Knowledge and attitudes of Spanish citizens towards big data and artificial intelligence
Patricia Sánchez-Holgado, Carlos Arcila Calderón, David Blanco-Herrero
Knowledge and attitudes of Spanish citizens towards big data and artificial intelligence
ICONO 14, Revista de comunicación y tecnologías emergentes, vol. 20, no. 1, 2022
Asociación científica ICONO 14
Conocimiento y percepción de la ciudadanía española sobre el big data y la inteligencia artificial
Conhecimento e atitudes dos cidadãos espanhóis sobre big data e inteligência artificial
Postdoctoral researcher, Department of Sociology and Communication of the University of Salamanca, Salamanca, Spain
Carlos Arcila Calderón
Associate Professor, Department of Sociology and Communication of the University of Salamanca, Salamanca, Spain
Pre-doctoral researcher, Department of Sociology and Communication of the University of Salamanca, Salamanca, Spain
Received: 13/june /2022
Revised: 17/september /2022
Accepted: 17/october /2022
Published: 08/december /2022
Abstract: Communicating information about data science is one of the greatest challenges facing society today, specifically big data and artificial intelligence, as the technology is constantly changing, and citizens need to understand it in order to make the best decisions. Traditionally, science communication has focused on the study of social attitudes and perceptions toward the most controversial topics, an example being the Survey of Social Perception of Science and Technology in Spain, which is carried out every two years by the Spanish Foundation for Science and Technology (hereinafter FECYT). This research is the first study carried out in Spain that focuses on the public's knowledge, understanding, and perception of big data and artificial intelligence. A national survey was conducted with a sample of 684 people. The results indicate that knowledge regarding big data and artificial intelligence is moderate, with a lower level of knowledge and interest among older people, and that artificial intelligence is more well-known and of more interest than big data. The way in which the public obtains information has not changed with regard to traditional surveys, so information can reach the public through these channels.
Keywords: Data science; big data; artificial intelligence; scientific communication; social perception of science; information.
Resumen: La comunicación de la ciencia de datos es uno de los mayores retos de la sociedad actual, y específicamente el big data y la inteligencia artificial, porque la tecnología está en constante cambio y la ciudadanía necesita comprenderla para tomar las mejores decisiones. Tradicionalmente la comunicación científica se ha centrado en el estudio de las actitudes y percepciones sociales hacia los temas más polémicos, como en el entorno español, con la Encuesta de Percepción Social de la Ciencia y la Tecnología, elaborada por la FECYT cada dos años. Este trabajo es el primer estudio realizado en España centrado en el conocimiento, la comprensión y la percepción del público hacia el big data y la inteligencia artificial. Se ha llevado a cabo una encuesta nacional a una muestra de 684 personas. Se observa que el conocimiento sobre el big data y la inteligencia artificial es moderado, con un menor grado de conocimiento e interés entre las personas de más edad y que la inteligencia artificial es más conocida y despierta mayor interés que el big data. La manera en que el público se informa no varía con respecto a las encuestas tradicionales, por lo que la información puede llegarles por esas vías.
Palabras clave: Ciencia de datos; big data; inteligencia artificial; comunicación científica; percepción social de la ciencia; información.
Resumo: A comunicação da ciência de dados é um dos maiores desafios da sociedade atual, e especificamente big data e inteligência artificial, pois a tecnologia está em constante mudança e os cidadãos precisam entendê-la para tomar as melhores decisões. Tradicionalmente, a comunicação científica tem se concentrado no estudo das atitudes e percepções sociais em relação aos temas mais controversos, como no ambiente espanhol, com a Pesquisa de Percepção Social da Ciência e Tecnologia, elaborada pela FECYT a cada dois anos. Este trabalho é o primeiro estudo realizado na Espanha focado no conhecimento, compreensão e percepção do público em relação ao big data e à inteligência artificial. Uma pesquisa nacional foi realizada em uma amostra de 684 pessoas. Observa-se que o conhecimento sobre big data e Inteligência Artificial é moderado, com menor grau de conhecimento e interesse entre os mais velhos e que a Inteligência Artificial é mais conhecida e desperta maior interesse e do que big data. A forma como o público é informado não varia em relação às pesquisas tradicionais, de modo que a informação pode chegar até eles por meio desses canais.
Palavras-chave: Ciência de dados; big data; inteligência artificial; comunicação científica; percepção social da ciência; informação.
Data is transforming our world, offering a new value to the market regarding increased efficiency and new opportunities for innovation. Society is constantly generating data. Thus, data is changing society, ranging from the way we shop, communicate, and get together, to the clothes we wear and the devices we use. Technology such as artificial intelligence are expected to have an impact on industry, productivity, labour, the environment, equality, and inclusion, among other areas (Vinuesa et al., 2020). Consequently, as the European Commission advances, citizens must be empowered to make better decisions based on knowledge from non-personalised data (European Commission, 2020a). This implies having adequate public understanding of the issue, which is seen as the evolution of a paradigm that has transformed the image of scientific concepts and processes needed for personal decision-making into matters that are essential in fostering economic productivity (Ballesteros-Ballesteros & Gallego-Torres, 2022).
However, there is a perceived gap between data-related knowledge (collection, use, and interpretation) and people's understanding of such knowledge, as society is unaware and lacks education on the basics of data science (Tomás et al., 2021). Moreover, in recent years certain negative attitudes toward advances in this discipline have been growing (Vodafone Institute for Society and Communications, 2016). Although terms related to data science are becoming part of the everyday language of citizens, there are still some difficulties in understanding concepts such as algorithm and machine learning, and there are growing concerns about the ethical aspects of these technologies (Luz Clara & Malbernat, 2021). From the perspective of the discipline's evolution, recognition of the challenges, opportunities and value of data is changing scientific fields that were already data-driven, but also others such as social science and business (Bakhshi et al., 2014; Khan et al., 2014; Labrinidis & Jagadish, 2012). Consequently, researchers and scientists play a key role in fostering the data science agenda (Cao, 2017).
In this context, in order to more appropriately guide communication on data science, big data and artificial intelligence, there are three fundamental parameters based on society's relationship with science that can help in this effort: social perception and attitudes toward science and technology; the scientific-technical culture of a society; and citizens' interest in science. Without social communication regarding science, it is difficult to achieve scientific literacy. One of the most relevant paradigms of scientific literacy is that of public engagement with science (Godin & Gingras, 2000), where two types of components are mixed: the level of scientific literacy of the public, and the positive or negative attitude toward science. With regard to the latter, studies on society's perception and knowledge of science and technology provide highly relevant information for identifying the impact of these disciplines, but also the degree of permeability of knowledge to society (Miller, 2012; Miller & Laspra Pérez, 2018).
Indicators of interest, perception, and information consumption are relevant in examining social perceptions of science and technology (Garretón Merino et al., 2018). However, there are several previous studies that focus on analysing citizen knowledge about data science. Some progress has been made on the issue in Arcila-Calderón et al. (2021), reflecting high public interest but low familiarity with the topic, and in Sánchez-Holgado et al. (2021), who show that there is a higher perception of benefits than risks, and that men perceive more benefits than women. In order to develop appropriate strategies to communicate the characteristics, benefits, and risks of big data and artificial intelligence, it is first necessary to address society's level of knowledge on the subject.
Therefore, the aim of this work is to survey the public's knowledge and attitude toward big data and artificial intelligence. Moreover, this is the first study of its kind to be carried out in Spain. This research is intended to fill the gap in studies on the social perception of science, with approaches that focus on big data and artificial intelligence. The work herein provides a more precise understanding of the current state of scientific culture regarding big data and artificial intelligence. Moreover, it examines whether there is societal interest in these technologies for individual use, as well as the perception people have of them, which will help in the development of future data science communication strategies to improve scientific culture and literacy.
1.1. Studies on the social perception of science
Science communication has been widely analysed, generally in national contexts, with a strong focus on public understanding and citizen attitudes (Ballesteros-Ballesteros & Gallego-Torres, 2022; Knight, 2006; Miller, 2004; Miller, 2001). There has also been academic research on the accuracy and complexity of science news (Mangione, 2021), the way this information is presented in the media (Moreno Castro, 2010; Summ & Volpers, 2016), and the role of specialised journalists (Sánchez-Holgado, Arcila-Calderón, et al., 2021).
Studies on the social perception of science aim to identify public support for science and technological policies. Moreover, such research is used as a tool for assessing citizens' scientific literacy. These issues have been approached from the following diverse perspectives: from the cognitive deficit model, which establishes a correlation between the level of knowledge and the social perception of science and technology; from an analytical perspective, which looks at the spaces of interaction between science and society; from the perspective of context, which states that it is necessary to perform an in-depth analysis of the concerns that arise and to separate knowledge and use, risk and ethics, as well as the technical and the cultural; and finally, from perspectives that are sociological, political, and institutional (Eizaguirre, 2009). On the other hand, these studies have been based on the assumption that the level of education, interest in scientific and technological issues, and the scientific information that individuals possess can explain their attitude toward science, which will be more positive as the values of these factors increase. However, the relationship between attitude, knowledge and interest is being reviewed, as high levels of understanding of science do not always correlate with a high degree of support for science: in post-industrialised countries, the higher the level of scientific literacy, the more aware society is of the benefits and risks, leading to more critical attitudes (Bauer, 2008).
Such studies address many questions, including the possibilities of increasing mutual understanding between social actors (both public and private sectors, research, media, education, and the scientific community), which can improve science dissemination and knowledge transfer, and in turn connect academia, science policy makers, and science communicators (Felt, 2007).
At the European level, the Eurobarometer regularly publishes perception surveys on different aspects of science. One of the most relevant reports is entitled, "Public perceptions of science, research and innovation in Europe", which suggests that the majority of Europeans believe that science and technological innovation will have a positive impact in tackling most of the problems that will face society in the next 15 years, with the key exception of reducing inequality (European Commission, 2014). More recently, in another report entitled, "Attitudes toward the impact of digitisation and automation on everyday life" (European Commission, 2017a), this idea is supported, as most respondents are positive about the effect that the latest digital technologies have had on society, the economy, and their quality of life, while ins the report known as "European attitudes toward cyber security" (European Commission, 2017b), citizen concern about privacy and online security is revealed as one of the biggest risks of new technology, but also confirms that the more knowledge citizens have about the issue, the better they can adapt their behaviour.
In Spain, the Spanish Foundation for Science and Technology (FECYT) is in charge of the main science indicators, as well as the most relevant study conducted, “Social perception of science and technology in Spain”, which provides relevant data (FECYT, 2020). Firstly, spontaneous interest refers to those who say they are mainly interested in science and technology topics, as opposed to other topics. Moreover, this figure has been growing in recent years, reaching its highest point in 2018 at 16.3%, although it dropped in 2020 to 14.2%, due to the advent of Covid-19 as the overshadowing topic. This decrease partly resulted from a widening gender gap as well: while interest among men retained similar figures (18.8% in 2020 vs. 18.9% in 2018), the number fell considerably among women (9.9% in 2020 vs. 13.9% in 2018), bringing the gap to 8.9%. In terms of age, the study indicates that interest decreases with advancing age: the highest level of interest in science and technology is found among 15-24 year olds (21.9%), whereas the lowest level resides among those aged 65 and over (9.2%).
Secondly, in line with European studies, 45.9% of the respondents consider that the benefits of science and technology outweigh their disadvantages. The areas where the greatest benefits are perceived are in fighting disease (62.9%), improving quality of life (52.3%), creating jobs (39%), protecting the environment (35.9%), increasing individual freedom (29%) and protecting personal data and privacy (25.6%). With the exception of the first two topics, the rest experienced a decrease in the perception of benefits in the 2020 survey. Finally, 22.3% respond that they feel quite or very informed about science and technology, while 38.8% consider themselves somewhat informed, and 38.7% believe they are only slightly informed, or not informed at all. The perception of being informed decreased when compared to 2018, when the highest value was obtained up until that year. These types of analyses and studies are common, and they are promoted by public institutions as well, yet research regarding the knowledge and perception of data science is just beginning in Spain, which is why it is one of the challenges of this study.
1.2. Public understanding of data science, big data, and artificial intelligence
Data science is a contemporary discipline that combines computer science with mathematical approaches in order to gain relevant knowledge about data. It partly includes big data and artificial intelligence techniques. Not only is the concept concise for the purpose of unifying statistics, data analysis, and methodologies, but it also involves their results, or in other words, it includes three distinct phases: data design, data collection, and data analysis (Hayashi, 1998). The lack of a formal definition of data science, big data and artificial intelligence has led to some efforts to establish the foundations of these concepts, such as those of Cao, 2017; Samoili et al., 2020; Ward & Barker, 2013.
Artificial intelligence is a combination of technologies that bring together data, algorithms and computing capabilities (European Commission, 2020b). In Spain, artificial intelligence strategy is defined as “the science and engineering that allows the design and programming of machines capable of performing tasks that require intelligence” (Ministerio de Ciencia Innovación y Universidades, 2019). On the other hand, the benchmark definition of big data is still that of Gartner (2001): “Big data is data containing greater variety, which increases in volume and at a higher velocity” (the 3 Vs: Volume, Velocity and Variety). This was later enhanced to include a fourth V (Veracity), which considers the factors of trust and uncertainty with regard to the data itself, as well as the outcome of the data analysis (Beyer et al., 2012; Chang & Grady, 2019; IBM, 2013). Thus, big data is a field that deals with massive amounts of complex data, which are included in the field of data science, while artificial intelligence is a more traditional discipline that studies how machines can solve complex problems by reasoning and choosing the best options. Although the latter is a broader discipline, it is partly included in the field of data science. In short, data science is an interdisciplinary field that includes other areas of study. As such, we can see how these fields are interrelated in the Venn diagram in Figure 1, which allows us to place it in the correct position (Rohlman, 2019).
Communicating data science is important. Moreover, governments, industry, academia and even private institutions are increasingly making efforts to use data for decision making, as well as to promote research and development in data science and analytics (Cao, 2017). Nevertheless, it is still partly overshadowed by a lack of understanding; Eurobarometer data regarding the impact of digitisation and automation on everyday life has revealed that attitudes towards robots and artificial intelligence are related to prior knowledge on the subject (European Commission, 2017a). Constant changes, scepticism about discoveries that could eliminate jobs or put our privacy at risk, and the enormous impact that these technologies can have on each person's life present new challenges that must be addressed by studying the nature of communication today, and what it could be like in the future.
In fact, people have a growing interest in the impact of data science on their daily lives, mainly due to the enormous media attention given to topics related to big data and artificial intelligence. An example of the significance of these technologies is the case of the Panama Papers, in which data science techniques were used to reveal global tax fraud (Woodie, 2016). This interest results in hopeful attitudes, but also in serious concerns, such as the use of personal data, machine-driven decision-making processes, the disappearance of jobs (Anderson et al., 2018), ethical considerations, and algorithmic bias (Cotino Hueso, 2019; Luz Clara & Malbernat, 2021). Together with these concerns, there is also a lack of public understanding about technologies that are still relatively new and constantly changing, which are linked to our daily lives, work, education, finances, and leisure. The communication of data science, and specifically big data and artificial intelligence, is taking place in a context in which the media are not the only groups in charge of transmitting such information, as those who produce this scientific knowledge can transmit it directly to society through digital channels in which new voices and dissemination outlets have made it more necessary than ever to guarantee the quality of these communications (Brondi et al., 2021; Taddicken & Krämer, 2021).
Thus, a cross-disciplinary field involving social science and computer science is starting to develop (Cao, 2017; Samoili et al., 2020), an example of which is the work of Lytras & Visvizi (2019), which identifies the social impact of big data by considering three factors: the intention to share personal data; individual concerns; and social impact. Based on these issues, the aim of this research is to explore how services based on big data influence individuals and societies, and the way in which peoples’ perception of data develops. An important contribution of this study revolves around the fact that users are innately motivated to protect their personal data and privacy, yet they also want to use services that make their lives easier.
With regard to Spain, another example of interest is a report by the Bankinter Foundation. Although it focuses exclusively on artificial intelligence, the report by this foundation, which involved the collaboration of experts from the Future Trends Forum, highlights the exponential growth and integration of big data and artificial intelligence into business decisions (Inteligencia Artificial, 2019).
Based on the above, before appropriate strategies can be developed to convey the characteristics, benefits, and risks of data science, it is necessary to address society's level of awareness of this discipline. Thus, the aim of this research is to survey the public's knowledge and perception of big data and artificial intelligence. The objective is to complement other studies, such as those mentioned above, by narrowing the existing gap regarding approaches that focus on data science, more specifically big data and artificial intelligence, which are two of its most relevant fields. To this end, the research questions to be answered are the following:
RQ1: What is the level of knowledge, familiarity, and interest in big data and artificial intelligence among Spanish citizens?
RQ2: What is the perception of big data and artificial intelligence by Spanish citizens?
RQ3: What are the channels through which Spanish citizens are informed about big data and artificial intelligence, and how do they perceive this information?
2.1. Sample and procedure
This research is based on a survey using closed-ended questions of a representative sample of Spanish citizens. The tool designed for this research is based on the Social Perception of Science and Technology questionnaire (FECYT, 2018; 2020), as well as on information from the Future Trends Forum report of the Bankinter Foundation (Artificial Intelligence, 2019), the European Commission's strategic reports on big data and artificial intelligence (European Commission, 2017a, 2020c), and the Planetic Interplatform Initiative in Spain (Sáez & Costa-Soria, 2019). After being validated by a team of 10 experts in the field through a pre-test/post-test to measure the stability of the responses, the distribution was subcontracted to the company Qualtrics, ensuring that the sample was adequate and representative, stratified by gender, age, and autonomous region. The sample size provides a confidence level of 95% and a margin of error of 5%. Responses were collected during the week of 20-24 January of 2020, with a sample of 684 people.
The measurements used to study the level of familiarity, interest, and the attitudes of citizens consisted of five-point Likert-type scales, with the exception of questions related to gender (men and women); age (five age brackets: 18-24, 25-34, 35-49, 50-64, and 65+); and knowledge of big data and artificial intelligence applications, which were measured with a dichotomous yes/no response regarding the main applications used in the household, such as virtual assistants, social networks, and audio/video streaming platforms. Thus, the measurement parameters of the study were as follows:
Familiarity with the concept of big data and artificial intelligence: between 1 (not at all familiar, I know nothing about it) and 5 (very familiar, I understand and use it). This measures the extent to which respondents are familiar with the subject.
Interest in big data and artificial intelligence, as well as knowledge and/or use: from 1 (not at all interested) to 5 (very interested).
Citizen perception of the concepts of big data and artificial intelligence was measured with a differential semantic scale of adjectives and perceived emotions.
The relationship to people's daily lives was measured on a scale from 1 (strongly disagree) to 5 (strongly agree) for the following items: the relationship to their behaviour as a consumer and user; to their understanding of the world; to relationships with other people; to their profession or work environment; and to shaping their political and social opinions.
About how the public is informed about big data and artificial intelligence, the following measures were used:
The type of media used for information, with a list of media to choose from, in which the frequency of reporting was measured from never (1) to always (5).
The amount of information received was measured on a scale of 1 to 5, from very little information (1) to a lot of information (5).
The quality of the information received was calculated using a scale of 1 to 5, from very poor (1) to very good (5).
Trust in the information received was measured on a scale of 1 to 5, from very distrustful (1) to very trusting (5).
Once the responses had been made anonymous and incorporated into the same data matrix, the analysis was carried out. Tests involving descriptive statistics were performed and inferential statistics tests were applied as follows: student's t-tests for paired samples to find out the different levels of knowledge and interest in big data and artificial intelligence; a one-factor ANOVA to investigate possible differences between age groups; and bivariate correlations to test the relationship between the different measures used to study the knowledge of these technologies.
In response to the first research question (RQ1) regarding the knowledge and degree of familiarity with and interest in big data and artificial intelligence, we started by analysing the specific knowledge of applications that utilise or are based on these technologies, examining the four most commonly used applications in people's daily lives: social networks, audio and video platforms, and virtual assistants. The total average rate of awareness of these applications is 92.14%, which is very high. Therefore, it can be assumed that these applications are widely used in society. If reviewed individually, the awareness rates vary slightly, ranging from 88.30% (audio platforms) to 95.76% (social networks), which can be seen in Table 1.
With regard to citizen knowledge of big data and artificial intelligence, we can observe that the surveyed population is more familiar with artificial intelligence than big data [MAI3.43; SDAI=1.276; MBD=2.85; SDBD=1.445; t(683)=-12.414; p<0.001]. The same is true for the levels of interest, which are higher for artificial intelligence, as well as its knowledge and/or use, than for big data, its knowledge, and/or use [MAI=3.89; SDAI=1.209; MBD=3.64; SDBD=1.248; t(683)=-6.948, p<0.001]. These data also show that interest in these new technologies or techniques is higher than the level of familiarity, which demonstrates that these novel and complex topics are of interest and considerable importance, yet they are still not widely understood by the public. Nevertheless, there is a significant and relevant correlation between the level of familiarity and interest. Thus, all four variables correlate positively and significantly, with a stronger correlation found between pairs that measure the same aspect, such as interest in big data and interest in artificial intelligence, r(684)=0.710, p<0.001), or familiarity with big data and familiarity with artificial intelligence, r(684)=0.600, p<0.001), as compared to pairs that refer to the same subject, such as interest in big data and familiarity with big data, r(684)=0.455, p<0.001), or interest in artificial intelligence and familiarity with artificial Intelligence, r(684)=0.552, p<0.001). Thus, it is conceivable that the increased interest that currently exists in these subjects will eventually be reflected in an increase in knowledge about them.
Age also proved to be a determining factor in the variables studied in terms of familiarity with and interest in big data and artificial intelligence. The one-way ANOVA test showed significant differences by age group for three of the variables, which are familiarity with big data [F(4, 679)=4.951, p<0.01], familiarity with artificial intelligence [F(4, 679)=4.594, p<0.01] (Figure 2), and interest in big data [F(4, 679)=3.813, p<0.01], as well as trend differences in the case of interest in artificial intelligence [F(4, 679)=2.353, p=0.053] (Figure 3).
The post-hoc tests performed were based on Tukey, as the homogeneity of variance test assumed equality of variances, and they indicate that older people are significantly less familiar with and less interested in both artificial intelligence and big data. It is among adults (25-34 years and 35-49 years) where significantly higher values of interest and knowledge are found. It is worth noting that the 65+ age group showed lower values than the three adult age groups in terms of familiarity with the concept of big data, a situation that was repeated in terms of interest in this concept, as people in the oldest age group also showed less interest in big data than the three adult age groups. In terms of familiarity with artificial intelligence, the oldest age group showed significantly lower values than the adult age groups from 25-34 and 35-49. Finally, people aged 65 and over showed significantly less interest in artificial intelligence than adults aged 35-49. In general, the 65+ age group is below the rest of the age groups, although interestingly, in all four cases the group with the next lowest level of familiarity and interest is the 18-24 age group (Table 2).
To answer the second question (RQ2) regarding the perception of big data and artificial intelligence by Spanish people, we sought to reveal the areas of daily life people consider to be most affected by these technologies and analysed a series of perceived emotions toward big data and Artificial Intelligence. Citizens feel that the area of their daily lives that is most related to big data and artificial intelligence is their consumer and user behaviour (M=3.93 SD=1.15), followed by their work environment (M=3.77 SD=1.26), their understanding of the world (M=3.70 SD=1.19), the shaping of their political and social opinions (M=3.52 SD=1.25), and finally, their relationships with other people (M=3.44 SD=1.29).
Regarding emotions felt, these are shown in alphabetical order in Table 3 for the two concepts. Regarding both technologies, it can be observed that interested is the most repeated qualifier (MBD=3.56; SDBD=1.40; MAI=3.64; SDAI=1.48), followed by active (MBD=3.46; SDBD=1.44; MAI=3.40; SDAI=1.54). This may lead us to consider a connection with the previously mentioned relations to areas of everyday life, as consumer and user behaviour, along with working life, are the domains where these emotions are clearly perceived. In the big data column, the next highest values are the following: satisfied (MBD=3.44; SDBD=1.46), enthusiastic (MBD=3.33; SDBD=1.48), and excited (MBD=3.32; SDBD=1.49). In the artificial intelligence category, the next highest values have been found as follows: enthusiastic (MAI=3.40; SDAI=1.58), satisfied (MAI=3.40; SDAI=1.60), excited (MAI=3.38; SDAI=1.57), and inspired (MAI=3.33; SDAI=1.62). A large majority of the terms coincide, as they are very personal emotions, and responding to them as one unit can lead to understanding them as a whole.
As for the differences in emotional perception, we can see that only two are significant: affected, which is higher in big data [MBD=2.99; SDBD=1.52; MAI=2.69; SDAI=1.64; t(682)=5.250; p<0.001; d=0.19], and unsettled, which is higher in big data [MBD=2.60; SDBD=1.58; MAI=2.28; SDAI=1.68; t(682)=5.858; p<0.001; d=0.20]. In the rest of the cases, we can observe which of the two concepts stands out. In the case of big data, the emotions with higher values are the following: Active; Aggressive; Distressed; Frightened; Embarrassed; Focused; Guilty; Determined; Alert; Irritable; Fearful; Nervous; and Self-satisfied.
In the case of artificial intelligence, the emotions that have higher values are as follows: Enthusiastic; Strong; Hopeful; Excited; Inspired; and Interested.
Finally, to answer the third research question (RQ3) regarding the channels through which Spanish people are informed about big data and artificial intelligence and how they perceive this information, it has been observed that the Internet is the priority medium, attaining an average of 3.33 (SD=1.20), following the usual trend in surveys on the Social Perception of Science and Technology. This is followed by personal environment (M=2.79 SD=1.25), television (M=2.62 SD= 1.19), and the work environment (M=2.58 SD= 1.32), all of which indicate that social influence is a variable of interest when it comes to finding out about these technologies (Table 4).
In those cases where information is provided by the Internet, we also determined which tools and resources are most used for this purpose. We found that instant messaging applications such as WhatsApp was the first choice (M=3.20 SD=1.36), followed by YouTube videos (M=3.19 SD=1.19), social media (M=3.01 SD=1.32), and digital media (M=3.77 SD=1.19), among others, as shown in Table 5.
Reviewing the perception of information that reaches respondents from any medium, we find that the amount of information people feel they receive has a mean of 2.94 (SD=1.39), which is close to an acceptable amount received. At least 40.35% consider that little or very little information is received, compared to 28.51% who consider it to be an amount that is quite large or large. The quality of the information is perceived as good (M=3.37 SD=1.24), which is a positive datum overall, yet the bulk of the responses are in the neither good nor bad range. On the other hand, 20.47% consider it to be bad or very bad, compared to 15.64% who consider it to be good or very good. Looking at these data, the balance is more negative than positive, despite the average. In terms of trust in the information, the mean is 3.51 (SD=1.16), which is also an optimistic figure. We have observed that around 42.84% trust information quite a lot or a lot, compared to 15.64% who distrust it quite a lot or a lot. As such, in this case the balance is more positive than negative, and trust in information has been verified by citizen perception.
4. Discussion and conclusions
Overall, the level of knowledge and interest in data science by Spanish citizens, specifically big data and artificial intelligence, is relatively high. However, given the relevance of these technologies in contemporary society, improvements in this area are advisable. It has been noted that citizens are more familiar with and interested in artificial intelligence than big data, which could be a result of the presence of each of them in the media and popular discourse. Consequently, new conclusions could be drawn in this regard. It is worth noting that interest in these subjects is higher than the degree of familiarity, which concurs with the national FECYT studies. As such, more focus could be placed on the media and communicators, so that non-specialised citizens would have access to more information about these technologies, and their digital literacy would be enhanced. In this respect, greater interest is directly related to increased knowledge, which is why, as proposed by Garretón Merino et al. (2018), “Policies that seek to promote a scientific culture insist on the need to increase the population's interest”. Moreover, it has been found that a high percentage of respondents believe they receive little or very little information on data science, despite the fact that their trust in the information is high. Citizens who are better informed will become consumers of scientific and technological information, an argument addressed in the paradigm of literacy (Miller, 2012), public understanding (Ballesteros-Ballesteros & Gallego-Torres, 2022), scientific culture (Pardo, 2014), critical participation (López Cerezo, 2010), and the adoption of science and technology (Cámara et al., 2016).
In terms of age, it is not surprising that older people have less interest and knowledge about this issue, although it is somewhat noteworthy that the younger age group (18-24 years) has less interest and knowledge than those in the adult age categories, in contrast to what has been observed in previous studies in Spain ( FECYT, 2018; 2020). In the year 2020, we experienced a significant change in the consumption of information based on large volumes of data, and the application of artificial intelligence to the field of medicine was crucial in the fight against the pandemic, so the comparisons between before and after are interesting, as suggested by Sánchez-Holgado, Marcos-Ramos, et al., 2021). Some authors have already suggested that the Covid-19 crisis has increased pessimism and negativity (Balluerka et al., 2020), which may have an effect on the social perception of these technologies. The growing presence and spread of disinformation, especially in association with scientific and technological issues, generates mistrust and social alarm (Salaverría-Aliaga, 2021); however, in this study we have seen that trust in information is high, yet it would be a variable to consider in analysing future changes of interests. In fact, the situation caused by the health crisis has highlighted the importance of data science, among many other issues. Consequently, this research trend is expected to undergo continued growth along the lines mentioned above regarding knowledge (Sánchez-Holgado, Marcos-Ramos, et al., 2021), disseminators (Sánchez-Holgado, Arcila-Calderón, et al., 2021), and especially literacy (Sánchez-Holgado, 2022), so that a comprehensive view can be obtained, thereby allowing appropriate strategies to be developed for the purpose of improving the knowledge of data science in Spain.
Acknowledgements and Funding
This work is part of the project entitled, Percepción Social de la Inteligencia Artificial en España” [Social Perception of Artificial Intelligence in Spain], funded by FECYT, reference FCT-21-17146.
This article is part of the doctoral thesis entitled, Communication of Data Science in Spain, written within the framework of the doctoral programme, Education in the Knowledge Society, University of Salamanca, Spain.
The dataset supporting the results of this study is available on request from the contact author at firstname.lastname@example.org. It is not publicly available as it is still in use by the research team.
Anderson, Janna, Rainie, Lee, & Luchsinger, Alex (2018). Artificial Intelligence and the Future of Humans. https://pewrsr.ch/3FZh98y
Arcila-Calderón, Carlos, Sánchez-Holgado, Patricia, Igartua Perosanz, Juan José, Ortega-Mohedano, Félix, González de Garay Domínguez, Beatriz, Frutos Esteban, Francisco Javier, Marcos Ramos, María, Cheng Lee, Lifen, Jiménez-Amores, Francisco Javier, Blanco-Herrero, David, & Frías Vázquez, Maximiliano (2021). Ciencia de Datos en España. Universidad de Salamanca. España. https://bit.ly/3hr1Uv8
Bakhshi, Hasan, Mateos–Garcia, Juan, & Whitby, Andrew (2014). Model workers: How leading companies are recruiting and managing their data talent. Engineering and Technology, 48. https://doi.org/10.1049/et:20080701
Ballesteros-Ballesteros, Vladimir, & Gallego-Torres, Adriana Patricia (2022). De la alfabetización científica a la comprensión pública de la ciencia. Trilogía Ciencia Tecnología Sociedad, 14(26), 2145–4426. https://doi.org/10.22430/21457778.1855
Balluerka, Nekane, Gómez, Juana, Hidalgo, Mª Dolores, Gorostiaga, Arantxa, Espada, Jose Pedro, Padilla García, Jose Luis, & Santed, Miguel Ángel (2020). Las consecuencias psicológicas de la Covid-19 y el confinamiento. https://bit.ly/3fAi1pr
Bauer, Martin W. (2008). Survey research and the public understanding of science. In Handbook of Public Communication of Science and Technology (pp. 125–144). Routledge.
Beyer, M., Laney, D., & Gartner. (2012). The Importance of “Big Data”: A Definition. Gartner. https://gtnr.it/3DSwAwT
Brondi, Sonia, Pellegrini, Giuseppe, Guráň, Peter, Fero, Martin, & Rubin, Andrea (2021). Dimensions of trust in different forms of science communication: the role of information sources and channels used to acquire science knowledge. Journal of Science Communications, 20 (03)(A08). https://doi.org/10.22323/2.20030208
Cámara, María Montaña, Laspra, Belén, & López-Cerezo, Jose Antonio (2016). Apropiación social de la ciencia en España. In Percepción Social de la Ciencia y la Tecnología en España 2016 (Issue December, pp. 19–49). http://hdl.handle.net/10651/54871
Cao, Longbing (2017). Data Science: A Comprehensive Overview. ACM Comput. Surv., 50(3). https://doi.org/10.1145/3076253
Chang, Wo & Grady, Nancy (2019). NIST Big Data Interoperability Framework: Volume 1, Definitions. NIST Special Publication, 1, 1–53.
Cotino Hueso, Lorenzo (2019). Riesgos e impactos del big data, la inteligencia artificial y la robótica. Enfoques, modelos y principios de la respuesta del derecho. Revista General de Derecho Administrativo, 50. https://fra.europa.eu/en/publication/2018/big-data-discrimination
Eizaguirre, Andoni (2009). Los estudios sobre percepción social de la ciencia. Acciones e Investigaciones Sociales, 27(27), 23–53.
European Commission (2014). Special Eurobarometer 419: Public perceptions of science, research and innovation (Issue October). https://doi.org/10.2777/95599
European Commission (2017a). Special Eurobarometer 460: Attitudes towards the impact of digitisation and automation on daily life (Issue May). https://doi.org/10.2759/835661
European Commission (2017b). Special Eurobarometer 464a: Europeans’ attitudes towards cyber security Fieldwork (Issue June). https://doi.org/10.2838/009088
European Commission (2020a). Comunicación de la Comisión al Parlamento Europeo, al Consejo, al Comité Económico y Social Europeo y al Comité de las Regiones. Una Estrategia Europea de Datos (COM(2020) 66 final). https://bit.ly/3h7s7hX
European Commission (2020b). Libro Blanco sobre la inteligencia artificial: un enfoque europeo orientado a la excelencia y la confianza. https://bit.ly/3Wp5oy2
European Commission (2020c). On Artificial Intelligence - A European approach to excellence and trust. https://doi.org/10.1017/CBO9781107415324.004
FECYT (2018). Percepción Social de la Ciencia y la Tecnología en España 2018. Fundación Española para la Ciencia y la Tecnología. https://bit.ly/3G3Rfk6
FECYT (2020). Percepción social de la ciencia y la tecnología en España 2020. Fundación Española para la Ciencia y la Tecnología. https://bit.ly/3UrT2TX
Felt, Ulrike (2007). O.P.U.S - Optimizing Public Understanding of Science. Final report project. Department for Philosophy of Science and Social Studies of Science, University of Vienna. https://bit.ly/3A0e30t.
Garretón Merino, Manuel Antonio, Muñoz Van den Eynde, Ana, Arancibia Gutiérrez, Marcelo, Camacho González, Johanna, Roberts Molina, Raimundo & Polino, Carmelo (2018). Ciudadanía: Ciencia y Tecnología. Reflexiones sobre la percepción de la ciencia y la tecnología en Chile. https://bit.ly/2WXM37d
Gartner (2001). Definition of Big Data. https://www.gartner.com/en/information-technology/glossary/big-data
Godin, Benoit, & Gingras, Yves (2000). What is scientific and technological culture and how is it measured? A multidimensional model. Public Understanding of Science, 9(1), 43–58. https://doi.org/10.1088/0963-6625/9/1/303
Hayashi, Chikio (1998). What is Data Science ? Fundamental Concepts and a Heuristic Example. In C. Hayashi, K. Yajima, H.-H. Bock, N. Ohsumi, Y. Tanaka, & Y. Baba (Eds.), Data Science, Classification, and Related Methods (pp. 40–51). Springer Japan.
IBM (2013). IBM What is big data? - Bringing big data to the enterprise. https://ibm.co/3WN3gk0
Khan, Nawsher, Yaqoob, Ibrar, Hasahem Ibrahim Abaker, Targio, Inayat, Zakira, Ali, Waleed Kamaleldin, Alam, Muhammad, Shiraz, Muhammad, & Gani, Abdullah (2014). Big data: Survey, technologies, opportunities, and challenges. The Scientific World Journal, 18. https://doi.org/10.1155/2014/712826
Knight, David (2006). Public understanding of science: A history of communicating scientific ideas..Routledge https://doi.org/10.4324/9780203966426
Labrinidis, Alexandros, & Jagadish, H. V. (2012). Challenges and opportunities with big data. Proceedings of the VLDB Endowment, Vol. 5, Issue 12, 2032–2033. https://doi.org/10.14778/2367502.2367572
López Cerezo, Jose Antonio (2010). Participación ciudadana y cultura científica. Arbor, CLXXXI(715), 351–362. https://doi.org/10.3989/arbor.2005.i715.417
Luz Clara, Bibiana Beatriz, & Malbernat, Lucía Rosario (2021). Riesgos, dilemas éticos y buenas prácticas en inteligencia artificial. XXIII Workshop de Investigadores En Ciencias de La Computación (WICC 2021, Chilecito, La Rioja), 155–159.
Lytras, Miltiades D., & Visvizi, Anna (2019). Big data and their social impact: Preliminary study. Sustainability, 11(18). https://doi.org/10.3390/su11185067
Mangione, Antonio (2021). La noticia sobre ciencia: sesgo hacia la comunicación de los resultados sobre los procesos de la investigación científica. SciComm Report, 1(1), 1–13. https://doi.org/10.32457/SCR.V1I1.660
Miller, Jon D. (2004). Public Understanding of, and Attitudes toward, Scientific Research: What We Know and What We Need to Know. Public Understanding of Science, 13(3), 273–294. https://doi.org/10.1177/0963662504044908
Miller, Jon D. (2012). The Sources and Impact of Civic Scientific Literacy. In M. Bauer, R. Shukla, & N. Allum (Eds.), The Culture of Science - How does the Public relate to Science across the Globe?. Routledge.
Miller, Jon D., & Laspra Pérez, Belén (2018). Los factores que influyen en la cultura científica. In Percepción Social de la Ciencia y la Tecnología 2018 (pp. 37–57).
Miller, Steve (2001). Public understanding of science at the crossroads. Public Understanding of Science, 10(1), 115–120. https://doi.org/10.3109/A036859
Ministerio de Ciencia Innovación y Universidades (2019). Estrategia Española De I+D+I En Inteligencia Artificial. https://bit.ly/3Ta8vY3
Moreno Castro, Carolina (2010). La construcción periodística de la ciencia a través de los medios de comunicación social: hacia una taxonomía de la difusión del conocimiento científico. Artefactos, 3(3), 109–130.
Paniagua, Esther (2019). Inteligencia Artificial. En Future Trends Forum (Vol. 52, Issue 55). Fundación Innovación Bankinter. https://bit.ly/3Tb3YEq
Pardo, Rafael (2014). De la alfabetización científica a la cultura científica: un nuevo modelo de apropiación social de la ciencia. In Culturas científicas e innovadoras. Progreso social (pp. 39–72).
Rohlman, Andrew (2019). What Is Data Science?. . Personal Page. https://bit.ly/3NOzk2G
Sáez, Daniel, & Costa-Soria, Cristóbal (2019). Whitepaper: Análisis de la estrategia Big Data e Inteligencia Artificial en España. https://bit.ly/3NCUUqO
Salaverría-Aliaga, Ramón (2021). Entender y combatir la desinformación sobre ciencia y salud (Informe GTM1). https://www.ciencia.gob.es/dam/jcr:44264922-8629-4567-b9b8-6d04aebb23d5/Entender_y_combatir_la_desinformacion_sobre_ciencia_y_salud.pdf
Samoili, Sofía, López Cobo, Montserrat, Gómez, Emilia, De Prato, Giuditta, Martínez-Plumed, Fernando, & Delipetrev, Blagoj (2020). AI Watch. Defining Artificial Intelligence. Towards an operational definition and taxonomy of artificial intelligence. En EUR 30117 EN, Publications Office of the European Union. https://doi.org/10.2760/382730
Sánchez-Holgado, Patricia (2022). La Ciencia de Datos como competencia transversal en Educación Secundaria en España. En S. Carrascal Dominguez & N. Camuñas Sánchez-Paulete (Eds.), Docencia y Aprendizaje. Competencias, identidad y formación del profesorado (pp. 419–450). Tirant lo Blanc.
Sánchez-Holgado, Patricia, Arcila-Calderón, Carlos, & Frías-Vázquez, Maximiliano (2021). El papel de los y las periodistas españoles ante la comunicación de la ciencia de datos en medios en línea. Revista Prisma Social, 32, 344–375. https://bit.ly/3fRUr7R
Sánchez-Holgado, Patricia, Marcos-Ramos, María, & González-de-Garay-Domínguez, Beatriz (2021). Diferencias de género en la percepción de la ciudadanía española sobre la Ciencia de Datos. Doxa Comunicación. Revista Interdisciplinar de Estudios de Comunicación y Ciencias Sociales, 235–256. https://doi.org/10.31921/doxacom.n33a1126
Summ, Annika, & Volpers, Anna-Maria (2016). What’s science? Where’s science? Science journalism in German print media. Public Understanding of Science, 25(7), 775–790. https://doi.org/10.1177/0963662515583419
Taddicken, Monika, & Krämer, Nicole (2021). Public online engagement with science information: on the road to a theoretical framework and a future research agenda. Journal of Science Communication, 20(3), 1–18. https://doi.org/10.22323/2.20030205
Tomás, David, Cachero, Cristina, Pujol, Francisco. A., Navarro Colorado, Borja, Caruana Ortuño, Maria Inmaculada, González Rico, Sergio, & Sempere Maciá, Natalia (2021). Identificación de sesgos y desinformación sobre la Inteligencia Artificial en el alumnado de Educación Superior. En Satorre Cuerda, Rosana (Ed.). Memorias del Programa de Redes-I3CE de calidad, innovación e investigación en docencia universitaria (pp. 2877-2897). Universidad de Alicante. http://rua.ua.es/dspace/handle/10045/121042
Vinuesa, Ricardo, Azizpour, Hossein, Leite, Iolanda, Balaam, Madeline, Dignum, Virginia, Domisch, Sami, Felländer, Anna, Langhans, Simone D., Tegmark, Max, & Fuso Nerini, Francesco (2020). The role of artificial intelligence in achieving the Sustainable Development Goals. Nature Communications 2020 11:1, 11(1), 1–10. https://doi.org/10.1038/s41467-019-14108-y
Vodafone Institute for Society and Communications (2016). Big Data: a european survey on the opportunities and risks of Data Analytics (Issue January). https://bit.ly/3NBQVej
Ward, Jonathan Stuart, & Barker, Adam (2013). Undefined by data: A survey of big data definitions. ArXiv. https://doi.org/10.48550/arXiv.1309.5821
Woodie, Alex (2016). Inside the Panama Papers: How Cloud Analytics Made It All Possible. Datanami. https://bit.ly/3EeC9XE
To cite this article
Sánchez-Holgado, Patricia; Arcila Calderón, Carlos; & Blanco-Herrero, David (2022). Spanish citizens' knowledge and perception of big data and artificial intelligence. ICONO 14. Revista científica de Comunicación y Tecnologías Emergentes, 20(2). https://doi.org/10.7195/ri14.v20i2.1908
Postdoctoral researcher, Department of Sociology and Communication of the University of Salamanca, Salamanca, Spain
Carlos Arcila Calderón
Associate Professor, Department of Sociology and Communication of the University of Salamanca, Salamanca, Spain
Pre-doctoral researcher, Department of Sociology and Communication of the University of Salamanca, Salamanca, Spain
Patricia Sánchez-Holgado 1, Carlos Arcila Calderón 2, David Blanco-Herrero 3