1 Introduction

Analyzing possession and use of digital devices and video games is essential to understanding how boys and girls in primary education begin their digital journey, and how the use of such devices can exert a psychosocial influence and affect their family life and education; research in this field also lends itself to proposals as to how to develop a more sustainable and educational use of such devices. A range of variables is involved in the supervised or unsupervised use of digital games by children, such as the digital, economic and skills divide in families, digital sustainability, data protection for minors, economic inequality, the possible adverse physical-emotional effects of constant screen-watching, and the supposed positives and negatives of using such devices in terms of students’ academic performance. It is essential to know the type of devices used and video games played by students at home and school in the earliest educational stages in order to detect unhealthy online habits that can affect psycho-emotional and educational development in boys and girls. Equally important is how to improve educational proposals for gamified activities both inside and outside the classroom, endowing them with a stronger pedagogical, disciplinary and psycho-emotional quality in order to provide a robust support for student learning and the feedback processes.

2 The Educational and Social uses of Gaming

Primary education is the first stage in compulsory education for children, and is essential for the development of competences, introducing students to the learning skills they need to acquire as the basis of the journey they will undertake in education. Since the start of this century, and especially in the 2010s, teachers have taken up a range of tools that combine games and computing to capture and hold the attention of their students, coalescing around the key element of gamification. The most gamified aspects of learning in primary education are in Mathematics, the Sciences, Language and the Social Sciences, as subjects taught throughout this educational stage; Literature also features in gamified form, to stimulate acquisition of knowledge and understanding of content (Hainey et al., 2016).

The emergence of video games and their positioning in the lives of young children has been studied from different perspectives. The literature on the issue (Bakker et al., 2015; Chiappe et al., 2013; Drummond & Sauer, 2014) indicates that when the same game is used at home and in school is more effective for students’ skills, furthermore, some action video games could increase students’ ability to take on additional tasks by increasing attentional capacity and video game use does not undermine the routine of studying or affect academic achievement. The parental factor was also influential in children’s behavior regarding video games, as children needed to be supervised while they played (Van Rooij et al., 2017). Parents with higher education qualifications tended to limit their children’s time spent viewing screens of all types (Christakis et al., 2004), however this discipline slackened when the adult was busy and needed to keep the children occupied (Domingues-Montanari, 2017; Olson, 2010) noted the emphasis on the playing of video games that involved the family, as a way to strengthen ties between parents and children, and which positively reinforced the values or content generated by the game (Ferguson & Olson, 2013). This situation also provided for children to do their homework in the company of their parents (Li & Chu, 2021). The results of a study by Hong & Masood (2014) found that students whose school work involved gamification rated highly aspects such as the parental support they received in their learning process, the importance of school work and their aspirations and future objectives.

3 Gamification and Academic Performance

In general, a large number of studies show that there is a negative impact of the use of video games and digital games on the academic performance of students. Students who spend more time playing video games obtain worse results in the different subjects (Adelantado-Renau et al., 2019; Drummond & Sauer, 2020; Hartanto et al., 2018; Vázquez-Cano et al., 2020–2022). Although recently, has been published different meta-analysis that show that the use of video games or digital games with curricular links begin to show a positive impact on the academic performance of students. In this sense, there is substantial evidence of the positive implications of gamification in primary education across a range of school subjects, as confirmed by the Bai et al., (2020a, b) meta-analysis of 30 independent interventions (3,202 participants) drawn from 24 quantitative studies that found a general trend towards improved academic results when gamification was used (Hedges’ g = 0.504, 95% CI [0.284–0.723], p < 0.001).

Different studies have ascribed correlation between more time spent on gamification activities in the classroom and improved academic results to the fact that the game itself stimulated concentration and boosted cognitive skills in children (Drummond & Sauer, 2014; Hartanto et al., 2018). In Mathematics, it was observed that gamification created significant relations for the learning process (Cunha et al., 2019), increased intrinsic motivation (Hallifax et al., 2020; Jagust et al., 2017), and was a more attractive option than traditional school work with pen and paper (Kickmeier-Rust et al., 2014). The same was found in Sciences, where gamification helped boost motivation (Hursen & Bas, 2019), in Geography, where higher levels of interest and enjoyment were noted in the learning process (Hong & Masood, 2014), and even in History, a subject traditionally prone to student disinterest (Martínez-Hita et al., 2021). From an interdisciplinary perspective seen through the transmedia narrative, Ruiz-Bañuls et al., (2021) postulated that gamification facilitates the use of valid strategies that enable students to acquire knowledge in different areas and see the positive results of their hard work, which in turn foments intrinsic motivation.

However, this positive trend can be undermined by addictive behaviors arising from gaming (Skoric et al., 2009) and by game mechanisms such as competition in the form of insignias or classification tables, which could have negative implications academically (Hanus & Fox, 2015; Toda et al., 2018) associated negative effects to the failure to implement appropriate methods and guidelines for gamification in educational contexts, most significantly failing to overcome student indifference, lack of motivation, undesirable behaviors and poor academic performance. Another criticism questions the validity of the pedagogical effects of gamification over time, since most research tends to measure the effects only in the short term (Li & Chu, 2021). These authors stated that in the case of reading and the willingness of students to participate in the gamified task, students not only got better academic results but also developed a more positive view of reading. Zhu et al., (2019), also interested on the long-term effects of gamified activities on reading competence, found that interest in reading rose even when gamified reading activities ceased to play a part in stimulating that competence; reading frequency, capacity and motivation also increased.

4 Student Profiles: Personality, Motivation and Gender

It is commonly accepted that for gamification to be effective, it must be tailored to users’ needs by matching gamified elements to the range of learner profiles. The literature on this issue has examined questions of gender and personality traits (Denden et al., 2021), though studies that analyze both variables together are few (Hallifax et al., 2020). Another aspect lies in the differences between types of player, according to their motivation to engage with the game, which generates categories of intrinsic and extrinsic motivation, the former relating to social connections, personal exploration or success (Kim, 2015). Player personality has been analyzed by the widely accepted Five Factor Model (FFM) (Chittaranjan et al., 2011) in Denden et al., (2021), which defines five areas of personality trait: (1) extraversion, (2) agreeableness, (3) conscientiousness, (4) neuroticism, (5) openness. These five traits are the basis on which to explain gaming preferences in Codish & Ravid (2014), who determined that extroverts found greater excitement in rewards, were motivated to show off their successes and did not modify their personalities in the gaming context, while introverts modified their behavior and tended to view the game as a competition, thus preferring prizes such as insignias. Feelings of discomfort with competition, such as progress league tables (Li & Chu, 2021), also seems to relate to personality traits, although Jagust et al., (2017) associated this to the lack of motivation in the slower students.

Gender is another factor that has been analyzed to describe player profiles, and is considered an important variable when studying the phenomenon of video games (Ferguson & Olson, 2013). Although some studies found no gender differences in gaming (Chapman & Rich, 2018), the majority have examined this variable from the perspective of frequency, motivation, preferences, styles and the emotions experienced during the game, among other aspects. Gender differences have been explained by the origins of the motivations that stimulate game playing. Males focus more on the usefulness of the technology and its “utilitarian use”, while females value the games ease of use, fun aspect and certain subjective characteristics known as “hedonistic use” (Codish & Ravid, 2015, 2017; Denden et al., 2021). Motivation is also explained by the fact that the competitive nature of the game is more attractive to males, which drives their desire to succeed, which is less evident in females (Hartmann & Klimmt, 2006; Wang & Wang, 2008), and was also found to be true for other ages, as in the case of primary school girls (Kickmeier-Rust et al., 2014). However, Hartmann & Klimmt (2006) found that such differences were not observed in games that were not competitive.

Taking the time spent on playing the game as reference, research has shown that males spend more time gaming than females (Ferguson & Olson, 2013; Hamlen, 2010; Van Schie & Wiegman, 1997) found that it was girls who opted to dedicate less time to playing, and neither did they seem to be less competent for doing so. In other words, if the boys reported more feelings of success generated by the game, it was because they spent more time playing it, but no difference was found in levels of competence between the genders in relation to game time. This author also found that boys preferred action games, and spent more time on this type of game, while girls liked simulation-type games. Girls’ choice to play less can be explained by low motivation for gaming in social situations and less orientation towards gaming genres (Gómez-Gonzalvo et al., 2020). Females excel in aspects such as “landmark memory”, “object displacement” and “perceptual speed”, while males are more comfortable with games that have clear objectives and role definitions (Gómez-Gonzalvo et al., 2020). These authors concluded that games do not generally satisfy specific needs attributable to females such as inclusion, affection and control.

These differences are equally apparent in young boys and girls. Boys are more motivated to play online games than girls, and they prefer physical and / or violent games, while girls opt for more traditional games or ones that require more reflection (Ferguson & Olson, 2013; Gómez-Gonzalvo et al., 2020). These results are similar to those in Del Moral et al. (2015; 2020). In a study on stimulation of different types of intelligence through gamification, they discovered that girls were more efficient at developing visual-spatial and logic-mathematical skills. Age is also relevant in gamification research in early educational stages, as there are differences in development that affect the processing of information and understanding (Greenberg et al., 2010). These authors found that a preference for competition manifests in the later stages of primary education, while younger boys and girls are still more motivated by fantasy in games.

5 Effects of Gamification at the Psychological and Emotional Level

The positive effects of gamification are evident at the motivational and psychological level (Fadhli et al., 2020). According to Van Rooij et al. (2021), these effects can be grouped in three categories: (1) the theory of active choice; (2) social-cognitive learning; (3) satisfaction of basic psychological needs. Gratification and extrinsic reinforcement are linked to the first category, and generally have a positive effect on users (Hakulinen & Auvinen, 2014; Sailer et al., 2017) explained that insignias, progress graphs and tables affect competition and satisfaction while avatars or gaming companions affect social experiences. Nevertheless, users’ characteristics and objectives, as well as context, are decisive in terms of the effect that these elements can provoke (Hakulinen & Auvinen, 2014). The second category refers to how learning happens through gaming; Noorhidawati et al., (2015) explained that this process can be measured in cognitive, psychomotor and affective terms, and Hong & Masood (2014) stated that greater effort is reported in students, enhancing the significance of what they do through gaming. Finally, in the third category, Olson (2010) stated that games can satisfy emotional, social and intellectual needs.

Gamification can stimulate behavioral changes in young children (Alsawaier, 2018) which, in education, can be an opportunity to develop stronger student commitment to their learning. Fun in gaming is fundamental for capturing children’s attention and developing conduct that lends itself to cooperation, and the development of capacities in young students for thinking (Da Seixas et al., 2016; Ferguson & Olson, 2013) stated that applying gamification to children is also motivated by social activity, the challenge of the game, potential reduction in stress levels, and as an alternative to boredom. It is also related to the opportunity the game provides to satisfy basic needs, such as chalking up successes and victories, acquiring control and a sense of autonomy even though this is not always attainable in real life (Ferguson & Olson, 2013). Despite the belief that playing video games isolates children, Olson (2010) found that boys and girls played games in a totally social way. Hamlen (2011) stated that boys’ and girls’ motivations have a psychological, physical and cognitive origin in primary school children, in which mental factors are more important than the actual qualities of the game.

6 Method

To identify the different typologies and profiles of primary school students according to their game-playing patterns, a latent class model was used with Latent Gold v4.5 software. A quantitative study was designed in which 242 participants filled out a questionnaire. Data has been collected and analysed to identify the typologies of individuals. Traditionally, cluster analysis is used to classify objects, but a major drawback is the lack of objective criteria to guide researchers when determining an optimum number of groups (Kaufman & Rousseeuw, 1990). In recent years, researchers have turned to latent class analysis to classify data as it requires fewer arbitrary criteria when determining the number of groups in the population. Latent class analysis is a statistical tool for modeling relations between observed variables, assuming that the structure of underlying relations is explained by a latent categorical variable (unobserved). This classification methodology is based on the estimate of conditional probabilities to enable an analysis of the variables measured in different metrics, especially categorical data (Magidson & Vermunt, 2004).

The first latent class analysis appeared in 1950 (Lazarsfeld, 1950) as a tool for constructing a typology in the analysis of a set of dichotomous variables. Lazarsfeld & Henry (1968) continued to develop the technique, and later authors went on to deliver the methodology. According to Vermunt & Magidson (2002), the latent class model can be expressed thus:

$$f({y}_{i}/\theta )= \sum _{k=1}^{K}{\pi }_{k}\prod _{j=1}^{J}{f}_{k}({y}_{ij}/{\theta }_{jk})$$

where \({y}_{i}\) represents an individual’s responses to a set of observable variables, K is the number of classes, \({\pi }_{k}\) indicates the probability of belonging to a latent class k, J indicates the total number of indicator variables, j is a particular indicator, and \({f}_{k}({y}_{ij}/{\theta }_{jk})\) is the univariate distribution function in each of the \({y}_{ij}\) elements conditioned by the set of j indicator variables of class k. In other words, the combined density function of the individual’s responses in a set of observed variables is equal to the sum of the probability of belonging to one of the classes by the product of the density function of each indicator conditioned by the class. The estimate of the model’s parameters is made using the maximum likelihood method.

A model’s fitness is usually measured by the Bayesian information criterion (BIC), which uses the number of parameters to determine goodness of fit. As a rule, the selection of the model that best fits the data is the one that has the lowest BIC value (Gill, 2002).

6.1 Participants

A probability sampling was implemented through a stratified selection process according to the different stage levels in Primary education (students from 6 to 12 years old). The sample was formed of 242 students with a mean age of 9.47 years, and a standard deviation of 2.20 years. The sample distribution was 61.98% girls, 38.02% boys, of whom 19.42% were year 1 of ESO (compulsory secondary education), 25.21% in years one and two of primary school, 27.27% in years three and four of primary school, and 28.10% in years 5 and 6 of primary school.

6.2 Instrument and Variables

A questionnaire was developed especially for the research based on the scientific literature and the age of the students for which it was designed (aged 6–12). The questionnaire was designed using the Quizz gaming app (https://acortar.link/K35T65) and distributed via e-mail to the schools and their corresponding school parents association (AMPA) during two months (September-October of 2021). Some school centers completed the questionnaire on paper (n = 89). The questionnaire was always delivered to the parents or legal guardians, never directly to the students. The questionnaire consisted of 32 items grouped in seven macrovariables: (1) sociodemographic variables, (2) possession and use, (3) places and times, (4) psycho-emotional variables, (5) social variables, (6) dynamic usability (flow) and (7) game features. The students had to respond to each item by scoring them on a 1–5 scale, 1 meaning “totally disagree” (lowest score) and 5 “totally agree” (highest score). The items selected had been adapted from works by a range of authors (Appendix).

The questionnaire also came with a video to explain the safety and ethical processes involved in the survey for the protection of minors, to be viewed by students, parents and legal guardians, and also included a request for informed consent. The questionnaire was designed using a gamified sequence in which each response was accompanied by a game dynamic or images associated to the digital world appropriate to the age of the students surveyed (Fig. 1).

Fig. 1
figure 1

Questionnaire design and gamified dynamic

Full size image

7 Results

Firstly, the questionnaire’s reliability was measured using the Bartlett’s test of sphericity and the KMO measure of sampling adequacy. The Bartlett test significance (p < 0.05) indicated that our matrix differed from the matrix unit with a 95% confidence level, thus significant correlations between the variables were found that indicated the possible existence of latent variables, the factors that explained them. The KMO test yielded a value close to 1 (0.865), such that the partial correlations of our variables were minor. Once the questionnaire’s structural validity had been confirmed, the latent class analysis was carried out, and the model’s goodness of fit was measured by BIC. We have used the BIC and the aBIC values for determining the number of classes. In this case, the model 2 presented the lower BIC, although the model 4 presented the lower aBIC. We run a simulation with more clusters, finding a decreasing trend in the aBIC values. Because of there is no common acceptance of the best criteria for determining the number of classes and following the parsimonious approach and the classification error we selected the model2 (Table 1).

Table 1 Selection of the model
Full size table

Table 2 presents the probabilities for each indictors and the Wald test. If the p-value is less than 0.05, then the indictor contributes to the creation of groups. In this case, we observe that all the p-values were significant, so all the variables contributed to the formation of the two groups. Table 2 also shows the R2 value that indicates each indicator’s variance proportion explained by the two-cluster model.

The variables related to gender and the best marks were used as covariates. The gender is a key factor for identification of the two clusters with a 95% confidence level. The subject with best marks can be considered discriminant with a 90% confidence level.

Table 2 Probabilities for each indicator and Wald test
Full size table

In Table 1 we can see the size of each cluster. The size of the first cluster is 50.82%, the second cluster 49.18%. The probabilities in Table 2 also describes the characteristics of the individuals of each group. Cluster 1 is formed of boys who usually get better marks in Social Sciences, play video games on desktop computers, usually play with friends online and spend more than two hours playing video games at weekends; they feel very happy when playing video games and tend to go directly to their gaming consoles whenever they have free time.

Cluster 2 is formed of girls who normally get better marks in Art and Biology, normally play video games on the tablet, and with a family member, spend little time playing video games at weekends; they do not feel very happy when playing video games, and do not tend to immediately pick up the tablet to play video games whenever they have free time.

The differences in the descriptions corresponding to the two groups are represented in Fig. 2:

Fig. 2
figure 2

Graph of group profiles

Full size image

8 Discussion

The aim of this research was to analyze and identify the different typologies of students who possess and play video games, and the results revealed two distinct clusters. The first group was formed of boys who usually get better marks in Maths, play video games on gaming consoles for TV, usually play with friends online and spend more than two hours playing video games at weekends; they feel very happy when playing video games and tend to go directly to their gaming consoles whenever they have free time. The desire to play video games in their free time, even for a few minutes, has been identified by researchers such as Moncada and Chacón (2012) and Sánchez-Domínguez et al., (2021), the latter in particular, as confirming that boys tend to “put on a videogame whenever I have a little free time”.

This user profile was also identified in previous research (Garcia-Continente et al., 2014; Baptista & Oliveira, 2019) in relation to the frequency and distribution of gaming time, with most time dedicated to gaming at weekends, and by more than two hours in the case of boys. The time spent on digital games during school days has been identified as more harmful for reading scores (-14%) (Vázquez-Cano, 2020); although other studies in the Australian context (Islam et al., 2020, p. 2) has showed that electronic gaming during weekdays tended to show a positive effect on reading scores, while internet use during weekdays showed a negative effect. The inverse relationship between time spent playing video games and academic performance can be related to the amount of time, when is than two hours, no differences can be found (Valencia-Peris et al., 2016). It seems more important the time of the day in which students play, if it is before going to school, it produces worst academic results (Drummond & Sauer, 2020).

Regarding the use of specific devices, boys use consoles as their main device for gaming. In this sense, the study of Gómez-Gonzalvo et al., (2020, p. 5) established that “more boys seem to be keener with traditional platforms (PC and console) than girls who are more engaged with new platforms (mobile phone and tablet).” Authors such as Kyriakou and Glentis (2019) confirmed that boys usually play video games on their gaming consoles, play online with other users and/or friends (Olson et al., 2008; Gonzálvez et al., 2016; Cha & Seo, 2018), and that this user profile remains constant throughout their development as individuals, hence game designers’ and developers’ focus on them as potential long-term consumers. This use of consoles can have positive educational implications since certain games can be used to improve the academic performance of students (i.e. National Geographic Challenge! Or Big Brain Academy, among others). Those video games which are built to use problem-solving strategies to approach challenges can have a positive impact on academic performance (Gros, 2008). Also, the informal learning acquire through this interaction can be positive for academic learning (Pereira et al., 2019). Boys in this study also manifest that the play online with other friends. Playing online has been also correlated with worst results in reading achievement (Drummond & Sauer, 2014). By contrast, “moderate” use of single-player games was associated with a performance advantage (Borgonovi, 2016). Other authors, like Gee (2003) and Schrader & McCreery (2008), stated that multiplayer online video games help to sharpen thought processes and competences that can lead to better results in Maths.

The second group was formed of girls who normally get better marks in Spanish Language and Literature, normally play video games on the tablet, and with a family member, and spend more than two hours playing video games at weekends; they feel very happy when playing video games though slightly less than boys but, unlike boys, do not tend to immediately reach for the tablet to play video games whenever they have free time. The data on frequency of gaming in our study contrast with those of other studies (Chamarro et al., 2014) that found that girls spent less time playing video games than boys, but these studies did not consider the range of devices that children use to play video games, focusing solely on gaming consoles (Cheema, 2015). Specifically for the Spanish context Gómez-Gonzalvo et al., (2020) showed that adolescents’ boys spent more time and money than girls did, and more time on weekends than on weekdays.

The results of our study make a difference on the results of other studies where the average time of gaming either in weekdays or weekends is positively associated with academic performance especially for reading scores. In this sense, we have identified that boys get better results on Maths and girls on reading and writing skills associated to Language subjects. These results contradict numerous studies that have showed a negative impact on academic performance (Wang et al., 2014; Wright, 2011).

In this sense, reports such as the one by the Spanish Video Game Association (AEVI) in 2020 analyzed the profile of the gamer in Spain and found that there were few differences between boys and girls in the use of video games between the ages of 6 and 14. However, use of video games and game-playing frequency diminished in girls from the age of 15, and was less than in adult males. Previous studies (Chóliz & Marco, 2011; AEVI, 2020) corroborate the data from our investigation, in that girls tend to prefer using the tablet to play video games. For example, Gómez-Gonzalvo et al., (2020) research, where authors established that more boys, compared to girls, play on console and computers, while more girls, compared to boys, play on mobile and tablet platforms. This and other studies found that girls were less likely to play video games online. Other studies (Ferguson & Olson, 2013; Hamlen, 2010; Gómez-Gonzalvo et al., 2020) found that girls prefer simulation games in which social interaction, cooperation or multiplayer participation in order to reach a higher level (online) are unnecessary, preferring instead games that require each player to take turns and/or for the device to be shared.

The profile of how primary education students use digital and video games is important to know in order to develop appropriate educational activities (Bai et al., 2020a, b). To understand how students play, what type of games they play, how long they play them for, and the sensations they feel when playing are essential when designing and recommending digital apps that can complement areas of the class curriculum (Wouters et al., 2013). Although the way a game is played for fun differs from how a student approaches a game with an educational purpose (gamification), many of the student user patterns of play are relevant for establishing the gaming dynamics for apps (digital games) and video games played inside and outside the classroom that can provide educational support and reinforcement, and help broaden and consolidate the content and competences being developed in the curriculum. Various studies have shown that gaming can contribute to students’ cognitive development, and collaborative gaming can improve students’ social skills and behavior relative to their classmates and academic performance (Lamb et al., 2018; Vlachopoulos & Makri, 2017). This integration of gaming in the classroom, taking advantage of the digital and videogame dynamics that students are most familiar with, can produce dynamics for solving problems and dealing with challenges in which the student has to manage content and competences prompted by clues, feedback, guidance or structures. When game design involves students, teachers and families by enabling them to consult the learning analytics on each student user profile, this can help improve cognitive development and lead to the inclusion of all the agents in the student’s learning process (Charitopoulos et al., 2020).

9 Conclusion

Determining the profile of children who begin to use digital games is an important aspect that allows us to identify patterns of use associated with social practices with implications in the educational context and also in the formation of the student’s identity. This research found two groups of students that differed in terms of gender, the school subjects in which they excelled, and gaming typology and formats. There is a trend among young boys and teenagers to favor digital and video games above all other types of leisure activity, and this can lead to problems with friends, family and school that need to be addressed in order to minimize the effects and risks of prolonged gaming. Knowledge of how students play digital and video games can be used to generate dynamics both inside and outside the classroom to improve gamification processes and to integrate digital games and devices in developing areas of the curriculum, in order to boost support, feedback and the development of content and competences associated to a range of school subjects. It is also necessary to understand the time dynamic and game formats, and whether these contribute to social and collaborative participation. These dynamics can affect the sustainability of the game and students’ psycho-affective and emotional development. Currently, there is no system that can guarantee protection of data and the safe and ethical use of games and devices by young children in primary education.