Table 4.
Summary of Results (Relationship between Ageism and Psychological Well-being).
| Author (year) | Sample | Methods | Measures | Key Findings | |
|---|---|---|---|---|---|
| Independent (Ageism) | Dependent | ||||
| Avidor et al. (2017) | Older adults aged 65–93 years drawn from the German ageing survey (n=615) | Hierarchical multiple regression analysis | A dichotomous variable: Participants were asked whether they had been discriminated against or placed at a disadvantage due to their age, in the past 12 months | Subjective well-being (1) the five-item satisfaction with life scale on a 5-point scale (Cronbach’s α for T1: .98 and T2: .84) 1 ; (2) the 10-item positive affect scale drawn from the positive and negative affect schedule on a 5-point scale (Cronbach’s α for T1: .99 and T2: .86) 1 | The direct path between age discrimination and the well-being of older adults (β = −.32, p < .01) was negative while except for demographic, economic, and physical health status variables |
| Bai et al. (2016) | Chinese adults aged 60 years and older in Jiangsu province (n=954) | Hierarchical multiple regression analysis | Participants’ burden views toward older people: Two items on a 3-point scale Cronbach’s α: 0.84 (1) the extent to which they perceived older people as a burden to family (2) the extent to which they perceived older people as a burden to society | 15-Item short form of the Chinese version of the GDS 2 short form. (Cronbach’s α: 0.80) | Those who held burden view had more depressive symptoms (β = .23, p < .001) controlling for sociodemographic variables, physical and functional health status and quality of family relations |
| Fernandez-Ballesteros et al. (2017) | Older adults aged 62 years and older from Germany, Mexico and Spain in 2005 (n=2005) | Structural equation modeling | Three questions with a 4-point Likert-type scale used to measure negatively perceived age discrimination | Active aging (a latent variable) was defined to include (1) life satisfaction: Five items with a 4-point Likert-type scale; (2) subjective health: Three items with a 4-point Likert-type scale; and (3) self-perception of aging: Five items from Levy, Slade, and Kasl, (2002) | A negative direct effect of perceived ageism on active aging that was found β = −.384, p < .001) |
| Garstka et al. (2004) | Older adults aged 64–91 years recruited from the community (n=60) | Structural equation modeling | Age discrimination: Four items by using a 7-point Likert scale Cronbach’s α: 0.77 (1) a victim of society; (2) deprived of the opportunities that are available to others; (3) victimized by society and (4) discriminated against more than members of other age groups | Psychological well-being (two dimensions) (1) self-esteem: Positive self-regard by using a 10-item personal self-esteem scale (Cronbach’s α: 0.77)<(2) life satisfaction using a 5-item scale derived from Schmitt et al. (2002) and two new items (Cronbach’s α: 0.57) | Perceived age discrimination had direct negative effects on the well-being of older adults (β = −.54, p < .05) |
| Kim (2015) | Older adults aged 65 and overdrawn from the health and retirement study (n=3991) | Hierarchical multiple regression analysis | A dichotomous variable based on two concepts: (1) perceived everyday discrimination, and (2) attributions of daily discrimination. Cronbach’s α: 0.82 | Eight questions from the CES-D³ scale (Cronbach’s α: 0.81) | Perceived ageism had a statistically significant effect on depression (β = .029, p < .05) controlling for demographic, socioeconomic, and physical health status variables |
| Kim et al. (2015) | Older adults aged 60–89 years in South Korea from the 2013 ageism and health study (n=812) | Hierarchical multiple regression analysis | The Palmore ageism scale (2001) Cronbach’s α: 0.83 | The CES-D³ scale (Cronbach’s α: 0.92)< | An increase in experiences of ageism was significantly associated with increased depressive symptoms (β = 0.12, p < .001) while controlling for age, sex, education, marital status, economic status, health condition |
| Lee and Kim (2016) | Older women aged 65 or above, who lived in rural areas and attended senior community centers (n=207) | Structural equation modeling | Used Palmore’s (2001) ageism survey to assess experienced ageism. It consists of 20 items with a 4-points Likert scale Cronbach’s α: 0.93 | (1)Depression: The Korean translation of GDS 2 -SF—15 items with a dichotomous scale (yes or no) (Cronbach’s α: 0.84) (2) stress—stress recognition scale (Lee & Lee, 2002)—21 items with a 4-points scale (Cronbach’s α: 0.85) | Ageism had a direct effect on stress (β = .49, p < .01) and had an indirect effect on depression through stress (β = .34, p < .01) |
| Lyons et al. (2018) | Australians aged 60 and over were recruited through various recruitment strategies both online and offline (n=2119) | Hierarchical multiple regression analysis | Used Palmore’s (2001) ageism survey to assess experienced ageism. It consists of 20 items with 4 points Likert scales Cronbach’s α: 0.83 | Depression, anxiety, and stress scale (DASS-21) (Antony, Bieling et al. 1998)—21 items with a 4-point Likert scale | Ageism experience had statistically significant effects on depression (β = .022, p < .001), anxiety (β = .023, p < .001), and stress (β = .026, p < .001) while controlling for demographic, socioeconomic, and physical health status variables |
| Sabik (2013) | European American and African American women in 60s (University of Michigan alumnae in 2008) (n=244) | Structural equation modeling | The five questions used to assess perceptions of age discrimination (adapted from measures of racial and gender discrimination) Cronbach’s α: 0.75 (1) deprived of the opportunities (2) my age group have been deprived of the opportunities (3) older people are excluded from many sectors of public life (4) after ending one’s working life, one is considered to be worthless (5) the achievements of older people are not appreciated in our society | The five-item mental health subscale from the MOS 4 36-item short-form health survey Cronbach’s α: 0.82 | Perceived age discrimination was directly associated with lower psychological well-being. (Direct effect: β = −.29, p <.001) |
| Shin et al. (2018) | Patients with cancer who were 60 years or older recruited from the National Cancer Center and 10 other regional cancer centers in Korea (n=439) | Multivariate linear regression | Yes/no dichotomous variables measuring seven discriminations based on their age in the course of their cancer care: disease information, treatment information, the daily mile (TDM) participation, attention from healthcare providers (HCPs), supportive care, and treatment Cronbach’s α: 0.82 | Mental health and quality of life were assessed with the Geriatric depression scale (GDS) 15 items with binary response options (yes = 1, No = 0) | Ageism experience was associated with a higher depression score (all p < 0.001) while controlling for age, gender, education, and income |
| Siguaw et al. (2017) | Retired seniors obtained from mostly an online survey (n=543) | Structural equation modeling | Garstka et al.’s (2004) 4-item perceived age discrimination scale | Diener, Emmons, Larsen, and Griffin’s (1985) 5-item SWL 5 scale | Ageism was significant in predicting SWL 5 and its direction was negative (β = −0.10, p < .05) |
| Zhang et al. (2018) | Recruited participants aged 60 or over from 17 neighborhoods in Beijing, China (n=279) | Hierarchical multiple regression analysis | 18-Item Image of ageing scale (Levy et al., 2004), with nine positive words and nine negative words-7-point Likert scale Cronbach’s α: 0.78 (PAS 6 ) Cronbach’s α: 0.87 (NAS 7 ) | Well-being was measured using 20-item life satisfaction Index-A (LSI-A) (Neugarten, Havighurst, and Tobin, 1961)—5-point Likert scale Cronbach’s α: 0.82 | Well-being was negatively influenced by NAS 7 (β = −0.14, p < .01) and positively influenced by PAS 6 (β = .21, p < .001) while controlling for age, gender, education, income, and marital status |
| Zhang et al. (2019) | Recruited participants aged 60 or over from 21 communities Chongqing, China (n=331) | Hierarchical multiple regression analysis | 18-Item Image of ageing scale (Levy et al., 2004), with nine positive words and nine negative words-7-point Likert scale Cronbach’s α: 0.85 (PAS 6 ) Cronbach’s α: 0.87 (NAS 7 ) | Older adults’ well-being was measured (1) morale (15-item Philadelphia Geriatric Center morale scale—Cronbach’s α: 0.89), (2) depression (10-item Center for Epidemiologic Studies depression scale—Cronbach’s α: 0.86), and (3) loneliness (the 8-item form of the University of California, Los Angeles loneliness scale—Cronbach’s α: 0.85) | PAS 6 had positive effect on well-being (β = .20, p < .001) and NAS 7 had negative effect on well-being (β = −.29, p < .001) while controlling for age, gender, education, income, marital status, and physical health |
1T1: Time 1, 2008; T2: Time 2, 2011
2GDS: Geriatric Depression Scale.
3CES-D: Center for Epidemiologic Studies Depression.
4MOS: Medical Outcomes Study.
5SWL: The Satisfaction with Life Scale.
6PAS: Positive Age Stereotypes.
7NAS: Negative Age Stereotypes.