 years and non-ser was 12.2 years. The respondents agreed that solar energy is the most appropriate source of energy when compared to fossil fuel with mean score of 3.83. High percentage of the respondents preferred solar energy (81.5%) to fossil fuel in the study area. It was revealed that the product is mostly preferred by the respondents based on its renewability with preference index of 0.91. However, solar energy product is least preferred because of its affordability and for being not easily accessible with an index of 0.45 and 0.48. The determinants of preference for solar energy was analyzed using tobit regression model. From the result, income and price of the product were significant at 1%, year of education and familiarity with climate change were significant at 5%. Household size, gender and main occupation were significant at 10%.
The study recommends that sensitization and awareness campaign of solar energy should be intensified. Respondents should be made aware of the benefits of solar energy while sensitizing them about the ill-effects of using fossil fuel, fostering a sense of personal environment responsibility, greater motivation and commitment towards the protection of the environment through media, promotions product and technology demonstration, houseto-house canvassing, trade fairs, educational outreach, shows at educational institutes, and public places.