1 Introduction

Nowadays, it is possible to game with a phone while attending a class at the university. Nevertheless, not in a surreptitious manner, but supported by the lecturer who designed the quiz according to the subject’s content. Hence, gamification in higher education is also of research interest (Swacha, 2021) and can influence learning (Zainuddin et al., 2020).

The European Higher Education Area (EHEA) has transformed teaching by introducing changes that lead to the search for student learning (Miñano, 2012). Within these new developments, the use of new technologies is configured as a good tool for teachers in the search for student motivation (Ulloa et al., 2020) as they represent an innovative and surprising aspect for students (Soler et al., 2018). Moreover, these technologies allow students to become the protagonists of their own learning (Sevillano García & Rodríguez Cortés, 2013).

It is therefore clear that Information and Communication Technologies (ICT) have brought about an unprecedented change in our society, having gained more and more importance in our lives (López et al., 2018), and also in education, has meant a great change in their environment and in the teaching-learning processes (Calabor et al., 2018; Fernández Cruz et al., 2018; Miralles Martínez et al., 2019; Montoya Alcacer, 2018) as well as teachers becoming familiar with their use (Castañeda et al., 2018; Fernández Cruz et al., 2018; López Belmonte et al., 2019).

Included within ICT are the applications of games, or gamification, as a strategy for teaching in the different fields of science. It should be noted that as a teaching tool it has positive effects on the motivation of students as it predisposes them positively for better assimilation of the contents and therefore for learning (Miralles Martínez et al., 2019). And not only on motivation, because literature has highlighted effects on student engagement, motivation, classroom dynamics and learning, and also a positive effect on student performance on exams (Wang, 2015).

One of the most popular game-based learning platforms is Kahoot!. In 2021 it reached 5,000 million of players accumulate in the whole world since it was set up in 2013 (El economista, 2022). Although there have been published many studies on the effect of using Kahoot! in the classroom (Aktekin et al., 2018; Tan Ai Lin et al., 2018; Tóth et al., 2019; Wang & Tahir, 2020), those that do study the case of higher education are mainly based on qualitative research and hence, on the opinion of students rather than on objective and quantitative data, as are the marks. One of the most used qualitative tools is the survey (this is the case of Jara Ulloa and Cancino Marentes (2020), Tan et al. (2018), Wang & Lieberoth (2016)). Other works that do use the grades obtained by the students usually make a comparison between a course in which Kahoot! is used and another control group in which it is not used, calculating differences between the different grades (such as Moris de la Tassa (2016), López et al. (2018), Ortiz-Martínez et al. (2022)d lez Osorio (2016)).

Hence, in this paper, as a new contribution to what has already been addressed in the existing literature, we propose to study, with a database of grades obtained by students in the university stage, the effect of the use of Kahoot! on their continuous learning, which is reflected in the grades obtained through continuous assessment, in the final exam and the final grade of the subject, in addition to the influence that other variables may have such as the student’s gender, or whether the subject of Financial Accounting is the basic or the higher one (Miñano, 2012). Therefore, this paper has been structured as follows: firstly, we highlight the background information on which it was based; then, we set out the objectives and primary hypotheses; we detail the sample used and the methodology; we continue with the results obtained; then the conclusions derived from the analysis; and, to close the paper, we set out the limitations and validity of the study.

2 Background

Through gamification, teachers have the possibility of combining traditional methodology with a more active one, establishing a less rigid relationship with students and broadening the methods for transmitting knowledge (Moris de la Tassa, 2016), also managing to increase the degree of student attention (Vázquez et al., 2018). Gamification is understood as the incorporation of engagement and enjoyment of the gameful process into activities outside of games. Therefore, it means employing game elements to afford the students gameful experiences in the education field (Koivisto & Hamari, 2019). Gamification in education is a subject of significant research interest, as is shown by the fast growth of publications for at least seven years (Swacha, 2021).

In this line, serious games explicitly have a thought-out educational purpose. Indeed, serious games are not played basically for amusement. Instead, the reason to use these gaming approaches in education is to improve learning with the different ways to do it (Laamarti et al., 2014). Therefore, as serious games, Game Based Learning (GBL) focuses on improving the entire learning process while playing. Both serious games and GBL are supposed to have a positive effect on cognitive development. The tool of mixing games and courses is handy in improving students’ learning and can considerably help enhance knowledge and abilities (Al-Azawi et al., 2016).

Audience response systems (ARS) or clickers were used in class in a variety of fields and at all levels of education as a management tool for engaging students in the large classroom. Their use generally had a positive effect on motivation and student performance on exams (Caldwell, 2007).

In a further step, there have been developed some other applications that allow the use of games in class with everyday devices such as smartphones, tablets or computers. These game-based learning tools are designed as student-response systems (SRS) but as a game or a game-based platform (Wang, 2015). There were different SRSs, such as Socrative, but Kahoot! was the first game-based SRS implemented as a game from ground-up, designed as a game initially, and not designed as a SRS then adding game elements later (Wang, 2015; Wang & Tahir, 2020). Kahoot! is a platform (kahoot.com) completely free of charge that allows teachers to carry out quiz-type games in which students become players. One of the advantages of using this option is that it is a totally intuitive application (Fernández-Mesa et al., 2016). It is enough with the use of a central device with a screen that allows students to follow the game and the use by students of devices with internet access, in addition to the necessary network connection. This application stands out for its easy use, as no knowledge of computers or technology is required for participation. Moreover, it only requires the teacher to have an open account in the application, and students do not need to register or download it beforehand. Some of the benefits of the use of Kahoot! are to provide interactivity, improve academic performance, and engagement, although there are some other inconveniences, such as waiting time (Aljaloud et al., 2015) or anxiety created in students (Lee et al., 2019). Kahoot! is a game-like student response system more dynamic than other tools that allow the use of video, pictures, music, scoring and ranking (Wang & Tahir, 2020).

The literature on the effects of gamification through the implementation of ICT in the classroom to improve academic performance, is quite prolific and suggests that it achieves positive results in learning. Hence, from the mere use of gamification, it goes further to GBL (Al-Azawi et al., 2016). Thus, Vélez Osorio (2016) compares the traditional teaching methodology, through lectures, with the gamification methodology, and concludes that the latter positively affects the learning process by stimulating motivation. For their part, Balta & Duran (2015) analyze the effects of the use of digital whiteboards in the classroom, concluding that they significantly improve academic performance, which coincides with previous studies such as that of Marzano (2009). Specifically for the teaching of Management Accounting, Calabor et al. (2018) conclude that the use of “serious games” (serious games SG) contributes significantly to the acquisition of skills, or Carenys et al. (2017) support the teaching of accounting in higher education through video games since it increases students’ motivation and therefore their acquisition of knowledge.

As for the specific use of Kahoot!, although it was first launched in September 2013, there are already many studies that have corroborated its effectiveness as an innovative method in teaching. Thus, Wang (2015) concludes that this application produces a positive influence on students, which is maintained over time thanks to the increased competitiveness it generates in them. Wang & Lieberoth (2016) also find that it has a very positive impact in terms of interaction, response and positive student spirit.

Dellos (2015) states that after implementing the use of this game in his classes and the feedback it allows, he has managed not only to increase the motivation, curiosity and results of his students towards his subject, but there has also been a positive effect on their confidence and the recognition of their peers. Castro López et al. (2018) highlight how the use of Kahoot! has led to an increase in motivation in 91% of students and how half of them reviewed the theory to get a higher score in the game. In the same sense, Ramirez Covarrubias et al. (2017) emphasize that this application commits students to document themselves and pay attention to the development of the subject. Moya et al. (2018) emphasize that it improves individual productivity by being able to self-evaluate and therefore know on what aspects to focus their studies.

Aktekin et al. (2018) conclude that it improves student participation, engagement and concentration. Also in higher education, research by Tan et al. (2018) found that 98% of students think that the use of Kahoot! is exciting and it motivates them to work hard throughout the course. In the case of Chaiyo & Nokham (2017), they compare the effects of Kahoot! and other similar utilities such as Quizizz and Google Forms on students’ concentration, engagement, learning and motivation, and highlight that Kahoot! and Quizizz obtain more positive results than Google Form when used in class.

Specifically, on the effects of Kahoot! on learning performance, Wang and Tahir (Wang & Tahir, 2020) define one of their research questions in this regard: How does Kahoot! affect learning performance? They conduct a literature review on the topic and find thirty studies investigating the learning outcome of using Kahoot! with university students, classifying most of them as quasi-experiments. Some of these studies cover business domains and mainly find that Kahoot! improves the final grade or exam results compared to other teaching approaches. Their main conclusion is that “Kahoot! can have a positive effect on learning compared to other tools and approaches and for various contexts and domains” (Wang & Tahir, 2020, p. 9).

However, there are some other studies that examine whether Kahoot! affects learning performance but without comparing the effect of learning in Kahoot! with other approaches. Wang (2017a, 2017b) examines whether students´ own design of Kahoot! has an effect on their academic outcomes, finding that it increases discussion and benefits those who have been the leaders. Another question in this field of research finds that there is a positive relationship between the number of Kahoot! games played by the students and their final grade (Tóth et al., 2019; Guardia et al., 2019) study grades but not official grades and do not establish a relationship between them and the results of Kahoot!. They compare the Kahoot! results with the ones obtained in a test reflection of the traditional evaluation methods. Studies analyzing Kahoot! effects on learning through motivation mainly use surveys, interviews, or other qualitative tools (Cárdenas-Moncada et al., 2020; Esteves et al., 2018; Minton & Bligh, 2021; Plump & LaRosa, 2017; Yapıcı & Karakoyun, 2017).

3 Objectives and hypotheses

In accordance with all of the above, we set out in this paper to analyze the impact that the use of Kahoot! has had on the performance of students in the university environment and, specifically, in the subject of Financial Accounting in the Faculty of Economics and Business. Therefore, our general objective is to determine whether the use of GBL improves student learning, specifically, if there are effects on the official marks obtained by the students. Moreover, as specific objectives, if the positive effects of GBL, are reflected in better academic results, reflected in all the official marks of each student: marks of the continuous assessment, the final exam and the overall mark of the subject. The contribution of this paper is to study the influence of Kahoot! on official marks through quantitative analysis. Although there is much research about its effectiveness, it is mainly based on qualitative tools. Therefore, the gap we fill with this paper is to check how GBL can determine the students’ marks and their learning performance.

In addition, the subject and gender of the students are also introduced as control variables, so that it is also possible to see whether these variables influence the average marks obtained.

Therefore, we put forward the following hypotheses:

H1: The scores obtained in Kahoot! influence the continuous assessment mark.

H2: The scores obtained in Kahoot! influence the mark of the final exam.

H3: The scores obtained in Kahoot! influence the overall mark of the subject.

4 Sample and methodology

In order to find out the effect of the use of GBL through Kahoot! on university students of financial accounting, we have created a database with the scores of the games played over a semester in two groups of students of two Financial Accounting subjects taught during the first semester of the 2018/19 academic year. The first subject is Financial Accounting I in the first year and the second is Higher Financial Accounting in the second year, both in the Bachelor’s Degree in Business Administration and Management.

In the sample, we took as valid only the marks of the students who had taken the Kahoot! tests during the term. These tests were conducted randomly and unannounced during lessons. Kahoot! was used in the lessons to review the content of the topics explained at the end of the lessons after introducing new concepts. The average length of the Kahoots! was ten multiple-choice questions with different difficulty levels, three different answers, and only one correct answer. Each question usually took 20 s, although some exceptions took 30 s because some results had to be calculated. The scores obtained in Kahoot! were displayed at the end of the game according to the scores calculated automatically by the application that uses the correct answer and the time to calculate the winners. In order to compare the Kahoot! scores with each student’s other official grades, we recalculated them on a scale of 0 to 10. The scores obtained in Kahoot! were not used for official assessment. Regardless of the number of Kahoot! tests taken, the average score of these tests was taken as the data to be studied. Students who played Kahoot! were asked to log in to the application with their ID number so that their scores could be compared with the other official marks.

The continuous assessment marks are part of the students’ official marks and are planned in advance and included in the teaching guides for each subject. The different continuous assessment tests are carried out throughout the course and can be of different formats (practical exams, virtual tests, multiple-choice tests…) and represent a percentage of the mark of the subject together with the mark of the final exam. The scores from Kahoot! are not included as a part of the final mark because it is the first year of its use and its use was carried out as a teaching innovation project with the aim of checking to what extent they motivate students and, furthermore, they may influence the rest of the marks. In the future, we could consider including the scores obtained from the games played with Kahoot! as one of the exercises that count towards the continuous assessment mark, and therefore towards the overall mark for the subject. Subsequently, we have completed the database by including all the official grades for the same students: their continuous assessment mark, the final exam mark and the overall mark of the subject in the first sitting of the course: January 2019. The final mark is obtained as the result of the corresponding weighting of the continuous assessment mark and the final exam. Thus, the sample analyzed is composed of 4 types of marks: the average of the Kahoot! scores, the continuous assessment, the final exam, and the overall mark of the subject, for 98 students in the 2018/19 academic year, which means that we have studied a total of 392 marks for the subject of Financial Accounting in the Bachelor’s Degree in Business Administration and Management. Taking into account that the average number of students enrolled in each group in accounting subjects is 70, and in this case, we have collected data from 2 groups, the total population of the study would be 140 students and 560 total marks (4 marks for each student). This leads to a sampling error of 5.44% with a confidence level of 95%.

In Table 1 we include the distribution of the sample according to subject and gender as categorical variables. We have also coded the students’ gender variable, as it can provide us with additional information on the grades obtained, as has been done in previous studies that include gender or, for example, the choice of studies (Caballero García et al., 2019). As can be seen, there is a higher presence of qualifications for the Higher Financial Accounting subject (61.2% of the sample, compared to 38.8% for Financial Accounting I), and a slightly higher presence of men compared to women in our database, 55% compared to 45%.

Table 1 Distribution of categorical variables in the sample
Full size table

Table 2 shows the most significant descriptive statistics of the numerical variables that are the object of this work, which are the four grades studied, all of them expressed on a scale of 0 to 10. In order to check whether the mean grades are affected by the control variables: subject or gender of the students, we have carried out a contrast of mean differences for independent tests, classifying the different grades according to the two subjects analyzed in the sample and gender, having obtained no statistically significant results, so that there is no a priori significant influence of these variables on the grades obtained in any of the 4 cases of grades analyzed.

Table 2 Descriptive statistics of the marks included in the sample
Full size table

The database with the collection of all the notes and their subsequent analysis has been carried out with the statistical package SPSS 25 for Windows.

In order to corroborate the existence or not of a significant influence of the effects of GBL on learning and therefore on the results obtained by the students in their official marks, we calculate the correlation matrix of all the grades. In order to further analyze the influence of the scores obtained in Kahoot! on the other three official grades, we also carry out a hypothesis test for differences in means in paired samples, applying a T-test to three pairs of marks:

Pair 1- The average Kahoot! scores and the continuous assessment mark.

Pair 2- The average Kahoot! scores and the final exam mark.

Pair 3- The average Kahoot! scores and the overall mark of the subject.

In order to test the hypotheses, three hypotheses of equal means between each pair of scores have been proposed to check whether or not the means in each pair are significantly different. If we obtain that they are significantly different, we can conclude that they do not follow the same pattern of behaviour, and viceversa.

Finally, we propose three regression models, in which the dependent variables will be successively each of the official marks. The independent variable in each model will always be the average scores obtained in the Kahoot! games, and subsequently we propose the same models but introduce the control variables subject and gender of the students. To avoid multicollinearity, we transform the categorical control variables into dichotomous variables and include only one of the two in the model. We include the six models in expressions (1), (2), (3), (4), (5), and (6).

Continuous assessment scorei = β0 + β1 Average Kahoot! scoresi + εit (1).

Final exam marki = β0 + β1 Average Kahoot! scoresi + εit (2).

Overall subject marki = β0 + β1 Average Kahoot! scoresi + εit (3).

Continuous assessment marki = β0 + β1 Average Kahoot! scoresi + β2 Higher Financial Accountingi+ β3Malei + εit (4).

Final exam marki = β0 + β1 Average Kahoot! scoresi + β2 Higher Financial Accountingi + β3Malei + εit (5).

Overall subject marki = β0 + β1 Average Kahoot! scoresi + β2 Higher Financial Accountingi + β3Malei + εit (6).

We intend to check whether the scores of the games carried out through the Kahoot! application influence the three regulated marks of the subject. Therefore, if they affect the continuous assessment mark (H1), the final exam mark (H2) and the overall mark of the subject (H3).

5 Results

In this section, in order to meet our objectives, we intend to check whether there is an influence between the average score obtained through Kahoot! and the rest of the official marks. The aim is to determine whether these exercises through games serve to improve the academic performance of students, who assimilate content in an enjoyable and motivating way and thus be able to accept the hypotheses established.

From the correlation matrix between the average Kahoot! scores and the rest (Table 3), we obtain a significant and positive correlation between them. These results take us to preliminary accept the three previous hypotheses. This relationship is in line with some other previous studies on learning effect from Kahoot! compared to traditional teaching (Bawa, 2019; Boboc et al., 2018; Kinder & Kurz, 2018). The correlations obtained are totally coherent with what is expected for the marks that make up the overall mark of the subject. The highest correlation is obtained between the final exam and the final mark of the subject, as the final exam constitutes 70% of the final mark of the subject, and therefore is determining it to a large extent and shows a correlation very close to unity and positive. This correlation shows that the influence of Kahoot! on the final exam mark is essential, as stated in H2.

Next in absolute values are the correlations between the continuous assessment marks and the final mark of the subject, and between the continuous assessment mark and the final exam. All these relationships would be included in H1, H2, and H3. The former is also consistent with the fact that continuous assessment accounts for 30% of the subject final mark, although the latter shows that there is a positive correlation between keeping the subject up to date and performing good continuous assessment tests and obtaining a better mark in the final exam, as shown in previous works such as Ortiz-Martínez et al. (2022; 2018, 2020) and Santos-Jaén et al. (2018).

Table 3 Correlation matrix of the marks with the average scores of Kahoot! (Pearson’s p)
Full size table

Finally, without having any weighting in any mark, a significant positive relationship is obtained, and quantitatively not negligible, between the average scores obtained in the Kahoot! games and the rest of the marks of the course, which would lead us to accept the three hypotheses raised. For greater consistency in our results, Table 4 shows the results of the hypothesis test for differences in means with paired samples. In the three pairs of marks analyzed, a p-value of less than 0.05 is obtained, so that the null hypothesis can be rejected and, therefore, the average scores obtained in Kahoot! influences the mark obtained in the continuous assessment tests (H1), the mark obtained in the final exam (H2) and the overall mark obtained in the subject (H3). In short, once again we can see that the scores obtained in the Kahoot! games have a positive influence on the rest of the regulated and official marks for the subject. This result is similar to the one obtained by Ortiz-Martínez et al. (2022) for the same subject but comparing the use of Kahoot! to a control group using traditional teaching .

Table 4 Paired samples test
Full size table

In Table 5 we include the results obtained for the six regression models proposed, obtaining statistically significant results for all of them. In all the models, the influence of the average score of the games is significant in any of the other grades that are regulated as stated in H1, H2, and H3.

Table 5 Estimation of regression models
Full size table

According to the results obtained (Table 5), we can affirm that we accept the three previous hypotheses (H1, H2 and H3), according to which Kahoot! in the classroom affects the official marks obtained by the students in the different regulated assessment tests (Nicolaidou, 2018; Wang & Tahir, 2020). Moreover, as we can see that all the coefficient values obtained are positive. Therefore, there is a positive influence between the score obtained in these games, played unannounced in the classroom, and the mark obtained by the students in any officially established marks (Nicolaidou, 2018; Wang & Tahir, 2020). Those students who assimilate the content of the subject on a day-to-day basis and work on it, who obtain higher scores on the Kahoot! podium, ultimately obtain better marks in any other grade for the subject. These positive relationships are in turn consistent with the correlations included in Table 3, which showed that when one mark increases the other does too and with the results of the contrast of means included in Table 4, which once again confirm the basic hypotheses to be tested in this study. It should also be taken into account that there is a greater similarity in the type of test, its temporal development and its objective, between the Kahoot! games and the continuous assessment tests, as their objective is to motivate the student throughout the course for continuous learning and continuous assimilation of the contents of the subject, they are carried out periodically throughout the semester, the former by surprise, and the latter fixed in advance in the group’s timetable, while the final exam is a single test, and the overall grade is obtained as a result of the weighting of the other two marks. Paying attention to Beta coefficients obtained in the estimation of the regression models (Table 5), the highest is the one that shows how average scores of Kahoot! affect continuous assessment marks, in both cases: with and without other control variables (in model (1) Beta coefficient is 0.49 and in model (4) it is 0.48). It shows that there is a quantitative effect of Kahoot! games on the continuous assessment marks and indirectly on the overall marks of the subject that also can be attributed to a training effect. If students do more Kahoot! games they are more trained for the official exams that are designed in a similar way (Tóth et al., 2019).

Finally, Table 6 includes a summary of the relationships and influences obtained to conclude the positive effect of Kahoot! on students’ performance. This table is helpful to clarify that the results obtained support the characteristics of game-based learning as included in the former background section.

Table 6 Hypotheses, results, and game-based learning characteristics
Full size table

When the control variables subject and gender are introduced, the goodness of fit of the model’s increases, although no statistically significant relationships are obtained between the grades, whatever the dependent variable, and the gender of the students, and no clear pattern of behaviour is observed for the subject.

6 Discussion

As the main theoretical implications of this study, our results have confirmed that there is a positive effect of Kahoot! scores in the learning process, as other studies confirmed (Wang & Tahir, 2020). In addition, however, the better performance of the students using Kahoot! is proved to positively affect their official marks, as we conclude.

Aktekin et al. (2018), Dellos (2015), Tan et al. (2018), Wang (2015), and Wang & Lieberoth (2016), among others, agree with our results - obtained in this specific case from a sample of university teaching. The use of Kahoot! in the classroom leads to an improvement in student performance, translated into better marks, but that is the result of the benefits of the use of Kahoot!: increase in participation, concentration and commitment that the holding of games of this game has provoked, especially not only because of the desire to surpass oneself and other classmates but also because of the possibilities for self-evaluation and feedback that the use of Kahoot! provides. As Wang and Tahir (Wang & Tahir, 2020) point out, “The results of this literature review may suggest that there is a relationship between engagement, motivation, and having fun and learning outcomes and classroom dynamics”.

In addition, we obtain an evident influence of the results of the Kahoot! games on those obtained in the continuous assessment tests, which represent essential variations in the final mark of the subject and can define it significantly. The importance of continuous assessment is checked in other workss, such as Santos-Jaén et al. (2018) or Ortiz Martínez et al. (2020), but in other aspects and not mainly when using GBL. Therefore, it is essential for the students´ final results to influence the mark of these continuous tests through GBL positively. This relationship may also be understood as a training effect: that the number of Kahoot! tests done are increasing the training of students to cope with the other official exams (Tóth et al., 2019).

Kahoot! is thus constituted as a working tool for the teacher in order to improve students´ involvement and learning, which will be reflected in better academic results, initially in the continuous assessment tests and subsequently in the final exam mark and an overall mark for the subject. Therefore, it should be taken advantage of Kahoot! bearing in mind that students´ and teachers´ perceptions of Kahoot! are positive and that, although there are still challenges in its use, they can be solved, and the main ones are technical (Wang & Tahir, 2020).

On the other hand, the additional control variables that we analyzed: student gender and the type of financial accounting subject, were not found to be significant determinants of differences in students’ academic performance as reflected in the different marks. Although one point of this research was to explore the use of GBL and its influence on this specific subject, it is checked that it is not significant. Al-Azawi (2016) argues that there can be a justification for using GBL when the subject is difficult to learn, which can not be said for Financial Accounting.

As management implications of this study, the most important one is the possibility of using Kahoot! as an assessment tool, which not only assesses but also has other side effects such as better motivation and so on. Teachers and students agree that the positive effects of using Kahoot!, outweigh its disadvantages, so the use of Kahoot! as a tool to include official grades in the overall marks of the subject should be considered.

7 Limitations and Validity

In this line of research, contributions can continue to the improvement of teaching in the university classroom with the introduction of pedagogical innovations, which would include increasing the sample analyzed, with data from more courses, subjects, and examinations, among other variables, which could be added to the analysis. Another improvement for future research is to apply longitudinal techniques since Kahoot! is used multiple times in different class sessions. Moreover, in this same sense, previous work could be replicated, establishing comparisons between groups of students with whom Kahoot! has been used and others with whom it has not, analyzing the marks and using the appropriate statistical tools to obtain significant conclusions, in such a way that originality would be introduced into the contribution. In such a way that, regardless of the future lines of research that are proposed, the fundamental practical application derived from the results of this research lies in the usefulness of Kahoot!, in general of GBL, to positively affect the students’ marks, so that a first step would be to include the scores obtained from Kahoot! as part of the continuous assessment so that it would represent a percentage of the overall mark for the subject and thus one more of the official and regulated marks. Another line of future research is to explore the use of other games to try to obtain differences in the results obtained for Kahoot!, and finally, to try to include in the sample different subjects to check if there is a use of GBL depending on the difficulty of the subjects.

A possible threat to the validity of this work is the learning context and how Kahoot! was used to obtain the sample data. In this case the design of the Kahoot! games and the fact that they were taken randomly and by surprise during the classes may have biased the results, as the students who attend the classes may be the ones who make the best use of the teaching, are the most serious and get the best marks. Thus, it could be argued that what this study shows that good students get high scores in the Kahoot! games and also higher official grades. But as shown in Ortiz-Martínez et al. (2022), this positive relationship between Kahoot! scores and the rest of the grades is also obtained when the analysis includes a control group where teaching is traditional, so it is independent of the way of using Kahoot!. Wang (2017a, 2017b) also concludes that Kahoot! games affect the learning outcome, in this case, students create their own games. Hence the results are similar, but the type of quizzes created in Kahoot! differs from our multiple-choice questions so that the different quizzes can be designed in Kahoot! do not affect the results.