Introduction

The future time perspective involves cognitive processes, emotions, and attitudes towards future events, and it may be positive or negative. The positive perspective toward future events is characterized by feelings of hope and a focus on planning and achieving future goals, thus promoting well-being and healthy behaviors. By contrast, the negative perspective is characterized by feelings of worry, fear, and anxiety. These two visions are not mutually exclusive but can be experienced simultaneously. However, when the negative perspective prevails, it may have a negative impact on physical and mental health (^{Carelli et al., 2015}; ^{Zaleski, 1996}, ²⁰⁰⁵; ^{Zaleski et al.}, 2019).

^{Zaleski (1996)} identified future anxiety as one of the basic elements of a negative future time perspective, defining it as a state of apprehension, uncertainty, fear, worry, and concern about unfavorable changes in the future. What distinguishes future anxiety from other types of anxiety is that it refers to a more distant personal future. Furthermore, it is a conscious form of anxiety that may influence thoughts, emotions, and behaviors, increasing the fear of future events and the anticipation of changes perceived as dangerous or adverse. Individuals with higher levels of future anxiety usually fear global disasters (e.g., earthquakes or wars) or painful personal experiences (e.g., losing a loved one, accidents, or illnesses). It is argued that the basis of future anxiety is personality traits that determine how a person responds to fear, personal experiences, and current events (^{Zaleski, 1996}, ²⁰⁰⁵; ^{Zaleski et al., 2019}).

Based on this conceptualization of future anxiety, Zaleski developed the Future Anxiety Scale (FAS; ^{Zaleski, 1996}), a self-report instrument that assesses the tendency to think about the future with uncertainty, anxiety, and fear. The FAS comprises 29 items, each rated on a 7-point Likert-type scale from 0 (decidedly false) to 6 (decidedly true), with higher scores indicating greater concern about the future. ^{Zaleski (1996)} reported satisfactory psychometric properties for the FAS, the scores on which were positively correlated with scores on anxiety, hopelessness, and neuroticism. These associations provide validity evidence based on the relationship of the FAS with other variables and highlight the relationship between future anxiety and personality traits related to anxiety.

More recently, ^{Zaleski et al. (2019)} developed a shorter version of the FAS, the Dark Future Scale (DFS), in a sample of Polish adults. The DFS comprises just 5 of the 29 items in the FAS, and it employs the same 7-point Likert-type rating scale (from 0, decidedly false to 6, decidedly true). Confirmatory factor analysis supported a single-factor structure, and reliability based on Cronbach's alpha was .90. Scores on the DFS were positively and strongly correlated with scores on the Future Negative scale (FN; ^{Carelli et al., 2011}), and positively and moderately associated with scores on the past-negative and present-fatalistic subscales of the Zimbardo Time Perspective Inventory (ZTPI; ^{Zimbardo & Boyd, 1999}). Analysis by gender showed that females scored higher than males on future anxiety.

Although developed only recently, the DFS has already been applied in different populations, including samples from Poland (^{Duplaga & Grysztar, 2021}; ^{Sobol et al., 2020}), Italy (^{Rapelli et al., 2021}; ^{Scandurra et al., 2021}), Germany (^{Dadaczynski et al., 2021}), Canada (^{Watson, 2020}), Turkey (^{Armağan & Durukal, 2021}), and other Asian countries, namely Iran, China, the Philippines, and Singapore (^{Leung et al., 2021}; ^{Shabahang et al., 2021}). It has also been used in a cross-national study involving emerging adults from China, Italy, Lithuania, Portugal, Slovenia, and the USA (^{Lanz et al., 2021}). The findings overall suggest increasing levels of future anxiety, with higher scores in females, and people who score higher on the DFS (indicating a more negative perception of the future) tend to have lower physical and psychological well-being.

Studies that have focused specifically on young adults have found that they report, among other things, psychological distress, uncertainty, anxiety, and future worries, with future anxiety being related to online news addiction and to an increased likelihood for multiple health complaints (^{Dadaczynski et al., 2021}; ^{Lanz et al., 2021}; ^{Shabahang et al., 2021}). The perception of the future is critical for young adulthood, because it is during this period that individuals make important life decisions such as choosing to embark on university studies, looking for their first job, or deciding where they will live. Accordingly, a fear of academic failure or worries about a lack of job opportunities may produce anxiety about what lies ahead. From the research perspective, there is also a growing interest in the study of future anxiety among young adults due to the health, social, and economic changes and challenges that society is facing (e.g., ^{Dadaczynski et al.}, 2021; ^{Lanz et al.}, 2021).

Given this interest, there is a need for adequate tools to measure future anxiety. The DFS is a brief and reliable instrument that, based on the aforementioned research, would appear to be suitable for this purpose. However, instrumental studies analyzing its psychometric properties are scarce, and validity evidence is limited and based solely on its relationship to other future perspective scales. In addition, measurement invariance across gender has yet to be examined. Although ^{Zaleski (1996)} found that scores on the FAS correlated with personality traits such as neuroticism, there is no published evidence of this association with DFS scores. Further studies are therefore warranted to provide more evidence for their reliability and validity. Another issue to consider, with regard to our cultural context, is that there is currently no validated Spanish version of the DFS.

In light of the above, the aim of the present study was to develop and validate a Spanish version of the DFS (hereinafter, the DFS-S), providing evidence of its psychometric properties in a sample of young adults. The process of adapting the DFS was carried out in accordance with the Standards for Educational and Psychological Testing (^{American Educational Research Association et al., 2014}) and the recommendations of the ^{International Test Commission (2017)}. First, confirmatory factor analysis was used to provide validity evidence based on the instrument's internal structure, including measurement invariance across gender. We then examined item homogeneity, the reliability of test scores, and obtained validity evidence based on the relationship with other variables. Given that some people have a predisposition to react with fear to a variety of life situations and that personality traits are postulated to be the basis of future anxiety (^{Zaleski, 1996}; Zaleski et al., 2019), we sought to obtain validity evidence based on the correlation between DFS-S scores and scores on the five core personality factors: extraversion, agreeableness, conscientiousness, neuroticism, and openness. Our main expectation here was that DFS-S scores would be positively associated with scores on neuroticism. Finally, gender differences in DFS-S scores were examined. In accordance with ^{Zaleski et al. (2019)}, we expected DFS-S scores to be higher among females.

Material and Method

Results

Descriptive Item Statistics

Table 1 displays mean scores, standard deviations, and skewness and kurtosis coefficients for DFS-S items. Some items showed slight deviation from the normal distribution.

Table 1. Mean score (M), standard deviation (SD), skewness, and kurtosis for DFS-S items, and corrected item-total correlations (N = 1019). 

Validity Evidence Based on Internal Structure

A CFA was first conducted with the total sample to examine the fit of the single-factor model of the DFS-S. The indices obtained indicated a good fit, with values of the CFI and NNFI above .95, the RMSEA below .06, and the SRMR below .08. We then tested the fit of the model for males and females separately. Both models yielded reasonable fit indices. Finally, the analysis of configural, metric, and scalar invariance also indicated a good fit, insofar as there was a decrement of less than .01 in the CFI and an increment of less than .015 in the RMSEA from the less constrained to the more constrained model (Table 2). The values of standardized parameters for the total sample were all statistically significant (Figure 1).

Table 2. Fit indices for the single-factor model of the DFS-S. 

^Note.N = 1019; Female, n = 585; Male, n = 434; χ² = chi-square statistic; df = degrees of freedom; CFI = comparative fit index; NNFI = non-normed fit index; RMSEA = root mean square error of approximation with 90% confidence interval¸ ◸CFI = CFI less constrained model – CFI more constrained model; Δ RMSEA = RMSEA less constrained model – RMSEA more constrained model.

Figure 1. Standardized parameter values for the single-factor structure of the DFS-S for the total sample. 

Item Analysis and Reliability

Corrected item-total correlations in the total sample ranged from .65 to .81 (Table 1). All values were therefore above .30, indicating adequate item homogeneity. Regarding the reliability of DFS-S scores, the values of McDonald’s omega coefficient were .92 for the total sample, .93 for males, and .90 for females, indicating satisfactory and similar reliability of test scores.

Validity Evidence Based on Relationships with Other Variables

The results of the correlation analysis showed that scores on the DFS-S were positively related with scores on agreeableness, neuroticism, and openness, and negatively related with scores on extraversion and conscientiousness. However, only the associations with neuroticism and extraversion were strong or moderate. The other associations were weak according to Cohen’s criteria (Table 3).

Table 3. Pearson correlation coefficients between DFS-S total score and scores on the IPIP-BFM-20 subscales. 

^Note.N= 1019.

^***p < .001;

^*p < .05.

Discussion and Conclusions

The aim of this study was to develop a Spanish version of the DFS (DFS-S) and to provide evidence of its psychometric properties in a sample of young adults. Validity evidence based on the instrument's internal structure, including measurement invariance across gender, as well as on relationships with personality traits (extraversion, agreeableness, conscientiousness, neuroticism, and openness) was obtained. Reliability and gender differences in DFS-S scores were also examined.

With regard to validity evidence based on internal structure, the results of the CFA supported, as expected, a single-factor structure for the DFS-S, with satisfactory fit indices. This structure is consistent with that reported by ^{Zaleski et al. (2019)} for the original version of the scale and it supports the use of the total score, with higher scores indicating a greater concern about the future. We also found configural, metric, and scalar invariance across gender, indicating that the DFS-S has a stable factor structure and the same measurement units and origins across male and female young adults. These results also indicate that latent factor means can be compared between these two groups and that potential differences in latent factors reflect true differences on the construct (^{Dimitrov, 2010}; ^{Han et al., 2019}).

Regarding item analysis, item-total correlation coefficients indicated satisfactory item homogeneity, with values above or equal to .65. The reliability of DFS-S scores was also adequate, with McDonald’s omega equal to .92, similar to the Cronbach’s alpha coefficient of .90 obtained with the original version (^{Zaleski et al., 2019}).

In terms of validity evidence based on relationships with other variables, the analysis showed that DFS-S scores correlated with scores on the IPIP-BFM-20 subscales. Overall, the most relevant findings were the strong positive correlation between future anxiety and neuroticism, as expected, and the moderate negative correlation with extraversion. The neuroticism subscale encompasses the facets of anxiety, depression, and anger, whereas the extraversion subscale captures the facets of cheerfulness and friendliness. These results therefore suggest that people who are anxious, nervous, and prone to anger and irritation, as well as those who are introverted, reserved, and socially inhibited tend to think about the future with anxiety, fear, and uncertainty. Although there is no previous evidence regarding the relationship between personality traits and future anxiety as measured by the DFS, our findings are consistent with studies using the Future Anxiety Scale (^{Zaleski, 1996}), in which a positive correlation was observed between neuroticism and future anxiety. They are also in line with the results of ^{Stolarski and Matthews (2016)}, who found that scores on the Future Negative scale (^{Carelli et al., 2011}) showed a particularly strong relationship with neuroticism. Mention should likewise be made of the empirical evidence showing that individuals who score high on neuroticism are more vulnerable to anxiety disorders, depression, and obsessions, and also that the combination of high neuroticism with low extraversion is associated with stress vulnerability, anxiety, and depressive disorders (^{Grant, 2011}; ^{Rosellini & Brown, 2011}; ^{Wenjuan et al., 2020}). Overall, our results suggest that the DFS-S may be useful in clinical settings to determine future anxiety, which in turn could be important for identifying individuals at potential risk of high levels of stress, anxiety or depression. However, more research is needed on the relationship between future anxiety and mental health.

The mean total score on the DFS-S in the present sample was 20.17 (out of a maximum possible of 30), with negative skewness, indicating that the distribution is concentrated on high scores. These results suggest that young Spanish adults are generally concerned about the future, which is consistent with recent reports for other countries (^{Dadaczynski et al., 2021}; ^{Lanz et al., 2021}). In line with our expectations and with other studies using the DFS (Dadaczynski et al., 2021; ^{Duplaga & Grysztar, 2021}; ^{Zaleski et al., 2019}), we also found higher levels of future anxiety among female participants. Empirical research has consistently shown that women are more susceptible than men to developing anxiety (^{Marques et al., 2016}), with some authors attributing this to a combination of social and biological factors (^{Hantsoo & Epperson, 2017}). Women are more likely than men to experience stressors and tend to use different coping strategies, and this, coupled with certain biological mechanisms, including hormonal fluctuation, may make them more vulnerable to anxiety disorders (^{Faravelli et al., 2013}; ^{Graves et al., 2021}; Hantsoo & Epperson, 2017; Marques et al., 2016). Overall, these results suggest the need to make psychological counseling available to young adults, especially for women, so as to teach them specific skills for managing and reducing anxiety about the future, thus improving their general well-being.

The present study has a number of limitations that need to be mentioned. First, the generalizability of results may be limited by our use of a convenience sampling strategy, as well as by the focus on young adults. Hence it would be interesting in further research to consider a wider age range and to analyze measurement invariance across stages of adulthood. Second, data were collected using self-report instruments, which can be affected by response bias. Third, test-retest reliability was not analyzed, and future research should provide information on this aspect. Fourth, the use of correlational analysis to examine relationships between variables means that causality cannot be inferred. Finally, validity evidence based on the relationship with other variables was only provided by means of associations with personality traits. Although personality traits have been considered the basis of future anxiety (^{Zaleski, 1996}, 2005; Zaleski et al., 2019), it would be interesting in future research to study the relationship between future anxiety and mental health problems such as stress or depression. The association between future anxiety and scores on other scales designed specifically to assess anxiety also needs to be established so as to analyze relationships with similar constructs.

Despite these limitations, the present study provides evidence of the psychometric properties of the Spanish version of the Dark Future Scale in a relatively large sample of young adults. The results indicated a single-factor structure which was invariant across gender, with a satisfactory reliability coefficient. The most relevant finding in terms of relationships with other variables was the strong positive correlation between future anxiety and neuroticism and the moderate negative correlation with extraversion. We therefore conclude that the DFS-S has satisfactory psychometric properties and that it is an adequate tool for measuring among young adults the tendency to think about the future with anxiety, fear, and uncertainty.