1. Background
Understanding death is one of the most important human queries, which is unresolved until the present (1). The fact that everyone will inevitably face death at one point in life can cause significant stress and anxiety in individuals (2). The anxiety about one’s own death might manifest in some individuals as a continuous and abnormal fear (3). This anxiety, as the initial component of death, includes a whole host of personal views, such as general fear, threat, worries, sense of discomfort, and other stressful reactions (4). With the advent of coronavirus disease 2019 (COVID-19), the overall anxiety in society is increased, which is to some extent due to the disease-related deaths reported to be 4.3% (5). Dadfar et al. described fear of death as a negative and threatening feeling that conjures up when the individual thinks about one’s own death. This is known as a fundamental fear and a broader part of psychiatric conditions, including clinical anxiety and depressive disorders (1).
The terms “fear of death” and “death anxiety” are often used interchangeably in the literature (6), and cross-cultural studies have proposed important methods for the determination of many unique phenomena, including death anxiety (7). Over the past 50 years, numerous advances have been made in personal views of death and assessment of individual worries about it, leading to the development and validation of some scales (8). Applying these scales enables us to improve upon the current views and applicability of the existing assessment tools to the concept of dying.
Death might be perceived differently in various cultures. These differences might be due to individual mental and emotional characteristics, culture or belief systems, which in turn form our views of death either consciously or unconsciously (9). Charmaz (10) believes that the role of culture or religion is to help individuals interpret the reality of death in a way to allay their fears and direct their attention away from the disabling sense of hopelessness. Therefore, it is logical to expect that fear of death might not be the same in various societies, considering their social and cultural characteristics. To date, several self-reporting questionnaires and scales have been developed and used for the assessment of death anxiety and its influence on psychological health (11); there are more than 10 measures regarding death studies currently available.
The scales for the assessment of death anxiety are either unidimensional or multidimensional. The most well-known scale for death anxiety is Templer unidimensional scale (12). Furthermore, several death anxiety scales have been used and validated in psychological studies in Iran, including the Templer scale, Collett-Lester Fear of Death Scale (CL-FODS), Reasons for Death Fear Scale, and Arabic Scale of Death Anxiety. Since death is a multidimensional concept, it is not easy to describe it briefly. A tool is required capable of addressing death and individuals’ reactions to it from different perspectives.
Recently, multidimensional scales of death anxiety have become more widely recognized (13). Currently, several multidimensional scales exist, with the most practical and applied ones including the CL-FODS (14), the Multidimensional Fear of Death Scale (15), and the Fear of Personal Death Scale (FPDS) by Florian and Kravetz (16). The FPDS by Florian and Kravetz is a self-report scale based on the multidimensional fear of death anxiety model. Multidimensional scales show the qualitative differences in the meaning individuals might attribute to their own mortality. Knowledge of such differences might be essential to comprehend the complexity of fear of death (17). Therefore, such scales are valuable instruments for cross-cultural studies and comparisons among Abrahamic religions and expansion of our knowledge on the factors that might be related to fear of personal death.
There are convincing reasons to select and translate the revised version of the Fear of Personal Death Scale (FPDS-R), modified by Burris and Bailey (18), into the Persian language and to study its psychometric properties. Firstly, the FPDS-R is multidimensional; however, its emphasis is on one’s own death, which is in contrast to the CL-FODS, in which half of the items are concerned with the death of others (19). The second reason for selecting the FPDS-R was that this questionnaire sees dying as the end of one’s own life (i.e., anxiety about self-annihilation or existential fear of death), which is distinctively different from other concepts associated with death (13), which was of our interest to measure this aspect of fear in the Iranian Muslim population. Thirdly, this scale offers a more accurate factor structure about the fear of personal death than the Holter scale. The Holter 8-factor questionnaire is also a multidimensional scale; nevertheless, it is not based on a specific theory (20). The Holter scale was used in Arab Muslim students who lived temporarily in the United States, which showed that being a Muslim might have prevented or minimized the fear of death in the believers of that religion. Long (21) have mentioned that “With our sample, Mohammedanism might simply preclude certain fears pertaining to death and dying that Americans have been found to hold”. Therefore, the Holter scale might not be appropriate for individuals belonging to either Arab or Iranian culture only because a common religion exists among them. Furthermore, individuals’ mentality is complex, containing a whole host of beliefs originating from their culture and religion, if not also other perspectives (22).
The model of the FPDS, originally proposed by Florian and Kravetz (16), suggests that the overt expression of fear of death is composed of three components and six factors. Components comprising the intrapersonal, interpersonal, and transpersonal consequences of death were considered the theoretical base. Two factors, namely deprivation and loss of self, correspond to the intrapersonal component of fear of death. The interpersonal component shows concerns related to attachment relationships, including fear of loss of one’s identity related to forgotten and other-oriented factors. The transpersonal component is reflected in fear of the unknown nature of death, represented by the mystery and punishment factor.
The FPDS revised by Burris and Bailey in 2009 (18), which is validated by the current study, minimizes the ambiguities that occurred initially during its translation from the original Hebrew version to English. Of note, Burris and Bailey (18) have pointed out that one factor (i.e., punishment or retribution in the hereafter) has been omitted in the English version since it has been considered an established fact; this might imply the assumption that humans’ awareness and most probably identities will endure after death, leading to the omission of the aforementioned factor. Upon a second review of the questionnaire, it was clear that the sixth factor was related to the acceptance of the idea of retribution in the hereafter, with a factor loading of 0.59; therefore, its omission was logically justified.
The Beck Depression Inventory (BDI) and Rosenberg Self-esteem Scale (RSES) questionnaires were used to assess the concurrent validity of the scale in the present study. A positive correlation was expected between depression and death anxiety. However, there was a negative relationship between self-esteem and death anxiety. Based on terror management theory, self-esteem is believed to serve as a buffer against death anxiety (23), and currently, a common agreement exists on the role of self-esteem (24, 25). Evidently, severe death anxiety is usually associated with psychiatric disorders, such as clinical depression and generalized anxiety (26, 27). In epidemiological studies, death anxiety has been confirmed in patients with severe clinical depression (28).
Considering the above-mentioned literature review regarding the paucity of such studies in the Iranian population, the present study aimed to evaluate the psychometric properties, factor structure, and concurrent validity of the FPDS-R in a typical Iranian population. For this purpose, the FPDS-R was systematically translated into the Persian language, and its validity, factor structure, and reliability were evaluated.
2. Objectives
To date, such a study has not been conducted on an Iranian population; therefore, the authors believe this study contributes positively to the available literature on individual personality and identity. The aforementioned measures would be helpful in clinical assessments and interventions.
3. Methods
3.1. Study Participants and Recruitment and Setting
This descriptive correlational study was performed on recruited volunteer subjects through convenient sampling from individuals residing in Tehran, Iran, including students of Tehran and Alameh Tabatabaee universities, through online contact due to the COVID-19 pandemic within April and November 2020. Initially, the scale was tested on six volunteers to ensure that the items were clearly understandable by the users and to detect potential ambiguities. For the purpose of studying construct validity, confirmatory factor analysis (CFA) was performed on the Persian version of the FPDS-R by AMOS software (version 22). The Pearson’s correlation and Cronbach’s alpha coefficients of the tool were calculated by SPSS software (version 21).
3.2. Sample Size and Sampling Method
The initial sample of 313 participants aged 18-71 years underwent the final implementation of the FPDS-R. This number of samples was taken into consideration in accordance with Meyers et al.’s advice (29). They believe that the number of 313 individuals is appropriate for the component analysis of this scale, which has 30 items, and that the subject should be taken into account roughly 10 times for each item.
The inclusion criteria included an age range of 18 - 71 years, undergrad students or above, and informed consent to participate in the study. The exclusion criteria included having a serious mental illness, such as psychosis or bipolar disorder, taking antipsychotic medications, having a suicide history, drug or stimulant addiction, the existence of several physical conditions (e.g., cancer, lupus, multiple sclerosis and acquired immunodeficiency syndrome)
The number of subjects was determined statistically based on the sample size recommended by Meyers et al. (29).
3.3. Translation Process
The translation of the FPDS-R questionnaire followed the World Health Organization protocol and went through a systematic translation and back-translation for accuracy and content validity. Firstly, the original English version was translated into Persian by the authors. Secondly, this version was back-translated into English by an Iranian psychology professor residing in the United States for the past 40 years. As part of the scrutiny, initially, two expert psychologists and academicians from the Department of Psychology of University of Tehran reviewed the questionnaire’s contents and verified the validity and the ease of comprehension of each item. Then, the semi-final questionnaire was reviewed by another five psychology professors from the same university (including two clinical psychologists, two general psychologists, and one health psychologist) for content validity and feedback. Upon receiving their feedback, the questionnaire was revised minimally, and the final version was saved on a computer. The concurrent validity of the final Persian scale was assessed using the RSES and BDI.
3.4. Instruments, Validity, and Reliability
Primarily, each subject reviewed the study protocol and signed an informed consent form. Subsequently, the subjects filled out the online study’s main questionnaire (i.e., the Persian version of the FPDS-R). Before responding to the questionnaire’s items, the subjects were asked to record demographic information, such as gender, economic status, and educational level. However, they were told not to mark their names or identities anywhere in the questionnaire at all; therefore, anonymity and confidentiality were guaranteed.
The reliability assessment of the data on death anxiety, performed among the university’s PhD students (n = 30) in psychology and undergraduate students from Tehran University of Arts, resulted in a Cronbach’s alpha of 0.9, demonstrating a good internal consistency. The obtained findings indicated that the questionnaire’s items had a high correlation with each other, representing their consistent underlying construct.
3.5. Revised Version of Fear of Personal Death Scale
This 30-item validated Persian questionnaire was the equivalent of the original FPDS-R. It asked the subjects to indicate their choice about their own death on a Likert scale of 1 - 7, with a 1 (totally incorrect) to 7 (totally correct) response format. The Cronbach’s alpha coefficients for the five subscales were excellent (deprivation = 0.93, forgotten = 0.87, other-oriented = 0.87, mystery = 0.91, and loss of self = 0.90) (18).
3.6. Rosenberg Self-esteem Scale
The RSES (1965) measures overall self-esteem and personal value, including 10 general items that measure life satisfaction and a good sense of self (30). The RSES is one of the most common scales for measuring self-esteem (30, 31). In this questionnaire, the subject is asked to answer the questions accurately on a 4-point Likert scale from strongly agree to strongly disagree. The scores on this scale ranged from 10 to 40, with higher scores indicating higher self-esteem. This scale includes five positive items (numbers 1 - 5) and five negative items (numbers 6 - 10). Recent discussions have focused on using the RSES as a single structure or two-factor structure. Sharratt et al. (32) believe the two-factor structure is a better way to apply this scale in social studies. A study by McKay et al. (33) also confirmed the two factors of this structure. Kielkiewicz et al. (34) reported Cronbach’s alpha coefficients of 0.66 and 0.79 for positive and negative self-esteem, respectively. In this study, internal consistency was calculated using Cronbach’s alpha coefficient for positive self-esteem (0.85) and negative self-esteem (0.80).
3.7. Beck Depression Inventory-Second Edition
This 21-item questionnaire evaluates the person’s depression severity with excellent validity, reliability, and Cronbach’s alpha coefficients of 0.92 and 0.93 when tested on outpatients and students, respectively (35). Each item is scored from 0 to 3, with the maximum potential score upon the completion of the questionnaire reported as 63 (36). The items are categorized under three sections, including emotional, cognitive, and physical symptoms. In the present study, internal consistency through Cronbach’s alpha coefficient was calculated at 0.88.
4. Results
4.1. Subjects’ Demographic Features
The current study was conducted mainly to determine the psychometric properties of the FPDS-R in a Persian-speaking population. This study was performed on 313 participants aged 18 - 71 years (mean: 35.12; standard deviation: 12.06), including 230 women (73.5%) and 83 men (26.5%). Slightly over half of the participants were single (n = 171; 54.6%), and the rest were married (n = 142; 45.34%). Regarding their educational status, 16% of the subjects (n = 50) had a high school diploma, and the rest completed their university education with a Bachelor of Arts or a Bachelor of Science (n = 111; 35.5%), master’s degree (n = 105; 33.5%), or doctoral levels (n = 47; 15%), respectively. Before implementing CFA, the assumptions of Keiser-Meyer-Olkin sampling adequacy and Bartlett’s test of sphericity were met (χ2 = 6202.915) (P < 0.001). The analysis indicated that the scale’s items were appropriate and capable of measuring the factors.
4.2. Confirmatory Factor Analysis
The CFA, consisting of five latent variables, was employed to determine the fitness of the scale for the experimental data. The fitness of the final CFA was acceptable (chi-square/df = 2.345; comparative fit index [CFI] = 0.913; incremental fit index [IFI]=0.914; Tucker-Lewis index [TLI] = 0.902; root mean square error of approximation [RMSEA] = 0.066). Table 1 shows details of the fitness indices.
| Goodness of Fit Indices | Chi-square | Chi-square/df | RMSEA | SRMR | CFI | IFI | TLI |
|---|---|---|---|---|---|---|---|
| Five-factor model | 905.260 | 2.345 | 0.066 | 0.080 | 0.913 | 0.914 | 0.902 |
| Acceptance cutoff | - | < 3 | < 0.08 | < 0.08 | > 0.90 | > 0.90 | > 0.90 |
Goodness of Fit Indices for Default Model
To determine the fitness of the hypothesized model, the absolute and comparative fit indices were used, of which RMSEA and standardized root mean square residual (SRMR) account for the most popular ones. The ideal model fitness value for RMSEA should be less than 0.1 and preferably less than 0.08. For SRMR fitness, the value must be less than 0.08 (37). For CFI, TLI, and IFI indices, values higher than 0.9 indicate an acceptable model; however, values higher than 0.95 account for good fitness (37). The CFA was carried out using AMOS software (version 22).
As observed in Table 2, based on the values obtained from the t-test, all of the items under the five factors (i.e., deprivation, loss of self, forgotten, other-oriented, and mystery) were higher than 1.96, which indicates the statistical significance of all paths (P ≤ 0.001) (Table 2).
| Factors and Items | β | b | Se | t | P-Value |
|---|---|---|---|---|---|
| Deprivation | |||||
| 9 | 0.597 | 0.780 | 0.075 | 10.346 | 0.001 |
| 4 | 0.603 | 0.842 | 0.081 | 10.450 | 0.001 |
| 8 | 0.583 | 0.778 | 0.077 | 10.155 | 0.001 |
| 7 | 0.682 | 0.941 | 0.079 | 11.893 | 0.001 |
| 11 | 0.788 | 1.115 | 0.081 | 13.831 | 0.001 |
| 3 | 0.809 | 1.240 | 0.087 | 14.224 | 0.001 |
| 5 | 0.742 | 1.042 | 0.080 | 12.987 | 0.001 |
| 10 | 0.798 | 1.156 | 0.082 | 14.027 | 0.001 |
| 6 | 0.858 | 1.223 | 0.081 | 15.122 | 0.001 |
| 2 | 0.711 | 0.979 | 0.046 | 21.096 | 0.001 |
| 1 | 0.730 | 1 | Reference indicator | ||
| Loss of self | |||||
| 30 | 0.544 | 0.532 | 0.049 | 10.868 | 0.001 |
| 29 | 0.491 | 0.528 | 0.055 | 9.526 | 0.001 |
| 28 | 0.743 | 0.770 | 0.043 | 17.789 | 0.001 |
| 27 | 0.944 | 1.021 | 0.033 | 31.122 | 0.001 |
| 26 | 0.958 | 1 | Reference indicator | ||
| Forgotten | |||||
| 13 | 0.685 | 1.291 | 0.149 | 8.683 | 0.001 |
| 25 | 0.641 | 1.163 | 0.139 | 8.390 | 0.001 |
| 15 | 0.393 | 0.689 | 0.119 | 5.795 | 0.001 |
| 14 | 0.669 | 1.334 | 0.108 | 12.399 | 0.001 |
| 12 | 0.568 | 1 | Reference indicator | ||
| 16 | 0.759 | 1.468 | 0.160 | 9.178 | 0.001 |
| Other-oriented | |||||
| 20 | 0.688 | 0.883 | 0.070 | 12.634 | 0.001 |
| 19 | 0.796 | 0.911 | 0.061 | 15.010 | 0.001 |
| 17 | 0.781 | 0.947 | 0.064 | 14.691 | 0.001 |
| 18 | 0.840 | 1 | Reference indicator | ||
| Mystery | |||||
| 21 | 0.688 | 0.728 | 0.051 | 14.337 | 0.001 |
| 22 | 0.764 | 0.844 | 0.050 | 16.922 | 0.001 |
| 24 | 0.900 | 0.969 | 0.043 | 22.402 | 0.001 |
| 23 | 0.903 | 1 | Reference indicator | ||
Standardized Coefficients and Significance of Factor Loadings
4.3. Evaluation of Validity and Correlation
The concurrent validity of the Persian version of the FPDS-R was evaluated. The correlation between each of the five factors (i.e., deprivation, loss of self, forgotten, other-oriented, and mystery) with depression or positive and negative self-esteem was determined. As shown in Table 3, there was a positive correlation between each of the four variables (i.e., deprivation, loss of self, forgotten, and mystery) and depression, with the correlation coefficients reported as r = 0.16, 0.25, 0.26, and 0.20, respectively (P < 0.01) (Table 3).
| Variables | Depression | Positive Self-esteem | Negative Self-esteem |
|---|---|---|---|
| Deprivation | 0.162** | -0.003 | 0.123* |
| Loss of self | 0.250** | -0.106 | 0.232** |
| Forgotten | 0.257** | -0.151** | 0.248** |
| Other-oriented | 0.045 | 0.028 | 0.029 |
| Mystery | 0.197** | -0.067 | 0.204** |
Correlations of Depression, Positive Self-esteem, and Negative Self-esteem with Death Concerns a
4.4. Positive and Negative Self-esteem
There was a negative correlation between forgotten and positive self-esteem (r = -0.15; P < 0.01). In addition, there was a positive correlation between the four variables and negative self-esteem, with r = 0.12, 0.23, 0.25, and 0.20, respectively (P < 0.01 and P < 0.05). Depression correlated positively with four factors of deprivation, loss of self, forgotten, and mystery, with r = 0.16, 0.25, 0.26, and 0.20, respectively (P < 0.01).
4.5. Reliability Assessment
In the next phase of the study, the reliability of the scale was evaluated on the main sample (n = 313) for each of the factors (r > 0.7). The Cronbach’s alpha values for deprivation, loss of self, forgotten, other-oriented, and mystery were 0.93, 0.88, 0.81, 0.86, and 0.88, respectively, which confirmed their appropriate reliability according to Hair et al. (38). Of note, the three versions of the questionnaire in Hebrew, English, and Persian languages resulted in similar Cronbach’ alpha values.
5. Discussion
This study was conducted to evaluate the validity, reliability, and factor structure of the Persian version of the multidimensional FPDS-R in an Iranian population. For the reliability assessment, this study used internal consistency through Cronbach’s alpha coefficients. Nevertheless, CFA was employed for the construct validity. Several other studies used Cronbach’s alpha coefficients and exploratory factor analysis (EFA) and CFA for the reliability and validity assessment of the questionnaire (39). Finally, concurrent evaluations were assessed using the BDI and RSES.
5.1. Consistency with Earlier Studies
The factor structure obtained in the current study was similar to that of Buriss and Bailey (18). Regarding fitness indices of the present study on five factors proposed by Burris and Bailey (18), it is concluded that the same factors are yielded. This finding can be interpreted as the similarity of the beliefs of Muslims, Christians, and Jews about personal death based on the commonality of Abrahamic religions.
The BDI and RSES questionnaires were used to evaluate the concurrent validity of the FPDS-R. Consistent with the findings of previous studies, the five factors and the components of the original scale (i.e., FPDS-R) have a positive relationship with depression (40), supporting the fact that psychological conditions might exist concurrently with severe death anxiety in individuals (41, 42). Earlier studies confirm the presence of death anxiety in individuals with depression. Individuals who were depressed showed higher than average scores on the Death Anxiety Scale. Researchers also found that fear of death is related to depression and mental suffering, especially in the elderly (43, 44). According to a study by Khaki et al., depression can also significantly affect the quality of life in the elderly. Decreased quality of life can have several adverse consequences, such as diminished life expectancy (45).
5.2. Self-esteem and Death Anxiety
Based on the clinical findings and terror management theory, cultural worldview and self-esteem serve as a buffer against anxiety about one’s own death (23, 46, 47). Furthermore, the research literature based on self-reports provides ample evidence that positive self-esteem is an effective protective mechanism against death anxiety (48). Nevertheless, individuals with low self-esteem often experience deep death anxiety. Experimental evidence supports the fact that individuals with a high sense of self-esteem do not show deep death anxiety upon watching murder videos and death scenes, unlike those with low self-esteem (49, 50). Additionally, Chung et al. have provided evidence that students with positive self-esteem rarely experience death anxiety (51).
The findings of the current study also demonstrated that negative self-esteem has a significantly positive correlation with deprivation, loss of self, forgetting, and mystery. The aforementioned findings are consistent with those reported by several earlier studies (52, 53). According to the study by Waters and Moore, deprivation is harmful to self-esteem in both men and women (54). It is clear that individuals with deprivation have to perform some works and activities that are not suitable for them and consequently have a negative effect on their self-esteem. In addition, Tice concluded that there is a positive correlation between self-esteem and mystery. Individuals with high self-esteem show no mystery; they think they are good at most things, and therefore they would like others to know about their activities. They would like to achieve outstanding, exceptional things. On the other hand, those with low self-esteem are cautious with a high fear rate and therefore are self-protected (55).
The present study also found a negative correlation between positive self-esteem and the factor of forgetting. The existence of this relationship is in line with the results of a study by Chung et al. (51) in this field which showed that students who had high self-esteem experienced less death anxiety. However, the absence of a relationship between other components of death anxiety and self-esteem is not in line with the results of other studies; that is, high and positive self-esteem in the present study’s sample did not have a protective effect on the individual’s sense of death anxiety. To explain this, the concurrence of this study with the COVID-19 pandemic and the elevated level of death anxiety subsequently might have affected the moderating effect of self-esteem on reducing death anxiety and the subjects’ responses. The findings reported by Soleimani et al. in an Iranian sample indicated that the subjects with positive self-esteem experience deeper death anxiety than those with negative self-esteem (56). To explain these inconsistencies, it should be considered that death anxiety is a multidimensional concept linked to individuals’ dominant culture, thereby influencing their belief about death (22, 57). To elucidate this phenomenon, further qualitative and quantitative studies are warranted in Muslim communities, such as Iran, which are distinctively different from the studies of the West.
5.3. Limitations of the Study
The fact that all of the subjects were recruited from the same city might be one of the limitations of the current study. Therefore, the findings might not be generalizable to other populations in Iran or neighboring countries. Another limitation was the lack of test and retest reliability of the findings in the study subjects. This study was conducted online due to the COVID-19 pandemic, which might have negatively affected the subjects and the accuracy of the responses since they could not ask questions regarding the items. The last limitation was not performing EFA on the Persian version of the FPDS-R in this study.
5.4. Recommendations for Future Studies
The results showed that there is only a significant inverse relationship between the component of “forgotten” and positive self-esteem (positive items); nevertheless, the lack of relationship between other components of death anxiety and self-esteem is not consistent with the results of other studies; that is, high and positive self-esteem in the present study’s sample did not have a protective effect against death anxiety. However, the results of Soleimani et al.’s (56) study in the Iranian sample indicated a positive relationship between these two variables, which is also contrary to the findings of other studies. Therefore, it is suggested to repeat this study in the Iranian sample because the relationship might be influenced by other factors, especially cultural and religious factors (57). Conducting only CFA on the sample could limit the generalizability of the results and scope of the study. As a result, it is recommended to perform EFA on another independent sample in the future to examine the factors’ stability. The current study was conducted on healthy participants. It is recommended to plan similar studies on individuals with clinical conditions and the elderly in a reproducible manner.
5.5. Conclusions
Based on the findings, the Persian version of the FPDS-R has the validity and capability required for the psychometric assessment of individuals’ views on their own death in an Iranian sample. The FPDS-R consists of 30 items. The analytical steps identified five subscales, including deprivation (11 items), loss of self (5 items), forgotten (6 items), other-oriented (4 items), and mystery (4 items). The translated version was found to have a similar structure to the revised version by the Burris and Bailey questionnaire and a high level of validity and reliability. The Persian version of the FPDS-R had a similar Cronbach’s alpha to those in the original Hebrew and revised English versions. This tool is also helpful for the identification and part of management of psychological disorders and cross-cultural comparisons of various communities. The five factors and the components of this version, including deprivation, loss of self, forgotten, other-oriented, and mystery, have a positive relationship with depression. In this study, negative self-esteem had a positive relationship with factors including deprivation, loss of self, forgetting, and mystery. However, positive self-esteem did not have a protective effect on the individual’s sense of death anxiety; nevertheless, positive self-esteem had a negative correlation with forgetting.
