3.1 Demographic results

With the intention of being able to carry out an analysis from a gender perspective in future research, all variables have been measured and contrasted both as a whole and ungrouped by gender. Thus, the Student's T test (parametric test) or the Mann-Whitney U test (non-parametric test) will be applied to compare the two independent samples. In all cases, a statistical analysis is carried out to determine whether parametric or non-parametric tests should be applied. To do this, we determined the values of normality, randomness and homoscedasticity. The normality test has been performed with the Kolmogorov-Smirnov test, since the sample is greater than 50. Randomness will be measured with the Rachas test and homoscedasticity with the Levenne test.

The work sample is based on the response of 456 individuals, of which 56.1% are women and 43.95% are men. As for the age of these, a large majority are those born in 1999 (18.4%), in 2000 (20.2%) and in 2001 (32.5%). In 1998 and 1997, 7.9% in each of them and, in 1996, 5.3%. The rest of the individuals - 7.8% - have a distribution in different years, the oldest being a student born in 1963. In addition, the vast majority of the students surveyed, 61.4%, began their studies in the 2019-2020 academic year, 27.2% in the 2017-2019 academic year and 11.4% before the 2018-2019 academic year.

Also, a large majority of the sample are third grade students, 71,9%; fourth grade are 27,2% and only 0,9% are second grade students. Regarding the distribution by courses in which they were in the 2019-2020 academic year (during confinement), the majority, 57,9%, were in the first year, 25,4% in the second, 14,9% in the third and 1,8% in the fourth. These data are consistent with the data on which course they were enrolled in the course following confinement (2020-2021 academic year). In you Table 2 can see the comparison of the confinement data and the post-confinement data.

In Table 3 can be seen that the number of credits enrolled and approved in the confinement semester (second semester of the 2019-2020 academic year), compared to the same semester of the following academic year (second semester of the 2020-2021 academic year), has very similar values, thus a large majority approved the 30 ECTS credits, which correspond to the five subjects that are taken in each semester, specifically 60,5% during confinement and 61,4% in the post-confinement.

Finally, question 1.8⁴ of the questionnaire evaluates whether or not the student had individual space for study during confinement. A large majority, 95,61%, did have that space compared to 4,39% who did not have it.

When analyzing these values from the point of view of gender and since non-parametric⁵ tests must be applied, specifically the Mann-Whitney U test must be done, the significance value is 0,028. Consequently, the null hypothesis is rejected. In this case, the null hypothesis is that there is no statistically significant difference in whether or not men or women have had an individual space for work. Therefore, there is a difference between men and women, with 6,3% of women not having individual work space during confinement, compared to only 2,0% of men.

The responses regarding the socio-demographic data of the students on the enrolled credits reflect that 1,8% of women enrolled during the pandemic in less than 30 ECTS credits, that is, less than five subjects, increasing to 0,9% after the pandemic. Men who enrolled in fewer than five subjects during the pandemic were 3,5%, rising to 2.6% after the pandemic. 42,1% of women enrolled in five subjects during the pandemic and 43% after it; men enrolled in 30 ECTS represent 28,10% during and after the pandemic. On the other hand, the percentage of enrollees of more than 30 ECTS credits (more than five subjects) was similar between sexes, with 12,30% of women enrolled, during and after the pandemic, the same percentage of men during the pandemic, rising to 13,20% of men after the pandemic.

3.2 Statistical results

The analysis of the students' opinion was carried out with the qualitative analysis software webQDA (^{Costa; Amado, 2018}). In the same way as ^{Arias et al. (2017}) and ^{Contreras et al. (2019}), the data were encoded in the group category structure shown in Table 4, in order to obtain the conclusions about each of the dimensions of the study: Teaching Methodologies, Technological Tools and Evaluation Systems. This table shows the data of the complete study of the three dimensions that are evaluated in the questionnaire of ^{Hidalgo et al. (2023}), although in this work only the results of the Evaluation Systems dimension are shown.

After coding the students' responses, each of them was assigned certain subcategories within each dimension, obtaining a tree structure with the frequencies obtained. Next, we analyze the data obtained by the dimension of evaluation systems.

To respond to the hypothesis related to the Evaluation Systems shown in Table 5, we will perform a hypothesis test using the SPSS software.

As the answers given by the students were on the Likert scale, they were transformed following the following criterion observed in Table 6:

The result of this transformation is to obtain two possible values in the students' responses: perception of improvement (value 0) of the Evaluation Systems or perception that there was no improvement (value 1) in the Evaluation Systems.

The normality test is performed in SPSS to know whether to perform parametric or non-parametric tests, the result being the one shown in the following Table 7.

As the Sig. value is 0,000, both in the Kolmogorov-Smirnov test and in the Shapiro-Wilk test, this indicates that there is no normality, so non-parametric tests will have to be performed.

The binomial non-parametric test is performed to test our hypothesis, taking into account that, if the significance result of the binomial test is less than or equal to 0,05, our hypothesis will be accepted and, otherwise, the null hypothesis (H₀) will be accepted, that is, the opposite of what we proposed.

The binomial test result is shown in Table 8 and 9:

As can be seen, the exact significance (bilateral) has a value of 0,000, indicating that the means are not equal. As the N value of the category of Group 1 (category with value 1) is higher than that of Group 2 (category with value 0), that is, the value of the perception of no improvement, our hypothesis is not accepted, that is: "The evaluation systems adapted during the confinement of the Covid-19 pandemic did not allow students to improve academic performance with respect to previous years".

Shows Table 10 the distribution of references by subcategories resulting from the qualitative analysis of question 4.5^6.

Regarding the data collected and analyzed in question 4.5, positive aspects, we observed that a large number of students do not value positively the aspects that have emerged about the evaluation during the pandemic ("None", with nine references). However, in contrast to these previous ones, there is also a large group of students who do consider that they have improved their grade or academic performance ("Improvement of the final grade or academic performance", with eight references), so there is no unanimity among the students and, therefore, our initial hypothesis cannot be verified.

They also acknowledge that less weight has been given to the value of the exams and more value has been given to the activities carried out in the continuous evaluation ("Less weight to exams and more to continuous evaluation", with eight references), which could have led to an improvement in final performance with these changes produced.

Another aspect that they value positively is the passage of the exams to an online model ("Online Exams (Safe Exam Browser)", with six references). This aspect was forced by the change to the teaching modality. However, they have considered that this aspect was positive for them.

Everything suggested that the changes produced in the methods and modalities of evaluation would improve the academic performance of the students, they have even commented that one of the positive aspects was that there was more possibility of copying in the exams ("It can be copied", with two references). Although all positive feedback indicated that there was improvement, the initial hypothesis ultimately could not be verified.

Table 11 displays the distribution of references by subcategories resulting from the qualitative analysis of question 4.6.⁷.

Regarding the data collected and analyzed in question 4.6, negative aspects that have emerged and remain with respect to the evaluation, we observe that a large number of students do not value negatively any of the aspects that have emerged about the evaluation during the pandemic ("None", with ten references). As a more negative aspect, they consider that their training has decreased in this pandemic period ("Decrease in training", with five references).

Other negative issues worth mentioning are: the communication and connection problems they have had ("Connection problems", with two references), some disagreements introduced by the teachers in the evaluation criteria ("Non-conforming evaluation criteria", with two references) and the uncertainty of how the exams would go in this new evaluation modality ("Uncertainty in assessment ", with two references).

There is another less significant range of negative aspects that they consider existed, which we can observe in Table 11.

3.3 Student suggestions

At the end of the questionnaire, students were given the opportunity to provide any suggestions they deemed necessary. To do this, section 5.2 was created in the questionnaire. Subsequently, the answers were analyzed and a new category was created that could encode the answers, calling this category "suggestions", and, in turn, it was subdivided into several subcategories, as shown in Table 12.

In this table it can be seen that the most referenced subcategory is "Improving teacher engagement ", with five references. Among some of the specific comments in this category, we have: "Greater involvement of teachers" or "More support from teachers". Among other interesting aspects that students have suggested, are: