Psychometric Evaluation of the Positive and Negative Suicide Ideation Inventory in Ecuadorian Population

Abstract

The Positive and Negative Suicide Ideation (PANSI) inventory is a widely used measure to assess risk and protective factors associated with suicidal ideation. Despite its use in various populations, its psychometric properties have not been evaluated in Ecuador. This study examines the reliability, factorial structure, and convergent validity of the PANSI in a sample of Ecuadorian university students. A total of 702 university students completed the PANSI, the Beck Hopelessness Scale, and the Rosenberg Self-Esteem Scale. Internal consistency was assessed using Cronbach’s alpha and McDonald’s omega. Confirmatory factor analysis (CFA) was conducted to evaluate the factorial structure of the PANSI. Convergent validity was examined through Pearson correlations between PANSI subscales and external measures. Both subscales of the PANSI demonstrated excellent internal consistency (Negative Suicide Ideation: ω = 0.95; Positive Suicide Ideation: ω = 0.82). CFA supported the original 2-factor model, with adequate fit indices (RMSEA = 0.092, SRMR = 0.08, CFI = 0.99, TLI = 0.99). The Negative Suicide Ideation subscale correlated positively with hopelessness (r = .518, P < .001) and negatively with self-esteem (r = −.529, P < .001), while the Positive Suicide Ideation subscale correlated negatively with hopelessness (r = −.557, P < .001) and positively with self-esteem (r = .530, P < .001), supporting convergent validity. The findings indicate that the PANSI is a reliable and valid measure for assessing suicidal ideation in Ecuadorian university students, and arguably, Ecuadorian population. Its ability to capture both risk and protective factors highlights its utility in clinical and educational settings for suicide prevention efforts.

Highlights

● The PANSI demonstrated excellent internal consistency in Ecuadorian university students.

● CFA confirmed the original two-factor model of the PANSI with strong fit indices.

● Negative ideation correlated positively with hopelessness and negatively with self-esteem.

● Positive ideation correlated negatively with hopelessness and positively with self-esteem.

● Findings support the PANSI’s use in suicide prevention within educational and clinical settings in Ecuador.

Introduction

According to the World Health Organization, ¹ approximately 720 000 people die by suicide each year. For every completed suicide, there are many more attempts. Among the general population, a previous suicide attempt is considered the most significant risk factor for future suicidal behaviour. Suicide ranks as the third leading cause of death globally among individuals aged 15 to 29, with 77% of suicides occurring in low- and middle-income countries. ¹ Methods such as pesticide ingestion, hanging, and firearms are commonly used.^2,3 In Ecuador, self-inflicted injuries are among the leading causes of death in young people, with rates of suicidal ideation, planning, and attempts notably higher among adolescents experiencing emotional crises. ⁴ Among adolescents hospitalised for suicidal behaviour, cutting and ingestion of psychotropic drugs were the most frequent methods, predominantly among females. Depression remains a primary contributor, as confirmed by findings from studies on adolescent populations in Ecuador, where depressive episodes were associated with 73% of suicidal cases. ⁵ These patterns emphasise the critical need for targeted interventions and prevention strategies to address this significant public health issue.

Suicidal ideation is a symptom related to thought content that may manifest as a suicide attempt. The Diagnostic and Statistical Manual of Mental Disorders ⁶ defines a suicide attempt as a self-initiated series of actions with some expectation of death as the outcome. The “initiation moment” refers to when an individual engages in suicidal behaviour with the intent to cause death.^6,7

Suicidal ideation is often accompanied by excessive worry and is associated with depression, somatic symptoms, irritability, sleep disturbances, restlessness, anxiety, and impairments in significant areas of the individual’s life. It is also frequently observed in conjunction with a range of mental and physical conditions. ⁸ Furthermore, early identification of these risk factors is crucial for suicide prevention efforts. ⁹

Measures for the detection of suicidal ideation can help in its diagnosis. Because of this, several instruments have been created for its assessment, such as the Beck Hopelessness Scale ¹⁰ (BHS), the Columbia-Suicide Severity Rating Scale ¹¹ (C-SSRS), and the Suicide Ideation Questionnaire ¹² (SIQ). These instruments primarily focus on the severity and frequency of suicidal thoughts but often lack an explicit assessment of protective factors against suicide. The Positive and Negative Suicide Ideation ¹³ (PANSI) inventory was selected for this study because it provides a unique 2-dimensional approach, measuring both risk and protective factors associated with suicidal ideation. This distinction allows for a more comprehensive understanding of suicidal thoughts, particularly in populations where protective factors play a crucial role in prevention efforts.

Te PANSI scale was originally developed in English for the detection of suicidal ideation. On its development, the inventory was meant to evaluate suicidal ideation as a single construct, however, the correlated 2-factor model presented a better fit of the data. Although both factors presented a significant correlation (r = −.51), it is clear from the data and the content of the items that the constructs are not the same, with positive items being more related to positive feelings about one-self, the future and life in general, while the negatively worded items are directly related to suicidal ideation. Items of this original version are presented in Table 1.

Table 1.

Items from the PANSI scale divided by factors.

Factor 1. Negative Suicide Ideation	Factor 2. Positive Suicide Ideation
7. Thought about killing yourself because you could not find a solution to a personal problem?	14. Felt confident about your plans for the future?
11. Felt so lonely or sad you wanted to kill yourself so that you could end your pain?	12. Felt confident about your ability to cope with most of the problems in your life?
10. Thought that your problems were so overwhelming that suicide was seen as the only option to you?	6. Felt hopeful about the future because things were working out well for you?
5. Thought about killing yourself because you could not accomplish something important in your life?	8. Felt excited because you were doing well at school or at work?
3. Felt hopeless about the future and you wondered if you should kill yourself?	2. Felt that you were in control of most situations in your life?
9. Thought about killing yourself because you felt like a failure in life?	13. Felt that life was worth living?
1. Seriously considered killing yourself because you could not live up to the expectations of other people?
4. Felt so unhappy about your relationship with someone you wished you were dead?

This questionnaire not only presented good psychometric properties in the original study by Osman et al., ¹³ but several other, and more recent studies, have also demonstrated the adequacy of the scale in different populations including adolescents ¹⁴ and in different languages. For instance, a study conducted in Malaysia ¹⁵ focussed on evaluating the psychometric properties of the PANSI in a sample of clinical outpatients. The study involved 283 psychiatric patients and 200 medical (non-psychiatric) patients. Results from confirmatory factor analysis supported a correlated 2-factor model, distinguishing between positive and negative suicide ideation. The internal consistency was strong, with Cronbach’s alphas of .93 for negative ideation and .84 for positive ideation. Moreover, the PANSI demonstrated high construct validity, correlating with 7 other self-report instruments and good discriminatory power between psychiatric and medical patients.

Good psychometric properties have also been found in Portuguese population. ¹⁶ In this case, the study included 259 young adults. Confirmatory factor analysis yielded excellent psychometric results, and internal consistency values were very high (0.96 for negative ideation and 0.84 for positive ideation). The Portuguese version of the PANSI also exhibited good discriminative ability, particularly in predicting previous suicide attempts.

In Latin America, however, the validation and application of the PANSI remain limited, though its adaptation and use have the potential to make significant contributions to suicide prevention in the region. Although the study by Muehlenkamp et al. ¹⁷ included some Latin American participants, these only represented a small proportion of the sample. However, psychometric properties of the PANSI can be found published in Spanish, including Colombian^18,19 and Peruvian ²⁰ population.

These studies underscore the instrument’s value for assessing both risk and protective factors for suicidal ideation across diverse clinical and non-clinical populations. However, most of these studies have primarily focussed on evaluating its factorial structure, often neglecting to compare its performance with other proxies of suicidal ideation. This limited scope may constrain the understanding of the construct validity of the instrument, as it is crucial not only to confirm its internal structure but also to establish its association with other relevant measures of suicidal ideation. Moreover, no study has yet investigated the psychometric properties of the PANSI in Ecuador, where suicide is a pressing public health issue, particularly among young people. The current study aims to address this gap by evaluating several psychometric properties of the inventory, namely, its reliability, factorial structure, and correlations with other scales in an Ecuadorian sample. Our results will provide valuable insights into the instrument’s functioning and contribute to the validity of future studies on suicide risk in this context.

Methods

The present study employed a cross-sectional, psychometric validation design to assess the reliability, factorial structure, and convergent validity of the PANSI inventory. Data were collected through an online survey from a sample of Ecuadorian university students. Confirmatory factor analysis (CFA) was conducted to evaluate the scale’s structural validity, and internal consistency was assessed using Cronbach’s alpha and McDonald’s omega. Pearson correlations were used to examine convergent validity with measures of hopelessness and self-esteem.

This study adheres to the Journal Article Reporting Standards (JARS) for quantitative research as outlined by the American Psychological Association. ²¹ The study design, data collection, analysis, and reporting procedures were conducted following these guidelines to ensure transparency, replicability, and methodological rigour.

Participants

The sample comprised a total of 702 Ecuadorian university students from the city of Guayaquil Ecuador between the ages of 18 and 35 (mean age = 22.1, SD = 4.27) from both a Public University (n = 332) and a Private University (n = 371). The sample included 250 male participants, 438 female participants and 14 individuals who preferred not to disclose their sex. All participants self-reported no previous diagnosis of a mental disorder, as assessed through a screening question. No other inclusion or exclusion criteria were applied.

A formal sample size calculation was not conducted prior to data collection. However, based on recommendations for psychometric validation studies, a minimum of 200 participants per factor is generally considered appropriate for CFA. ²² Given that the PANSI comprises a 2-factor structure, the final sample size of 702 exceeds these recommendations, providing adequate power for statistical analyses. Nevertheless, the absence of a formal a priori power analysis is acknowledged as a limitation, which is further discussed in the Discussion section.

Instruments

To evaluate convergent validity, the BHS and the Rosenberg Self-Esteem Scale ²³ (RSES) were selected as criterion measures due to their well-established theoretical and empirical associations with suicidal ideation. The BHS assesses negative expectations about the future, a known predictor of suicide risk, making it relevant for examining the relationship between hopelessness and negative suicidal ideation. The RSES, which measures global self-esteem, was included to assess its association with both risk and protective factors of suicidal ideation, given that lower self-esteem has been linked to increased suicidal thoughts, while higher self-esteem serves as a protective factor.

Positive and Negative Suicide Ideation (PANSI)

This questionnaire was designed and validated by Osman et al. ¹³ to assess suicidal ideation through a 2-factor structure, consisting of 14 items. The scale includes 6 items measuring positive suicidal ideation, which assess protective factors such as a sense of control over one’s life, self-confidence, positive expectations for the future, feelings of joy, and overall life satisfaction. The remaining 8 items assess negative suicidal ideation, capturing risk factors related to suicidal thoughts and emotions, including feelings of failure, hopelessness, low self-efficacy, frustration, and sadness. Participants respond to each item based on how often they have experienced these thoughts over the past 2 weeks, using a 5-point Likert scale ranging from 0 (never) to 4 (always). Higher scores on the negative ideation subscale indicate a greater presence of suicidal thoughts, whereas higher scores on the positive ideation subscale reflect stronger protective factors. The 2 subscale scores are calculated separately and interpreted independently. The PANSI has demonstrated strong psychometric properties across different populations and has been adapted into various languages. For the present study, we used the Spanish adaptation by Villalobos-Galvis, ¹⁹ which has been validated in Latin American samples.

Beck Hopelessness Scale

The BHS is a 20-item self-report inventory developed to measure negative expectations about the future. ¹⁰ Each item is a statement that participants rate as true or false, reflecting their thoughts and feelings over the past week. The BHS assesses 3 major aspects of hopelessness: feelings about the future, loss of motivation, and future expectations. The total score ranges from 0 to 20, with higher scores indicating greater levels of hopelessness. The scale has demonstrated good internal consistency, with Cronbach’s alpha coefficients typically ranging from .82 to .93. In our case, we found both α and ω to be .84. It has also shown good test-retest reliability and has been validated across various populations, including clinical and non-clinical samples. The BHS is widely used in both clinical practice and research to assess hopelessness, which is a critical predictor of suicidal ideation and behaviour. It is an essential tool for identifying individuals at risk and for evaluating the efficacy of therapeutic interventions aimed at reducing hopelessness.

Rosenberg Self-Esteem Scale

The RSES is a widely used measure to assess self-esteem in adolescents and adults. It consists of a 10-item scale that explores the individual’s perception of himself, including aspects such as self-assessment, self-affirmation, and self-acceptance. Each item presents 4 response options on a scale between 0 and 3, from totally agree to totally disagree. Total scores can range from 0 to 30, with higher scores indicating more positive self-esteem. This scale has been validated across diverse populations, demonstrating strong psychometric properties, including high internal consistency and factorial stability. ²⁴ For the current study, we obtained 0.83 for both α and ω coefficients. Additionally, cross-cultural research has confirmed its construct validity, making it a reliable measure of global self-esteem across different linguistic and cultural contexts. ²⁵

Procedure

The study was advertised through targeted email campaigns sent to potential participants within the participating universities from Guayaquil between 14 March 2023 and 29 April 2023. The emails included a brief description of the study’s aims, information about the voluntary nature of participation, and a hyperlink to the online survey platform hosting the questionnaires. Upon accessing the link, participants were presented with an informed consent form outlining the study’s objectives, procedures, confidentiality measures, and the right to withdraw at any time without consequences. Only participants who provided explicit consent by clicking “Accept” were granted access to the questionnaires.

The questionnaires were administered online and could be completed at the participants’ convenience. To ensure privacy, no identifiable information (eg, names, emails, or IP addresses) was collected. All responses were securely stored in an encrypted online database accessible only to the principal investigator. This database complied with data protection regulations to safeguard participants’ confidentiality and anonymity.

Analyses

The psychometric properties of the PANSI were investigated in several steps. First, internal consistency was evaluated to determine the reliability of the 2 PANSI subscales (Negative and Positive Suicide Ideation) and the overall score using Cronbach’s α and McDonald’s ω coefficients. Reliability was interpreted based on standard guidelines, with values above 0.70 considered acceptable, 0.80 good, and 0.90 excellent.

Next, the factorial structure of the PANSI was assessed using CFA to examine the original 2-factor correlated model. The analysis employed the diagonally weighted least squares (DWLS) estimator, which is appropriate for ordinal data. Model fit was evaluated using multiple fit indices, including the chi-square statistic ²⁶ (X²), the root mean square error of approximation ²⁷ (RMSEA), the standardised root mean square residual ²⁸ (SRMR), and incremental fit indices such as the Tucker-Lewis Index ²⁹ (TLI) and Comparative Fit Index ³⁰ (CFI).

Following conventional guidelines, acceptable model fit was determined based on RMSEA values below 0.08 for acceptable fit and below 0.05 for good fit, ³¹ SRMR values below .08, ³⁰ and TLI and CFI values above 0.90 for acceptable fit and above 0.95 for good fit.^28,30 Factor loadings and the correlation between the 2 latent factors were also examined to assess the adequacy of the model. Items were retained if their factor loadings were above 0.50, indicating strong item-level contributions to their respective factors. ³²

Finally, convergent validity was evaluated by examining the Pearson correlation coefficients between the PANSI subscales and 2 external constructs: the BHS and the RSES. These measures were selected as proxies of suicidal ideation and protective factors, respectively. The strength and direction of the correlations were used to evaluate the theoretical relationships between the constructs, providing evidence for the validity of the PANSI in capturing both risk and protective dimensions of suicidal ideation.

Results

Internal Consistency

Results from the internal consistency analyses demonstrated good reliability for both individual scales (Negative Suicide Ideation’s ω = 0.95; Positive Suicide Ideation’s ω = 0.82) and overall score (ω = 0.9). Cronbach’s α coefficients were identical to those obtained from the ω in all cases.

Factorial Structure

We first tested the original structure of the PANSI as a 2 correlated factors model. This analysis was performed using a CFA with the DWLS estimator. Results from the chi-square analysis indicated a poor fit of the model (X² = 522.041, df = 76, P < .001). However, other fit indices provided evidence of a good fitting model, with the RMSEA = 0.092, SRMR = 0.08, CFI = 0.99 and TLI = 0.99. All factor loadings were above 0.5 for both factors (see Figure 1), with the first factor, negative suicide ideation, presenting the highest loadings. The loadings of items on the non-target factor, that is, the factor not primarily intended for evaluation, were all below 0.4.

Figure 1.

Two Correlated Factors Model.

Note. The figure shows a CFA with 2 latent factors, Fc1 and Fc2. Factor loadings range from 0.78 to 0.94 for Fc1 and from 0.54 to 0.89 for Fc2. Error variances for each item are displayed at the base of the items. Fc1 and Fc2 have a correlation of −.45. The latent factors are standardised with a variance of 1.00 each.

Convergent Validity

Convergent validity was evaluated by calculating Pearson correlation coefficients between both scales from the PANSI and the total scores of the BHS and the RSES. The results showed that the PANSI Negative Suicide Ideation scale had a significant positive correlation with the BHS (r = .518, P < .001). Conversely, the PANSI Positive Ideation scale demonstrated a significant negative correlation with the BHS (r = −.557, P < .001). Regarding self-esteem, the PANSI Negative Suicide Ideation scale showed a significant negative correlation with the RSES (r = −.529, P < .001). Similarly, the PANSI Positive Ideation scale showed a significant positive correlation with the RSES (r = .530, P < .001).

Discussion

The present study aimed to evaluate the psychometric properties of the PANSI in a sample of Ecuadorian university students, providing evidence for its use in Spanish-speaking contexts and, specifically, in Ecuador, where suicide represents a significant public health issue. The results obtained support the use of the questionnaire in this population, both in terms of internal consistency, and construct validity.

First, the internal consistency analyses showed that both PANSI subscales, as well as the overall score, exhibit excellent reliability. This finding is consistent with previous studies conducted in different cultural contexts, such as Malaysia ¹⁵ and Portugal, ¹⁶ where similar levels of internal consistency were reported. Regarding the construct validity, it was assessed both by analysing its factorial structure and its convergence with constructs expected to be related to suicide risk and protective factors.

Results from the confirmatory analyses supported the original model of 2 correlated factors, despite a poor fit indicated by the chi-square statistic. However, other fit indices, such as RMSEA, SRMR, TFI, and TLI, suggested that the model is adequate. These results are congruent with previous studies^13,15 and underscore the validity of the negative and positive suicidal ideation dimensions as distinct yet related constructs. The high factor loadings of the items in both dimensions reinforce the model’s stability in this population. Additionally, the negative correlation between the factors (−.45) supports the hypothesis that protective and risk factors operate inversely.

With regard to convergent validity, evaluated through correlations between the PANSI subscales and other psychological measures such as the BHS and the RSES, the results showed significant and moderate correlations in the expected directions. The negative subscale of the PANSI showed a moderate positive correlation with the BHS, suggesting that higher levels of negative suicidal ideation are associated with greater hopelessness, a well-known predictor of suicide risk. ¹⁰ Conversely, the positive subscale of the PANSI demonstrated a moderate negative correlation with the same scale, indicating that protective factors are associated with lower levels of hopelessness. Regarding self-esteem, the negative subscale of the PANSI showed a moderate negative correlation with the RSES, while the positive subscale showed a moderate positive correlation with the same scale, highlighting that greater protective factors are linked to higher self-esteem and that negative suicidal ideation is associated with a lower sense of self-worth. These results support the convergent validity of the PANSI and its utility for measuring both risk and protective factors associated with suicidal ideation.

This study has important practical implications for suicide prevention in Ecuador. The validation of the PANSI in this population enables its application in educational and clinical settings, where it could be used to identify individuals at risk and design preventive interventions. Additionally, the focus on protective factors, through the positive subscale, highlights the importance of promoting positive psychological resources as part of prevention efforts. This approach complements traditional strategies focussed solely on risk identification, offering a more comprehensive perspective to address the complexity of suicidal ideation.

Limitations and Future Directions

Despite its contributions, this study presents some limitations that should be considered. First, the sample consisted exclusively of university students from a specific city in Ecuador, which limits the generalisation of the findings to other populations and contexts. Future studies should include more diverse samples that encompass different regions and age groups to assess the cross-cultural validity of the PANSI in other subgroups of the Ecuadorian population. Furthermore, although convergent validity measures were assessed, it would be useful to explore the discriminant validity of the PANSI against other scales that evaluate related constructs, such as depression or anxiety. Additionally, a formal power analysis was not conducted prior to data collection, which may limit the precision in estimating the required sample size. However, the final sample size aligns with standard recommendations for psychometric validation studies, providing sufficient power for the analyses performed.

Supplemental Material

sj-docx-1-inq-10.1177_00469580251337665 – Supplemental material for Psychometric Evaluation of the Positive and Negative Suicide Ideation Inventory in Ecuadorian Population

Supplemental material, sj-docx-1-inq-10.1177_00469580251337665 for Psychometric Evaluation of the Positive and Negative Suicide Ideation Inventory in Ecuadorian Population by Daniel Oleas, José Alejandro Valdevila Figueira, Rocio Valdevila Santiestevan, Rodrigo Moreta-Herrera, Guido Mascialino and Jose A. Rodas in INQUIRY: The Journal of Health Care Organization, Provision, and Financing

Not applicable.

ORCID iDs

Daniel Oleas https://orcid.org/0000-0002-7347-814X

Jose A. Rodas https://orcid.org/0000-0001-8217-0238

Ethical Considerations

The current study was reviewed and approved by the research and ethics committee from Universidad Ecotec (CERTIFICATE No. 10-03-03-2023) on 3 March 2023.

Consent to Participate

All participants provided written informed consent prior to participation.

Consent for Publication

The consent form specified that all data would be used for research purposes and may be used for the publication of an academic article.

Author Contributions

DA conceived the study, developed the methods, analysed the data, and contributed to the original draft of the manuscript. JAVF and RVS participated in drafting the original manuscript. RM was responsible for data analysis and reviewed the final version of the article. GM contributed to writing the final version of the article. JAR worked on the methodological aspects of the study, as well as drafting and revising both the original manuscript and its final version.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The publication of this article was funded by Universidad de las Américas, Ecuador.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Supplemental Material

Supplemental material for this article is available online: https://doi.org/10.17605/OSF.IO/BVK8X