Exploring the Dynamics of Suicidal Ideation, Negative Emotional States, Uncertainty, Work Overload, Illusion for Study, and Persistence among University Students during COVID-19: A Comprehensive Study

Abstract

This study explores the complex interplay of suicidal ideation, negative emotional states, uncertainty due to COVID-19, remote work overload, illusion for study, and persistence among university students during the initial peak of the COVID-19 pandemic. Utilizing a sample of 17,057 Mexican university students. This research uses a structural equation modeling to determine the influence of uncertainty due to COVID-19 as a latent variable in the negative emotional states and persistence (as mediating variables) vs. suicidal ideation as a dependent variable. The six variables are related through hypotheses and tested using partial least squares and path analysis. We used an adapted questionnaire sent by email. Findings show that uncertainty due to COVID-19 had a direct and significant influence on Negative Emotional States and a significant inverse effect on persistence; in the trajectory, suicidal ideation is explained. Our findings highlight the indirect impact of persistence on suicidal ideation through the lens of COVID-19-induced uncertainty. Additionally, we emphasize the substantial relationships between suicidal intent on negative emotional states and remote work overload. These results underscore the need to integrate these factors into the design of prevention and therapeutic interventions.

1. Introduction

The outbreak of COVID-19 (Coronavirus disease 2019) caused the emergence of a severe acute respiratory syndrome type-2 (SARS-CoV-2), posing a significant global threat. In March 2020, the World Health Organization (WHO) officially designated the situation as a pandemic, urging all nations to implement urgent measures. Mexico witnessed its first cases in April 2020, prompting an immediate healthcare system response that included quarantine measures. Initially, there was an expectation that “normal activity” would resume without risks by mid-2020. However, the situation remained on high alert. While cases initially subsided after a substantial portion of the population received vaccination, subsequent waves of infections have maintained a state of continuous global vigilance [1]. Many families lost dear members or had to deal with COVID-19 recovery. Faculty students were anxious and worried, having to spend more than a year in solitude and isolation, going through depression and Negative Emotional States, many of them with suicidal ideation [2].

This crisis has garnered substantial media attention, propelling it to unprecedented levels of coverage [3]. A search on Scopus yields 137,969 results with the term “COVID-19” in the title and 191,045 results with the term in the abstract, title, or keyword sections. Despite the volume of information available, the majority of publications have taken the form of guides, manuals, and clinical reports. Few technical and scientific papers have focused on the realm of suicidal ideation and its association with negative emotional states, uncertainty due to COVID-19, remote work overload, illusion for study, and persistence among students.

In the year 2019, the global count of lives lost to suicide surpassed a troubling mark of 700,000, a concerning statistic that predates the onset of the pandemic. Paradoxically, in the tumultuous year of 2020, defined by the impact of COVID-19, suicide rates demonstrated a modest reduction across various nations, including the United States. This intriguing trend echoes historical patterns observed during earlier crises such as World War II and instances of terrorist attacks, challenging conventional assumptions about suicide’s prevalence during periods of upheaval [3]. However, in 2021, suicide rates raised again. Research in the field of suicidal ideation involving young individuals serves the purpose of shedding light on the true extent of this issue. Suicidal thoughts and actions in adolescents and young people are the result of intricate processes deeply ingrained in various aspects of their lives and functioning [4, 5]. Several factors contribute to suicidal ideation, including anxiety disorders, which are diagnosed in a significant proportion of adolescents who attempt suicide [6], as well as depressive disorders [7, 8]. It is important to note that environmental factors such as early-life trauma, neglect, inadequate parenting, and persistent stress, elevate the risk of developing anxiety, depression, and other stress-related disorders [9]. Of particular concern are the prospects for adolescents with depressive disorders. Current observations suggest that a substantial portion of these young individuals may be at a heightened risk of attempting suicide in adulthood [10]. Adolescents are especially vulnerable to depressive and anxiety disorders due to dynamic biological and emotional changes (e.g., emotional volatility and altered perception of stimuli) as well as social factors (e.g., a perceived lack of support and impulsive behaviors). These factors collectively make it challenging for young people to adapt to the evolving demands of their surrounding world [11].

COVID-19 pandemics exacerbated factors associated with depression, anxiety, and PTSD symptomatology in young adult mental health, mainly in university students [12, 13, 14, 15]. This pandemic developed a tendency to worry and fear uncertainty, and students had to copy a big deal of topics during the quarantine [16, 17, 18].

1.1 Objective

To determine the impact of COVID-19-induced uncertainty on factors such as persistence, negative emotional states, illusion for study, and remote work overload, and to explain its connection to suicidal ideation, we employed structural modeling with latent variables in a sample of Mexican students comprising 17,057 individuals.

1.2 Research hypothesis

On Table 1, the expected influence of the exogenous variables related to the endogenous variables is shown as proposed below in the hypotheses:

Hypothesis	Exogenous variables	Influence	Sign	Endogenous variables
H1:	Uncertainty during COVID-19	====>>	—	Persistence
H2:	Uncertainty during COVID-19	====>>	—	Illusion for study
H3:	Uncertainty during COVID-19	====>>	+	Work overload
H4:	Uncertainty during COVID-19	====>>	+	Negative emotional states
H5:	Persistence	====>>	—	Suicide ideation
H6:	Negative emotional states	====>>	+	Suicide ideation
H7:	Illusion for study	====>>	—	Suicide ideation
H8:	Work overload	====>>	—	Suicide ideation

Table 1.

Expected influence of the independent variable (exogenous) in connection with the dependent variables (endogenous).

Source: Self-elaboration.

It is observed in Table 1 that there is an inverse and direct influence of exogenous variables toward endogenous variables according to their expected theoretical correspondence.

1.2.1 Negative emotional states

The pandemic-induced social distancing measures led to various psychosocial effects characterized by uncertainty, a sense of threat, and confinement. Consequently, individuals have experienced a range of negative emotional states, including anger, frustration, insomnia, stress, anxiety, and depression [19]. Prolonged periods of concern and uncertainty have been linked to heightened levels of anxiety and depression [21]. It has been observed that among university students, these negative emotional states are more prevalent among women than men; however, literature mentions that males have a tendency to suicide [22]. Therefore, it is of utmost importance to assess the current mental well-being of university students and understand how various factors impact their mental health. These factors may include isolation, family pressures, exposure to violence, overcrowding, excessive academic and work demands, individual traits, physical living conditions, and available financial resources.

For these reasons, we have chosen to examine potential predictors of suicidal ideation such as persistence, negative emotional states, remote work overload, and illusion for study in our sampled student population. Previous research has indicated that uncertainty played a significant role in triggering Negative Emotional States among university students during COVID-19 [23]. Recent studies have shown that over 50% of the students in our sample have experienced a decline in psychological well-being as a result of the lockdowns and social isolation [24]. Furthermore, more than 80% of the students have reported symptoms of depression [25]. These findings underscore a notable surge in depression and anxiety among students, largely attributed to prolonged unemployment, financial instability, and family-related uncertainties. In light of these empirical findings, we propose Hypotheses 1–4 as follows:

H1: “There is a statistically negative significant influence of Uncertainty during COVID-19 on Persistence”.

H2: “There is a statistically negative significant influence of Uncertainty during COVID-19 on Illusion for Study”.

H3: “There is a statistically positive significant influence of Uncertainty during COVID-19 on Remote Work Overload”.

H4: “There is a statistically positive significant influence of Uncertainty during COVID-19 on Negative Emotional States”.

These hypotheses are observed in Table 1 and Figure 1.

Figure 1.

Structural equations model with latent variables (SEM).

1.2.2 Suicidal ideation

Suicidal ideation has become a topic of global concern [26]. Currently, almost 1,000,000 individuals take their own lives each year [27, 28]. Notably, among those aged 15–29, suicide ranks as the second leading cause of death worldwide [29], with males being particularly susceptible [30, 31]. Research has indicated that various factors contribute to suicidal thoughts among university students. These factors include academic overload, inadequate rest, the pressure to meet responsibilities, depressive and anxiety disorders, stress, and other psychosocial risk factors they encounter. Additionally, family issues, socioeconomic constraints, and substance and alcohol abuse, among others, are often linked to suicidal ideation [32].

Suicidal ideation is characterized by recurrent thoughts and planning related to taking one’s own life, although these thoughts are not acted upon [33]. Scholars have reached a consensus that suicidal ideation typically progresses through stages, starting with a desire to die, followed by passive thoughts of suicide, and ultimately leading to concrete suicidal ideation without a specific method [34].

In the scientific literature, it is observed that Latin American countries, including Mexico, report lower rates of suicidal ideation, suicide attempts, and completed suicides compared to Europe and the United States. Generally, in Latin America, the rate of suicidal ideation among non-medical students ranges from 10 to 15%. However, medical students tend to exhibit higher rates, ranging from 17 to 22% [35]. Given the current circumstances of isolation and uncertainty stemming from the pandemic, it is crucial to assess whether suicidal ideation has increased and to identify potential risk profiles that warrant attention and support [36].

Despite the extensive body of research on suicidal ideation, there is a notable dearth of studies in Mexico. Therefore, there is a pressing need to investigate suicidal ideation among students by analyzing various variables, including gender, substance and alcohol abuse, other sociodemographic characteristics, and mediating factors that may help explain higher rates of suicidal ideation. Recent studies have revealed elevated levels of somatization, obsessive-compulsive disorder, anxiety, phobic anxiety, paranoia, and suicidal ideation, as well as a general severity index among sampled students during the COVID-19 pandemic [37]. Research also demonstrates a direct relationship between uncertainty and suicidal ideation, anxiety, and obsessive-compulsive disorders [38]. Furthermore, negative emotional states have been found to contribute to suicidal ideation [39]. Building upon this research, we propose Hypothesis 4–9 as follows, as seen in Table 1 and Figure 1:

H5: “There is a statistically negative significant influence of Persistence on Suicidal Ideation”.

H6: “There is a statistically positive significant influence of Negative Emotional States on Suicidal Ideation”.

H7: “There is a statistically negative significant influence of Illusion for Study on Suicidal Ideation”.

H8: “There is a statistically positive significant influence of Remote Work Overload on Suicidal Ideation”.

H9: “There is a statistically positive significant influence of Uncertainty during COVID-19 on Suicidal Ideation”.

A structural modeling (SEM) with latent variables is proposed under the method of Partial least squares (PLS) considering the research hypothesis, which is presented in Figure 1.

Figure 1 shows the expected trajectories of each of the exogenous variables comprehensively toward each of the endogenous variables according to the hypotheses of the SEM model.

This article is divided into four sections. The first section is a general introduction to the research topic, variables, and context, as well as the objective and hypotheses proposal. The second section describes the methods and sample procedure followed. In the third section, we present the results, and finally, in the fourth section, we analyze the discussion of all the implications, limitations, and areas of opportunity for future research and conclude with some recommendations.

2. Methods

2.1 Study design

This study is a cross-sectional analysis of data. Ex post facto, non-experimental, explanatory design that was conducted through an online survey. Structural equations with latent variables under the method of partial least squares were used for the analysis.

2.2 Unit of analysis

Mexican University Students n = 17,046. Their sociodemographic data was: gender 63.9% (f = 10,255 women), whereas 36.1% (f = 6791 men); marital status 91.3% (f = 15,575 Single), 3.8% (f = 646 civil union), 3.8% (f = 655 married), 1% (f = 170 other); scholarity 90.3% (f = 15, 396 Bachelor’s degree), 6.4% (f = 1088 specialization), 2.3% (f = 392 Master’s degree); 1% (f = 170) other; age 73.2% (f = 12,488 aged 17 to 22; 19.4% (f = 3314 aged 23 to 27 years old); 7.3% (f = 1244 28 to more than 63). All students received detailed information regarding the purpose of the study and provided online informed consent to participate. No money or fees were paid to the students who participated in the study.

2.3 Data collection

The instrument was sent electronically from the beginning of April 2020 to the end of May 2020 to University students enrolled. We sent the questionnaire to the total population of Students and Professors of the Mexican Northern Universities with a resulting sample of Students of n = 17,057. Professors were not included in this manuscript. The information from the answered questionnaires was uploaded into a database that was edited and analyzed in the Statistical Package for the Social Sciences (IBM SPSS) version 23 for Windows and the Smart PLS version 3.

2.4 Instruments

The instrument used for data collection included the following sections:

• Student’s questions regarding their sociodemographic characteristics including gender, marital status, age, scholars, and academic demographic information such as academic program and scholarship status to categorize the demographic variables.

• Covid-19 pandemic uncertainty: we developed this part completely using five questions to measure uncertainty as a unidimensional variable with a Likert scale of five points. Examples of the questions used are: “I was afraid when facing the pandemics”, “I felt that life is very fragile”, “I feared to be infected”, “I was terrified to imagine someone dear was going to die”. While this instrument was never validated in Mexico before since it was our creation, we did test the Internal Consistency with Cronbach’s Alpha Value as well as Content Validity, Construct Validity and Concurrent Validity obtaining reliable results. The reliability of the constructs measured by the instruments was scientifically fulfilled: all the Alpha and Rho c coefficients of the subscales were higher or 0.70, so they are consistent. For the validity, the AVE score (Average Explanation Variance) considering convergent validity; all the subscales obtained scores equal to and greater than 0.50, while the discriminant coefficients obtained scores above 0.70, as well as consistency with the Fornell-Larcker criterion.

• Resilience scale CD-RISC25 [Connor-Davidson]. Focused on determining resilience. Resilience is the ability of human beings to adapt and overcome adverse situations, measured through their level of positive response to risk situations as a multidimensional construct. This instrument consists of 25 items that must be answered on a 5-point Likert scale (1 to 5). We used only persistence/tenacity/self-efficacy dimension (items 10,11, 16, 23–25). This scale was tested and validated in Latin America and Mexico with acceptable results (Also, the reliability of the constructs measured by the instruments was scientifically fulfilled: all the Alpha and Rho c coefficients of the subscales were higher or 0.70, so they are consistent. For the validity, the AVE score (Average Explanation Variance) considering convergent validity; all the subscales obtained scores equal to and greater than 0.50, while the discriminant coefficients obtained scores above 0.70, as well as consistency with the Fornell-Larcker criterion.

• SBI (Gil-Monte) questionnaire of burnout. This scale has four dimensions. We used only the illusion for the study dimension (items 1,5,10,15,19). This scale was tested and validated in Latin America and Mexico with acceptable results [20]. Also, the reliability of the constructs measured by the instruments was scientifically fulfilled: all the Alpha and Rho c coefficients of the subscales were higher or 0.70, so they are consistent. For the validity, the AVE score (Average Explanation Variance) considering convergent validity; all the subscales obtained scores equal to and greater than 0.50, while the discriminant coefficients obtained scores above 0.70, as well as consistency with the Fornell-Larcker criterion.

• Adaptation of the Scale for Suicide Ideation SSI-W. Focused on determining the suicide ideas of students in the face of the pandemic. It is based on the Scale for Suicide Ideation (SSI-W), and it consists of six questions that must be answered in a range of one to two [21]. While this suicidal ideation scale in the United States has seen widespread use ever since it was standardized for use with psychiatric patients [21], as well as in outpatient contexts [21]. Furthermore, this scale has been barely applied to university students in the Mexican context. We found a few studies that applied the scale in Mexico with good results [22]; however, the reliability of the constructs measured by the instruments was scientifically fulfilled: all the Alpha and Rho c coefficients of the subscales were higher or 0.70, so they are consistent. For the validity, the AVE score (Average Explanation Variance) considering convergent validity; all the subscales obtained scores equal to and greater than 0.50, while the discriminant coefficients obtained scores above 0.70, as well as consistency with the Fornell-Larcker criterion.

3. Results

The expected results include revealing the intricate relationships between suicidal ideation, study excitement, remote work overload, negative emotions, uncertainty during COVID-19, and persistence. The above implies the analysis of positive aspects such as Persistence and illusion for studying, as opposed to negative aspects such as negative emotional states and remote work overload, being influenced by uncertainty during COVID-19 wanting to explain suicidal ideation. The results of structural modeling are seen in Figure 2 and Table 2.

Figure 2.

Restructured structural equations modeling. Source: Own elaboration.

Hypothesis	Exogenous variables	Influence	Loadings of the regression coefficients	Endogenous variables	R-Square	Decision
H1:	Uncertainty during COVID-19	====>>	−0.124	Persistence	0.015	Accept
H2:	Uncertainty during COVID-19	====>>	0.073	Illusion for study	0.005	Decline
H3:	Uncertainty during COVID-19	====>>	0.39	Work overload	0.152	Accept
H4:	Uncertainty during COVID-19	====>>	0.692	Negative emotional states	0.479	Accept
H5:	Persistence	====>>	−0.197	Suicide ideation	0.173	Accept
H6:	Negative emotional states	====>>	0.353			Accept
H7:	Illusion for study	====>>				Decline
H8:	Work overload	====>>	−0.075			Accept

Table 2.

Results of hypotheses testing.

Source: Self-elaboration.

Figure 2 and Table 2 show the results of the Structural Equation Model where the nine hypotheses were verified, as follows:

As noticed in Table 2, the hypotheses were tested and results are commented on below:

Hypothesis 1: The results show a significant inverse influence of uncertainty during COVID-19 toward persistence with a standardized regression coefficient of −0.125; explaining approximately 0.016% of the variance of the R-square. Therefore, this hypothesis was accepted.

Hypothesis 2: The results showed a significant direct influence of uncertainty during COVID-19 toward illusion for study with a standardized regression coefficient of 0.027, explaining approximately 1% of the variance explained from its R-square. Therefore, this hypothesis was rejected. It is important to note that although it was statistically correct being significant, a significant inverse influence was expected from a theoretical point of view.

Hypothesis 3: The results demonstrated a significant direct influence of uncertainty during COVID-19 toward remote work overload with a standardized regression coefficient of 0.390, explaining approximately 15% of the variance explained from its R-square. Therefore, this hypothesis was accepted.

Hypothesis 4: The results demonstrated a significant direct influence of uncertainty during COVID-19 toward Negative emotional states with a standardized regression coefficient of 0.693, explaining approximately 48% of the variance explained from its R-square. Therefore, this hypothesis was accepted.

Hypothesis 5: The results demonstrated a significant inverse of persistence toward suicide ideation with a standardized regression coefficient of - 0.192. Therefore, this hypothesis was accepted.

Hypothesis 6: The results demonstrated a significant direct influence of negative emotional states toward suicide ideation with a standardized regression coefficient of 0.249. Therefore, this hypothesis was accepted.

Hypothesis 7: The results demonstrated a significant inverse of illusion for study toward suicide ideation with a standardized regression coefficient of – 0.021. Therefore, this hypothesis was accepted.

Hypothesis 8: The results demonstrated a significant inverse of remote work overload for the study toward suicide ideation with a standardized regression coefficient of – 0.077. Therefore, this hypothesis was accepted.

Hypothesis 9: The results demonstrated a significant direct influence of uncertainty during COVID-19 toward suicide ideation with a standardized regression coefficient of 0.148. Therefore, this hypothesis was accepted.

The influence of each of the variables from hypotheses H5: to H9: toward suicidal ideation, allowed us to explain approximately 18% of the variance from its R-Square.

Complementing the validation of the model and its hypotheses, it was necessary to run Bootstrapping with 500 samples, as seen in Table 3.

	Original sample (O)	Sample mean (M)	Standard deviation (STDEV)	T statistics (|O/STDEV|)	P values
Illusion for study - > Suicide ideation	−0.021	−0.022	0.01	2.2	0.028
Negative emotional states - > Suicide ideation	0.249	0.249	0.01	24.161	0
Persistence - > Suicide ideation	−0.192	−0.192	0.011	18.252	0
Remote work overload - > Suicide ideation	−0.077	−0.077	0.008	9.868	0
Uncertainty during COVID-19 - > Ilusion for study	0.027	0.027	0.009	3.009	0.003
Uncertainty during COVID-19 - > Negative emotional states	0.693	0.693	0.004	163.462	0
Uncertainty during COVID-19 - > Persistence	−0.125	−0.125	0.008	14.942	0
Uncertainty during COVID-19 - > Remote work overload	0.39	0.39	0.007	58.875	0
Uncertainty during COVID-19 - > Suicide ideation	0.148	0.148	0.009	16.51	0

Table 3.

Bootstrapping of the SEM models.

Source: Own elaboration.

The bootstrapping of 500 samples in Table 3 confirms all statistically significant paths or influences of the SEM model at a confidence interval of 0.95.

3.1 Descriptive statistics, reliability, and validity of the model

3.1.1 Descriptive statistics

Table 4 presents the descriptive statistics, reliability, and validity indexes, and the Pearson Moment-Product bivariate correlations between the subscales.

Interval confidence
Subscales	Mean	Standard deviation	Lower	Upper	Cronbach’s alpha	Composite reliability (rho_c)	Average variance extracted (AVE)	Negative emotional states	Suicide ideation	Illusion for study	Uncertainty during COVID-19	Persistence	Work overload
Negative emotional states	2.80	1.16	2.78	2.81	0.92	0.93	0.67	0.82
Suicide ideation	1.21	0.47	1.20	1.21	0.87	0.90	0.65	0.36	0.81
Illusion for study	4.17	0.92	4.16	4.19	0.82	0.91	0.84	−0.04	−0.12	0.91
Uncertainty during COVID-19	2.88	1.16	2.86	2.90	0.87	0.90	0.65	0.69	0.31	0.07	0.81
Persistence	4.01	0.83	3.99	4.02	0.86	0.90	0.64	−0.25	−0.28	0.46	−0.12	0.80
Remote work overload	3.61	1.00	3.60	3.63	0.89	0.91	0.64	0.54	0.13	0.02	0.39	−0.07	0.80

Table 4.

Descriptive statistics, convergent and discriminant validity.

Source: self-elaboration.

As noticed in Table 4, significant bivariate correlations are observed. It is important to highlight these correlations, since they support the expected theoretical coherence with empirical evidence in the present study. Correlations are:

Significant inverse correlations that correspond to the theoretical foundation: illusion for studying with negative emotional states (−0.10), persistence with negative emotional states (−0.25); suicidal ideation with illusion for study (−0.15), suicidal ideation with persistence; suicidal ideation with remote work overload (−0.13); illusion for study with remote work overload (−0.02); uncertainty during COVID-19 with persistence (−0.13); as well as persistence with remote work overload (−0.07).

Significant positive correlations that correspond to the theoretical foundations: negative emotional states with suicidal ideation (0.36); negative emotional states with uncertainty during COVID-19 (0.69); negative emotional states with remote work overload (0.54); suicidal ideation with uncertainty during COVID-19 (0.31); illusion for study with uncertainty during COVID-19 (0.03); illusion for study with persistence (0.54); as well as uncertainty during COVID-19 with remote work overload (0.39).

4. Reliability and validity

Also, as seen in Table 4, the reliability and validity indexes are as follows:

Uncertainty during COVID-19: Alpha = 0.87, Composite Rho_c = 0.90, Average variance extracted (AVE) = 0.65, Discriminant = 0.81.

Persistence: Alpha = 0.86, Composite Rho_c = 0.90, Average variance extracted (AVE) = 0.64, Discriminant = 0.80.

Remote work overload: Alpha = 0.89, Composite Rho_c = 0.91, Average variance extracted (AVE) = 0.64, Discriminant = 0.80.

Negative emotional states: Alpha = 0.92, Composite Rho_c = 0.93, Average variance extracted (AVE) = 0.67, Discriminant = 0.82.

Illusion for study: Alpha = 0.89, Composite Rho c = 0.93, Average variance extracted (AVE) = 0.76, Discriminant = 0.87.

Suicide ideation: Alpha = 0.87, Composite Rho c = 0.90, Average variance extracted (AVE) = 0.65, Discriminant = 0.81.

As noticed in Tables 2, 4, and 5, the reliability and validity of the constructs measured by the instruments were scientifically fulfilled; all the Alpha and Rho c coefficients of the subscales were higher or 0.70, so they are consistent. For the validity, the AVE score (Average Explanation Variance) considering convergent validity; all the subscales obtained scores equal to and greater than 0.50, while the discriminant coefficients obtained scores above 0.70, as well as consistency with the Fornell-Larcker criterion. In the main diagonal of Table 4, the discriminant coefficients of each of the subscales of the SEM model are observed.

Items	Remote work overload	Negative emotional states	Uncertainty during COVID-19	Illusion for study	Suicide ideation	Persistence
Q20R1	0.80
Q20R2	0.77
Q20R3	0.85
Q20R4	0.82
Q20R5	0.74
Q20R6	0.81
Q21R1		0.75
Q21R2		0.87
Q21R3		0.79
Q21R4		0.84
Q21R5		0.84
Q21R6		0.80
Q21R7		0.85
Q22R1			0.82
Q22R2			0.87
Q22R3			0.82
Q22R4			0.78
Q22R5			0.73
Q27R2				0.84
Q27R3				0.89
Q27R4				0.88
Q27R5				0.87
Q36R1					0.75
Q36R2					0.81
Q36R3					0.85
Q36R4					0.85
Q36R6					0.77
Q45R10						0.77
Q45R11						0.85
Q45R23						0.75
Q45R24						0.82
Q45R25						0.80

Table 5.

Standardized loadings of the regression coefficients by factors.

Source: Self-elaboration.

In addition, an analysis of the standardized regression loadings was made to check the congruence of each of the items with respect to their group or theoretical construct of Structural Equation Modeling. Standardized regression coefficients are observed in all items being equal to or greater than 0.70; which confirms the validity of each construct or factors of the model. These standardized regression coefficients are presented in order for each of the subscales of the SEM model in Table 5.

The main findings derived from the specific SEM hypotheses were:

• Uncertainty during COVID-19 had very little significant inverse influence on the positive aspects of the study such as persistence (−0.125) and illusion for study (0.027), explaining very little of its variance from its R-square respectively (0.016) and (0.001). These results were observed for Hypotheses 1 and 2.

• On the contrary, uncertainty during COVID-19 greatly influenced negative emotional states (0.693) and remote work overload (0.152), explaining high scores of their variance from their R-squared (0.480) and (0.152), respectively. We affirm that uncertainty during COVID-19 increased the levels of negative emotional states and remote work overload, relevantly and significantly influencing suicidal ideation. These results were observed in Hypotheses 3 and 4.

• Persistence had greater influence (− 0.192), compared to illusion for study (− 0.021) to reduce suicidal ideation, counteracting the direct effects of uncertainty during COVID-19 (0.148), as well as the indirect effects of negative emotional states (0.249). On the other hand, remote work overload (− 0.077) had a very small influence in reducing suicidal ideation. These results were observed in Hypotheses 5, 6, 7, 8 and 9.

• Comprehensively, uncertainty during COVID-19 (0.148), persistence (0.192), illusion for study (−0.021), negative emotional states (0.249), and remote work overload (− 0.077) explain approximately 18% of suicide ideation. Considering that suicide ideation is a psychological aspect prior to suicide, this finding is pertinent and relevant because it allows us to intervene before it is too late.

Concerning our first hypothesis, consistent with the literature, uncertainty during COVID-19 in students was negatively related to persistence, but positively related to illusion for study; contrary to the theory and to what we were expecting, it seemed that uncertainty boosted illusion for study in students.

5. Discussion

Individuals afflicted with mental disorders face an elevated susceptibility to psychological distress amid the COVID-19 pandemic. Nonetheless, there exists a dearth of comprehensive research scrutinizing the determinants linked to suicidal ideation, recognized as the most influential precursor to suicidal actions, particularly within the demographic of university students. Anticipations regarding the substantial influence of the COVID-19 pandemic and its associated lockdown measures on suicidal tendencies were made. Nevertheless, extant studies have yielded incongruous results, and there is a scarcity of longitudinal data tracking these trends. While this study was not longitudinal, it aimed to investigate the factors perceived to be linked with suicidal ideation among university students during the COVID-19 pandemic. This research underscores the heightened vulnerability of individuals with mental disorders during the COVID-19 crisis and highlights the significant gap in the literature concerning the determinants of suicidal ideation, a critical precursor to actual suicidal behaviors, specifically among university students. The text emphasizes the initial predictions regarding the pandemic’s impact on suicidal tendencies, which have yielded contradictory results and lacked temporal context. It articulates the study’s intention to explore the factors associated with suicidal ideation within the context of the pandemic among university students as a very important topic.

The COVID-19 pandemic brought about unprecedented challenges for the education sector worldwide, and Mexico was no exception. Universities in the country had to quickly adapt to the new reality by shifting to online teaching to ensure the safety of students and staff. This abrupt transition exposed students to a level of uncertainty that they had never experienced before.

One critical aspect examined in this research was the influence of uncertainty during COVID-19 on several psychological variables, including persistence, negative emotional states, illusion for study, and remote work overload, with the ultimate goal of understanding its potential connection to suicide ideation among students.

One of the immediate consequences of the pandemic and the shift to online learning was the disruption of campus life. The physical separation from the university environment, fellow students, and the daily routine was a significant change that impacted students differently. Many students faced challenges such as a lack of access to necessary equipment and a suitable study environment at home. As a result, attending classes online became a challenge for some, which further contributed to the overall uncertainty they were experiencing.

The concept of “illusion for study” is an important dimension to consider in this context. It is essentially a measure of how motivated and enthusiastic students are about their studies. The study finds that this dimension was affected by the uncertainties brought about by the pandemic, as expected; however, Students’ enthusiasm and motivation to study were not easily dampened by the sudden changes and the challenges they faced in adapting to the new learning format. The observation that uncertainty somewhat enhances students’ illusion for study, suggesting heightened motivation to engage with their studies, is somewhat in line with certain psychological theories. Self-determination theory posits that individuals can be intrinsically motivated to pursue activities, such as learning, when they perceive them as personally meaningful. Some previous research has found that individuals may engage more deeply in learning when faced with challenges or uncertainties, as they seek to make sense of and adapt to new situations. However, it is important to note that this finding contradicts the conventional notion that uncertainty generally hampers motivation, and further exploration is needed to understand the underlying mechanisms and boundary conditions.

This research utilizes a structural equation model to explore the complex relationships between uncertainty during COVID-19, psychological variables, and suicide ideation among Mexican university students. It highlights the significant impact of the pandemic on students’ lives, from the challenges of remote learning to the potential negative consequences on their psychological well-being. Understanding these dynamics is crucial for developing strategies to support students during and after the pandemic, as they navigate the “new normal” in the realm of education.

The findings of this study provide valuable insights into the complex interplay between various psychological variables and their impact on students’ well-being during the COVID-19 pandemic. Our structural equation model (SEM) analysis revealed several important relationships and implications for both academia and policy development.

However, the same uncertainty had a substantial and significant influence on negative emotional states and remote work overload. This suggests that the psychological toll of uncertainty was felt more strongly in terms of increased negative emotions and the burden of adapting to remote learning environments. Importantly, these negative emotional states and remote work overload significantly contributed to suicidal ideation, emphasizing the critical need for mental health support for students facing such challenges. The substantial and significant influence of uncertainty on negative emotional states aligns with a body of literature suggesting that uncertainty and ambiguity can trigger stress and anxiety. Numerous studies conducted during the pandemic have reported elevated levels of anxiety, depression, and other negative emotions among students and the general population. This finding reinforces the idea that the psychological toll of uncertainty during the pandemic has had far-reaching effects on individuals’ emotional well-being. The link between negative emotional states, remote work overload, and suicidal ideation is consistent with broader research on mental health during the pandemic. Several studies have highlighted the heightened risk of mental health challenges, including suicidal ideation, among university students as a result of the disruptions caused by COVID-19. The multifaceted impact of the pandemic, encompassing social isolation, academic stress, and economic uncertainties, has created a fertile ground for mental health struggles.

Our study also revealed that persistence played a crucial role in reducing suicidal ideation. It had a greater influence compared to illusion for study, highlighting the importance of students’ determination and resilience in the face of adversity. Persistence acted as a counteractive force, mitigating the direct and indirect effects of uncertainty and negative emotional states.

In summary, while some aspects of the findings are consistent with prior research, such as the association between uncertainty and negative emotional states, the observation that uncertainty enhances the illusion for study is somewhat novel and intriguing. These results underscore the complexity of human responses to crises like the COVID-19 pandemic and emphasize the need for multifaceted support systems for students, encompassing both academic and mental health components. Further research is warranted to delve deeper into the mechanisms underlying these relationships and to explore potential interventions to mitigate the adverse effects of uncertainty and negative emotions on students’ mental well-being.

5.1 Overall implications and opportunities for further research

These findings have significant implications for universities and policymakers in their efforts to support students during the ongoing pandemic and future crises. The results emphasize the importance of comprehensive student support systems that address not only academic challenges but also the emotional and psychological well-being of students. Institutions should consider strategies to enhance students’ persistence and resilience while providing resources for coping with negative emotional states and remote work overload.

Furthermore, this study opens up several avenues for future research. The adaptive mechanisms developed by students during the pandemic are an intriguing area of study, as these mechanisms may have long-term implications for their academic and mental health outcomes. Additionally, tracking the changes in these variables over time through repeated assessments can provide valuable insights into the evolving challenges faced by students as they continue to navigate the uncertainties of the pandemic.

Also, our findings have significant implications for universities and policymakers striving to bolster student support systems during the pandemic. By identifying pivotal factors that impact student well-being and academic engagement, institutions can tailor interventions that address the unique challenges posed by the ongoing crisis; for example, Certainly, supporting student well-being and academic engagement during the pandemic is crucial for maintaining their mental health and educational success.

The present COVID-19 pandemic context presents various research avenues centered around the adaptive mechanisms’ students are acquiring. It is crucial to assess the levels of uncertainty and persistence students have cultivated as part of their adaptation and how these factors influence their coping strategies and overall response to this pandemic. Additionally, it is advisable to administer the instrument a second time to examine how these variables change over time.

In conclusion, this study underscores the importance of understanding the psychological experiences of students during the COVID-19 pandemic and the potential consequences on their mental health. By addressing these issues proactively, universities and policymakers can better support students and intervene in a timely manner to prevent and mitigate suicidal ideation, ultimately safeguarding the well-being and success of their student populations.

5.2 Limitations to this research

One drawback, as mentioned above, is conducting cross-sectional research at a single point in time, which prevents the comparison of how variables may evolve at various stages of this pandemic. On the other hand, not having inclusion and exclusion criteria for the study was a limitation. For example, those receiving psychiatric treatment were not excluded from the study. Future research should exclude any students receiving treatment.

6. Conclusions and recommendations

This study contributes to the expanding realm of research regarding the impact of COVID-19 on university students. It delves into the intricate relationships between suicidal ideation, illusion for study, work overload, negative emotions, uncertainty, and persistence. Through a comprehensive approach, we aim to illuminate the distinctive challenges faced by students during these unprecedented times. Ultimately, our findings seek to guide the development of strategies that foster resilience, well-being, and academic achievement within the context of the pandemic.

Students who take their own life impact their faculty members, families, friends, and the local community deeply. It goes beyond the immense pain and sorrow experienced by loved ones; it extends to have far-reaching and enduring effects on all those connected. Everybody around them feel guilt and sorrow for not noticing on time the signs of suicidal ideation and intent [40, 41]. It is crucial to recognize that suicide is not an inevitable outcome. Suicidal deaths typically result from a complex history of suicidal ideation caused by distress, trauma, bullying, loneliness and isolation, lack of social support, experiences of abuse, experiences of disaster, discrimination, gender (male), chronic pain, self-harm behaviors, feelings of hopelessness and despair – key symptoms of suicidal ideation – and adversity. As stated before, young people who faced mental and emotional health difficulties during COVID-19 were at higher probability of having suicidal ideation. Young men are at higher risk than young women. They occur not because someone desires to die but rather because they believe they can no longer endure their circumstances [41, 42]. While no single initiative or organization can single-handedly prevent suicide, there are numerous ways in which our services, communities, individuals, and society as a whole can collaboratively strive toward this goal [43, 44, 45].

Here are some interventions that universities and institutions can consider implementing:

• Mental Health Services and Counseling: Provide accessible and virtual mental health services, including counseling, therapy, and support groups, to help students cope with stress, anxiety, and other mental health challenges.

• Online Well-Being Workshops: Offer virtual workshops and webinars that focus on building resilience, managing stress, improving sleep, practicing mindfulness, and promoting overall well-being.

• Peer Support Programs: Establish peer mentorship or support programs where experienced students can connect with and offer guidance to their peers, creating a sense of community and belonging.

• Flexible Academic Policies: Implement flexible academic policies, such as extending assignment deadlines, providing options for pass/fail grading, and allowing students to withdraw from courses without academic penalties.

• Adaptive Online Learning: Provide training and resources to instructors to help them adapt their teaching methods for online learning, making use of engaging and interactive tools to enhance student engagement.

• Digital Social Spaces: Create virtual spaces for students to connect with their peers, participate in online clubs or interest groups, and engage in social activities to combat feelings of isolation.

• Physical and Mental Wellness Challenges: Organize wellness challenges that encourage students to engage in physical activities, practice mindfulness, and share their experiences with their peers.

• Regular Communication and Updates: Maintain transparent communication with students through regular updates on safety measures, academic changes, and available support services.

• Online Academic Support: Provide online tutoring, academic coaching, and study skills workshops to help students adapt to the online learning environment and succeed academically.

• Virtual Career Services: Offer virtual career counseling, resume workshops, and networking opportunities to help students plan for their future despite the uncertainty.

• Inclusive Technology Access: Ensure that all students have access to the necessary technology and internet resources to participate in online learning and access support services.

• Promote Self-Care Strategies: Educate students about self-care strategies, time management, and maintaining a healthy work-life balance to prevent burnout.

• Wellness Apps and Resources: Share resources, apps, and online tools that can help students manage stress, anxiety, and maintain their mental well-being.

• Peer Support Networks: Encourage students to form study groups, virtual study sessions, and peer accountability partnerships to stay connected and motivated.

• Faculty–Student Interaction: Encourage professors to maintain open lines of communication with students, providing opportunities for virtual office hours and one-on-one discussions.

• Holistic Health Initiatives: Promote healthy habits through initiatives that focus on nutrition, exercise, sleep hygiene, and overall wellness.

Remember that each institution and student population is unique, so it is important to tailor these interventions to your specific context and the needs of the students. Regular feedback and evaluation of the effectiveness of these interventions will help refine and improve the support provided.

Conflict of interest

The authors declare no conflict of interest.

Institutional review board statement

The study was conducted according to the guidelines of the Declaration of Helsinki. The final questionnaire was presented to the Ethics and Bioethics Commission of the Universidad Autonoma de Baja California for its evaluation and authorization, according to NOM-0035.