Validity and Reliability of Persian Version of the 12-Item Expectations Regarding Aging Survey Iranian Older Adults Population during COVID-19 Pandemic.

Background: The current pandemic has classied older people into the most vulnerable category. Prior to the pandemic, they led active social lives, but are now cautioned to adopt more precautionary and preventive behaviors. Older people are advised to change their lifestyle given the challenges and expectations regarding aging. Therefore, the present study is designed to assess the construct validity and reliability of the Persian version of the 12-Item Expectations Regarding Aging Survey among older adult Iranian population during COVID-19 pandemic. Methods: The Persian version of this scale was developed using translation and revision in current study. The construct validity was assessed through Exploratory Factor Analysis (EFA) and Conrmatory Factor Analysis (CFA). Also, the reliability was assessed through through its internal consistency (Cronbach’s alpha and McDonald’s omega), composite reliability (CR), and maximal reliability (MaxR). The data compiled online was based on a sample of 400 older adults aged above 65 years. Results: The results of the study indicate that the Persian version of this scale comprises three factors with 11 items (Factor 1: Physical Health; Factor 2: Mental Health; Factor 3: Cognitive Function) that explaining 65.764% of the total variance. The CFA results, after reviewing the model modication indices, shows the revised three factors model ts the data well with excellent internal consistency and reliability. Conclusion: The nding indicates that the Persian Version of ERA is reliable and valid. It can be used for aging, educational, and clinic research, to develop and evaluate on suitable intervention to enhance the quality of life among Iranian aging population.


Introduction
With the onset of Covid 19 disease, the World Health Organization (WHO) has identi ed quarantine and social isolation as one of the safest ways to control and combat the disease. However, the quarantining a large segment of society will not be without the much dreaded psychological, social and economic effects (1). As a result, older people with active social life are probably experiencing major changes in their normal routine of life, leading to feelings of isolation, loneliness, or low mood (2).
Aging is a critical process of life where physical and mental needs are considered social necessity, and close attention on health-promoting behaviors is essential (3). As part of the aging process, older people would experience change in personality and behaviour, with incidences of loneliness and depression reported as common problems among the elderly (4,5). Studies indicate that loneliness is a painful and nettlesome feeling for older people (6,7), with possibility of multiple negative physical and mental health concerns (8). Needless to say, loneliness and depressions an existing concern among older people is exacerbated during the pandemic. Studies highlight that concerns regarding COVID-19 are higher among older (9), expressing higher vulnerability which impacts their overall attitude to aging (10). The concerns expressed by the elderly is likely justi ed as the fatality rate of COVID-19 is higher among older compared to younger adults. The health concern is connected to the level of expectations regarding aging, hence the importance of addressing this concern among the elderly (11).
Aging expectations (ERA) are beliefs that people have about their level of physical and cognitive health with age (12). The concept of "successful" aging as an alternative to "usual" aging was rst mooted by Rowe and Kahn (1997) focusing on importance of health promotion and disease prevention. The proposition was for the elderly to play an active role in their health to enjoy full life as much as possible while maintaining full function (13).
Studies show that there is a positive relationship between attitude and tendency to participate in healthpromoting behaviors such as physical activity, stress management, interpersonal relations, increased longevity and a better quality of life in older adults (14)(15)(16)(17)(18)(19)(20). Correspondingly, negative expectations were associated with lower ability to engage in physical activity, accepting a more sedentary lifestyle, a higher likelihood of experiencing acute medical events, negative health outcome, cognitive decline, and emotional responses to stress (21)(22)(23)(24) and feelings of loneliness (25).
The importance of expectation in achieving and maintaining a high level of physical and mental function was explored by Sarkisian et al (2002) with a 38-item survey to measure expectations regarding aging  in the USA. The scale incorporated items of general health, cognitive function, mental health, functional independence, sexual function, pain, sleep, fatigue, urinary incontinence and appearance. Higher scores on the ERA-38 expressed an increased attainment of high physical and mental functioning with aging (12). It was interesting to note that functional status and the psychosocial components of health, such as depressive symptoms, social functioning, emotional well-being, and biological and cultural status have a more signi cant impact on perceptions compared to others aspect (26). It was also noted that older adults' age expectations differed across race/ethnic groups (27). Since the risks of hospitalization or mortality resulted COVID-19 are higher in older adults than in young adults, the use of preventive approach in this group is more emphasized. An approach which is relatable to their view of aging and expectation regarding aging. The attitude adopted by older adults with concerns about COVID-19 can be seen from two perspectives. Firstly, a prescribed isolation may result in older adults feeling lonely or alternatively, Alternatively, the prescribed isolation may result in them feeling safe and more positive (29). If the expectations regarding aging remained relatively stable, a higher resilient was expressed in facing COVID-19 (30).
In justifying the use of ERA-12, it is noted that this instrument showed acceptable validity and reliability when applied to a sample of middle aged Singaporeans, middle aged Koreans and elderly women (31,32). The focus of the present study is driven by two important considerations, (1) the increase in the elderly population as a vulnerable group in Iran, and (2) insu cient resources on aging expectations among Iranians which is key in developing informed and effective policies in this area. The present study, therefore, aims to evaluate the construct validity and reliability of Persian version of the 12-Item Expectations Regarding Aging Survey among an Iranian older adult population during COVID-19 pandemic.

Design
A methodological cross-sectional approach is adopted to evaluate the psychometric properties and the feasibility of a Persian version of the "ERA-12". The sample consist of Iranians older adults aged above 65 years, uent in Persian and residing in Tehran. Additionally, the respondent's criteria to participant in this study include the use of social networks and informed consent is provided to be participated in this study.

Measures
There were two sections to the questionnaire. First section captures the demographics of the respondents. The following section contains items from the ERA-12 instrument to capture expectations on physical health, mental health, and cognitive function (28) of the elderly in Iran. All items assessing respondents' expectations to aging applies a 4-point Likert scale to be consistent with the original study by Sarkisian et al. (2005).

Procedure Translation
The original ERA-12 instrument developed in English needed to be translated and revised for the Iranian respondents. In order to develop the Persian version of this scale, rstly, a written approval was obtained from Catherine Sarkisian, the developer of the original scale. Upon obtaining the approval, two independent English-Persian translators worked on translating the items. Subsequently, both the translated versions were checked and evaluated by a group of experts (including some of this paper's authors (H.SH and L.H). In the nal process, two professional translators proceeded to create a single Persian version of this questionnaire. All of their professional suggestions were included in the translations.

Construct Validity and Reliability Assessment
In this study, we randomly divided the dataset into two parts to examine the construct validity and reliability. First, the maximum likelihood exploratory factors analysis (EFA) with Varimax rotation was conducted using the rst dataset (n = 200). The Kaiser-Meyer-Olkin (KMO) measure of sampling adequacy test and the Bartlett's test of sphericity were used to ensure the adequacy of the sampling and suitability of the data for conducting factor analysis. Extraction of factor structure was referring the criteria of (1) eigenvalues > 1; (2) communalities > .3, and (2) scree plots (33)(34)(35). Next, we conducted the maximum likelihood con rmatory factor analysis (CFA) in accordance with the factor structure obtained from the EFA using second dataset (n = 200). The assessment of model t was based on several t indices such as Chi-square (χ 2 ) test, Chi-square/degree of freedom ratio (χ 2 /df) < 4, goodness-of-t index (GFI) >.90, comparative t index (CFI) >.90, incremental t index (IFI) >.90, normed t index (NFI) >.90, Tucker-Lewis index (TLI) >.90, relative t index (RFI) >.90, root mean square error of approximation (RMSEA) <.08, and standardized root mean square residual (SRMR) <.09 (36). Moreover, construct validity was assessed through convergent validity and discriminant validity. To achieve convergent validity of the construct, Average Variance Extracted (AVE) should be greater than .5 and composite reliability (CR) should be greater than .7 (37). In this study,, we used the Heterotrait-monotrait ratio of correlations (HTMT) in assessing the discriminat validty and all values of the HTMT matrix should be less than .85 (38).
The reliability of the Iranian version of ERA-12 was examined through its internal consistency (Cronbach's alpha and McDonald's omega), composite reliability (CR), and maximal reliability (MaxR). Cronbach's alpha and McDonald's omega of higher than .7 are considered as acceptable internal consistency, CR and MaxR should be greater than .7 to demonstrate satisfactory reliability (37,39).

Multivariate Normality and Outliers
We tested both univariate and multivariate distributions in this study. The outliers, and the values of the skewness and kurtosis were determined and tested for univariate distributions. In terms of multivariate distributions, Mardia's coe cient of multivariate kurtosis was used to check the multivariate normality, the result of Mardia's coe cient was greater than .8 indicating deviation of multivariate normality (40). Also, Mahalanobis distance was used to identify the Multivariate outliers, and multivariate outliers can be identi ed through a Mahalanobis distance of p < .001 (41).

Ethical Considerations
The Ethics Committee of Mazandaran province, north of Iran approved the Ethical Considerations of this study (Code: IR.MAZUMS.REC.1399.6682). Additionally, we noti ed all participants regarding the aim and procedures of current study. The participation in this study was voluntary, where informed consent was acquired before participating in this study from all participants. In reporting the ndings, the study ensured that the anonymity of the individuals' information was maintained.

Pro ling of Respondents
The sample of this study consists of 242 (60.5%) males and 158 (39.5%) females with the mean age of 71.32 (SD ±6.09) year. Majority of the participants were married (82.75%) university graduates (55.25%) with good economic status (56.75%). The details of this results have been shown in Table 1.

Reliability and Validity
The results of the maximum likelihood EFA with varimax rotation on Persian version of the EARS are shown in Table 2 based on the rst dataset (n = 200). The results showed that there are three factors are extracted in explaining of 65.764% of the total variance, containing 11 items. One item was excluded because of low factor loading (< .5). The result of KMO test was .874, and the Bartlett's test of sphericity was signi cant (p < .001, 995.472, df = 66), indicating the adequacy of the sampling and appropriateness of the data to conduct factor analysis. See details in table 2 and gures 1.
Next, we performed maximum likelihood CFA using the second dataset (n = 200) to ensure and validate the factor structure extracted from EFA. Figure 2 shows the nal measurement model, three sets of measurement errors (between item 7 and 8, item 9 and 10, and item 9 and 12) were free to covary to adjust the initial measurement model in a better t according to modi cation indices. The results showed that after reviewing the model modi cation indices, the revised three factors model ts the data well, as evidenced by goodness-of-t indices (χ 2 (38) = 77.272, p < .05, χ 2 /df = 2.033, GFI = .967, CFI = .981, NFI = .963, IFI = .981, RFI = .947, TLI = .972, SRMR = .028, RMSEA (90% C.I.) = .051 [.043, .067]), and factor loadings for all items were more than .6 and signi cant. Moreover, as presented in Table 3, the result of Cronbach's alpha (ranged from .788 to .845), McDonald's omega (ranged from .791 to .845), CR (ranged from .772 to .846), and MaxR (ranged from .781 to .863) were all greater than .7 revealing satis ed internal consistency and reliability. Also, the results indicated that AVEs of all three factors were more than .5 and less than their respective CR, demonstrating convergent validity. In terms of discriminant validity, all values in the HTMT matrix analysis were smaller than .85, indicating discriminant validity established all for constructs (Table 4).

Discussion
The analysis of psychometric properties of the ERA-12 Persian version carried out on a sample of older adults Iranians based on CFA indicated a three-factor solution of the Expectation Regarding Aging Survey has a good t after one item was dropped (item 5: I expect that as I get older, I will spend less time with friends and family). The modi ed model including covariances between three factors (physical health, mental health, and cognitive functions) with total of 11 items showed an improved t. The ndings of the current study demonstrate an acceptable level of reliability, as indicated by good item discrimination scores and Cronbach's alpha values of the ERA-11. An outcome that is supported by previous ndings on the three factors solution to the Expectations Regarding Aging questionnaire (31,42). The ndings of a systematic review of existing ageism scales showed that among 11 scales included in the study, only the Expectations Regarding Aging Questionnaire had adequate ratings on the psychometric properties (43).
Despite the mentioned nding, the authors have concluded that none of the available ageism scales have both adequate scop and psychometric validity (43). Considering the three dimensions of ageism that include stereotypes (cognitive component), prejudice (emotional component), and discrimination (behavioral component) (44), the ERA concerns to assess the stereotypes (43).
In justifying the removal of item 5 from the nal ERA Persian version, several factors were considered. According to Yao (2020), some factors such as personal values and traits, role models of aging, personal experiences with growing older, and stereotypes about older adults in uence on subjective aging. In addition, subjective aging is also in uenced by cultural values and social interaction (45) allowing the present study to draws parallels with the cultural values of Iranians, speci cally the elder adults' position within the family. Although socio-economic changes have impacted family institution during the last decades, the Iranian families have preserved many of their traditional values, speci cally a support system for family members and respect for elder adults. The cultural value system has ensured Iranians maintain a strong relationship with their elderly parents after marriage, including providing care and support for them. Given the contextual sensitivity of subjective age, culture can be considered as the moderator variable in the relationship between views of aging and subjective age (46).
The ndings indicate that expectation regarding aging is attributable to physical and mental health, in addition to elder adults' cognitive function that is congruent with the existing knowledge. ERA plays an essential role in implementation of physically active lifestyles for older adults, in uencing health conditions, such as physical functions (15). In addition, positive ERA provides social support through network of friends (19), which improves quality of life (47), leading to a positive in uence on elderly peoples' mental health (48). Conversely, negative age stereotypes, highlights higher instances of developing psychiatric conditions and impaired coping skills (49). Studies have also noted a de cit in cognitive functioning among negative age stereotypes (50,51).
Our results also demonstrated three pairs of measurement errors. Firstly, between item 7 (as people get older, they worry more) and item 8 (its normal to be depressed when you are old). Next, between item 9 (I expect that as I get older, I will become more forgetful) and item 10 (It's an accepted part of aging to have trouble remembering names). Finally, between item 9 and item 12 (It is impossible to escape the mental slowness that happened with aging). The ndings indicate a semantic equivalence between worrying and depressing symptoms among older Iranians. Semantic equivalence is also observed between trouble remembering, being forgetful, and mental slowness. In order to minimize the response burden and increase response rate among older people, shorter questionnaires maybe a possible solution (31). However, further investigations are needed before shortening the next Persian version of the ERA. On the component of age-related health changes among the Iranians, the ndings reveal high priority for economic elements such as health insurance and income (52,53). Iranian older people expect that they face by a range of age-related health changes such as memory de cit, to have trouble in remembering names (54), or depressive symptoms(55, 56).

Implication
Base on evidence, the expectation regarding aging contributes to physical and mental health and cognitive function in elder adults. Therefore, the positive ERA improves older adults' life quality. It is a suitable scale for assessing ERA of older adults and improving target interventions to improve their conditions. It can be used for aging, educational, and clinic research to develop and evaluate on suitable intervention to enhance the quality of life among Iranian aging population.

Study limitations
This study, however, not without limitation. The sample respondents are from one geographical urban location, Tehran. It is likely, respondents from other regions may have responded differently. As such, a broad generalization of the results to the rural population maybe not be applicable. The e cacy of the Persian ERA can be strengthened by replicating this psychometric testing in other settings and by including comparison with other indicators of aging expectation.

Conclusion
The present study provides the appropriate Persian version of ERA-11 among Iranian elder adults that can detect and distinct three dimension of ERA in this age group. But the remarkable thing is that this scale can identify the Cognitive Function more powerfully than the Physical and Mental Health in this population.

Declarations
Ethics approval and consent to participate In this study, the ethical points of the Helsinki Declaration were observed, including informing all participants of the objectives and procedures of this study, ensuring the voluntary participation of the study, anonymity of the participants, and obtaining informed consent from all participants.

Not applicable
Availability of data and materials The datasets generated and analysed during the current study are available from the corresponding author on reasonable request.

Competing interests
The authors did not have any competing of interest in this work.

Not applicable
Authors' contributions HSN and LH designed the study and collected the data and drafted the manuscript. LS analysed the data and contributed to the interpretation of the result and writing the result . FKF and OK contributed to the interpretation of the result and writing the manuscript. All authors read and approved the nal manuscript. .599 .589 9. I expect that as I get older I will become more forgetful.