1. Introduction
In recent years, social media platforms such as Facebook, Twitter and YouTube have provided an accessible, affordable and ubiquitous space for the flow of ideas and opinions [
1]. Social media have also been an incredible tool for individual empowerment in developing countries, where citizens have increasingly turned to such platforms to express their opinions and perceptions of issues that impact their lives [
2]. The emergence of social media has also generated new opportunities for researchers and policymakers, specifically, opening interesting avenues for accessing citizens’ perceptions of socioeconomic, environmental, scientific and political issues.
Citizens’ perceptions can be defined as the ways in which the citizens of a given country or region (as individuals or a group) perceive specific issues or events that affect them, i.e., how they think about issues or events, understand them and form an impression of them [
3]. Citizens’ perceptions can have significant implications regarding the design and successful implementation of policy initiatives [
4,
5], and can be an important tool for increasing transparency, efficiency, accountability, good governance and trust [
6,
7]. Opinions shared on social media are often expressed by a generally active citizenry, and social media opinions are becoming increasingly important, emerging as a legitimate, influential expression of citizens’ opinions [
8]. Scholars in several disciplines have examined social media opinions in order to assess citizens’ perceptions of, for example, public health policy [
9], healthcare services [
10], energy [
11], and vaccinations [
12]. Surprisingly, this emerging source of citizens’ perceptions has yet to be adequately accessed for the understanding of citizens’ perceptions of land-related policies.
Land-related policies are those associated with the land, such as land-use policy, agricultural policy and housing policy. Land-related policies are of critical importance in the achievement of sustainable development, good governance and economic growth, and in the mitigation of climate change and the alleviation of poverty, particularly in developing countries [
13,
14]. Through land-related policies, governments express their perception of the ways in which to solve land-related problems or challenges, such as unsustainable land-use management, inadequate land-use planning and land–natural resource conflicts [
15]. Citizen’s perceptions of land-related policies matter, as citizens are the major stakeholders in most policies, including land-related policies. In sub-Saharan Africa, for instance, the people are the main landholders, with around 90% of the land being under customary tenure, in which land is held by people in the community [
16]. From this customary standpoint, most people consider land to be the basis not only of their economic livelihood but also their heritage, and part of their ancestral, spiritual and social identity [
17]. These emotional and psychological attachments to land affect people’s perceptions of and resulting reactions to land-related policies and reforms. For example, Shipton [
18] notes that, in the context of African countries, attachment to land makes it impossible to mortgage or alienate land, as the citizens of these countries view the cultural features of the land as an inalienable heritage that cannot be sold. Consequently, a policy that promotes mortgages for credit accessibility may be unable to achieve its objectives. This highlights the importance of understanding citizens’ opinions regarding land-related policies.
This study showcases the potential of social media as a data source for accessing citizens’ perceptions of land-related policy in the context of a developing country. The study builds upon the case of a land-related policy in Nigeria, the 2019 Rural Grazing Area (RUGA) settlement programme (hereafter ‘RUGA policy’), which attracted great attention and reactions from Nigerians. The RUGA policy was introduced to curb the persistent conflicts between farmers and herders over property rights and natural resource exploitation in Nigeria, but it did not receive the expected support of the people, and was suspended six weeks after its introduction due to the overwhelmingly negative citizen outcry. This study examines perceptions from that intense six-week period. With the RUGA policy, the government intended to use eminent domain to take over private/community land in some parts of the country for animal husbandry purposes, thus placing the policy on a wider scale with the potential to impact Nigerians as a whole (including farmers and herders). Foreseeing the potential impacts that the policy could have on their lives, the citizens approached social media platforms to express their opinions regarding the policy. We focused on YouTube, one of the more widely used social media and video platforms [
19,
20], on which Nigerians increasingly voice rich narratives that express opinions on political issues, providing a unique window into the perceptions of citizens.
We systematically collected YouTube videos of Nigerians expressing their perceptions of the RUGA policy during its implementation phase between 21 May 2019 and 3 July 2019, and we took an inductive thematic analysis approach to analyse the data and gain insights into the key recurring themes of the citizens’ perceptions of the policy. The study’s main contribution lies in showcasing the potential of the use of social media to access citizens’ perceptions, even in the case of sensitive interventions such as land-related policies.
The remainder of the article proceeds as follows. The next section briefly discusses the use of social media as a data source, noting the associated benefits, biases and concerns. Next, the methodology section describes the case study policy and the study’s data collection and analysis approach.
Section 4 presents a descriptive analysis of the emergent themes and recurring themes of the citizens’ perceptions of the policy, followed by a discussion in
Section 5. Finally,
Section 6 provides the concluding remarks.
2. Social Media as a Data Source for Citizens’ Perceptions
The assessment of citizens’ perceptions is a challenging task [
21], as perceptions can be complicated and distinctive. A person’s perception at a given time does not necessarily represent their view at other times, making “simple analysis and generalisation problematic” [
22] (p. 273). Consequently, there are no straightforward ways to access citizens’ perceptions on a particular topic; as such, scholars have relied mainly on indirect methods, such as focus groups, surveys and content analysis [
21]. Such methods have many strengths, but they can also be expensive and labour-intensive ways of collecting information.
Recently, researchers’ attention has shifted towards the use of social media as a data source to accesscitizens’ perceptions. Citizens are becoming actively engaged in policy issues, and social media provide a networked space for the formation, circulation and expression of citizens’ opinions [
1]. Because of the vast amount of information shared on social media, researchers now see social media as a fertile ground for accessing citizens’ opinions [
23]. In fact, some scholars suggest that social media data have the potential to augment or replace the traditional survey methods of understanding citizens’ opinions; see, for example, [
24,
25].
The opinions expressed on social media are generally from the more active citizenry [
8], and these opinions often generate great interest from various stakeholders (in private and public institutions) because they can reveal insightful information that otherwise may not be observed [
23]. For instance, Tumasjan et al. [
26] used citizens’ opinions expressed on Twitter as an indicator of political sentiment. They found that the opinions expressed on social media closely aligned with election results; see also [
27,
28]. Similarly, on policy issues, Barisione and Ceron [
8] examined citizens’ opinions on social media regarding austerity policies in three European countries. They note that social media provided a space for citizens to contest and mobilise against the governments’ austerity policies. The study emphasises that social media empower the citizenry and afford citizens the opportunity to make their voices heard, even regarding policy issues.
Aside from uncovering insightful information, several other benefits may accrue from the gathering of citizens’ opinions on social media. Social media data are timelier and less expensive to access than the traditional survey methods, generally comprising publicly available data that are easily accessible at close to zero cost, thus reducing the time, effort and money spent by researchers using social media content [
24,
25]. Additionally, studies have shown that people are becoming unwilling to participate in surveys—e.g., see [
29]—suggesting that the traditional survey method may become increasingly challenging due to changes in people’s communication habits [
30].
Social media data provide opportunities to study many topics—including citizens’ perceptions—that may be hard to study by conventional research methods. There are, however, several limitations to be considered when using social media as a data source. The first is the methodological concern regarding the generalisability of social media data. While the number of social media users has increased over the years, many people still lack access to social media platforms [
31]. In addition, most users are assumed to be passive and inactive [
32], such that those users who do express opinions may not be representative of the population, making it necessary to be cautious in drawing inferences and conclusions from social media data [
33].
The validity and reliability of social media data present additional challenges. The validity challenge is summarised, for example, by Mahrt and Scharkow [
34], who note that big data, such as social media data, are useful to reveal what users do but not why they do it, thus limiting the questions that can be asked or answered using social media data [
35]. Likewise, there are limitations related to the reliability of social media data. Social media data are governed by private companies, and not all of the interactions on a social media platform are shared. Social media platforms evolve in their settings, functionality and popularity (e.g., the composition of their user population fluctuates with current events and the rise and fall of other platforms), which affects the ways in which data are collected and analysed [
33].
Aside from these methodological challenges, the use of social media as a data source brings legal and ethical concerns. Sometimes, social media data may contain sensitive personal information that must be handled carefully according to privacy laws (e.g., the EU countries’ General Data Protection Regulation), and the nature of such legal and ethical guidelines may make it difficult to collect and analyse social media data [
35]. Furthermore, from an ethical point of view, informed consent and anonymisation must also be considered. Typically, in qualitative research, it is essential for researchers to seek the informed consent of the participants whose data are used, but there is a lack of consensus regarding the ethical question of whether one must seek informed consent when using social media data, as those data are publicly available on social media platforms [
36]. Furthermore, the anonymisation of a study’s participants is important but may be impossible to fully ensure with social media data, as a simple text string might suffice to expose a participant’s identity [
37].
Because of these concerns and potential biases, a researcher’s well-reasoned judgement is required when using social media as a data source [
33,
38]. Consequently, the highlighted biases and concerns were taken into consideration when collecting and analysing the data in this study. Most importantly, as in the case of all research, all of the limitations and considerations are reported along with the findings in order to prevent misinterpretation and enable the appropriate use of this study’s findings.
3. Methodology
This section presents an overview of the case study’s land-related policy, and describes the study’s data collection and analysis.
3.1. Case Study: Rural Grazing Area Settlement (RUGA) Policy
Strife between farmers and herders remains a cause of persistent destructive conflict in Nigeria. Nomadic herders in search of pasture and water routinely clash with farmers whose crops are destroyed by their herds. These clashes occur in various part of the country, especially the north, and have often resulted in the loss of lives and property. According to the 2017 International Crisis Group report, thousands of civilians have lost their lives and tens of thousands have been displaced because of these persistent conflicts. The economic toll of the conflicts has been similarly huge [
39]. Various solutions have been explored in order to resolve the persistent conflict, but none have yielded a desirable outcome. However, of great interest to this study is the recent policy introduced by the federal government of Nigeria, i.e., the RUGA policy, which attracted a great deal of attention from Nigerians.
On 11 May 2019, without consultation or the involvement of citizens in the policy process, the office of the Federal Government of Nigeria approved the RUGA policy, which sought to establish settlement areas for farmers involved in animal husbandry, particularly herders. The government intended to compulsorily acquire land from private/community landholders in various parts of the country, and to transform the land into settlements. These settlements were expected to be organised colonies with adequate basic infrastructure and amenities, such as road networks, veterinary clinics, markets, schools and hospitals. Starting with 12 pilot states, the programme was expected to be implemented in various states in Nigeria.
The office of the president of Nigeria envisioned the RUGA policy as a panacea for the persistent problem of conflict between herders and farmers, and that it would also boost the country’s economy, but it did not foresee the degree of confrontation and mass rejection that the proposed programme met. The government touted several benefits that the policy could provide for Nigerians, including a drastic reduction in clashes between herders and farmers; a boost in animal production, quality and hygiene; accessible animal care; job creation; the reduction of cattle rustling; and access to credit facilities and private sector participation through investment. Despite the supposed benefits, the programme was suspended two months after its introduction. A concise timeline of the events surrounding the RUGA policy is given in
Figure 1. Citizens’ social media opinions during the policy’s implementation are of particular interest in the context of this study.
3.2. Data Collection
This study used YouTube-generated video clips as a data source, which served two major functions: (1) it captured people’s perceptions during the policy implementation process, and (2) it provided naturally occurring data, which are believed to be preferable to researcher-generated data, which often suffer from “researcher effects”, whereby the presence of a researcher influences the findings [
40].
A systematic approach was taken in order to identify relevant videos generated on YouTube during the implementation phase of the RUGA policy in Nigeria (see
Figure 2). Firstly, the YouTube database was queried using the following search terms: “RUGA policy in Nigeria”, “RUGA settlement policy Nigeria” and “RUGA Nigeria”. The initial search yielded 788 videos. The videos were then screened on the basis on their titles and when they were uploaded. Our focus was on perceptions expressed during the implementation phase of the RUGA policy (see
Figure 1); as such, videos uploaded before 21 May 2019 or after 3 July 2019 were excluded, along with duplicate videos and videos with titles that were not relevant to the RUGA policy. This screening excluded 601 videos, reducing the number of eligible videos to 187.
The resulting 187 videos were then fully assessed by watching each of them in order to identify those in which people expressed their opinions on the RUGA policy. This led to the exclusion of 153 more videos, including videos that had “RUGA” in their title but were not related to the RUGA policy (65 videos), duplicate videos and videos that were partial or full repetitions of other videos (48 videos), videos that did not include expressions of the perceptions of RUGA and videos that were not motivated by the RUGA policy (37 videos), and videos deleted by the creator during the analysis (3 videos). Ultimately, 34 YouTube videos were included in this study. The data were collected in February 2021.
The included YouTube videos comprise two types: personal reaction videos and interview videos. In personal reaction videos (11 videos), YouTube users speak directly to viewers to share their opinions of the RUGA policy, while street and studio interview videos (23 videos) depict interviews conducted to elicit other people’s opinions of the RUGA policy. The majority of the people in the videos are regular citizens (75) who wanted to air their opinions regarding the RUGA policy. In addition, there is a total of four government officials in the videos. These government officials are introduced in the videos as people holding political positions in the country. A summary of the properties of the included videos is given in
Appendix A (
Table A1).
Given the ethical concerns associated with social media data (see
Section 2), we ensured the informed consent of the video creators by notifying the selected creators of the intention to use their videos in the study. The notifications were sent through email or by posting a comment when no email address was available. None of the creators asked to be excluded from the study.
3.3. Data Analysis
The included YouTube videos consist of Nigerians expressing their opinions of the RUGA policy at the time of the posting of the video. A playlist of the videos was generated, and they were studied intensively in order to identify each person’s expressions regarding the RUGA policy. These expressions were carefully collected and transcribed verbatim. The transcripts were imported into Atlas.ti™ (version 8) for analysis.
The collected data were analysed with inductive thematic analysis, which offers a theoretically flexible approach to qualitative data analysis [
41], and allows the data to directly inform the themes. The videos were watched, and the transcripts were read in order to understand the data, and to generate relevant codes that characterised and summarised people’s expressed opinions and perceptions of the RUGA policy. The generated codes were then grouped by collating the related codes within a group. The codes within each group were organised and summarised into themes (subthemes), and related subthemes were further aggregated to form the major themes.
In order to ensure the trustworthiness and credibility of the analysis, researcher triangulation was used to refine the codes and themes. The generated themes were compared between the researchers, and no major differences were identified, but minor differences were resolved through discussion and consensus. Finally, the resulting themes were further re-examined against the codes in order to ensure consistency.
5. Discussion
This study demonstrates that social media data can provide valuable insights into citizens’ perceptions regarding land-related policies. Our findings portray a spectrum of citizens’ perceptions of the RUGA policy. While some people perceived the RUGA policy to be beneficial to Nigeria and its future development, most people voiced the opposite view, i.e., that the RUGA policy was unacceptable. This finding of mixed citizen perceptions reflects one of the challenges and complexities which is unique to devising land-related policies, i.e., an inability to adequately balance the interests of all stakeholders—see [
42,
43]—which often leads to land-related policies that benefit some people to the detriment of others [
44].
The social media data also provided access to rich narratives of citizens’ opinions. We found that, beyond explicitly expressing their objections or support of the policy, the people in the selected videos suggested various potential alternatives and approaches by which the government could solve the farmer–herder crisis. These findings suggest that, by assessing the perceptions expressed by the active citizenry on social media, it is possible to identify sensitive opinions and feedback that policymakers could use to improve the policy process. Additionally, the findings further point out the need for policymakers to endeavour to engage active citizenries in the governance of land, which can be crucial to the success of land policies, and can promote the ownership of the policy and encourage transparency and accountability, among other things [
45].
While this study’s approach appears promising, certain limitations must be noted concerning the use of social media data, particularly in this study. Firstly, the results revealed that some citizens believed that the RUGA policy was made for the benefit of only a certain ethnic group in the country. In a country with various ethnic groups, this finding suggests that ethnicity may have played a significant role in the perception of the RUGA policy, with members of the allegedly “favoured” ethnic group accepting the policy while members of other ethnic groups mobilised (albeit through social media) to reject the policy. This shows that the citizens’ opinions expressed in this study may have resulted from a cascade of influences in which individuals’ opinions were influenced by others in their immediate social media circle [
46]. Social media can be used to propagate widespread (mis)information that may be used to manipulate users’ emotions and influence their perceptions [
47], and this limitation could be a source of bias in studies using social media data. Hence, subsequent studies using social media data could suggest ways of controlling for this possible source of bias.
Secondly, this study used the expressed perceptions of people who appeared in videos uploaded to YouTube. The sample size was relatively small, and, consequently, the citizens’ perceptions presented in this study cannot be generalised to the entire Nigerian population. This notwithstanding, the study did not aim to generalise its findings but instead to offer insights on citizens’ perceptions expressed on social media regarding the RUGA policy. Similarly, we note that there is very little background information available on the citizens whose expressed perceptions were presented in this study (see
Table 1). Therefore, there is no way to fully determine the motivations or ideologies that drove their expressed opinions. Although this type of concern is inherent in qualitative research and cannot be fully avoided, more detailed information about the citizens would have enabled a more informed interpretation of the results.
Finally, we stress that the perceptions considered in this study were expressed around the time of the implementation phase of the RUGA policy. At that phase, citizens were becoming aware of precisely what the policy entailed, and were reacting and responding to the changes. Consequently, it is important to point out that it is uncertain whether the method used in this study could be applied at the planning phase of a policy. At that phase, presumably, people are to some extent unaware of the form the policy is taking, such that they would be unlikely to express their views as actively as during the implementation phase. Hence, data on citizens’ perceptions may not be as easily available in the planning phase as in the implementation phase of a policy. In addition, while this study has presented the potential benefits of the use of social media data to access citizens’ perception regarding the RUGA policy, we suggest that future studies should expand the scope of analysis by combining social media data analysis with additional methods, such as interviews with government officials, in order to enable the evaluation of the implementation effect of the policy.
6. Conclusions
This study demonstrates that social media data can be used to gain valuable insights into citizens’ perceptions of land-related policy, providing access to organic data and rich narratives of citizens’ perceptions of policies. With the growing use of social media, particularly in developing countries, we may assume that, in due course, it will be increasingly possible for both for scholars and policymakers to use social media to access citizens’ perceptions for the purposes of land-related policymaking. Hence, this study can serve as a reference for future studies using social media to access and collect citizens’ perceptions of land-related policies. However, the social media data must be accessed and used with caution due to the limitations inherent in the use of such data.
This study contributes to the literature on land-related policies in developing countries. In addition, the study showcases the potential of the use of social media as a source of data for accessing citizens’ perceptions, even in the case of sensitive interventions such as land-related policies. The method proposed in this study could provide practical means by which policymakers can access information on citizens’ perceptions, which could serve as a catalyst for policymakers to improve existing policy processes and deepen their insights into the ways in which people perceive land-related policies, and into the factors that contribute to those perceptions, providing policymakers with information that will help them target services in the most appropriate manner.