*2.3. Procedure*

The filling of the questionnaire was done in person and anonymously, with the study researchers present in the classroom to help answer any possible doubts of the students in the understanding of the items in the instrument. In the case of the Colombian students, a videoconference system was established, which allowed for the resolution of problems in real time during the completion of the instrument.

### *2.4. Analysis Conducted*

In this study, the following analysis were utilized to reach the objectives set:

First, a descriptive analysis of the 26 variables of the questionnaire was performed, through the measurements of central tendency (mean) and dispersion (standard deviation). And the variable of the self-perception of the problematic use of the Smartphone.

In second place, a descriptive analysis was performed of the six dimensions of the questionnaire, which was similar to the analysis described above. In third place, an analysis of variance was conducted to verify the existence or not of differences in each of the dimensions of the questionnaire as a function of the independent variables (country, gender, age and macro area), through Student's *t*-test and ANOVA (both with n.s = 0.05). Afterwards, the relationship between the dimensions that comprised the questionnaire were verified through bivariate correlations.

Lastly, binary logistic regressions were utilized with a forward selection Wald's test and considering the goodness-of-fit of Hosmer-Lemeshow and the Confidence Interval (CI) for exp(B), to explain the influence of the six factors studied on the self-perception of the problematic use of the smartphone and the influence of the country.

All of these analyses were performed with the SPSS statistical package, version 23 (IBM Corp., Armonk, NY).

#### **3. Results**

#### *3.1. Descriptive Study*

In first place, the descriptive results (mean and standard deviation) of the 26 items that compose the *MPPUSA Questionnaire* used in this research study are listed (see Table 2).


**Table 2.** Frequency distribution of the items from the MPPUSA Questionnaire.

Notes: M = mean; SD= standard deviation. Source: Author created.

It can be observed that the items 5, 9, 10 and 16 point to being more in agreement with the statements as compared with items 4, 14, 21 and 22, which point to being more in disagreement.

Likewise, the results obtained in the self-perception on the problematic use of the Smartphone, point that 79.9% (*n* = 3205) considered that they did not have a problematic use, as compared to 20.1% (*n* = 804) who did.

#### 3.1.1. Tolerance

The results obtained highlight that the students were indifferent as for the tolerance in the use of the smartphone (M = 3.00, SD = 0.87).

With respect to the comparison of means related with country and gender, the Student's *t*-test for independent samples showed statistically significant differences in the first case, as *t* = 6.167, *p* < 0.050, with the Colombian students being the ones who obtained a greater score in this dimension (M = 3.04 vs. M = 2.87).

Lastly, for age and the macro area, the analysis of variance (ANOVA) only pointed to the existence of statistically significant differences in the second case, as [F (4, 4003) = 5.373; *p* < 0.050]. The post hoc multiple comparisons using the Games-Howell test showed within which macro area the differences in means were found. These macro areas were Experimental Sciences along with Engineering and Architecture, with respect to Social and Judicial Sciences, which had the greatest score (μ = 3.13 and y μ = 3.05 vs. μ = 2.93).

#### 3.1.2. Escape Route

Once the analysis was performed, the results pointed out that the students were partially in disagreement as for the use of the smartphone as an escape route (M = 2.90, SD = 0.94).

As for the comparison of means, the Student's *t*-test for independent samples provided evidence of the existence of statistically significant differences for the country, exclusively (*t* = 2.262, *p* < 0.050), with the mean for the Colombian subjects being higher than the Spanish ones (2.91 vs. 2.85).

Lastly, for age and macro area, the analysis of variance (ANOVA), showed statistically significant differences in the first case [F (3, 4005) = 20.038; *p* < 0.050], with the difference of means found in students older than 26 with respect to those aged 18 to 20, 21 to 23, 24 to 26, respectively (μ = 2.74 vs. μ = 3.01, μ = 2.97 and μ = 2.89).

#### 3.1.3. Disconnection

The results obtained showed that the students were partially in disagreement with respect to the disconnection of the mobile phone (M = 2.63, SD = 1.09).

The Student's *t*-test for independent samples, performed to detect possible differences with respect to the country and gender, highlighted statistically significant differences in both cases. In the first (*t* = −3.577, *p* < 0.050), the Spanish students obtained a greater score with respect to the Colombian students (M = 2.73 vs. M = 2.60). As for gender (*t* = −3.456, *p* = 0.001), women were the ones who obtained a greater score with respect to the men (M = 2.68 vs. M = 2.56).

On the other hand, the analysis of variance (ANOVA) utilized to find statistically significant differences between age and the macro area, showed statistically significant differences in both cases, with the students older than 26 obtaining the lowest score [F (3, 4005) = 6.778; *p* = 0.000], as compared to those aged 18–20, 21 to 23 and 24 to 26, respectively (μ = 2.54 vs. μ = 2.71, μ = 2.70 and μ = 2.58). On the other hand, as for the macro area [F (4, 4003) = 2.634; *p* < 0.050], differences were found in the area of Social Sciences with respect to Health Sciences, with this last obtaining the lowest score (μ = 2.54 vs. μ = 2.68).

#### 3.1.4. Anxiety

The results obtained highlight that the students had an opinion that was partially in disagreement for the anxiety dimension of the use of the smartphone (M = 2.19, SD = 1.04).

With respect to the comparison of means related with country and gender, Student's *t*-test for independent samples showed statistically significant differences only in the second case, as *t* = −2.225, *p* = 0.020, with the women receiving a higher score in this dimension (M = 2.22 vs. M = 2.14).

Lastly, for age and the macro area, the analysis of variance (ANOVA) only pointed to the existence of statistically significant differences in the first case (age), as [F (3, 4005) = 7.180; *p* = 0.000]. The multiple post-hoc comparisons with the Games-Howell test allowed us to find exactly in what age range were these differences in means found, with those older than 26 being the ones who had a lower score with respect to students with ages between 18 and 20, and 21 and 23, respectively μ = 2.09 vs. μ = 2.28 and μ = 2.21)

#### 3.1.5. Negative Consequences

Once the analysis was performed, the results showed that the students were partially in disagreement about the negative consequences derived from the use of the smartphone (M = 2.31, SD = 0.84).

For the comparison of means, Student's *t*-test for independent samples showed the existence of statistically significant differences for country, exclusively (*t* = 8.891, *p* = 0.000), with the means for the Colombian students being higher than for the Spanish ones (2.36 vs. 2.14).

Lastly, for the age and the macro area, the analysis of variance (ANOVA) indicated statistically significant differences in both cases. In the first case [F (3, 4005) = 13.181; *p* = 0.000], finding the differences in means of the students aged between 18 and 20, with respect to the rest (μ = 2.42 vs. μ = 2.30, μ = 2.29 and μ = 2.21), and in the second [F (4, 4003) = 9.024; *p* = 0.000], the differences in means were found in the students belonging to the macro area of Social and Judicial Sciences with respect to those in Arts and Humanities and Engineering and Architecture, respectively (μ = 2.23 vs. μ = 2.48, μ = 2.39).

#### 3.1.6. Social Motivations

The results obtained showed that the students were partially in disagreement with respect to the social motivations derived from the use of the mobile phone (M = 3.22, SD = 1.03).

Student's *t*-test for independent samples, performed to detect possible differences with respect to country and gender, highlighted statistically significant differences only in the first case, *t* = 5.079, *p* = 0.000, with the Colombian students the ones who had the highest scores as compared to the Spanish ones (M = 3.27 vs. M = 3.09).

Also, the ANOVA performed to verify statistically significant differences between age and macro area, showed that there were no statistically significant differences in any of the cases.

#### *3.2. Correlational Analysis*

This section addresses the correlational study between the six dimensions of the questionnaire. The data obtained, after the use of a Spearman's correlation test to observe the relationship between the six dimensions of the scale, can be observed below (see Table 3):


**Table 3.** Results of the bivariate correlations of the items from the 6 dimensions of the questionnaire.

Note: T = Tolerance; E = Escape; D = Disconnection; A = Anxiety; C = Consequences; M = Motivations. \*\* The correlation is significant at 0.01 (2-tailed).

As a function of the data obtained, it is observed that all the dimensions were correlated amongst themselves, with these correlations being high and moderate [36], using as a reference the value of their respective coefficients. In this sense, the following had a high correlation: the dimension Disconnection with the dimension Anxiety, as Rho = 0.754 and *p* = 0.000; as well as this last with the dimension Negative consequences (Rho = 0.678 and *p* = 0.000). As for moderate correlations (as defined by the authors cited previously), these were observed in the rest of the dimensions, as they obtained values between 0.300 and 0.600.

#### *3.3. Binary Logistic Regression Analysis*

The best fitting model with respect to the self-perception shown by the university students from Spain and Colombia on the problematic use they display, was produced through a binary logistic regression. As this independent dependent variable is dichotomous in nature, with Yes/No values, the forward selection Wald test was utilized [37]. This model had a high specificity (98.8%) and high sensitivity (95%), with an overall percentage of 98% for the cut-off value of 50%, which indicates that the model classifies equally well those who consider themselves to have a self-perception on the problematic use of the Smartphone as well as those who do not.

The results shown from the analysis point out, for a sample of 4009 subjects, where 3205 answered No (79.9%) and 804 Yes (20.1%), that the goodness-of-fit of the best fitting model shows: a -2log of likelihood = 354.509, a Cox and Snell R<sup>2</sup> = 0.599; and an R2 of Nagelkerke = 94.6% (0.946); aside from the value of the Hosmer and Lemeshow test which shows a good fit, as its significance is 0.988 (≥0.05 according to [38]). Likewise, the values obtained in the Chi-square test for step (χ<sup>2</sup> = 3663.81, gl = 6 and *p* < 0.05); block (χ<sup>2</sup> = 3663.81, gl = 6 and *p* < 0.05); and model (χ<sup>2</sup> = 3663.81, gl = 6 and *p* < 0.05), shows that the fitted model is significantly differentiated from the initial or base model.

As shown in Table 4, it can be observed how all the factors that appear in the MPPUSA intervene in the explanation in a significant manner. From them and tending to their relevance according to the value of B, these would be ordered as Consequences; Tolerance; Disconnection; Escape; Motivations and Anxiety.

Thus, the logistic equation would be:

$$\text{Y (self-percentage in the problematic use)} = -89.07 + 1.13 \text{Tolerance} + 1.11 \text{Escape} + 1.12 \text{Disconrection} + 0.54 \text{Anixety} + 1.18 \text{Consquences} + 0.99 \text{Motivations}$$

Thus, it can be observed that all the factors resulting from the MPPUSA positively and significantly influence the self-perception of considering that a problem with the use of the smartphone exists (reference category).


**Table 4.** Binary logistic regression of the self-perception on the problematic smartphone use <sup>a</sup> of Spanish and Colombian university students with respect to the MPPUSA.

Note. <sup>a</sup> = Dependent variable: Self-perception of having a problematic use of the smartphone Yes and No.

When analyzing the explanatory model of the self-perception of the problematic Smartphone use, having in mind the country of origin as the selection variable, it is observed, for 2965 Colombian students, of which 2266 answered No (76.4%) and 699 Yes (23.6%), that the goodness-of-fit had the following values: a -2log of likelihood = 290.826; a Cox and Snell R<sup>2</sup> = 0.630; and a, R<sup>2</sup> from Nagelkerke = 94.8% (0.948); with the Hosmer and Lemeshow value = 0.999, which demonstrates a good fit, as its significance is ≥0.05 [38]. Also, it has a high specificity (98.5) and a high sensitivity (95%), with an overall percentage of 97.7% for the cut-off value of 50%, which indicates that the model classifies those who consider themselves to have a self-perception on the problematic use of the Smartphone as well as those who do not equally well.

Table 5 shows how all the factors obtained from the MPPUSA significantly intervene in the explanation, as it occurred in the general model, in the same order according to their relevance (B value).


**Table 5.** Binary logistic regression of the self-perception of the problematic smartphone use <sup>a</sup> of Colombian university students with respect to the MPPUSA.

Note. <sup>a</sup> = Dependent variable: Self-perception of having a problematic use of the Smartphone Yes and No.

In this occasion, the logistical equation is the following:

Y (self-perception on the problematic use of Colombian students) = −88.13 + 1.12Tolerance + 1.06Escape + 1.09Disconnection + 0.50Anxiety + 1.19Consequences + 0.95Motivations (2)

As a result, the self-perception of the Colombian students when considering if they have a problematic smartphone use (reference category), all the factors from the MPPUSA have a positive and significant influence.

Lastly, Table 6 shows the results of the explanatory model of the self-perception on the problematic use of the smartphone of Spanish university students (selection variable). Of the 1044 students, 939 who answered No (89.9%) and 105 Yes (10.1%). The goodness-of-fit of the best fitting model obtained the following values: a -2log of likelihood= 54.446; a Cox and Snell R<sup>2</sup> = 0.451; and a Nagelkerke R2 = 94.2% (0.942); with the Hosmer and Lemeshow test obtaining a value of 1.0, which demonstrates a good fit, as its significance is ≥0.05 [38]. The specificity is high (99.4%), as well as its sensitivity (94.3%), with an overall percentage of 98.9%) for the cut-off value of 50%), which indicates that the model classifies those who consider themselves to have a self-perception on the problematic use of the Smartphone as well as those who do not equally well.

**Table 6.** Binary logistic regression of the self-perception on the problematic smartphone use <sup>a</sup> of Spanish university students with respect to the MPPUSA.


Note. <sup>a</sup> = Dependent variable: Self-perception of having a problematic use of the smartphone Yes and No.

Thus, the equation is:

Y (self-perception of the problematic use of the smartphone of Spanish university students) = −108.94 + 1.46Tolerance + 1.48Escape + 1.08Disconnection + 1.25Anxiety + 1.27Consequences + 1.46Motivations. (3)

On Table 6, it is observed that the factors that appear in the MPPUSA significantly intervene in the explanation; however, when taking into account their relevance (B value), it is observed that the order is different from the general model and the order found for the Colombian students, being: Escape; Motivations; Tolerance; Consequences; Anxiety; Disconnection.

In summary, for the self-perception of the Spanish students who consider if they have a problem with the use of smartphones (reference category), all the factors found from the MPPUSA have a positive and significant influence.

When comparing the Spanish students with the Colombian ones, it is observed that although all the factors contribute to the best fitting models, in Spain the factor Escape had more weight than in Colombia, with Consequences being more significant for the latter.

The analysis of the explanatory model of the self-perception on the problematic Smartphone use, with sex as the selection variable, shows that for 1601 male students, of which 1270 answered No (79.3%) and 331 Yes (20.7%), the goodness-of-fit had the following values: a -2log of likelihood = 134.565; a Cox and Snell R<sup>2</sup> = 0.607; and a R2 from Nagelkerke = 95% (0.950); with a Hosmer and Lemeshow value = 0.994. This demonstrates a good fit, as its significance is ≥0.05 [38]. Also, it had a high specificity (98.8%) and a high sensitivity (94.3%), with an overall percentage of 97.9% for the cut-off value of 50%, indicating that the model classifies both of those who consider themselves to have a self-perception on the problematic use of the smartphone as well as those who do not, equally well.

Table 7 shows that all the factors obtained from the MPPUSA significantly intervened in the explanation, as in the general model, in the same order according to their relevance (B value).


**Table 7.** Binary logistic regression of the self-perception of the problematic Smartphone use <sup>a</sup> of male university students with respect to the MPPUSA.

Note. <sup>a</sup> = Dependent variable: Self-perception of having a problematic use of the smartphone Yes and No.

On this occasion, the logistical equation is the following:

Y (self-perception on the problematic use of male students) = −93.51 + 1.26Tolerance + 1.22Escape <sup>+</sup> 1.16Disconnection <sup>+</sup> 0.60Anxiety <sup>+</sup> 1.22Consequences <sup>+</sup> 0.90Motivations (4)

As a result, all the factors from the MPPUSA have a positive and significant influence on the self-perception of the male students when considering if they have a problematic use of the smartphone (reference category).

Table 8 shows the results of the explanatory model for female university students (selection variable). Of these 2408 students, 1935 answered No (80.4%) and 473 Yes (19.6%). The goodness-of-fit of the best fitting model had the following values: a -2log of likelihood = 215.28; a Cox and Snell R2 = 0.594; and a Nagelkerke R2 = 94.5% (0.945); with the Hosmer and Lemeshow test obtaining a value of 0.897, which demonstrates a good fit, as its significance was ≥0.05 [38]. The specificity and the sensitivity were high (98.7% and 94.3%, respectively), with an overall percentage of 97.8% for the cut-off value of 50%, indicating that the model classifies both of those who consider themselves to have a self-perception on the problematic use of the Smartphone as well as those who do not, equally well.

**Table 8.** Binary logistic regression of the self-perception on the problematic smartphone use <sup>a</sup> of female university students with respect to the MPPUSA.


Note. <sup>a</sup> = Dependent variable: Self-perception of having a problematic use of the smartphone Yes and No.

Thus, the equation is

Y (self-perception of the problematic use of the smartphone of female university students) = −88.4 + 1.09Tolerance + 1.03Escape + 1.11Disconnection + 0.48Anxiety + 1.19Consequences + 1.08Motivations. (5)

In Table 8, it is observed that the factors from the MPPUSA significantly intervened in the explanation; however, when taking into account their relevance (B value), it was observed that the order was different from both the general model and the one for the male students. Thus, for females, the order was: Consequences; Disconnection; Tolerance; Motivations; Escape; Anxiety.

Having in mind age as the selection variable, the explanatory model for 1237 university students aged from 18 to 20 years old, of whom 979 answered No (79.1%) and 258 Yes (20.9%), had the following goodness-of-fit values: a -2log of likelihood = 146.052; a Cox and Snell R<sup>2</sup> = 0.596; and a R2 from Nagelkerke = 93% (0.930); with the Hosmer and Lemeshow value = 0.993, which demonstrates a good fit, as its significance was ≥0.05 [38]. Also, it had a high specificity (98.1%) and a high sensitivity (93.8%), with an overall percentage of 97.2% for the cut-off value of 50%, which indicates that the model classifies those who considered themselves to have a self-perception on the problematic use of the Smartphone as well as those who do not, equally well.

Table 9 shows how all the factors obtained from the MPPUSA significantly intervene in the explanation, as it occurred in the general model, in the same order according to their relevance (B value).


**Table 9.** Binary logistic regression of the self-perception of the problematic smartphones use <sup>a</sup> of university students aged between 18 and 20 years old with respect to the MPPUSA.

Note. <sup>a</sup> = Dependent variable: Self-perception of having a problematic use of the smartphone Yes and No.

On this occasion, the logistical equation is the following:

Y (self-perception on the problematic use of smartphone of students between 18 and 20 years) = −76.99 + 0.88Tolerance + 0.95Escape + 0.99Disconnection + 0.49Anxiety + 1.03Consequences + 0.83Motivations (6)

As a result, all the factors of the MPPUSA have a positive and significant influence on the self-perception of university students aged between 18 and 20 years old when considering whether they have a problematic use of smartphones (reference category).

Table 10 shows the results of the explanatory model of students aged from 21 to 23 years old (age selection variable). Of these 860 students, 696 answered No (80.9%) and 164 Yes (19.1%). The goodness-of-fit of the best fitting model obtained the following values: a -2log of likelihood= 45.820; a Cox and Snell R<sup>2</sup> = 0.602; and a Nagelkerke R<sup>2</sup> = 96.7% (0.967); with the Hosmer and Lemeshow test obtaining a value of 0.995, which demonstrates a good fit, as its significance is ≥0.05 [38]. Both the specificity and sensitivity were high (99.1% and 97%, respectively), with an overall percentage of 98.7% for the cut-off value of 50%, which indicates that the model classifies those who consider themselves to have a self-perception on the problematic use of the smartphone as well as those who do not equally well.


**Table 10.** Binary logistic regression of the self-perception on the problematic smartphone use <sup>a</sup> of university students between 21 to 23 years of age with respect to the MPPUSA.

Note. <sup>a</sup> = Dependent variable: Self-perception of having a problematic use of the smartphone Yes and No.

Thus, the equation is:

Y (self-perception on the problematic use of smartphone of students between 21 and 23 years) = −134.75 + 1.32Tolerance + 1.34Escape + 1.59Disconnection + 0.77Anxiety + 1.95Consequences + 1.94Motivations. (7)

For university students between the ages of 24 and 26, the results of the explanatory self-perception model had the following aspects (see Table 11): of the 546 students, 442 answered No (81%) and 104 Yes (19%). The goodness-of-fit of the best fitting model obtained the following values: a -2log of likelihood = 56.168; a Cox and Snell R2 = 0.581; and a Nagelkerke R2 = 93.4% (0.934); with the Hosmer and Lemeshow test obtaining a value of 0.876, which demonstrates a good fit, as its significance is ≥0.05 [38]. The specificity and sensitivity were high (99.1% and 94.2%, respectively), with an overall percentage of 98.2% for the cut-off value of 50%, indicating that the model classifies those who consider themselves to have a self-perception on the problematic use of the smartphone as well as those who do not, equally well.



Note. <sup>a</sup> = Dependent variable: Self-perception of having a problematic use of the smartphone Yes and No.

Thus, the equation is:

Y (self-perception on the problematic use of smartphone of students between 24 and 26 years) = −77.27 + 1.06Tolerance + 0.92Escape + 0.79Disconnection + 0.47Anxiety + 1.06Consequences + 0.83Motivations. (8)

Table 12 shows the results of the explanatory model of students older than 26 (age selection variable). Of the 1366 students, 1088 answered No (79.6%) and 278 Yes (20.4%). The goodness-of-fit of the best fitting model obtained the following values: a -2log of likelihood = 86.276; a Cox and Snell R2 = 0.612; and a Nagelkerke R2 = 96.3% (0.963); with the Hosmer and Lemeshow test obtaining a value of 0.987, which demonstrates a good fit, as its significance is ≥0.05 [38]. Both the specificity and its sensitivity were high (99.2% and 96.4%, respectively), with an overall percentage of 98.6% for the cut-off value of 50%, which indicates that the model classifies those who consider themselves to have a self-perception on the problematic use of the Smartphone as well as those who do not, equally well.


**Table 12.** Binary logistic regression of the self-perception on the problematic smartphone use <sup>a</sup> of students older than 26 years old with respect to the MPPUSA.

Note. <sup>a</sup> = Dependent variable: Self-perception of having a problematic use of the smartphone Yes and No.

Thus, the equation is:

Y (self-perception on the problematic use of smartphone of students older than 26 years) =−117.61 + 1.72Tolerance + 1.56Escape + 1.49Disconnection + 0.75Anxiety + 1.48Consequences + 1.22Motivations (9)

On Table 12, it is observed that the factors that appeared in the MPPUSA significantly intervened on the explanation; However, having in mind its relevance (B value), it was observed that the order was different from the general model and the order found for different ages, being Tolerance; Escape; Disconnection; Consequences; Motivations; Anxiety for those over 26 years old.

#### **4. Discussion**

We are in agreement with [39] in that the technologies themselves are not elements that can provoke a problematic behavior in its use, but it is the individuals who develop this behavior, which may or may not be problematic [9].

In this sense, after utilizing the MPPUSA, the results obtained provided us with a profile of a young Spanish and Colombian university student who has a problematic behavior with the mobile device as a function of 6 factors relative to the tolerance towards the time spent using the device, its regard as a means of escape or the disconnection to the world that surrounds them, linked to the anxiety that could be caused if they do not use it, just as the results reached by [17], who showed that the university students with whom the same instrument was utilized, felt better when they utilized the smartphone to evade or avoid situations they found themselves in. The results also showed that the participating students felt that not having the device available gave them anxiety and stress, the negative consequences its use could have due to their using it most of their time in different ways, and lastly, the social motivations related to its use of not, in line with data from [12,17,18,34]. It is meaningful that these results coincide with those from [40], who reached similar results after the use of an instrument called Questionnaire of Experiences Related to the Mobile Phone (Spanish acronym: CERM) with university students, although its short length (10 items) did not analyze in detail aspects such as feelings of avoidance or FOMO.

Taking into account the objectives established, it was verified that the students self-perceived not having a problematic use of the device. This result should be pondered, given that the study by [13] pointed that the university students have control over the addiction to the Internet (85.17%), and in our case, the perception of not having a problematic use was 79.9%. This shows that the university students younger than 35 years old do not consider a problematic or addictive use of the Smartphone or the Internet.

As opposed to results reached by [17,18,21], after the use of the same instrument along the same line, and also as opposed from results reached by [41], who reported an increasing trend of cell phone use after the use of MMPUS. Nevertheless, it is significant that in general, the Colombian students were more aware of the negative consequences that its use could imply, pointing out that this is linked to the time they spend connected or using it in a generalized manner or having to hide this usage time from family and friends, data that coincides with those from [42], who after the application of the HUTL, pointed out the same aspects for this study population. They are also more aware that their state of rest and their academic performance will the affected due to the excessive use of the Smartphone [18,43], with this aspect being common to all the research studies performed with other instruments such as MMPUS, CERM or HUTL [17,40–43]. Nevertheless, these behaviors that are detrimental, in one way or another, to daily life, are also observed in the use of substances o addiction to the Internet, where the study by [44] points that the altered behaviors related to the Internet, taking into account the sex of the students, could be due to the use or contents consumed. Therefore, we should ask ourselves if the devices are understood as mere devices or associated to the consumption of contents they enjoy.

They also indicated that when they are bored, sad or alone, they use the mobile phone. On other hand, they also commented that if they did not have a device such as this, it would be difficult for their friends to locate them and that they did not like it when the turned their phones off, and this is where the importance a smartphone has in their social relations is derived from just as the results found after the use of the MMPUS [42]. On the other hand, the Spanish students have features of the FOMO syndrome, manifesting that they are worried about not having the device with them, and they would miss calls or messages, and that it is difficult for them to turn the phone off [12,18,29,43].

If we center our attention to the age of the participants, it is verified, as in the work by [12], that the younger students from both countries had a PSU as compared to those who were older, as opposed to the results from [45], given that significant differences were found in the factors found that referred to the use of the mobile phone as a means of escape, to disconnect, anxiety and the negative consequences due to its use.

Gender could be pointed out as an element that determines the attitude towards the mobile phone [32,43] an aspect that coincides with other tests applied [40–42]. As opposed to the work by [12,41–43], it was the women, and more specifically Spanish women, the ones who had a problematic attitude with the device, such as the shown in the results from works by [18,20,46–49], as the participating sample expressed feeling anxiety and disconnection with their surroundings if they did not have a working phone [47–50].

On the other hand, if attention is placed on the macro area of study, as referred to the awareness of the consequences, the Arts and Humanities students, as well as the Engineering and Architecture ones, as compared to the other macro areas, such in the case of Social Sciences, are ones who had features of the so-called FOMO [10,29,45]. As for the elements that comprise the factor Tolerance, it was verified that the students from the macro areas of Experimental Sciences, Engineering and Architecture, were more aware of having less tolerance (being in a bad mood, using the device due to boredom or for feeling alone, and to make superfluous calls) [46,50]. It should be highlighted that the students from the area of Health Sciences did not have any prevalence of problematic behavior.

With respect to the factors reached in this study, it can be confirmed that the previous works [30,34,49,50] have shown the existence of 6 factors, and that the grouping of the items, aside from the correlations between themselves, determine the elements that define a problematic behavior towards smartphones, and about which work models should be developed to be able to perform an intervention.

Lastly, it was verified that the model of 6 factors created in line with the one from [35], follows the trail of the data from [17,30,49,50], and points to the Colombian students being closer to this model as compared to the Spanish students, as it describes the factor Consequences as being more significant in the model, as compared to Escape from the Spanish group. On the other hand, age and sex do not exactly follow the general model.

The section should not end without indicating the limitations of this international study, where the Latin American context, as well as the geography of the terrain, could define uses, trends, and views. One of these limitations could language used. Although in this case the language used was Spanish, it should be point out that just as with other languages, it has local variations depending on the country we find ourselves in. Therefore, the initial instrument developed in the Spanish utilized in Spain had to be adapted to Colombian Spanish, resulting in a delay of the research study.

On the other hand, although the sample was large, the Spanish representation should be increased, not only for the generalization of the data, but also for the performing a comparison that is not catalogued as cross-sectional, and that contributes a PSU analysis model for the entire Latin American context, which entails the intention of widening the study to other countries in the American continent.
