Psychometric properties of the Persian version of the ambivalent ageism scale (benevolent and hostile) in the adult population in Iran

Abstract

Background

With the growing population of older adults and the prevalence of negative attitudes towards them, the issue of ageism and its health and economic impacts in both benevolent and hostile contexts warrants special attention. It is crucial to examine the attitudes of other age groups towards older adults across different societies. Particularly, the benevolent dimension of ageism, which has been less explored in research, requires more focus. Therefore, this study aims to conduct a psychometric evaluation of The Ambivalent Ageism Scale among the adult population in Iran.

Methods

This methodological study was conducted in comprehensive health centers in Gorgan city in 2023. A total of 381 eligible adults participated. The Ambivalent Ageism Scale (AAS) was utilized, and the psychometric assessment included translation, face validity, and content validity. Additionally, exploratory factor analysis and confirmatory factor analysis were performed. The reliability of the scale was evaluated using the internal consistency method. The research findings were analyzed using SPSS and AMOS software version 24.

Results

Qualitative face and content validity assessments led to textual and editorial modifications of the items. The Content Validity Ratio (CVR), Item-Content Validity Index (I-CVI), and Kappa (K*) scores were acceptable for all items. In the exploratory factor analysis (EFA), similar to the original questionnaire, three factors were extracted, accounting for approximately 54% of the total variance. The fit indices in the confirmatory factor analysis (CFA) indicated an acceptable model fit. During CFA, four items were eliminated. The reliability of the entire questionnaire was deemed acceptable with a Cronbach’s alpha coefficient of 0.763. Consequently, the Persian version of the Ambivalent Ageism Scale was confirmed with nine items.

Conclusion

The Persian version of The Ambivalent Ageism Scale demonstrates sufficient validity and reliability for measuring attitudes towards aging within Iranian society. Given the cultural adaptation of this tool, the questionnaire can be utilized to assess adults’ views and attitudes towards older adults in both hostile and benevolent dimensions. Furthermore, it can aid in formulating family-oriented policies for older adult care and facilitate improvements in the quality of care for this population group.

Introduction

The global population of older adults, defined as those aged 60 and above, is steadily increasing [1]. In Iran, a significant upward trend in this demographic is projected between 2015 and 2050 [2]. Older adults constitute a heterogeneous group with diverse needs, and the aging process is influenced by genetics, lifestyle, interpersonal relationships, cultural contexts, and societal attitudes towards aging and older adults [3,4,5].

Despite the growing number of older adults, negative perceptions of this group persist in many countries [6–7]. These perceptions include viewing older adults as non-contributors to the economy, reminders of death and vulnerability [8], and as a group from which younger individuals seek to differentiate themselves [9]. Additionally, older adults are often judged by their appearance, deemed less attractive [10], and stereotyped as kind but incompetent [11]. This stereotyping, prejudice, and discrimination based on chronological age is termed “ageism”, a phenomenon that is pervasive worldwide [12], and ranks as the third most common form of discrimination, following sexism and racism [13]. Despite its prevalence, ageism remains largely unrecognized [14].

Older adults are particularly vulnerable to the adverse effects of ageism, which can have significant consequences [15]. A recent systematic review examining the impact of ageism on health across 11 dimensions in 45 countries, based on 25 years of empirical evidence, found that 95.5% of studies reported negative mental health outcomes associated with ageism [16]. Beyond health issues, older adults may also face limitations in employment, housing, and access to social services [17, 18].

Negative actions and attitudes do not encompass all existing behaviors towards older adults [19]. Stereotypes can be either positive or ambivalent [20]. While negative stereotypes often attribute characteristics such as illness, forgetfulness, moodiness, and depression to older adults [17, 21], positive stereotypes such as wisdom, kindness, trustworthiness, and generosity also exist in society [17, 22].

According to the Stereotype Content Model, two fundamental dimensions—“warmth” and “competence”—guide people’s perceptions and emotional and behavioral responses towards others or social groups. The perception of older adults is shaped by the combination of these dimensions, resulting in views of low warmth and low competence or high warmth and low competence [23, 24]. Stereotypes characterized by high warmth and low competence evoke feelings of pity and active facilitation, such as direct assistance, as well as passive harm, including neglect and contempt, and paternalistic prejudices. These views are referred to as benevolent or compassionate ageism, which are supportive and helpful in nature. In contrast, the hostile dimension of ageism leads to active harm, such as verbal harassment or discriminatory policies, and passive harm that results in feelings of humiliation [25, 26].

Benevolent stereotypes are subtle and often imperceptible, frequently being conflated with positive stereotypes. However, they ascribe unfavorable characteristics such as incompetence, making their acceptance in society easier compared to overtly negative forms of ageism [25, 27, 28]. Although this type of ageism is often applied with good intentions [10], supportive attitudes such as unwanted help, overprotection [25], and the use of “Elder Speak” [29] can diminish self-esteem, motivation, independence, performance, and social participation, ultimately leading to behaviors associated with helplessness [30].

Most scales for measuring ageism focus on the hostile dimension, which is more overt. However, the benevolent dimension also warrants measurement [25, 30]. The Ambivalent Ageism Scale (AAS) is the first scale to simultaneously measure both dimensions, which predict different outcomes. In other scales, these dimensions are often combined, obscuring the distinction between them. The Ambivalent Older Adults Ageism Instrument is the first scale specifically designed to measure both hostile and benevolent dimensions concurrently, recognizing that these dimensions do not predict the same outcomes and should not be conflated [31].

The Ambivalent Ageism Scale (AAS) was developed in Canada between 2016 and 2017 by Carey, Chastain, and Remedios, based on the Ambivalent Sexism Inventory and The Stereotype Content Model [11, 32]. The scale consists of 13 items: the first nine items measure the benevolent dimension, and the remaining four items measure the hostile dimension. Responses to each item are recorded on a 7-point Likert scale, ranging from “completely disagree” (score 1) to “completely agree” (score 7). The AAS has undergone psychometric testing in various countries, including Slovakia [33], Poland [34], India [35], Turkey [6], and Japan [36]. The Turkish version of the scale demonstrated sufficient reliability and validity. However, there is limited psychometric information available for the other versions.

To date, no study has been conducted in Iran to evaluate the AAS instrument in Iranian society among adults, despite the cultural and social differences of Iran compared to other countries, it is essential that this scale be comprehensively and systematically examined with appropriate and adequate psychometrics with Iranian culture. On the other hand, In Iran, instruments measuring ageism in older adults, including Kogan’s attitude toward aging [37], Palmer’s ageism [38], perception of aging [39], and attitude toward aging [4], have been validated based on the country’s culture.

The Kogan Attitudes to Aging Questionnaire was psychometrically validated by Rejeh et al. in 2012. However, this instrument is limited in scope as it only measures the attitudes of nursing students towards older adults [37]. Sharif Nia et al. conducted a psychometric evaluation of the Palmore Ageism Questionnaire for Iranian older adults in 2021. This instrument focuses solely on the negative aspects of ageism experienced by older adults, with the respondent population being older adults themselves [38]. Mir-Emadi et al. performed a psychometric evaluation of the Ageism Questionnaire in Tehran in 2020, which concentrates on the psychological component of ageism experienced by older adults, specifically its internal type and the older adults’ perception of their own aging process [39]. Rejeh et al. (2017) examined the psychometric properties of the Persian version of the Ageism Questionnaire among Iranian older adults, which also measures only the negative attitudes of older adults towards themselves [4].

Therefore, given the gaps in studies conducted in Iran, which indicate that psychometric instruments focus on explicit attitude scales measuring hostile attitudes alone, without considering benevolent dimensions, and given the increasing older adult population and the discrimination they face even in normal circumstances, as well as the consequences of such discrimination, this study aimed to determine the psychometric properties of the Persian version of the Ambivalent Ageism Scale (benevolent and hostile) in the adult population in Iran.

Methods

Study design

This methodological study, conducted in 2023, focused on the translation and investigation of the psychometric properties of the Ambivalent Ageism Scale (benevolent and hostile) among the adult population in Iran.

Study population/sampling

The study population consisted of 381 adults (aged 18–59 years) covered by urban health centers in Gorgan city. According to the 10-to-1 rule [40], which requires at least 10 samples per questionnaire item, a minimum of 260 samples were needed to conduct this study, accounting for potential sample dropouts. A total of 381 samples were included for factor analysis, with 191 samples allocated for exploratory factor analysis (EFA) and 190 samples for confirmatory factor analysis (CFA). Inclusion criteria for the study were: being aged 18 to 59 years, residing in Gorgan city, not suffering from obvious mental disorders as self-reported, and having at least basic reading and writing literacy. The exclusion criterion was providing the same answer to all questions.

Study instrument

Demographic information questionnaire

The demographic information collected in this study included the age, sex, education, occupation, and marital status of the eligible adult population.

Ambivalent ageism scale (AAS)

The AAS is a 13-item scale designed by Carey et al. between 2016 and 2017 in Canada to simultaneously measure hostile and benevolent attitudes towards older adults. The scale consists of 13 items that evaluate benevolent ageism (cognitive assistance/protection and unwanted help) with 9 items, and hostile ageism with 4 items. Responses to this questionnaire are recorded on a 7-point Likert scale, ranging from “completely disagree” (score 1) to “completely agree” (score 7). The total scores of the questionnaire range from 13 to 91, with benevolent scores ranging from 7 to 63 and hostile scores ranging from 7 to 28. Higher scores indicate higher levels of ageism in that dimension or overall [25].

Translation procedure

Following the model by Wild et al., the translation process was conducted in a forward-backward manner [41]. First, an email was sent to Chasteen, the primary designer of the Ambivalent Ageism Scale (AAS), requesting permission to validate the instrument. Permission was granted, allowing the work to commence. The original version of the instrument was then provided to two translators who were fluent in English and knowledgeable in Persian. These translators worked separately, and their translations were reviewed in a meeting with the researchers. Through consensus, an initial joint translation was produced. In the back-translation stage, the common Persian translation prepared in the previous stage was retranslated into English by two native speakers who were fluent in both Persian and English but unaware of the original version. This resulted in an English version that was approved by the researchers during a meeting. The final version was then sent to the primary designer of the instrument for conceptual review and approval. The stages of cultural adaptation and other psychometric properties were subsequently carried out as follows.

Face validity

Face validity was assessed using both qualitative and quantitative methods. For qualitative face validity, 15 eligible adults were asked to express their understanding, difficulty, relevance, and any ambiguity they perceived in the items. For quantitative face validity, the Item Impact Method was employed. The target group of 15 eligible adults evaluated each item of the translated version of the questionnaire on a five-point Likert scale, ranging from “not important at all” (score 1) to “very important” (score 5). The Impact Score was calculated using the formula: Impact Score = frequency × importance, where frequency is the percentage of adults who rated the item as 4 or 5, and importance is the average score of importance (the average score obtained from adults’ responses to that item). Items with an Impact Score of 1.5 or higher were considered appropriate and retained for further analysis [40, 42].

Content validity

In this study, content validity was evaluated using both qualitative and quantitative methods. In the qualitative method, 15 experts, including 2 psychologists, 3 geriatric nurses, 4 faculty members from the Geriatrics Department, 2 gerontologists, 2 methodologists, and 2 individuals working in care centers for older adults, reviewed the questionnaire and provided corrective feedback on grammar, wording, item allocation, simplicity, clarity, and scaling. For quantitative content validity, the Content Validity Ratio (CVR) and Content Validity Index (CVI) were calculated. To calculate the CVR, experts rated the necessity of each item on a three-point Likert scale ranging from “not necessary” (score 1) to “necessary” (score 3). The average CVR scores for all items were then calculated using the formula: CVR = (nE - N/2) / (N/2), where nE is the number of experts who rated the item as necessary, and N is the total number of evaluating experts. The numerical value of the CVR was determined according to Lawshe’s table. The minimum acceptable value for CVR with 15 experts is 0.49, with a statistical significance level of p < 0.05.

To calculate the CVI, experts rated the relevance of each item on a four-point scale ranging from “not relevant” (score 1) to “completely relevant” (score 4). The Item-Level Content Validity Index (I-CVI) and modified kappa statistic (K) * were then calculated. For I-CVI, a score higher than 0.79 is considered suitable, and for kappa, a value higher than 0.74 is considered excellent, between 0.6 and 0.74 is good, and less than 0.6 is considered poor [40, 42, 43].

Construct validity

This study was conducted with 381 adults (aged 18–59 years) covered by urban health centers in Gorgan city. For the implementation of exploratory factor analysis (EFA) and confirmatory factor analysis (CFA), 191 samples were randomly selected for EFA, and the remaining 190 samples were allocated for CFA. Initially, 17 households were randomly selected from each of the 26 urban health centers in Gorgan city. Subsequently, using a table of random numbers, one person from each family was chosen to complete the questionnaire. The questionnaire link was sent to the selected individuals, and they were included in the study based on the entry criteria after receiving explanations and consenting to participate in the research.

To perform the exploratory factor analysis (EFA), sampling adequacy was assessed using the Kaiser-Meyer-Olkin (KMO) test and Bartlett’s test. For the KMO test, values of 0.7–0.8 were considered good, and values of 0.8–0.9 were considered excellent. Bartlett’s test required a p-value less than 0.05 [44]. Factor extraction was then conducted using principal component analysis (PCA) with Promax rotation and a scree plot. The presence of an item in a factor was determined based on the formula CV = 5.152 ÷ √(n-2), which approximates to 0.37 in this study (where CV is the amount of factors that can be extracted and n is the sample size) [45]. Stevens (2012) suggests that a valid latent variable is a factor that has at least 10 items with a loading of 0.4, despite having 150 research units [46]. According to the rule of three indicators, each latent variable [47] must have at least three observed variables (items). Items with a loading value less than 0.5 were removed from the EFA [48].

Subsequently, 190 individuals were included in the study according to the inclusion criteria for confirmatory factor analysis (CFA). The most common goodness-of-fit indices for the proposed model were obtained using maximum likelihood estimation, based on accepted thresholds. The assumption of normality was assessed with skewness values between − 3 and + 3 and kurtosis values between − 7 and + 7 [49].

Hooper et al. (2008) state that there is no golden rule for evaluating model fit, and it is advisable to report several indicators [50]. To determine the model’s goodness of fit, the following fit indices were assessed: root mean square error of approximation (RMSEA) < 0.08, comparative fit index (CFI) > 0.9, normed fit index (NFI) > 0.9, adjusted goodness of fit index (AGFI) > 0.8, and the ratio of chi-square to degrees of freedom (CMIN/DF) 3 > [51, 52].

Reliability

To assess the internal consistency of the AAS, Cronbach’s alpha coefficient was first estimated for the entire questionnaire and then for each extracted factor. Internal consistency addresses the fundamental question of how well each item examines the desired content or structure, meaning whether the items in a scale are conceptually compatible. It is expected that the correlation between the answers for each test will be highly correlated with the total test [40]. In the fields of psychology and social sciences, a Cronbach’s alpha value greater than 0.6 is considered acceptable [42].

Data analysis

Statistical analyses were performed using SPSS and AMOS software, version 24. Descriptive statistics for quantitative variables were reported as mean (standard deviation [SD]), and for qualitative variables, as frequency (percentage). Construct validity of the scale was assessed through exploratory factor analysis (EFA), reporting the Kaiser-Meyer-Olkin (KMO) measure of sampling adequacy and Bartlett’s test of sphericity. Reliability of the scale was evaluated by calculating Cronbach’s alpha coefficient for internal consistency. AMOS software was used for confirmatory factor analysis (CFA).

Ethical considerations

The present study has been approved by the ethics committee of Golestan University of Medical Sciences under the ethics code IR.GOUMS.REC.1402.149. After providing the necessary explanations to eligible participants, informed consent was obtained. After providing the necessary explanations to eligible participants, informed consent was obtained. Participants were assured that their information would remain confidential.

Results

Demographic information is given in Table 1.

The average age of the participants in this study (39.70 ± 11.330), the minimum age was 18 and the maximum age was 59 years. 61.2% of people (233 people) were female, 42.8% had bachelor’s education (63 people), 34.1% (130 people) were employees, 67.5% (257 people) were married.

Table 1 Characteristics of the participants

Face validity

During the qualitative face validity process with 15 participants, items 1, 4, 5, 10, and 12 underwent editorial and textual changes based on the opinions of the target group. In the quantitative phase, all items had an impact score greater than 1.5 and were retained.

Content validity

Based on the opinions of the 15-member panel of experts for qualitative content validity, items 1, 2, 3, 4, 9, 10, 12, and 13 underwent editorial and textual changes. For quantitative content validity, the CVR score for all items was higher than 0.49 according to Lawshe’s table, the CVI score for all items was higher than 0.79, and the Kappa score for all items was higher than 0.74, indicating acceptability. No items were deleted.

Construct validity

In construct validity, hidden factors were extracted based on exploratory factor analysis (EFA). The Kaiser-Meyer-Olkin (KMO) index was 0.806, indicating the adequacy of the sample size of 191 participants. Additionally, the Bartlett’s test value was less than 0.001, with an approximate Chi-Square value of 520.676 and 78 degrees of freedom. The rotation method used was Promax, suggesting that factor analysis is suitable for identifying the structure of the factor model. Using the maximum likelihood method, Promax rotation, and a scree plot in EFA, three factors were extracted based on eigenvalues greater than one, accounting for 53.989% of the total variance (Table 2). Items with loadings below the threshold of 0.5 were removed, and no items were omitted in EFA. The scree plot confirms these three factors (Fig. 1). The first factor, labeled “cognitive assistance/protection,” includes items 1, 2, 3, and 13. The second factor, labeled “unwanted help,” includes items 4, 6, 7, 8, and 9. The third factor, labeled “hostile,” includes items 5, 10, 11, and 12. In this study, similar to the original questionnaire [25], two factors—cognitive assistance/protection and unwanted help—were included in the subscale named “benevolent.” However, in the original questionnaire, item 13 (“Older adults are a burden on the health and economic system”) was in the hostile dimension, but in this study, it appeared in the benevolent dimension. Item 5 (“People should avoid giving sad news to older adults because they cry easily”) appeared in both the second and third factors, but due to its higher factor loading, it was placed in the hostile factor.

Table 2 EFA results of AAS

Fig. 1

Scree plot

In the confirmatory factor analysis (CFA), The results indicate that the minimum and maximum skewness among the items are − 1.28 and 0.18, respectively, while the lowest and highest kurtosis values are − 1.32 and 1.7, respectively. Chi-square results and goodness-of-fit indices were obtained as follows: χ² = 42.282, p = 0.016; RMSEA = 0.062; NFI = 0.886; CFI = 0.947; IFI = 0.949; AGFI = 0.918; CMIN/DF = 1.720. In the CFA, four items (12, 13, 4, and 6) were eliminated, and the Persian version of the Ambivalent Ageism Scale was confirmed with 9 items. The results indicated that the CFA based on the two-factor model extracted from the EFA with the obtained data has a good fit (Fig. 2).

Fig. 2

Confirmatory factor analysis of the AAS

In this study, the reported values for Cronbach’s alpha were as follows: 0.763 for the entire instrument, 0.705 for the benevolent dimension consisting of 6 items, and 0.603 for the hostile dimension consisting of 3 items. These values indicate appropriate and acceptable internal consistency for both the individual dimensions and the overall instrument.

Discussion

This study aimed to investigate the psychometric characteristics of the AAS. Following the translation process, various psychometric properties, including face validity, content validity, construct validity, and reliability, were examined. The results obtained support the sufficient validity and reliability of the Persian version of the AAS. The reliability of the instrument was assessed using the internal consistency test, yielding a Cronbach’s alpha coefficient of 0.763, which indicates appropriate reliability for the entire instrument. The Persian version of the scale consists of 9 items distributed across two dimensions: benevolent, with 6 items, and hostile, with 3 items. The translation process was meticulously conducted to produce a final Persian version of the AAS. Subsequently, the validity and reliability of the scale were evaluated within the Iranian context. Understanding the psychometric characteristics of an instrument is crucial for researchers selecting appropriate tools in the field of health, and ensuring validity is a vital component of an instrument’s psychometric evaluation [53].

In this study, following the translation process, face validity was assessed using both quantitative and qualitative methods involving 15 individuals from the target group who were qualified for the study. The qualitative assessment of face validity led to the editing and correction of several items, while the quantitative assessment, evaluated using an impact score higher than 1.5, was deemed appropriate [54, 55]. In contrast, the qualitative face validity of the AAS instrument in the Japanese version was examined with only 5 Japanese speakers [36].

The content validity of the scale was evaluated by 15 experts. During the qualitative review, several phrases were edited. The Content Validity Ratio (CVR) of the scale was found to be suitable according to Lawshe’s table, and no items were deleted. Additionally, the Content Validity Index (CVI), Item-Level CVI (I-CVI), and Kappa (K*) were all deemed appropriate, with no items requiring deletion. In comparison, the Turkish version of the narrative instrument did not undergo content validity assessment, and there is limited information available regarding the psychometric properties of other scales [6].

In the present study, exploratory factor analysis identified three factors: cognitive help and protection, unwanted help, and hostile help, with eigenvalues above one, collectively explaining 54% of the total variance of the instrument, which is considered acceptable. According to Hair et al. (2010), the acceptable cumulative variance should fall within the range of 50–60% [56]. In the original version [25], the details of the EFA were not reported, and neither the Turkish [6] nor the Japanese [36] versions provided information about the EFA, making direct comparisons difficult. However, similar to the findings of Carey et al. [25] and Ozturk et al. [6], the three-factor structure of the AAS was also observed in the Persian version of the scale. Consistent with these studies [6, 25], the two factors of cognitive assistance/protection and unwanted help, which both exhibit a benevolent approach towards older adults in terms of content, were combined into a single benevolent factor.

Ageism is a discriminatory process in the treatment of older adults, which can manifest as either positive or negative discrimination, either implicitly or explicitly, across cognitive, emotional, and behavioral dimensions [57]. In the present questionnaire, ageism was identified through two distinct factors: hostile and benevolent. The hostile factor pertains to the negative dimension of discrimination, encompassing prejudices, stereotypes, and behaviors that aim to discredit individuals based on their age [58]. Several studies have highlighted negative stereotypes towards older adults, which can lead to negative feelings, prejudice, and discrimination against them [59, 60]. Conversely, the benevolent factor represents the positive dimension of discrimination, involving prejudices, stereotypes, and behaviors that aim to show benevolence based on age. Some studies have noted that such positive stereotypes can also be present in the treatment of older adults [9, 58].

In this study, the fifth item (“People should avoid giving sad news to older adults because they cry easily”) was placed in the hostility subscale due to its higher factor loading. While this item was categorized under the benevolent dimension in the original questionnaire [25] and the Turkish version [6], it is inconsistent with the results of the present study. To justify this discrepancy, it can be stated that this item is conceptually consistent with other items in the hostility dimension, including item 11 (“Older adults are easily offended”), which also refers to negative characteristics such as the fragility and emotionality of older adults.

In the present study, item 13 (the elderly are a burden on the health system and the economy) was placed in the benevolent subscale in the exploratory factor analysis with a factor loading of 0.777. However, this item was categorized in the hostile dimension in the original instrument [25] and the Turkish version [6]. The difference in the placement of this item could be related to several factors. One of these factors is the cultural and religious background. In Iranian culture, the elderly are seen as sources of knowledge and experience, and this attitude may lead to a more positive interpretation of this item. Also, Iran is a Muslim country where Islam recommends respect for the elderly, and most people in society consider respect for the elderly obligatory and do not consider them a burden on the family and the health system. The second factor is the way individuals perceive it, as respondents may look at this item from different angles. For some, this phrase may be interpreted as attention and support for the elderly, not as a negative burden. The third factor is the personal experiences of individuals, which can influence how they respond to this item. Individuals with different experiences of interacting with the elderly may have different perceptions of the item. Someone who has lived with elderly people who need care may feel that they are a burden, while others may see the elderly as a source of emotional or social support.

The similarities and differences found in the exploratory factor analysis are noteworthy, but due to the limited number of psychometric studies on this scale, there is a lack of comparative data across different cultures and conditions. The only available study is the Turkish version of the scale, which, although Turkey shares some cultural similarities with Iran, still exhibits different psychometric results due to the general suitability of most ageism scales to Western culture [61]. To verify the model obtained from EFA, confirmatory factor analysis was conducted on a separate sample of 190 participants. The results confirmed the goodness of fit of the Ambivalent Ageism Scale model with 9 items.

The findings from the confirmatory factor analysis indicated that items 4, 6, 12, and 13 were removed due to factor loadings of less than 0.5. In justifying the removal of these items, it can be stated that items 4 and 7 refer to hearing problems in older adults. Additionally, item 6 has content similarity with item 5. Therefore, considering the similarity and conceptual fit of these two items, the removal of items 4 and 6 does not create a gap in the overall framework of the questionnaire.

In justifying the elimination of item 12, it can be stated that, firstly, during the formal and content validity phases, most participants and experts did not agree to retain it, and it was deemed inappropriate for Iranian culture.

In this study, the reported value for Cronbach’s alpha for the Persian version was 0.763, indicating suitable and acceptable internal consistency for the entire instrument. In comparison, the Cronbach’s alpha coefficient for the original version was 0.91 [25], and for the Turkish version, it was 0.89 [6]. For the subscales in the Persian version, Cronbach’s alpha was 0.705 for the benevolent dimension and 0.603 for the hostile dimension. In the Turkish version [6], these values were 0.89 and 0.79, respectively; in the original version [25], they were 0.89 and 0.84; and in the Japanese version [36], they were 0.83 and 0.77. These findings are consistent with the current study.

Conclusion

The Persian version of The Ambivalent Ageism Scale has demonstrated sufficient validity and reliability for measuring attitudes towards aging within Iranian society. Given its cultural adaptation, this questionnaire can be effectively utilized to assess adults’ views and attitudes towards older adults across two dimensions: hostile and benevolent. Additionally, it can aid in the formulation of family-oriented policies for older adult care and facilitate improvements in the quality of care provided to this population group.

Limitation

This questionnaire focuses on the views of adults aged 18–59 towards older adults and is not generalizable to individuals outside this age range. Another limitation of the research is the lack of concurrent validity of this instrument because there was no exactly similar instrument available for this stage, considering Iranian culture.