The Assessment of Attitudes towards Retirement from a Psychosocial Approach

Abstract

The aim of this research is, on the one hand, to develop a scale that can be used for assessing attitudes towards retirement as a prevention and intervention tool for improving the life of people in retirement, thereby enhancing a sustainable world that offers quality of life, as well as personal, social, and community well-being, and the efficient use of available materials and socio-sanitary resources. On the other hand, the aim of this research is to analyze the psychometric properties of such a scale, i.e., the reliability and validity of a sample of people at an age that is close to retirement age. Hence, the factorial validity was tested using the confirmatory factor analysis (CFA) technique, and the criterion validity was tested by considering general self-efficacy, self-regulation, state of irritation, and certain dimensions of health and personality. The obtained results confirmed the existence of four factors in the scale of attitudes towards retirement as follows: (i) leisure, (ii) economy, (iii) status, and (iv) health. The obtained correlations showed that attitudes towards retirement are positively linked to variables such as self-esteem and self-regulation, whereas significant and negative correlations related to irritation and fatigue were found. Hence, according to the obtained results, the proposed scale is an easy and relevant tool for working on a better and more profitable psychological adaptation to retirement from work. Thus, a society where tools of psychological evaluation, such as the Attitudes toward Retirement Scale (ARS) proposed in this study, are utilized allows for the detection of problems among people facing retirement, which will lead to a more sustainable and evolved society that provides quality of life as well as personal, social, and community well-being.

1. Introduction

Currently, research related to retirement is constantly increasing, thereby reaching a place of relevance to the challenges that developed countries are facing [1]. Retirement from work is a psychosocial process that is currently of major relevance to society. According to the ageism report by the World Health Organization (WHO), one of the most important milestones of the XXI century is the increment in the life expectancy of the world population [2]. In Spain, people older than 65 years old comprise 20% of the total population [3]. Thus, work life is extended for many workers and, as a consequence, retirement is considered to be a rupture from the role that is played during working life, thereby causing consequences at a psychosocial and emotional level. Research regarding developmental psychology has increased too in recent years, which describes and studies the diverse stages of human life, thus allowing us to obtain scientific information regarding longevity [4]. Hence, the main attention is focused on studying the changes that a human being undergoes when old age is reached, which are as follows: retirement [5], the impact of retirement on aging [6], and the quality of life of elderly people [7].

Elderly people face retirement in many different ways. For some people, retirement simply means the end of their employment period, a new role in society, and the beginning of old age [8]. For others, retirement means an everlasting release from stressful work [9]. However, for an increasing number of people, retirement occurs during a productive and fulfilling stage of professional life and, hence, is considered to be a personal loss [10].

Retirement is a process that is influenced by personal attitudes. Diverse authors have revealed this influence. For example, Lupien and Wan [11] make a clear plea in favor of positive attitudes toward retirement in their review of successful aging. In their review of attitudes towards aging, Levy et al. [12] include a reference to a series of scientific works that confirm that the attitude towards retirement plays a key role in positive aging, with an influence on longevity too. An inappropriate evaluation of retirement can affect both people in retirement and people with an age that is close to retirement; it can even affect those who consider work to be a source of value and personal identity [13].

Retirement includes changes at an inter- and intrapersonal level, but the approach to this transition is not adequate nowadays since it is only focused on an administrative, juridical, and organizational process [14]. Quality of life covers different dimensions, among which physical and emotional wellness, social inclusion, social participation, and self-development deserve to be highlighted [15,16,17]. Within this approach, attitudes are a factor that will influence the three levels of a person, which are as follows: the physical one, the psychological one, and the social one.

The phase that is previous to retirement must be analyzed by considering the psychosocial context of the person instead of exclusively considering the legal aspects that give him/her the tools for facing this new stage of life and that let him/her positively adapt to their new life out of work. The phase that is previous to retirement is a process where attitudes, perceptions, and beliefs about retirement influence a worker’s psyche. Thus, this stage is the right time for preparing him/her for retirement and for achieving a positive adaptation accordingly.

Different authors support the idea that responses against retirement are influenced by dispositional and situational variables that drive the responses of people [18]. Brajsa et al. [19] point out that positive self-esteem enhances psychological wellness, leading to a more generally positive perception. Haro and López [20] highlight that preconceived stigmas about retirement cause negative pressure in facing retirement and living during it. The new circumstances to be faced can cause an attitude in a person that is full of mental, emotional, and compartmental responses [21]. Regarding the financial factor, it deserves to be highlighted that diverse studies show that many people at retirement age are not economically prepared [22,23]. Only 24.8% of people around the world have savings set aside for when they are elderly [24], and this obviously supposes the risk that is due to the lack of economic resources during retirement [25].

The end of a person’s work role is often linked to a duel that is influenced by the study level of the person who faces retirement [20]. Thus, workers who have undertaken university studies plan their retirement and restructure their time according to their personal projects in a better way, leading to an emotional balance and a lower level of stress [26]. Socializing is another relevant factor to be considered in the retirement process since a high level of socialization is related to a low level of familiar disarrangement and a minimal level of depression. These things considered, people with hobbies exhibit a lower level of psychopathology [26]. Reviews carried out on satisfaction with life [27] only found significant relations between neuroticism, responsibility, and openness to experience.

Retirement from work is an inevitable stage of life for many people at the end of their working career. Retirement is a complex process that significantly affects the person facing retirement, and this process is influenced by different psychosocial factors that will influence the way a person faces retirement. Thus, a person’s attitudes towards retirement are key for the worker’s adaptation to their new stage of life. Hence, people with a positive attitude and a high level of satisfaction about their lives will achieve an effective adaptation to retirement.

Several studies are focused on the analysis of attitudes toward retirement. However, there are no evaluation instruments for evaluating attitudes towards retirement and, hence, reveal the potential problems that the future retiree will face. This way, research developed in this field using tools such as the one proposed in this study will lead to improving the quality of life of people in retirement and will make easier the development of intervention programs. Thus, the purpose of this work is to delve further into the study of retirement from a psychosocial point of view by developing an instrument to be used for evaluating a series of key attitudes toward retirement from work for people with an age that is close to retirement or people already in retirement. From the revealed attitudes, a personalized intervention program can be developed that will enhance the adaptation process to retirement, showing how to face and solve problems that will appear during this process. Thus, the well-being and the quality of life of people in retirement will be improved during this long stage of life. Obviously, the psychological preparation for retirement must be considered within social policies that support retirees, and hence, the proposed tool can be used to reveal the needs of this part of society. The intervention program must be developed from an interdisciplinary point of view, considering economic factors and the health of both retirees and his/her partners. The tool presented in this work can also be used after the application of such programs. This way, the attitude changes appearing after the implementation of the program and the effectiveness of the intervention program will be revealed. In addition, the scale developed in this work is useful for human resource departments of companies since it allows revealing the attitudes towards the retirement of their workers with an age that is close to retirement and, this way, such companies could develop measurements that enhance the adaptation of their workers to this new stage of life.

In addition, cognitive reserve is a key factor that must be considered in analyzing the adaptation to retirement by using implicit strategies of compensation in the programs for preparation for retirement. Such strategies must be focused on a reorganization process where changes are compensated by strategies of behavioral and cognitive character.

Taking this into account, the general aim of this research is to develop a scale to assess attitudes towards retirement and to analyze its psychometric properties of reliability and validity. Taking into account previous studies and the bibliographic review, four dimensions are considered especially relevant for elaborating such a scale, as follows: (i) leisure, (ii) economy, (iii) status, and (iv) health.

The first one, leisure, is a key factor since retirement supposes a significant increment of leisure and leads to a reorganization of family relationships [28]. Becchetti et al. observed a link between the practice of leisure activities and the quality of life of retirees [28]. Diverse studies revealed that the use of leisure and status are related. Several experimental studies have shown that a lack of activity in old age can negatively influence health and personal well-being [29].

The second dimension, the economy, deals with financial changes caused by retirement [24,30]. Retirees with (i) an adequate income during retirement (commonly, economic stability is lost), (ii) good health, and (iii) a social net of social support are more satisfied with retirement than those who have no such conditions.

The third dimension, status, is essentially supervened due to the loss of employment. Finally, regarding the last dimension, health, unfortunately for many people, retirement is linked to a deterioration of their general health state, but it does not have to be this way since the general health state depends on fitness and healthy life habits [31,32,33].

This paper also seeks to delve deeper into previous research by studying the relationship between states of psychological health (irritation, physical, and mental fatigue) and attitudes toward retirement. Thus, high levels of irritation and fatigue would be linked to negative attitudes towards retirement. However, high scores in self-esteem, self-efficacy, and self-regulation (positive concepts about oneself) would be linked to positive attitudes toward retirement since the subject has a more positive view of him/herself, and a positive adaptation to the retirement process will be achieved [34,35]. Finally, the roles of agreeableness and extroversion are also analyzed because, from a theoretical point of view, people with a high level of these personality factors will have better social relations and a better adaptation to retirement.

Within this framework, the present study has been developed according to the following objectives: (i) to design an instrument to assess attitudes towards retirement with adequate psychometric properties of reliability and validity; (ii) to study the relationship between attitudes towards retirement and certain positive aspects, such as general health, self-esteem, self-efficacy, and self-regulation, and other negative aspects such as fatigue; and (iii) to analyze the relationship between attitudes towards retirement and certain personality factors, such as those contemplated in the Five-Factor Model: extraversion, neuroticism, conscientiousness, agreeableness, and openness to experience.

2. Hypothesis

As it was previously stated, the general objective of this research is to study attitudes toward retirement by developing an evaluation instrument. In addition, this study tries to reveal the relationship between attitudes toward retirement and, on the one hand, positive variables such as self-esteem, self-efficacy, and self-regulation and, on the other hand, negative variables such as irritation and fatigue. Finally, this study reveals the relationship between attitudes toward retirement and psychosocial factors and the dimensions included in the Five-Factor Model of personality, i.e., extroversion, emotional stability, responsibility, kindness, and openness to experience.

The following general hypothesis is considered in this study: the ARS scale measures attitudes towards retirement, considering that a high score on this scale is related to a positive attitude towards retirement. A positive relation with self-esteem, self-efficiency, and auto-regulation is expected, and in addition, a negative relation with irritation and fatigue is also expected.

From this general hypothesis, the following hypotheses can be derived:

(1)

The scale toward retirement must exhibit an adequate index of self-internal consistency.

(2)

It is expected to confirm the existence of the four dimensions proposed in the ARS scale: leisure, economy, status, and health by means of confirmatory factor analysis (CFA) with adequate reliability for all the factors (internal consistency).

The following hypotheses are formulated in terms of correlations between variables regarding attitudes toward retirement.

(3)

People with a high level of self-esteem will achieve higher scores in the ARS, and consequently, a significant relationship between results given by the ARS scale and self-esteem will exist.

(4)

People with a high level of self-efficiency will achieve a better score in the ARS, and hence, a positive and significant relationship between the obtained score in the ARS and general self-efficiency will exist.

(5)

People with a high level of self-regulation will achieve a high score in the ARS. Therefore, a positive and significant relationship between the obtained score in the ARS and self-regulation will exist.

(6)

People with a good general state of health will exhibit better attitudes toward retirement. Hence, a negative and significant correlation between the obtained score given by the GHQ and the score given by the ARS will be obtained since a high score in the GHQ means a worse health state.

(7)

People with a high level of job irritation will exhibit a worse attitude towards retirement. Therefore, a negative and significant correlation between irritation and the obtained ARS score will be obtained.

(7.1.)

A negative and significant correlation between cognitive irritation and the obtained score in the ARS will exist.

(7.2.)

A negative and significant correlation between emotional irritation and the obtained score in the ARS will exist.

(8)

People with a high level of fatigue will exhibit a worse attitude towards retirement. Therefore, the fatigue state will have a negative and significant relationship with the ARS score.

(8.1.)

The physical fatigue state will have a negative and significant correlation with the ARS score.

(8.2.)

The mental fatigue state will have a negative and significant correlation with the ARS score.

(8.3.)

The activity reduction state will have a negative and significant correlation with the ARS score.

(8.4.)

The motivation reduction state will have a negative and significant correlation with the ARS score.

(9)

A positive and significant relationship between the personality dimension included in the FFM and the ARS scores will exist.

(9.1.)

People with a high level of extroversion will exhibit a better attitude towards retirement.

(9.2.)

People with a high level of emotional stability will exhibit a better attitude towards retirement.

(9.3.)

People with a high level of responsibility will exhibit a better attitude towards retirement.

(9.4.)

People with a high level of kindness will exhibit a better attitude towards retirement.

(9.5.)

People with a high level of openness to experience will exhibit a better attitude towards retirement.

(10)

A positive and significant correlation between life satisfaction and the personality dimensions included in the FFM will exist.

(10.1.)

A positive and significant correlation between life satisfaction and extroversion will exist.

(10.2.)

A positive and significant correlation between life satisfaction and emotional stability will exist.

(10.3.)

A positive and significant correlation between life satisfaction and responsibility will exist.

(10.4.)

A positive and significant correlation between life satisfaction and kindness will exist.

(10.5.)

A positive and significant correlation between life satisfaction and openness to experience will exist.

3. Methods

This research is based on a non-experimental design with intentional sampling and instrumental scope, whose aim is to review the psychometric properties of a measurement tool for a psychological construct. The following phases were followed to carry out this study: (i) preparation of the initial version of the questionnaire, (ii) validation of the questionnaire, and (iii) analysis of the psychometric properties. Firstly, the initial version of the questionnaire was obtained from a review of bibliographic sources. Later, the Angoff method was applied to ensure the validity of the initial test. This way, the participation of leading experts in the field was needed. The selected experts specialized in social psychology and the psychology of aging, and in addition, an expert in test design was also included. The Angoff method was applied in two sessions, where experts rated vocabulary, clarity, wording, and conceptual structure for each one of the items considered on a scale from 1 (minimum score) to 10 (maximum score). Each of the experts justified the rate given to each item and proposed potential changes to the items considered. Each change was proposed to the team of experts, and finally, the change was only included if a consensus of all the experts was reached. Finally, the psychometric properties of the ARS instrument were reviewed, and the procedure for obtaining the test reliability indicators was carried out in the third phase of the study.

3.1. Procedures and Measurements

The research was developed in two phases: (1) the preliminary study and (2) the final study. Once the study was planned, different companies, public institutions, professional associations, self-employed professionals, and retirees or people at an age close to retirement were invited to participate in the study. After acceptance, the battery of questionnaires, consisting of independent booklets with instructions, was individually answered in presential group sessions of 30–45 min. A talk was given before completing the questionnaires, including the general aims of the research and instructions for completing the questionnaires, the approximate duration, the inclusion criteria, etc. In this talk, we highlighted the voluntary nature of participation and the possibility of dropping out of the study at any time, the anonymity and confidentiality of the obtained data, the importance of answering honestly, and, in addition, it was explained that no answer would be considered correct or incorrect since any answer is considered appropriate and good. Regarding the inclusion criteria in the study, participants with an age close to retirement (5 years or less) or recent retirees (less than 5 years) were selected. Participants who did not meet these requirements were excluded from the study. For the sake of clarity, all the questionnaires were written in Spanish.

3.1.1. Phase 1—Preliminary Study—Elaboration of the Scale

A theoretical review of the subject was first made to develop the attitude evaluation scale. Forty people in retirement collaborated in the elaboration of the items of the first scale. Thus, participants wrote down in a table the words linked to the term “retirement from work” (both a list of positive and a list of negative terms). Afterward, a selection and filtering process was carried out with all the obtained answers. The terms were grouped by synonyms, and a representative term of each group (highest frequency) was chosen. Once the items were developed, they were submitted for validation by expert reviewers. The experts comprised 10 research and teaching specialists in the field of occupational retirement. After collecting all the information, the content and recommendations of the reviewers were analyzed. As a result of this process, the first version of the scale was obtained, consisting of 80 Likert-type items, scored on a scale from 1 (total disagreement) to 7 (total agreement).

Once the preliminary scale was prepared, the scale was applied to a pilot sample of 50 people aged between 51 and 86 years old to confirm the feasibility and applicability of the scale, as well as to start the psychometric validation process. Participants were contacted through directors of different institutions who agreed to collaborate in the research. This group was initially informed about the aims of the study in a simple way to avoid bias in the obtained results and, in addition, participants signed the informed consent after being informed about their voluntary participation in the study. The importance of answering honestly on the scale was highlighted, and it was pointed out that obtained data from the study and his/her collaboration would be anonymously treated. This work follows the ethical procedures in accordance with the Declaration of Helsinki [36].

3.1.2. Stage 2—Final Study

From the results obtained in the preliminary study, a reduced final scale consisting of 20 items (selected from the initial scale) was applied, together with other measuring instruments, to a larger sample than the one used in the first phase, specifically to 201 subjects.

The study was proposed to different directors and coordinators of diverse Spanish labor sectors such as institutions in the education field, federations of retiree people, banks, public administration, service companies, professional associations, self-employed professionals, etc. Thus, both workers and managers were considered for the scale. This way, a wide typology of job fields and job positions was included in the study to obtain a representative sample.

3.2. Sample

Participation in the research was voluntary, confidential, and anonymous. This study follows the ethical procedures consistent with the Declaration of Helsinki [36]. The sample used for this work was composed of 291 participants. A total of 40 people were involved in the preliminary study, where the scale was prepared (phase 1.1 of the study); 50 participants were included in the pilot study (phase 1.2), where the feasibility of the study was tested; and finally, 201 people were involved in the final study (phase 2).

Regarding the final sample, 201 participants (Figure 1) were 58.2% women and 41.8% men. The ages included were within the range of 51 and 86 years old, with a mean distribution of 61.36 years (SD = 8.08). Regarding marital status, 70.6% of the subjects were married, 10.9% were widowed, and the same percentage was single; finally, in the sample, there were 7.5% separated or divorced participants. In relation to the place of residence, 59.2% lived in the urban core, and 40.8% lived in rural areas. Finally, regarding the level of studies, 33.3% of the people only studied at primary school, whereas 35.3% obtained a university degree; the remaining people had intermediate studies.

3.3. Instrument

The battery of questionnaires was delivered to participants in a face-to-face meeting where the objectives of the study and the importance of their collaboration were explained. The questionnaires used in this study, as well as the variable to be measured and its psychometric properties, are the following:

The Attitudes toward Retirement Scale (ARS). This scale, the ARS, is an instrument designed with the purpose of assessing attitudes towards retirement. In the initial phase of preparing the questionnaire, a review of aspects related to attitudes towards retirement was carried out. Initially, 80 items were developed, from which the 20 most appropriate were selected to measure four dimensions: (i) leisure, (ii) economy, (iii) status, and (iv) health. Subjects were asked to answer the items on a Likert-type scale with options ranging from 1 (strongly disagree) to 7 (strongly agree). To obtain the final score, the items of the economy, status, and health dimensions must be inverted. This way, a high score in each of the dimensions, as well as on the global scale, indicates a better attitude towards retirement. Although different questionnaires or scales exist that evaluate the relevant factors related to retirement, unfortunately, none of them focus on attitudes toward retirement. Attitudes, together, include the cognitive and emotional issues causing a given behavior. Thus, the analysis of attitudes by using an instrument such as the one proposed in this study offers relevant information regarding the thoughts, feelings, and future behaviors that a person will have after retirement. If the attitudes are negative, adaptation to this new stage will not be satisfactory, causing health problems at physical, social, and physiological levels.

The General Self-Efficacy Scale (GSES) [37,38]. This scale, initially developed by Jerusalem and Schwarzer in 1992 [39], aims to measure the person’s belief in their ability to complete a task or achieve a goal, or in other words, the feeling of personal aptitude to efficiently manage different stressful situations that a person faces throughout his/her life. In this research, the Spanish version of Baessler and Schwarzer [37] was considered. The version of the instrument used in this study consists of 10 items with a Likert-type scale from 1 (false) to 4 (true). A high score indicates a high overall self-efficacy.

The Self-Regulation Scale (SRS) [40]. This scale offers a general score on the subjects’ ability to self-regulate their daily lives. The instrument used consists of seven items. Regarding the psychometric data, an alpha coefficient of 0.76 was found. Regarding the stability of the scale with samples from different countries, values of 0.57 (workers from Costa Rica), 0.73 (teachers from Germany), and 0.62 (students from Colorado in the United States) have been found. The items are included in a Likert-type scale with scores from 1 (strongly disagree) to 4 (strongly agree). A high score indicates high self-regulation.

The Rosenberg Self-Esteem Scale (RSES) [41]. This scale is used to evaluate the satisfaction that a person has with him/herself. Personal self-esteem can be understood as the feeling towards oneself, positive or negative, built up by means of self-evaluation of their own characteristics. The scale is one-dimensional and consists of 10 items, with five positive statements and five negative, with Likert-type responses, considering four options ranging from 1 (strongly disagree) to 4 (strongly agree). A high score on the scale indicates a high self-esteem of the subject.

The General Health Questionnaire (General Health Questionnaire, GHQ Spanish Form) [42]. This questionnaire includes 93 items in its original version (1972) and, later, shorter versions have been developed. The one applied in this study contains only 12 items, and it was validated in Spain by Sánchez-López and Dresch [42]. This questionnaire allows the detection of cases of psychiatric morbidity or psychological pathologies, although this questionnaire should not be used to make clinical diagnoses. The 12 items are included in a Likert-type four-point scale. The rating of the scale is obtained by giving (for each item of the scale) 0 points if the scale points 0 or 1 are selected and 1 point if the scale points 2 or 3 are selected. The total score is obtained by adding the points given for each item. Therefore, the total score is within the range of 0 to 12. When interpreting the scores, it must be considered that 0 indicates better health and 12 points indicate worse health. Internal consistency values of alpha = 0.78 have been found. Regarding the internal structure of the questionnaire, three factors were found: (i) successful coping, (ii) self-esteem, and (iii) stress. For this study, only a global score of the state of health was considered.

The Irritation Scale (IS) [43]. The objective of this instrument is the evaluation of stress at work. The scale consists of eight items distributed via two factors: cognitive irritation (CI) and emotional irritation (EI). On the one hand, CI is described as the tendency to recurrently dwell upon work-related thoughts when one is not at work. On the other hand, EI is a state of progressive psychological exhaustion that cannot be alleviated with normal breaks, the emotional tendency to experience and respond with anger or aggressiveness when the incentives to achieve an important goal or objective for the person are lost. The way to respond to these items is through this scale, which is composed of a Likert-type response scale with scores ranging from 1 (strongly disagree) to 7 (strongly agree). Total irritation is the sum of CI and EI. This instrument is currently the property of the German publisher Hogrefe.

The Multidimensional Fatigue Inventory (MFI) [44]. This inventory assesses asthenia or decay in terms of 20 items divided into five dimensions: (i) general fatigue, (ii) physical fatigue, (iii) mental fatigue, (iv) reduced activity, and (v) reduced motivation. The scale is composed of a Likert-type scale ranging from 1 (strongly agree) to 7 (strongly disagree) to the fatigue experienced during the previous days. Studies carried out with the scale in different types of samples (from clinical patients to university students) show adequate internal consistency (alpha values higher than 0.80) and validity indices, both for the global scale and for each one of the scale dimensions. It is an instrument that serves to detect differences in fatigue between groups and has been used mainly in clinical contexts [45].

Personality Scale “Overall Personality Assessment Scale” (OPERAS) [46]. This instrument is based on the model of the five major personality factors, also known as FFM [47], which consider the existence of five personality traits that condition human behavior. In the FFM, the term trait is defined as an enduring personality characteristic that describes or determines an individual’s behavior across a range of situations. The five traits are as follows: (i) extraversion, (ii) responsibility, (iii) emotional stability, (iv) kindness, and (v) openness to experience. The scale used in this study consists of 42 items (the latest version is 40) with a Likert-type response scale ranging from 1 (strongly disagree) to 5 (strongly agree). This way, personality factors included in the FFM, such as extroversion, emotional stability, responsibility, kindness, and openness to experience, were included in the study.

The purpose of using both validated scales and the ARS in the battery of instruments of this study is to analyze the ARS criterion validity. To achieve this goal, potential correlations of each one of the factors included in the ARS were analyzed, with other psychological variables simultaneously evaluated. This way, results will support the external aspect of construct validity and, in addition, will verify the discriminant validity of the ARS, proving that the proposed instrument measures different traits that are not considered in the other scales.

3.4. Statistical Analysis

Statistical analyses were performed using the IBM SPSS Statistics 27.0 and IBM-SPSS-AMOS 26 applications. The statistical tests used in this study are the following:

• Tables of frequencies and percentages for qualitative or categorical variables, with the Chi-square test of homogeneity. This test checks if two or more samples are obtained from homogeneous populations according to certain criteria. It was used to check the homogeneity in different variables such as gender, educational level, leisure, etc.
• Contingency tables with a Chi-square test of non-dependence between two qualitative variables were used to determine the existence of a relationship between the two variables. However, this test does not provide the percentage of influence of one variable on the other.
• Exploratory and descriptive analyses of quantitative variables with a goodness-of-fit test were also used. This analysis was used to examine the discrepancy between the observed values and the expected values of the normal Gaussian model (Kolmogorov–Smirnov test), and box plots were used to detect outliers.
• Cronbach’s alpha to calculate the reliability or internal consistency of the measurement instruments, particularly the ARS scale proposed in this study, was used.
• The bivariate correlation test between variables using the Pearson correlation coefficient was used since quantitative variables can be fitted to a normal distribution.
• Significance tests of mean differences:

  ▪

  Student’s t-test: this test is used to estimate the means and proportions of quantitative variables, specifically for the comparative study of sociodemographic data.

  ▪

  ANOVA. This is an analysis of variance to compare several groups in a quantitative variable. This test was applied to contrast the equality of means of three or more independent populations with a normal distribution. In addition, this analysis was used to check the significance of the consistency analysis.

In addition, the following non-parametric statistical tests were used when necessary:

• Mann–Whitney: This test is the non-parametric version of the Student’s t-test, and it was used to confirm the heterogeneity of two samples.
• Friedman: This test is equivalent to the ANOVA test for repeated measurements. This test consists of organizing the data by blocks or rows, substituting them for their respective location, and once ordered, the existence of equal data is examined.
• Exploratory Factor Analysis (EFA). This technique is used to find out the internal structure of the variables analyzed. As a result, the weight of the items on the a priori proposed factors is obtained. Specifically, this technique is applied to verify the internal structure of the ARS.
• Confirmatory Factor Analysis (AFC). This test is used to find out if the number of factors obtained and their weights are in agreement with what could be expected according to the previously proposed theory in terms of the data obtained through AFE. This multivariate technique aims to corroborate the factors obtained so far through the contrast of hypotheses.

4. Results

4.1. Evidence of Internal Validity

Firstly, internal validity was checked. Two models with different dimensions were tested: a one-dimensional model (1 factor) and a multidimensional model (4 factors).

Table 1. Goodness-of-fit indices in theoretical models in CFA.

Model	χ2	Gl	χ2/df	RMSEA	RMSEA 90% I.C.	CFI	NNFI
4 factors	252.917 **	164	1.542	0.052	0.039–0.064	0.913	0.899
1 factor	692.869 **	170	4.075	0.124	0.114–0.133	0.490	0.430

Note. ** p < 0.001; χ² = chi-cuadrado; df = degree of freedom; RMSEA = root-mean-square error of approximation; CFI = comparative fit index; NNFI = Bentler Bonnet non-normed fit index.

shows the goodness-of-fit indices for both models. According to these results, the four-factor model offers satisfactory fit indices (χ²/df = 1.54; RMSEA = 0.052; CFI = 0.913; and NNFI = 0.899), whereas the values of the one-factor model are unacceptable (χ²/df = 4.08; RMSEA = 0.124; CFI = 0.490; and NNFI = 0.430). Consequently, the four-factor model is more suitable.

The final ARS is composed of 20 items grouped in four dimensions as follows (Figure 2): (i) 20% for the health dimension, (ii) 25% for the economy dimension, (iii) 20% of the items are focused on the status dimension and, finally, (iv) 35% for the leisure dimension. The statement of each item considered in such a scale is included in Figure 2. Items were evaluated through a Likert-type scale ranging from 1 (strongly disagree) to 7 (strongly agree). To obtain the final score of the items corresponding to the economy, status, and health dimensions, the answers must be inverted. Thus, a high score on each one of the factors means a better attitude towards retirement.

Table 2. Configuration matrix with factor weights and correlation of factors with the global score on the scale.

	Factor 1 Status	Factor 2 Leisure	Factor 3 Economy	Factor 4 Health
Item 9	0.85
Item 8	0.78
Item 10	0.72
Item 18	0.63
Item 20		0.78
Item 12		0.77
Item 11		0.71
Item 4		0.70
Item 7		0.70
Item 16		0.66
Item 5		0.66
Item 19			0.81
Item 14			0.82
Item 3			0.68
Item 2			0.64
Item 17			0.59
Item 13				−0.80
Item 15				−0.77
Item 1				−0.63
Item 6				−0.49
Overall score on ARS	0.73 **	0.53 **	0.65 **	0.70 **

Note: ** p < 0.001.

shows the items that are included in each one of the factors and their factor weights. The obtained coefficients are high, and they are in the expected direction. This way, the following correlations of the global score of the scale and empirical scores are obtained, as follows (Table 2 and Figure 3): (i) status (r = 0.73, p < 0.001), (ii) leisure (r = 0.53, p < 0.001), (iii) economy (r = 0.65, p < 0.001), and (iv) health (r = 0.70, p < 0.001). Regarding correlations between the analyzed factors, the leisure dimension only exhibits a positive correlation with status (r = 0.15, p = 0.032), whereas a non-significant correlation was obtained with the other two factors, economy (r = −0.02, p = 0.800) and health (r = 0.11, p = 0.112). However, the economy factor exhibits a positive and significant relationship with both status (r = 0.36, p < 0.001) and health (r = 0.38, p < 0.001). Finally, the relationship between status and health is also significant (r = 0.45, p < 0.001). The use of leisure is different according to cultural, economic, and social factors. Thus, the activities carried out in leisure are an expression of identity and social position. Obviously, the economy has a strong influence on the type of leisure activities and hobbies. This way, the economy positively affects the status of a person and his/her health in terms of the availability of resources; thus, economy, status, and health are correlated.

4.2. Internal Consistency Analysis

Secondly, an analysis of the reliability of the four dimensions considered in the instrument was carried out. The reliability coefficients (Cronbach’s alpha) are acceptable, with the highest value corresponding to the leisure dimension (0.84), whereas the lowest value corresponds to the health dimension (0.67), whereas in the economy dimension, the value is 0.79, and for status, it is 0.78. The internal consistency of the global scale reached a value of 0.82.

Table 3. Descriptive statistics and reliability values, Cronbach’s alpha.

Variable	Minimum	Maximum	M	SD	Cronbach’s Alpha
Retirement Attitudes	3	7	4.65	0.89	0.82
Free time/Leisure	1	7	5.78	1.12	0.84
Economy	1	7	3.45	1.43	0.79
Status	1	7	4.26	1.66	0.78
Health	1	7	4.56	1.45	0.67
Self-efficacy	1	4	2.97	4.96	0.87
Self-regulation	1	4	2.84	0.49	0.83
Self-esteem	2	4	3.35	0.46	0.73
General Health	0	2	0.95	0.38	0.79
Fatigue Inventory	1	7	2.98	1.09	0.91
General fatigue	1	7	3.41	1.52	0.79
Physical fatigue	1	7	3.41	1.41	0.78
Mental fatigue	1	7	2.87	1.32	0.74
Activity reduction	1	7	2.75	1.20	0.67
Motivation Reduction	1	7	2.69	1.18	0.56
Extroversion	19	66	43.44	8.13	0.86 *
Emotional stability	17	71	47.18	8.71	0.86 *
Responsibility	16	69	46.32	9.22	0.77 *
Kindness	10	69	47.09	9.308	0.71 *
Opening experience	27	66	49.00	8.311	0.81 *

Note: * p < 0.05.

shows the internal consistency coefficient values of the remaining instruments used, as well as the descriptive ones. All of them present acceptable reliability coefficients, except for the motivation reduction dimension of the fatigue inventory.

4.3. Evidence on External Validity

The external validity was checked (

Table 4. Correlations between the ARS and the rest of the variables.

	ARS Total	ARS Leisure	ARS Economy	ARS Status	ARS Health
Self-efficacy	0.12	0.11	0.00	0.06	0.15 *
Self-regulation	0.13	0.13	0.04	0.07	0.10
GHQ-Health	−0.16 *	−0.10	−0.11	−0.08	−0.13
Self-esteem	0.18 *	0.08	0.03	0.15 *	0.23 **
Cognitive irritation	0.27 **	−0.06	−0.28 **	−0.14	−0.25 **
Emotional irritation	−0.34 **	−0.25 **	−0.30 **	−0.12	−0.20 *
Total Irritation	−0.36 **	−0.20 *	−0.34 **	−0.15	−0.26 **
General fatigue	−0.13	−0.09	−0.16 *	−0.03	−0.05
Physical fatigue	−0.06	−0.04	−0.05	0.01	−0.08
Mental fatigue	−0.23 **	−0.10	−0.19 **	−0.10	−0.22 **
Activity Reduction	−0.30 **	−0.16 *	−0.14 *	−0.19 **	−0.29 **
Motivation Reduction	−0.12	−0.10	−0.10	0.03	−0.15 *
MFI Total	−0.23 **	−0.14	0.17*	−0.08	−0.21 **
Extroversion	−0.04	0.03	−0.10	−0.02	−0.00
Responsibility	0.20 **	0.26 **	0.08	0.07	0.07
Kindness	0.19 **	0.15 **	0.13	0.08	0.13
Emotional stability	0.21 **	0.23 **	0.05	0.08	0.17 *
Opening experience	0.16 *	0.11	0.14	0.01	0.017 *

Note: * p < 0.05, ** p < 0.01; ARS = Scale of Attitudes toward Retirement; GHQ = General Health Questionnaire; MFI = Multidimensional Fatigue Inventory.

). The score of the global ARS shows significant and positive correlations with most of the variables considered; a negative correlation with the health variable was found because a high score indicates worse health in the measurement instrument (GHQ). According to these values, people who exhibit a high level of job irritation have a negative attitude toward retirement. Regarding self-esteem, self-efficacy, and self-regulation (positive concepts about oneself), high scores are linked to positive attitudes towards retirement since the subject has a more positive view of him/herself, leading to a positive adaptation to the retirement process.

The highest correlation was obtained for the total irritation variable (r = −0.36, p < 0.001), and the lowest correlation was obtained for the extraversion variable (r = −0.04, p = 0.614). There is a relationship between states of psychological health (irritation, physical, and mental fatigue) and attitudes towards retirement, with high levels of irritation and fatigue linked to negative attitudes towards retirement.

The factor with the most significant correlation is health. The highest correlation is presented with the variables reduction of activity (r = −0.29, p < 0.01), total irritation (r = −0.26, p = 0.06), cognitive irritation (r = −0.25, p = 0.08) and mental fatigue (r = −0.22, p = 0.01). No relationship was found with self-esteem, kindness, openness to experience, self-efficacy, or responsibility. The leisure factor exhibits a high correlation with emotional stability (r = 0.26, p < 0.001), kindness (r = 0.23, p < 0.01), responsibility (r = 0.15, p = 0.36), and self-regulation (r = 0.13, p = 0.142). No correlation was found with emotional irritation, total irritation, activity reduction, or total MFI. The economy factor is correlated with opening to experience (r = 0.14, p = 0.56), responsibility (r = 0.13, p = 0.74), emotional stability (r = 0.08, p = 0.248), and kindness (r = 0.05, p = 0.498). No relation was found with the total MFI, total irritation, emotional irritation, cognitive irritation, or mental fatigue. Finally, the results show a high correlation of status with the variables self-esteem (r = 0.15, p = 0.30), responsibility (r = 0.08, p = 0.242), kindness (r = 0.08, p = 0.248), self-regulation (r = 0.07, p = 0.350), and emotional stability (r = 0.07, p = 0.316). However, no correlation was found with activity reduction, total irritation, cognitive irritation, emotional irritation, or mental fatigue. The correlations between variables are included in Table 4.

4.4. Analysis of Criterion

Firstly, the variables ranging from self-efficacy to self-esteem are analyzed. The total ARS shows significant and positive correlations with all the variables considered. A negative correlation with the health variable was found due to a high score in the measurement instrument (GHQ), indicating worse health. The highest correlation was obtained for the variable self-esteem (r = 0.183, p < 0.01) and the lowest one for the self-efficacy variable (r = 0.119, p < 0.05). The only variable where no significant correlation was found is life satisfaction (p < 0.10).

Among the four factors included in the ARS, the health factor exhibits the highest performance, obtaining significant correlations with self-efficacy (r = 0.146, p < 0.05), health (r = −0.131, p < 0.05), and self-esteem (r = 0.234, p < 0.01).

In addition, the correlation between the factors related to fatigue, measured with the MFI and the ARS, was obtained. Negative and significant correlations were found between the global scores of the two scales (r = −0.132, p < 0.05). Furthermore, the total ARS also exhibits negative and significant correlations with the following MFI factors: mental fatigue (r = −0.231, p < 0.01), reduced activity (r = −0.295, p < 0.01), and reduced motivation (r = −0.122, p < 0.05). Regarding the ARS factors, health and economy are highlighted. The health factor correlates negatively and significantly with mental fatigue (r = −0.218, p < 0.01), reduced activity (r = −0.291, p < 0.01), and reduced motivation (r = −0.145, p < 0.05). The economy factor correlates negatively and significantly with general fatigue (r = −0.160, p < 0.05), mental fatigue (r = −0.188, p < 0.01), and reduced activity (r = −0.144, p < 0.05).

The total ARS correlates positively and significantly with the following factors: emotional stability (r = 0.199, p < 0.01), responsibility (r = 0.190, p < 0.01), kindness (r= 0.208, p < 0.01), and openness to experience (r = 0.162, p < 0.05). Regarding the ARS factors, leisure and health stand out. Leisure correlates positively and significantly with emotional stability (r = 0.264, p < 0.01), responsibility (r = 0.147, p < 0.05), and kindness (r = 0.232, p < 0.01). The health factor correlates positively and statistically significantly with responsibility (r = 0.131, p < 0.05), kindness (r = 0.168, p < 0.01), and openness to experience (r = 0.165, p < 0.01).

To complete the analysis regarding personality factors, the results corresponding to the relationship between life satisfaction and the personality factors measured with the Operas instrument are also included. This way, life satisfaction correlates positively and significantly with extroversion (r = 0.194, p < 0.01), emotional stability (r = 0.323, p < 0.01), and kindness (r = 0.216, p < 0.01). However, no significant correlations were obtained for the responsibility and openness to experience factors.

5. Discussion

5.1. Hypothesis-Driven Discussion

Hypothesis 1.

“The Scale of Attitudes toward Retirement (ARS) must exhibit an adequate index of internal consistency”.

According to the obtained results, this hypothesis is validated. The internal consistency of the scale evaluated using Cronbach’s alpha coefficient yields positive results. Specifically, the complete scale presents an index higher than 0.80. In addition, the following factors considered exhibited acceptable values: leisure (alpha = 0.836), economy (alpha = 0.791), and status (alpha = 0.783). Finally, the value obtained for the health factor is the lowest one, but the obtained value is considered acceptable (alpha = 0.673).

Hypothesis 2.

“It is expected to confirm the existence of the four dimensions proposed in the Attitudes Toward Retirement Scale (Leisure, Economy, Status, and Health) by means of a confirmatory factor analysis (CFA), with an adequate reliability for all the factors (internal consistency)”.

This hypothesis is validated according to the obtained results. The EFA and CFA analysis revealed the existence of the four expected factors: leisure, status, economy, and health, as it was initially proposed. The statistical indices obtained with both types of analysis are satisfactory. Therefore, the structure of a scale with 20 items (see Appendix A) distributed in four dimensions was confirmed with 56.31% of the total variability and a total Cronbach’s alpha of 0.815. The leisure dimension has a Cronbach’s alpha of 0.836 and explains 18.48% of the total variability. This dimension consists of the following seven items: (i) retirement allows us to do pleasurable activities; (ii) retirement is a time to enjoy; (iii) when we retire, we have more time for our hobbies and interests; (iv) retirement allows us to spend time with our loved ones, (v) the best thing about retirement is the freedom from schedules, responsibilities, and demands; (vi) retirement means spending time with family and friends; and (vii) retirees travel more than working people. The status dimension exhibits a Cronbach’s alpha of 0.783 and explains 12.39% of the total variability. The dimension consists of the following four items: (i) In our society, only those who produce and consume are valued, so retirees are undervalued; (ii) society undervalues pensioners; (iii) when people retire, they lose the social prestige they once had; and (iv) pensioners are useless to society. Regarding the economy dimension, a Cronbach’s alpha of 0.791 was found, and it explains 14.20% of the total variability. This dimension consists of the following five items: (i) retirement causes negative changes in our economic status that we have to deal with; (ii) a pensioner has more difficulties in making ends meet; (iii) when we retire, we have more time to do activities, but we do not have enough money to afford them; (iv) retirement causes a considerable income drop; and (v) retirement means economic decline. Finally, the health dimension presents a Cronbach’s alpha of 0.673 (the lowest one) and explains 11.24% of the total variability. The dimension consists of the following four items: (i) After retirement, people suffer more from illnesses; (ii) with retirement, we are forced to grow older; (iii) retirement causes people to become physically weaker; (iv) anxiety and depression are normal after retirement. In summary, the existence of the factors’ structure proposed in the initial hypothesis is confirmed, and all dimensions exhibit adequate reliability.

Hypothesis 3.

“People with a high level of self-esteem will achieve higher scores in the ARS and, consequently, a significant relationship between the results given by the ARS scale and self-esteem will exist”.

The third hypothesis is also validated since a positive and significant correlation is found between the level of self-esteem and the score obtained in the total ARS. Retirement is one of the basic challenges that people face after an active working life. Self-esteem deals with the vision that we have about ourselves of being able to lead a meaningful life and meet its demands; that is, it is confidence in the ability to successfully face life’s challenges. On the other hand, attitudes are internal beliefs that affect personal acts and show personal characteristics, and at the same time, attitudes are stable organizations of beliefs and desires towards something. If a person auto-evaluates positively his/her abilities to face retirement, the obtained results verify that the attitude he/she will show towards the retirement process will also be positive. The person with a positive self-esteem will consider that he/she can satisfactorily face this new stage of life, and the evaluation of retirement and the retirement process will also be favorable (retirement is considered a stage of opportunities). If a person believes and feels that his/her state of health is good and he/she has a medium-high status, he/she will have more tools to face a new situation, feeling more capable of facing it and with a more positive emotional position (positive self-esteem). Thus, a more detailed analysis of the obtained data reveals that the health factor and the status factor positively and significantly correlate with the level of self-esteem. Therefore, positive self-esteem will be linked to a positive attitude toward retirement. On the other hand, low self-esteem will be one of the causes that push towards an unfavorable attitude towards retirement.

Hypothesis 4.

“People with a high level of general self-efficiency will achieve a better score in the ARS and, hence, a significant and positive relationship between the obtained score in the ARS and the general self-efficiency will exist”.

The fourth hypothesis is also validated. According to the obtained results, a positive and significant correlation exists between the level of self-efficacy and the ARS total score. This result seems to be reasonable because self-efficacy is the constant feeling that a person must efficiently manage stressful situations. Retirement is a time of change, causing feelings of insecurity and fear that lead to stress. If the person has a high level of self-efficacy, he/she will feel that he/she is ready to face this new situation, believing in his/her knowledge and abilities to effectively manage the retirement stage. Therefore, the attitude towards this moment of change will be positive since the evaluation of retirement will be acceptance and not rejection. Self-efficacy and attitudes have a bidirectional relationship; that is, self-efficacy leads the person to favorably value retirement, and attitudes can also increase a person’s confidence to face a new situation. Going deeper into the correlations, it was observed that health is a high influencing factor, so if the person feels that his/her health is optimal, he/she will think that he/she can face new situations and that he/she has the tools to do so, whereas if he/she feels weak, his/her motivation and his/her interest in facing a new situation will be lower.

Hypothesis 5.

“People with a high level of self-regulation will achieve a high score in the ARS. Therefore, a significant and positive relationship between the obtained score in the ARS score and self-regulation will exist”.

According to the obtained results, the fifth hypothesis is confirmed and validated, with a positive and significant correlation between the level of self-regulation and the ARS total score. Considering the definition of self-regulation, i.e., the person’s ability to modify their behavior due to the demands of specific situations, a positive relationship with attitudes towards retirement exists. Thus, if the level of self-regulation is high, the person will be prepared to change and he/she will modify their behavior according to changes linked to retirement. This factor interacts with environmental factors (retirement) with individual predispositions (the ability to adapt oneself to changes). Thus, if a person can modify his/her behavior and adapt him/herself to a new situation that arises from retirement, he/she will feel capable of facing it. Hence, attitudes towards retirement will be positive. A significant correlation was found between the leisure factor and self-regulation. This result can be explained by considering leisure as one of the resources for facing retirement. Thus, persons with self-regulation use leisure to motivate themselves, regulate their behaviors and thoughts, and provide an adaptive response to retirement. According to the research regarding self-regulation, some domains coexist in emotional regulation that give rise to self-regulation, such as attentional processes, characteristics of the stimuli, understanding the internal emotion signals, the regulation of demands, and the selection of an adaptive response. Leisure time, as it was previously pointed out, can be a resource for facing retirement, both enhancing adaptation and self-regulation to retirement.

Hypothesis 6.

“People with a better state of general health will exhibit a better attitude towards retirement. Hence, a negative and significant correlation between the obtained score given by the GHQ and the score given by the ARS, since a high score in the GHQ means a worse state of health”.

The sixth hypothesis is also validated. A negative and significant correlation between health and status was found by means of both the GHQ and the total ARS. As the theoretical model predicted, a negative and significant relationship is obtained between the health status measured with the GHQ and the health factor obtained from the ARS. Attitudes interact with various factors, causing a person to act in a certain way or another, depending on the moment and situation. Taking this interaction into account, the obtained result seems to be logical. If the person’s state of health is optimal at retirement time, the person will have the strength to face it because they do not perceive themselves as physically weak, and health is not an obstacle to facing this new challenge. On the contrary, if the person feels down, their attitude towards retirement will be negative because retirement will be considered a time of changes that he/she cannot face or cannot enjoy due to the limitations derived from his/her poor physical condition and his/her weakness. With a poor state of health, the attitude towards retirement will be unfavorable, causing an earlier retirement from work for this reason. As a practical application, it would be interesting for a future study to analyze the relationship between attitudes toward retirement by using the ARS and the reasons why a person decides to retire and the time when he/she makes the final decision.

Hypothesis 7.

“People with a high level of work irritation will exhibit a worse attitude towards retirement. Therefore, a negative and significant correlation between irritation and the obtained ARS score will exist”.

This hypothesis is validated because the negative and significant correlation between the ARS and the global irritation scale is confirmed. A more detailed analysis confirmed this correlation for all the ARS factors, with the exception of the status factor.

Hypothesis 7.1.

“A negative and statistically significant correlation between cognitive irritation and the obtained score in the ARS will exist”.

This sub-hypothesis is confirmed because a negative and significant correlation between the ARS and cognitive irritation was found. In this case, the economic and health factors are the causes of this relationship.

Hypothesis 7.2.

“A negative and statistically significant correlation between emotional irritation and the obtained score in the ARS will exist”.

This sub-hypothesis is also confirmed. In this case, the leisure factor is also added to the economic and health factors. These results are in agreement with the theoretical model since a person who presents high levels of irritation, where irritation is defined as a state of progressive psychological exhaustion that cannot be alleviated with normal breaks, will generally have a negative mood that makes him/her pessimistic, including their attitude towards retirement. In both cognitive and emotional irritation, two ARS factors exist (the economic factor and the health factor). This fact is in agreement with the theoretical model. Therefore, the economic and health factors have the highest negative relationship with a state of irritation in the work context. The regression analysis carried out in this study supports these conclusions.

Hypothesis 8.

“People with a high level of fatigue will have a worse attitude towards retirement. Therefore, the obtained results in the fatigue state will have a negative and statistically significant relationship with the ARS scores”.

According to the obtained results, a negative and significant correlation between the ARS and the total MFI score is confirmed. In a more detailed analysis, factors can be found as follows:

Hypothesis 8.1.

“The state of physical fatigue will have a negative and statistically significant relationship with the ARS scores”.

This sub-hypothesis is not confirmed since no negative and significant relationships are obtained between physical fatigue and the total ARS scores. In addition, no correlation between physical fatigue and the four ARS factors was found in the following: leisure, status, economy, and health.

Hypothesis 8.2.

“The state of mental fatigue will have a negative and statistically significant relationship with the ARS scores”.

This sub-hypothesis is confirmed since a negative and significant correlation between mental fatigue and the total ARS scores was found. The economic and health factors present a negative and significant relationship, supporting this result. However, the factors of leisure and status have no correlations with mental fatigue.

Hypotheses 8.3.

“The activity reduction state will have a negative and statistically significant relationship with the ARS scores”.

This sub-hypothesis is confirmed since a negative and significant correlation between the total ARS score and the activity reduction status was found. In addition, a negative and statistically significant relationship was obtained between the state of reduced activity and all the factors included in the ARS: leisure, health, status, and economy.

Hypothesis 8.4.

“The state of reduced motivation will have a negative and statistically significant relationship with the ARS scores”.

This sub-hypothesis is also validated. A negative and significant correlation between reduced motivation status and the total ARS score was found. The health factor supports this result, presenting a negative and significant correlation with reduced motivation.

According to the results, three out of the four fatigue factors measured using the MFI exhibit a significant relationship, and consequently, these results are in agreement with the theoretical model, considering that a higher state of fatigue is linked to a worse attitude toward retirement. This is due to people who, when in a state of weakness or asthenia, feel less capable of facing a new situation, such as retirement. Hence, the factors of mental fatigue, reduced activity, and reduced motivation are significantly related to low scores on the ARS. Although three out of the four fatigue factors show significant results, physical fatigue is not related. This may be because the cognitive component plays a more relevant role in attitudes toward retirement than the physical or psychological components.

Hypothesis 9.

“A positive and statistically significant relationship between the personality dimensions included in the FFM and the ARS scores will exist”.

The ninth hypothesis is validated in four of the five factors included in the FFM personality model, specifically in the factors of emotional stability, responsibility, kindness, and openness to experience.

Hypothesis 9.1.

“People with a high level of extroversion will exhibit a better attitude toward retirement”.

This sub-hypothesis is not fulfilled. No relationship was found between the high levels of extraversion and the obtained score in the ARS. In addition, no significant correlations with the factors included in the ARS were found. This is a surprising result because a significant relationship between extroversion and attitudes was initially expected according to other studies, where it is pointed out that extroversion causes people to remain more active in retirement and socially committed, and hence, extroversion was initially considered a positive attitude that leads to consider retirement as something positive.

Hypothesis 9.2.

“People with a high level of emotional stability will have a better attitude toward retirement”.

This sub-hypothesis is validated since a positive and significant relationship appears between emotional stability and the total score obtained in the ARS. The leisure factor enhances this result, presenting a positive and significant relationship with emotional stability. The factors of economy and health do not support these data.

Hypothesis 9.3.

“People with a high level of responsibility will have a better attitude toward retirement”.

This sub-hypothesis is validated. A positive and significant correlation between levels of responsibility and the total score in the ARS was obtained. In addition, a positive and significant relationship between responsibility and the factors of leisure, economy, and health was found. The status factor shows a different behavior than the other three factors since no significant correlations appear.

Hypothesis 9.4.

“People with a high level of agreeableness will have a better attitude toward retirement”.

This sub-hypothesis is validated according to the obtained results. A positive and significant relationship between agreeableness and the total score in the ARS was obtained. Leisure and health factors support this result, with a positive and significant relationship. On the other hand, economic and leisure factors do not show correlations.

Hypothesis 9.5.

“People with a high level of openness to experience will have a better attitude toward retirement”.

High levels of openness to experience are positively and significantly related to scores in the ARS. Health and economic factors show positive and significant relationships that support this result. On the other hand, the factors of leisure and status do not exhibit correlations with openness to experience. This result has a high practical impact, specifically for designing programs that encourage retirees to engage in activities that allow them to have deeper contact with the designed experience and take advantage of the opportunities around them.

Four out of the five personality factors considered obtained significant relationships. This result reveals that personality, as evaluated by the FFM model, is related to attitude toward retirement.

Hypothesis 10.

“A positive and statistically significant relationship between life satisfaction and the personality dimensions contemplated in the FFM will exist”.

This hypothesis is validated in the case of extroversion, emotional stability, and agreeableness. However, it is not supported in the case of responsibility and openness to experience.

Hypothesis 10.1.

“A positive and statistically significant relationship between life satisfaction and extroversion will exist”.

This sub-hypothesis is validated since a positive and significant correlation between life satisfaction and extroversion was found.

Hypothesis 10.2.

“A positive and statistically significant relationship between life satisfaction and emotional stability will exist”.

The obtained results confirm that a positive and significant relationship between life satisfaction and emotional stability exists and consequently, this sub-hypothesis is validated.

Hypothesis 10.3.

“A positive and statistically significant relationship between life satisfaction and responsibility will exist”.

This sub-hypothesis is not confirmed since no significant positive correlations were found between life satisfaction and responsibility.

Hypothesis 10.4.

“A positive and statistically significant relationship between life satisfaction and agreeableness will exist”.

This sub-hypothesis is confirmed since a positive and statistically significant correlation was found between life satisfaction and agreeableness.

Hypothesis 10.5.

“A positive and statistically significant relationship between life satisfaction and openness to experience will exist”.

This sub-hypothesis is not confirmed. No positive and significant correlations were found between life satisfaction and openness to experience. The results are not in agreement with those found in the bibliography related to life satisfaction. In this study, significant relationships were found between life satisfaction and neuroticism and responsibility and openness to experience. However, the obtained results show other types of relationships, suggesting a future research line for clarifying this discrepancy.

This study enlarges the knowledge of attitudes toward retirement, providing a measuring instrument to study and investigate these attitudes. By using this tool, the elimination of stigmas towards retirement and a satisfactory adaptation to the retirement process can be obtained.

5.2. Discussion-Driven Objectives

The first objective of the study was to create a measurement instrument to assess attitudes towards occupational retirement. The 20-item ARS was developed considering four factors: (i) leisure, (ii) economy, (iii) status, and (iv) health (see Figure 2 and Appendix A). The answers are given through a Likert-type scale with response options ranging from 1 (strongly disagree) to 7 (strongly agree). To obtain the final score of the items of the economy, status, and health dimensions, answers must be inverted. Once this is done, a high score on each of the factors, as well as on the global scale, indicates a better attitude towards retirement. The obtained values of the reliability and validity indices are adequate.

Retirement is a crucial moment in workers’ lives, and the worker’s perception of retirement highly influences the way he/she faces and adapts to retirement. This way, if the attitudes that a person has towards this moment are known, the retirement adaptation can be prepared from a psychosocial approach in order to improve the quality of life of the future retiree. The ARS is an important tool to analyze this change and to design a program according to the obtained results, helping to face one of the challenges that developed countries have now, the occupational retirement, as pointed out by [1].

In this line, previous studies pointed out the importance of studying retirement due to the rupture that occurs in terms of attachment to the work role, causing consequences at psychosocial and emotional levels [48,49]. This way, if the attitude to retirement is positive, an effective adaptation to retirement and a higher life satisfaction is reached, and, consequently, a better quality of life can be achieved. If attitudes are analyzed through instruments designed for this purpose, and those attitudes of the person who is retiring are evaluated by a specialist, successful aging and high longevity are achieved since attitudes influence people’s longevity, as different authors revealed [11,12,50].

The second objective of the study was to analyze the relationship between attitudes toward retirement and positive aspects of general health, self-esteem, self-efficacy, and self-regulation. In this study, it has been observed that the higher scores in the ARS were reached by people who presented a higher level of self-esteem; hence, the health and status factors are correlated. Self-esteem is related to the sense of esteem, worthiness, and ability. A positive self-esteem will enhance psychological well-being, leading to a more positive perception [19]. All this will allow obtaining a healthier quality of aging to face the different changes in the life cycle.

Regarding general health status, people who show a better attitude towards retirement have a better score on the health-related items of the questionnaire. On a psychological level, the changes that occur during aging will influence the way aging is perceived in terms of self-realization and leisure activities. This way, a person who perceives him/herself as an active person is an autonomous and independent person [51].

The obtained results show that people who exhibit a high level of job irritation have a negative attitude toward retirement. In addition, the state of health influences how retirement is experienced. On the one hand, health influences delaying or advancing the retirement age, and on the other hand, health enhances or limits the development of activities and social contacts [29,48,49]. If the attitude towards this new stage is positive, important health triggers are prevented in the retirement process. In aging, phenomena such as the loss of friends, social roles, and mandatory retirement will have a social impact [18]. An inadequate assessment of attitudes toward retirement can affect the retiree or person with an age close to retirement, especially those who consider work as a source of value and personal identity [13]. The results of this study indicate that attitudes towards retirement are related to positive variables such as self-esteem, which enhances people’s abilities to face this new change during this new stage of life. However, when attitudes toward retirement are linked to negative variables such as irritation or fatigue, such attitudes can cause a lack of adaptation to the retirement process, and therefore, negative consequences appear, affecting the person on physical, psychological, and social levels, or in other words, affecting his/her health.

According to the WHO [52], health is a complete state of bio-psychosocial well-being and not only the absence of pathologies. Therefore, this study is not only a tool to analyze attitudes toward retirement but also, this study provides a series of measurements to improve the health of people who are going to face retirement and, at the same time, active aging [53]. Through the study of attitudes, measurements are proposed to improve the quality of life by optimizing the physical, psychological, and social potentials of a person with an age close to retirement or a retiree. Therefore, these results are supported by previous studies that indicate that it should not be forgotten that active and healthy aging are focused on quality of life.

Finally, the third objective was to analyze the existence of a relationship between attitudes toward retirement and certain personality factors, such as those contemplated in the FFM model. The study found a positive and statistically significant relationship between positive attitudes toward retirement and emotional stability, responsibility, kindness, and openness to experience. Consequently, these results confirm that personality is related to attitudes toward retirement.

In recent years, research on developmental psychology, where the different stages that comprise the life of human beings are described and studied, has provided scientific information about longevity [4]. At the present time, different branches of science have focused on studying the changes that a human being goes through when reaching old age, such as retirement [5], the impact of aging [6], and the quality of life in the elderly [7]. As it was revealed in this work, these three changes are influenced by attitudes.

Retirement is a moment in life that all workers around the world must face during his/her life cycle. Three key moments appear during the retirement process: (i) the years previous to retirement, (ii) the end of the work life, and (iii) the retirement life. The first one is developed during the last years of active work when the worker´s age is close to retirement, and he/she starts to think about the end of his/her working life. At this time, worker faces many uncertainties and the attitudes toward retirement play a key role in the adaptation to retirement. Thus, for those who have negative attitudes toward retirement or those who have an attachment to work and, hence, are not prepared for this life change, many psychological problems affecting mental health, such as social isolation, mental fatigue, apathy, anxiety, and depression, appears, leading to an increase in the use of medication or medical/psychological assistance. This lack of well-being can also cause health problems, resulting in diverse illnesses in the elderly. However, those who have a positive attitude towards retirement are prone to face this period of time in a mentally and physically healthier way.

The second one, retirement time, is when mental health problems arise and become worse for those workers who are not well prepared for retirement. At this time, a worker must face a sharp change in his/her life in which the worker has to fill the former working time in a day (8 h in a day) with other activities. Many people have a sense of loss in social status due to their attachment to work, a loss of economic resources, and, in some cases, a sense of being a non-productive member of society. However, for those workers with high study levels and positive attitudes towards retirement, this time is an opportunity to spend time with family and friends or to carry out many activities that they could not develop in the past due to a lack of time.

Regarding the final one, retirement time is increasing due to the increment of life expectancy in the nowadays world. Thus, the way workers prepare in the previous stages for this long period of time will cause well-being and a healthier retirement or, otherwise, a difficult time, where people must face and overcome diverse mental health problems linked to a wrong adaptation to this new life stage.

This way, the assessment of attitudes toward retirement is a key issue for improving the adaptation of workers to retirement. Thus, the ARS is a promising tool that allows for individually identifying the psychosocial factors (leisure, health, status, and economy) influencing the retirement process of a person. Accordingly, the information obtained from the ARS will be very useful for identifying potential problems and establishing an individual intervention at the psychological level for all the people who face retirement, especially those with negative attitudes toward retirement. This intervention must be applied with an educational program in terms of a retirement preparation course, where changes produced during retirement are implemented with psychological tools and the use of new information and communication technologies (ICT). This way, an adequate adaptation can be achieved, allowing one to enjoy a healthier retirement. Thus, the quality of life of people and their sense of well-being are improved and, in addition, it can serve as a way of prevention of certain illnesses. The proposed ARS can be used at any one of the key moments of the retirement process, but it is considered to be more effective at the early stage of retirement. This way, workers with personalized psychological aid can know better from the beginning those changes and difficulties that he/she will face in the near future and the psychological tools that will help him/her overcome them. Anyway, the ARS can also be useful in the other two moments of the retirement process. At the time of facing the end of work life and during retirement life, the ARS allows to reveal the causes why people will experience problems derived from retirement, and this way, a personalized intervention based on the results of the analysis of the ARS can be helpful to solve such problems. This way, a society that prepares workers for retirement is a society that enhances sustainability, on the one hand, by taking care of the personal, social, and community well-being and, on the other hand, by the efficient use of available materials and socio-sanitary resources.

Finally, retirement must be considered from an interdisciplinary point of view, and hence, this process cannot be separated from social, economic, and personal aspects (psychosocial factors) since these aspects will determine the adaptation to retirement. Attitudes toward retirement play a key role in cognitive and emotional levels in facing this new life stage. The person who is going to retire must assume a series of expectations and attitudes that will benefit or harm him/her. For all these reasons, it is important to address considering different aspects from an interdisciplinary approach, the wide network of attitudes and beliefs towards retirement determined by society and by oneself, working on them through the ARS that has been developed in this study, and the programs for retirement preparation.

Government policies are needed to improve the four influencing factors on retiree people, lowering the discrimination and inequality that are risk factors linked to cognitive deterioration. In addition, not only should the age factor be considered, but other factors that influence stigmas and discrimination towards certain collectives, such as women, people with certain incapacity, or people with a low economic state, should also be contemplated.

5.3. Limitations of the Study

The first limitation of this study is the sample size that was planned for the different phases of our study. In the first stage of the study, 50 people answered the 80-item ARS scale. Although this is a very small sample, at this stage, it was a pilot study whose aim was to test the items’ performances.

A second limitation of this study is the extension of the battery of questionnaires used in this study (eight questionnaires). For this reason, most of the people refused to participate in the study, but all of the questionnaires were necessary to achieve reliable results.

A third limitation was that the sample was taken from various geographical zones in Spain, so it would be appropriate in future works to extend the sample to include people from other countries around the world.

In future studies, it is intended to increase the study sample and to add medical health data for verifying the consequences of retirement from work and its relationship with attitudes. Regarding the future lines of research that will be covered in future works, it would be a great advancement to study the effect that intervention programs within the framework of active aging may have on attitudes toward retirement. It is also possible to analyze the mediating nature of some of the variables included in this study, such as self-regulation or self-esteem, within the development of attitudes toward retirement.

As future research lines, the following is proposed:

• To delve further into the analysis of the relationship between personality and life satisfaction. The results found in this present study are not in agreement with those found in the literature.
• To advance the research to verify the effectiveness of intervention programs by using the ARS.
• To use the ARS with other samples to check the generalizability of the scale.
• To investigate the relationship between attitudes towards retirement, through the ARS, and the motivation, reasons, and moment in which the retirement decision is taken, comparing the obtained results with others carried out in this line.
• To include cognitive reserve into the variables to study.

6. Conclusions

The obtained results confirm the existence of four factors in the scale of attitudes toward retirement, as formulated from a theoretical level: (i) leisure, (ii) economy, (iii) status, and (iv) health. The obtained correlations between the scale and the variables considered show that the attitudes toward retirement are linked, in a positive way, to aspects such as self-esteem, whereas significant and negative correlations related to irritation and fatigue were found. These results show, on the one hand, that when retirement attitudes are related to variables of a positive type, such as self-regulation, self-esteem, and self-efficacy, then the personal capabilities to face this new life stage are enhanced. However, on the other hand, when attitudes towards retirement are linked to negative variables, such as irritation or fatigue, attitudes can cause a lack of adaptation to retirement and, hence, a series of negative consequences on his/herself, affecting their physical, psychological, and social levels, i.e., affecting his/her health.

The scale elaborated and validated in this work constitutes an effective tool to determine attitudes toward occupational retirement. In this way, attitudes toward retirement are a key pillar in the study of active aging, defended by the WHO. This way, the evaluation and analysis of attitudes toward retirement with the ARS scale provide relevant information to improve the quality of life of elderly people, fight against false beliefs and stereotypes about retirement, and help to optimize the physical, psychological (including cognitive reserve), and social potential of the person who faces retirement, since, as it has been verified in the validation of the scale, aforesaid attitudes correlate with the person’s health, status, economy, and leisure.

Retirement from work cannot be separated from social, economic, and personal aspects. In particular, attitudes toward retirement are important on cognitive and emotional levels to face the new stage of life and live fully. It is important to tackle the wide web of attitudes and beliefs towards retirement, determined by society and by oneself from different points of view. It is relevant to work with people who are going to face retirement so that they can successfully adapt and live this new stage. Attitudes towards retirement should be analyzed by means of the ARS and worked on through the retirement preparation programs. These types of programs are of the highest relevance since they improve the quality of life of the future retiree and modify the negative perception of the moment of retirement, thus preventing negative emotions, cognitions, and feelings toward retirement. Government policies are needed to improve the four influencing factors on retirement people, lowering discrimination and inequality, which are risk factors linked to cognitive deterioration. In addition, not only should the age factor be considered, but other factors that influence stigmas and discrimination towards certain collectives, such as women, people with certain incapacity, or people with low economic status, should be considered.

Finally, an attitude evaluation and educational program, where the changes produced during retirement are worked through with psychological tools and the use of new ICTs to face the new stage of life, are key issues. As a consequence, the impact on society of such retirement preparation programs will lead to a more sustainable world since, on the one hand, the quality of life and well-being at the community, social, and personal levels during a long time period of the life cycle of people are improved and, on the other hand, the use of available materials and socio-sanitary resources will be more effective, leading to a more sustainable world. Thus, a society where tools of psychological evaluation and intervention, such as the ARS proposed in this study, allows the detection of problems in people facing retirement and leads to a more sustainable and evolved society that provides quality of life and personal, social, and community well-being.