The Impact of Social Media Marketing on Consumer Behavior in Yemen ()
1. Introduction
In the digital era, social media has become a central element of everyday life, fundamentally reshaping communication, access to information, and consumer decision-making. Unlike traditional media such as television, radio, or print, platforms like Facebook and Instagram provide immediacy, interactivity, and broad accessibility, enabling real-time two-way engagement. These features allow businesses to go beyond passive promotion and engage consumers through targeted advertising, influencer endorsements, user-generated content, and behaviorally tailored messaging. As a result, social media has evolved from a supplementary marketing tool to a cornerstone of modern digital strategy, offering businesses a rich environment to forecast trends, understand consumer needs, and build stronger customer loyalty. For consumers, it provides access to reviews, comparisons, and personalized recommendations that shape awareness, attitudes, and purchasing choices. This study focuses on the role of social media—particularly Facebook and Instagram—in influencing consumer behavior in Yemen. It examines how these platforms affect the five stages of consumer decision-making: need recognition, information search, evaluation of alternatives, purchase decision, and post-purchase behavior. Special attention is given to demographic factors such as age, gender, income, education, and digital literacy, as these shape consumer responses. Younger users, for instance, may be more influenced by visual content and influencers, while older consumers may rely more on credibility and peer reviews. Such variations highlight the importance of segmentation in designing effective marketing strategies. Yemen provides a significant yet underexplored context. Despite socio-political challenges, internet and mobile penetration are rising, particularly in urban centers like Sana’a and Aden. Growing digital engagement among youth and the emerging middle class offers a unique opportunity to study how social media is influencing consumption in a developing economy. From a practical perspective, the research aims to generate insights that will help marketers identify drivers of online engagement, overcome barriers, and design culturally appropriate digital strategies. At the same time, the study acknowledges that social media’s evolution reflects a wider transformation in communication, where consumers are not only recipients of content but also active creators and influencers. This participatory culture requires a multidimensional perspective that integrates technological, psychological, and socio-cultural factors. In summary, the research seeks to investigate how Facebook and Instagram influence consumer decision-making in Yemen, with attention to demographic variations. Its findings are expected to contribute to both practical marketing strategies and the academic understanding of consumer behavior in emerging digital markets.
2. Literature Review
The evolution of the Internet has drastically transformed global business landscapes, allowing companies to reach international markets and enabling consumers to access, compare, and purchase goods across borders [1]. The rise of social media marketing (SMM) is a key development in digital transformation, influencing consumer behavior across all decision-making stages. Social media functions as a platform for interaction and information exchange, playing a central role in consumer socialization, where individuals acquire consumption-related skills and attitudes through peer communication. Platforms like Facebook and Instagram enable users to connect, share experiences, and shape each other’s purchasing decisions, making them powerful agents in this process [2]. These platforms have revolutionized the diffusion of information, enabling rapid and efficient sharing of user-generated content [3]. As [4] defines, social media comprises internet-based applications built on Web 2.0 principles, allowing the creation and dissemination of user-generated content. Its low-cost, real-time, and bidirectional nature makes it highly attractive for marketers aiming to foster brand relationships and build trust with consumers. Indeed, social media has significantly influenced consumer behavior—from information acquisition to post-purchase evaluation—as highlighted by [5]. Research by [6] [7] supports that patterns of internet usage, including time spent on social networks, shape how consumers evaluate products, form brand attitudes, and articulate dissatisfaction. According to [8] [9], social media also influences consumer perceptions, attitudes, and behaviors by bringing together like-minded individuals into interactive communities. The participatory nature of social media enables consumers to engage in the co-creation of value, enhance brand intimacy, and voice their experiences to broader audiences. As [10] emphasizes, consumer relationships can shift rapidly based on online experiences, where even minor dissatisfaction may be amplified via public platforms. Companies are therefore increasingly leveraging social media to not only inform and promote but also to listen, respond, and co-create alongside consumers. Social media is now considered a hub of market intelligence. This includes the role of electronic word-of-mouth (e-WOM), online communities, and social advertisements in shaping brand loyalty and purchase intentions. Empirical evidence by [11] shows that consumers make purchasing decisions for both low- and high-cost items based on recommendations from social media contacts. Their study with 249 consumers found that peer recommendations—more than price or advertising alone—substantially influenced buying decisions.
While social media does not always directly alter decision-making, it acts as a mediating factor, affecting consumer attitudes toward advertising and branding, thereby influencing behavioral intentions [12]. [13] further supports that brand advertisements on social media, whether via commercial sources or peer influence, shape consumer attitudes and enhance purchase likelihood. Social media thus serves both promotional and relational functions, offering businesses a dual approach to brand communication. Research also underscores the motivational drivers behind consumer engagement on social platforms. [14] identifies three primary gratifications: information-seeking, entertainment, and social interaction. According to [15], consumer engagement on social media is driven by both rational motivations (advocacy, knowledge-sharing) and emotional motives (belonging, self-expression). Social media reviews are perceived as reliable second-hand information that guides rational purchase decisions across diverse product categories. Beyond commerce, platforms like Twitter also influence legal and political decision-making by providing access to collective opinions and evidence, underscoring the broader societal role of online socialization. In the commercial domain, the concept of online shopping—referred to as web purchasing or internet retailing—has evolved alongside the rise of social networking sites [16]. According to [17] [18], these digital platforms have accelerated the diffusion of consumer-generated content, which significantly influences purchasing patterns. With the rise of smartphones and constant connectivity, consumers are now permanently embedded in digital networks [19], producing reviews and feedback that serve as valuable inputs for prospective buyers [20] [21].
As noted by [22], social media now takes centre stage in digital marketing environments, enhancing brand visibility and enabling real-time engagement. Studies such as [23] suggest that unlike traditional promotional tools, social media offers cost-effective and efficient methods for small and medium enterprises (SMEs) to reach their target audiences. The relevance of SMM is reflected in corporate strategies, reporting that by 2012, 73% of Fortune 500 companies had a Twitter presence, and 66% were active on Facebook. This increasing adoption highlights the recognition of social media as a vital marketing channel, especially in enhancing brand perception and shaping consumer expectations.
In summary, the literature consistently supports the idea that social media marketing significantly impacts consumer behavior, particularly in shaping perceptions, influencing attitudes, and supporting decision-making across all stages—from need recognition to post-purchase feedback. Given the widespread use of platforms like Facebook and Instagram in Yemen, understanding their influence provides both theoretical and practical insights into online consumer behavior and modern marketing strategy.
3. Research Methodology
The selection of an appropriate research methodology is critical to ensuring the validity, reliability, and overall credibility of the findings. In the context of this study—“The Impact of Social Media Marketing on Consumer Behavior in Yemen: A Study of Facebook and Instagram across the Decision-Making Process”—a mixed-methods approach has been adopted to comprehensively investigate how social media platforms influence consumer behavior. This methodological framework enables the integration of both quantitative and qualitative data sources, thereby providing a richer and more nuanced understanding of the complex relationship between social media marketing and consumer decision-making.
3.1. Research Design
The study adopts a descriptive exploratory research design to examine the influence of social media marketing—specifically Facebook and Instagram—on consumer behavior in Yemen. The descriptive component allows for a systematic analysis of behavioral trends across the five stages of consumer decision-making: need recognition, information search, evaluation of alternatives, purchase decision, and post-purchase behavior. This provides a factual representation of consumer behavior without manipulating variables. The exploratory component is used to address the limited understanding of Yemen’s digital consumer landscape. While exploratory research often relies on qualitative methods, this study employs a quantitative approach to enhance generalizability. Quantitative methods enable the collection and statistical analysis of numerical data, ensuring objectivity, replicability, and the ability to identify patterns and correlations across demographic groups. By combining a descriptive exploratory design with quantitative analysis, the research establishes a comprehensive framework for investigating how social media impacts consumer engagement in Yemen’s emerging digital market. This design is particularly suited to contexts where empirical data is scarce but growing in importance.
3.2. Population and Sampling
The population for this study consists of Yemeni social media users in urban areas such as Sana’a, Aden, and other digitally active cities. They differ in demographics but share active engagement on Facebook and Instagram. The sample is a subset of this population, with 250 individuals selected to ensure statistical reliability and feasibility. This sample allows the study to analyze consumer behavior patterns and generalize findings to the broader population of Yemeni social media users.
3.3. Sampling Technique
The study used a non-probability voluntary response sampling technique due to practical constraints in reaching the population. Participants were recruited via invitations on email, Facebook, and Instagram, targeting active Yemeni social media users. To reduce bias, recruitment was spread across different demographic groups and regions, ensuring diverse representation despite the lack of true random sampling.
3.4. Data Collection
The study uses both primary and secondary data. Primary data will be collected through a structured questionnaire of 40 closed-ended questions covering demographics and the influence of Facebook and Instagram on consumer decision-making. Participants must have an active social media account, and the survey will be distributed electronically via Facebook, Instagram, WhatsApp, and email. Secondary data will come from academic sources such as journal articles, books, databases, and reports to support the theoretical framework.
3.5. Data Analysis
The quantitative data collected from the questionnaire will be processed and analyzed using descriptive statistical techniques. The responses will be tabulated and visually presented through charts, graphs, and tables, utilizing Microsoft Excel for data visualization and interpretation. The analysis will focus on identifying patterns, frequencies, and distributions that illustrate how social media influences consumer behavior in various stages of the decision-making process.
3.6. Measurement Scales
Table 1 shows ensure clarity and consistency, all questionnaire items were measured using standardized scales. Constructs relating to consumer attitudes and perceptions (e.g., the influence of social media marketing on need recognition, information search, evaluation of alternatives, purchase decision, and post-purchase behavior) were assessed using a five-point Likert scale anchored from 1 = Strongly Disagree to 5 = Strongly Agree. This scaling allowed respondents to indicate varying levels of agreement with each statement. Behavioral frequency items, such as intensity of social media usage for purchasing, were coded using a five-point ordinal frequency scale: 1 = Rarely, 2 = Sometimes, 3 = Often, 4 = Very Often, and 5 = Always. This structured categorization enabled quantitative assessment of online shopping patterns. Demographic variables were grouped into ordered categories for comparability. For instance, age was classified as 16 - 21, 22 - 27, 28 - 33, and 34+; monthly income was coded into <$200, $200 - $250, $300 - $450, ≥$500; and education level was recorded as High School, Bachelor’s (ongoing/held), and Master’s/Postgraduate.
Table 1. Socio-demographic characteristics of the study sample (N = 250).
Characteristic |
Category |
Frequency (n) |
Percentage (%) |
Gender |
Female |
240 |
80% |
Male |
60 |
20% |
Age Group (years) |
16 - 21 |
75 |
25% |
22 - 27 |
129 |
43% |
28 - 33 |
66 |
22% |
34 and above |
30 |
10% |
Geographic Residence |
Southern Districts |
141 |
47% |
Eastern Districts |
108 |
36% |
Northern Districts |
30 |
10% |
Western Districts |
21 |
7% |
Marital Status |
Single |
180 |
60% |
Engaged |
62 |
21% |
Married |
58 |
19% |
Educational Level |
High School |
78 |
26% |
Bachelor’s (ongoing/held) |
177 |
59% |
Master’s (ongoing/held) |
45 |
15% |
Employment Status |
Employed |
219 |
60% |
Unemployed |
81 |
40% |
Monthly Income (dollars) |
<200 |
24 |
8% |
200 - 250 |
33 |
11% |
300 - 450 |
90 |
30% |
≥500 |
153 |
51% |
Note: Categories were coded as follows-Age (1 = 16 - 21, 2 = 22 - 27, 3 = 28 - 33, 4 = 34+); Income (1 = <$200, 2 = $200 - $250, 3 = $300 - $450, 4 = ≥$500); Education (1 = High School, 2 = Bachelor’s, 3 = Master’s/Postgraduate).
4. Data Analysis and Interpretation
4.1. Consumer Patterns of Social Media Usage for Online
Purchasing in Yemen
The charts titled “Consumer Patterns of Social Media Usage for Online Purchasing in Yemen” provide a comprehensive overview of how social media platforms, particularly Facebook and Instagram, are utilized by consumers in Yemen for online shopping activities. The data reveals that an overwhelming majority of respondents (93%) engage in online purchasing through social media, highlighting the platform’s dominant role in modern consumer behavior. Only a small portion (7%) indicated that they do not use social media for purchasing, suggesting that online shopping via these platforms is now a mainstream practice within the population studied (see Figure 1).
Figure 1. Use of social media for online purchasing.
Among the platforms, Instagram emerges as the most preferred for online purchases, accounting for 48% of the responses. Facebook follows with 29%, while 20% of respondents use both platforms. A minimal 3% rely on other platforms, indicating a concentrated preference for these two dominant networks (see Figure 2). This preference likely stems from Instagram’s visual appeal and interactive nature, making it particularly effective for showcasing products and influencing buying decisions. In terms of user experience, the data shows that most consumers have a long-standing history with social media purchasing. Specifically, 56% have been engaged in such activities for 25 to 36 months or longer, demonstrating sustained usage and familiarity with online buying processes. A smaller but still significant portion has been involved for 19 - 24 months (22%), 7 - 18 months (13%), and 1 - 6 months (9%) (see Figure 3). This indicates a well-established behavior among consumers, with most having more than a year of experience in social media purchasing. The frequency of online purchases also reflects strong engagement. Nearly half of the respondents (48%) reported making purchases every month, while 27% shop weekly and 23% shop daily. Only 2% fall into the “other” category, suggesting irregular purchase patterns (see Figure 4). These figures underscore the high level of integration of social media into consumers’ shopping routines, with many making purchases regularly.
![]()
Figure 2. Preferred social media platform for purchasing.
Figure 3. Duration of using social media for purchasing.
Figure 4. Frequency of online purchasing.
Overall, the charts illustrate a digitally engaged consumer base in Yemen that actively uses social media platforms—particularly Instagram and Facebook—not just for social interaction, but as a primary channel for discovering, evaluating, and purchasing products. The data suggests high platform preference, long-term use, and frequent purchasing behavior, emphasizing the central role that social media marketing plays in shaping consumer behavior across different stages of the decision-making process.
4.2. Consumer Motivations and Perceptions of Social Media
Influence on Purchasing Behavior
The charts titled “Consumer Motivations and Perceptions of Social Media’s Role in Online Purchasing Decisions” provide a detailed insight into how consumers perceive the influence of social media platforms—especially Facebook and Instagram—on their online purchasing behavior. The data highlights several motivational and perceptual dimensions related to social commerce. One of the main findings is that the ease of search is the leading reason consumers use social media for shopping, accounting for 42% of responses. This is followed by motivations related to personal needs or wants (27%), affordability of prices (19%), and perceived product quality (12%). These results suggest that consumers are primarily drawn to the convenience, relevance, and value offered by social media platforms (see Figure 5).
In terms of awareness, the majority of respondents acknowledge that social media plays a significant role in increasing their awareness of needs, with 45% stating it “always” increases their awareness and another 35% saying it does so “most times.” Only 20% believe it “sometimes” increases awareness, and none stated that it never does, indicating that social media is nearly universally perceived as a tool that stimulates recognition of consumer needs and desires. This dynamic reinforces the role of social media as a proactive channel that shapes rather than merely responds to consumer intent (see Figure 6).
Figure 5. Reason for using social media for purchasing.
Figure 6. Does social media increase need awareness?
Regarding the alignment of social media content with evolving consumer preferences, 50% of consumers agree and 32% strongly agree that what they see on social media reflects their changing needs and wants. This suggests that most users perceive a strong consistency between their evolving expectations and the content promoted on these platforms. Only a small portion disagreed (16%) or strongly disagreed (2%), indicating minimal dissatisfaction with the adaptability of social media marketing (see Figure 7).
Figure 7. Consistency with changing needs/wants.
Finally, when it comes to motivational influence, social media appears to be highly persuasive. A majority of consumers (56%) stated that it always motivates them to make purchases, while 28% reported it often does. Only a small fraction (14%) are sometimes influenced, and just 2% said they are rarely motivated by social media. These results affirm the idea that social media functions not only as a space for product discovery but also as a strong trigger for purchase intention. Overall, the charts reflect that consumers view social media as an essential and highly effective component of their shopping journey, influencing everything from the recognition of needs to the final act of purchasing (see Figure 8).
Figure 8. Does social media motivate you to purchase?
4.3. Consumer Feedback and Post-Purchase Behavior Influenced by Social Media Marketing
The charts titled “Consumer Feedback and Post-Purchase Behavior Influenced by Social Media Marketing” present a cohesive picture of how social media impacts consumer decisions after making a purchase. It emphasizes the importance of online reviews and user feedback in shaping not only the consumer’s initial decision but also their post-purchase perceptions and future behavior. The data reveals that a substantial proportion of consumers—specifically 38%—have made purchases based on social media reviews more than eleven times, while 32% have done so between seven and ten times, indicating that many consumers consistently rely on reviews as part of their buying routine. An additional 19% have been influenced by reviews three to six times, and 11% one to two times, underscoring that reviews are a powerful and recurring influence in digital consumer behavior (see Figure 9). Furthermore, 48% of respondents strongly agree and 38% agree that feedback found on social media affects their purchase decisions, suggesting that peer opinions and user experiences have a higher level of influence than traditional advertising (see Figure 10). In this context, 81% of consumers also acknowledged that online reviews can shift their attitude toward a product or brand, demonstrating how social proof can alter brand perception (see Figure 11). However, there is a noticeable disparity between product presentation and actual consumer experience; only 12% of consumers stated that product features always match advertisements, while a majority reported only occasional or rare consistency. Specifically, 36% said the match occurs sometimes, 27% said rarely, and 25% said most times (see Figure 12). This indicates a gap in expectation versus reality, which could erode consumer trust over time. Lastly, the responsiveness of brands to post-purchase feedback appears limited, with 53% of consumers believing that their opinions are not taken into consideration, while 47% felt that brands do acknowledge feedback. Collectively, this data illustrates the dominant role of social media in influencing consumer attitudes and satisfaction, the strong value placed on peer feedback, and the critical need for brands to not only deliver on their promises but also to actively engage with consumer input in order to foster loyalty and improve post-purchase experiences.
![]()
Figure 9. How often do you purchase based on reviews?
Figure 10. Does feedback on social media affect purchase decision?
Figure 11. Do reviews change your attitude toward a product/brand?
Figure 12. Do product properties match advertisements?
4.4. Consumer Perceptions of Social Media Influence on
Trend-Based Purchasing, Product Evaluation, and
Decision-Making Stages
The charts present a detailed overview of how consumers perceive the influence of social media on their purchasing behavior, particularly concerning trend-based decisions, product evaluations, demographic targeting, and its role across different stages of the consumer decision-making process. A significant finding is that 45% of consumers reported making purchases based on trends one to two times, followed by 29% who did so three to six times, 16% seven to ten times, and 10% more than eleven times (see Figure 13). This indicates that while trend-driven behavior exists, it is more occasional than habitual for most users. Regarding the role of platform-generated metrics such as likes, shares, and views, only 46% of respondents found them helpful in evaluating products, while a slightly larger portion—54%—did not (see Figure 14). This suggests that although social validation indicators are visible, they may not always translate into meaningful influence on consumer judgment. On the other hand, the presence of user queries and comments on brand pages proved far more impactful, with 95% of consumers confirming that such interactions help them evaluate products, and only 5% disagreeing (see Figure 15). This highlights the value of authentic, peer-generated content in shaping consumer trust and decision-making. In terms of forming perceptions before purchasing, 31% of respondents strongly agreed and 53% agreed that social media gives them a positive prejudgment toward a product, while only a small portion expressed disagreement (13% disagreed and 3% strongly disagreed) (see Figure 16). This demonstrates social media’s powerful role in setting favorable expectations before actual consumption.
![]()
Figure 13. How often do you purchase based on trends?
Figure 14. Do ratings (likes-shares-views) help evaluate products?
Figure 15. Do queries/comments on brand pages help evaluation?
Figure 16. Does social media give a positive prejudgment before consumption?
Conversely, when it comes to negative prejudgment, only 4% strongly agreed and 10% agreed that social media fosters negative expectations, while 42% disagreed and 44% strongly disagreed (see Figure 17). This implies that consumers generally perceive social media as a more positive and encouraging influence rather than a discouraging one. Additionally, 86% of the participants believe that social media content is targeted at specific demographic segments, reinforcing the platform’s use of tailored algorithms and audience segmentation to enhance marketing effectiveness (see Figure 18).
Figure 17. Does social media give a negative prejudgment?
Figure 18. Does social media target a particular demographic?
When asked to rate social media as an effective marketing tool, the responses formed a descending gradient: 36% rated it a 10, 24% a 9, 14% an 8, and smaller percentages rated it between 7 and 4 (see Figure 19). These high scores collectively illustrate a strong endorsement of social media’s effectiveness in contemporary marketing. Finally, when evaluating which stage of the decision-making process is most influenced by social media, 36% chose the information search stage, followed by 24% who selected need awareness, 19% for evaluation of alternatives, 15% for purchase decision, and 6% for post-purchase (see Figure 20). This ranking confirms that social media plays its most crucial role early in the consumer journey, particularly by helping consumers become aware of products and gather relevant information that shapes their purchasing trajectory. In summary, the charts demonstrate that while social media-driven trends may only modestly impact purchasing frequency, the platform’s true strength lies in facilitating product evaluation through peer interaction, establishing positive expectations, targeting specific user segments, and supporting early-stage decision-making with credible and accessible information.
![]()
Figure 19. Rate social media as an effective marketing tool.
Figure 20. Most impacted stage of decision-making.
5. Drief Inferential Analyses (Chi-Square Tests or Logistic Regression)
Table 2. Inferential results of demographic characteristics and consumer behavior outcomes (N = 250).
Test Type |
Dependent Variable (Outcome) |
Independent Variable (Demographic) |
χ2/OR |
df |
p-value |
Interpretation |
chi-square |
Information Search |
Age Group |
χ2 = 11.42 |
3 |
0.009** |
Significant: younger groups rely more on social media for product info |
chi-square |
Purchase Decision |
Education Level |
χ2 = 9.63 |
2 |
0.022* |
Significant: higher education increases reliance on social media ads |
chi-square |
Post-purchase Satisfaction |
Gender |
χ2 = 0.84 |
1 |
0.360 |
Not significant |
Logistic Regression |
Purchase Decision (Yes/No) |
Monthly Income |
OR = 1.72 |
– |
0.004** |
Higher income increases likelihood of social
media-influenced purchases |
Logistic Regression |
Post-purchase Behavior |
Education Level |
OR = 1.45 |
– |
0.031* |
Education positively predicts
post-purchase engagement |
Logistic Regression |
Purchase Decision |
Gender |
OR = 0.91 |
– |
0.420 |
Not significant |
Note: ***p < 0.001, **p < 0.01, *p < 0.05.
As shown in Table 2, several demographic characteristics were found to significantly influence consumer behavior outcomes in response to social media marketing. Age was significantly associated with information search behavior (χ2 = 11.42, p < 0.01), with younger respondents—particularly those aged 22 - 27—demonstrating greater reliance on social media platforms to evaluate products. Education level also exhibited a significant effect on purchase decision-making (χ2 = 9.63, p < 0.05), indicating that respondents with bachelor’s and master’s degrees were more likely to consider social media advertisements when making purchases. By contrast, gender was not significantly related to post-purchase satisfaction (χ2 = 0.84, p = 0.36).
Logistic regression analyses further supported these findings. Monthly income emerged as a strong predictor of purchase decisions, with higher-income respondents being 1.72 times more likely to have their purchases influenced by social media marketing (p < 0.01). Education level also positively predicted post-purchase behavior (OR = 1.45, p < 0.05), suggesting that higher educational attainment enhances the likelihood of engaging with brands after purchase. Gender and marital status, however, did not yield significant predictive effects. Overall, these results highlight the importance of age, education, and income in shaping consumer responses to social media marketing strategies in Yemen.
6. Conclusion
This study confirms that Facebook and Instagram significantly influence consumer behavior in Yemen across all decision-making stages—from product discovery and information search to evaluation, purchase, and post-purchase engagement. Instagram’s visual appeal and Facebook’s wide reach make them particularly effective, with peer reviews, user-generated content, and interactivity exerting stronger influence than traditional advertising. The information search stage was identified as the most impacted, highlighting social media’s crucial role in shaping early consumer choices. Post-purchase behaviors reveal both opportunities and challenges: while consumers rely on reviews and share experiences, gaps remain between brand promises and consumer experiences, particularly regarding responsiveness to feedback. The study acknowledges limitations, including urban bias, overrepresentation of female respondents, reliance on voluntary participation, and self-reported behavior, which restrict generalizability. Future research should adopt more representative samples, integrate behavioral tracking, and ensure demographic balance to strengthen validity. Overall, social media has become a central ecosystem for Yemeni consumers, requiring marketers and policymakers to prioritize targeted engagement, responsiveness, transparency, and trust-building to sustain consumer confidence and loyalty.
Conflicts of Interest
The authors declare no conflicts of interest.