Demographic and Socioeconomic Determinants of Self-Efficacy: An Empirical Study of Korean Older Adults

Abstract

This study conceptualizes the dimensions of older adults' self-efficacy and identifies its determinants by using panel data collected by the National Pension Research Institute in Korea. The analysis supported the conceptualization of older people's self-efficacy as a multidimensional construct and found strong support for the three subscales. For older adults in Korea, general self-efficacy was condensed into a sense of competence. Older adults' social self-efficacy was delineated into two subscales: inward social self-efficacy and interactive social self-efficacy. Inward social self-efficacy could be characterized as a set of inward-looking social capabilities, whereas interactive self-efficacy could be considered as a set of relation-oriented capabilities with various groups of individuals. Of the three regression models representing dimensions of older people's self-efficacy, each model uncovered its own set of variables that played a decisive role in perceived self-efficacy. Demographic variables were more effective determinants of older adults' self-efficacy than socioeconomic factors.

Keywords

older adults self-efficacy competence construct of self-efficacy determinants of self-efficacy

Introduction

Demographic aging across the world continues to present major challenges for practitioners and academics to identify robust roles for older people in their families and communities. Older adults have long been considered as wards, rather caregivers or guardians, even though they can frequently continue to actively participate in familial, civic, socioeconomic, and political activities. These same challenges exist for Korean older adults. In fact, their position as a protector in families and communities is quite tenuous (Lee, Ko, & Lee, 2006). Globally, most older adults desire active aging (Lim & Thompson, 2016), which is the process of optimizing opportunities for health, work, social participation, and security to enhance their quality of life (Plouffe & Kalache, 2010).

Various studies (Holahan & Holahan, 1987; Johnson, Stone, Altmaier, & Berdahl, 1998) in the field of gerontology have recently shown that active aging, or sustained activity in old age, is closely associated with self-efficacy or beliefs that they can successfully perform a particular behavior (Rowe & Kahn, 1997). Self-efficacy has numerous benefits for active aging, including giving people the ability to make positive changes in their living standards and lifestyles (Callander & Schofield, 2017). In addition, older people who believe that they can perform well on a task do better than those who think they will perform poorly. Older adults with stronger self-efficacy beliefs also contribute more to their families and communities than those with weaker ones. For this reason, self-efficacy has become a critical variable in human behavior (Bandura, 1997, 2006). However, most works on self-efficacy have focused on how self-efficacy influences subsequent thought and motivational mechanisms, which, in turn, influence performance.

Self-efficacy perceptions may be influenced by differences in demographics, personality, and motivation (Albert et al., 1995; Gatti, Brivio, & Galimberti, 2017; Gist & Mitchell, 1992). Self-efficacy can also be affected by external variables, such as marital status, family background, and socioeconomic status (Gist & Mitchell, 1992; Plouffe & Kalache, 2010). Theoretical analyses of the determinants of self-efficacy have been performed in the extant literature (Bandura, 1982; Gist & Mitchell, 1992; Zimmerman, 2000). Determinants may vary cross-culturally, because cultures are diverse and dynamic social systems, and may affect the various sources of self-efficacy belief systems (Bandura, 2002; Oettingen, 1995). In Korea, old adults' perceived self-efficacy has become an important research interest in gerontology since the middle 1990s. However, empirical studies from a comprehensive perspective have rarely been conducted on the determinants of older people's self-efficacy. The present study conceptualizes the dimensions of Korean older adults' self-efficacy and identifies its determinants by using panel data collected and managed by the National Pension Research Institute (NPRI) in Korea. Appropriate change strategies for the enhancement of older adults' perceived self-efficacy are also identified and discussed.

Theories of Self-Efficacy

Construct of Self-Efficacy

Self-efficacy, introduced as a key component in social cognitive theory by Bandura (1977), constitutes one's beliefs in perceived capabilities to exercise control over challenging demands and one's own functioning and to organize and execute courses of action to attain designated goals. Bandura (1997) demonstrated that self-efficacious individuals participate more readily, work harder, persist longer, and have fewer adverse emotional reactions when they encounter difficulties than do those with less self-efficacy. In terms of activity choice, self-efficacious individuals more frequently undertake difficult and challenging tasks than do inefficacious ones (Zimmerman, 2000). Therefore, the stronger a person's perceived self-efficacy, the higher the goal challenges a person will set and the firmer will be his or her commitment to them (Bandura, 1993, 2006).

Self-efficacy beliefs do not constitute a single disposition, but rather are multidimensional and differ on the basis of the domain of functioning. Self-efficacy is usually understood as being either task-specific or domain-specific because no comprehensive measure of perceived self-efficacy currently exists (Bandura, 2006; Zimmerman, 2000). However, numerous researchers have conceptualized a generalized sense of self-efficacy, and other researchers (Chen, Gully, & Eden, 2001; Luszczynska, Scholz, & Schwarzer, 2005) have empirically demonstrated this concept by using multicultural validation and conceptualized it as a broad and stable sense of personal competence to deal effectively with a variety of problems (Sherer & Maddux, 1982). In these studies, a sense of competence was found to be the core component of general self-efficacy.

In addition, a sense of social self-efficacy, which is conceptualized as a willingness to initiate behavior in a social situation, was frequently extracted as a subscale from the general self-efficacy scale (Sherer & Adams, 1983; Sherer & Maddux, 1982). The items in the subscale were characterized as a set of social skills, including communication skills and interpersonal problem-solving skills (Smith & Betz, 2000). Empirical studies have also revealed that people with high social self-efficacy use more effective ways to solve problems because they possess self-confidence about their ability to initiate social contact, develop new friendships, and work well with others (Bandura, 2006; Elozkan, 2013). Social self-efficacy has also been frequently employed as a key research topic in the fields of behavioral science (Connolly, 1989), psychology (Constantine, Okazaki, & Utsey, 2004; McFarlane, Bellissimo, & Norman, 1995; Wei, Russell, & Zakalik, 2005), education (Elozkan, 2013; Raskauskas, Rubiano, Offen, & Wayland, 2015), aging (Burnett, Cully, Achenbaum, Dyer, & Naik, 2011; Peggi & Jopp, 2015), and gerontology (Tovel & Carmel, 2016).

Determinants of Self-Efficacy

Earlier self-efficacy theorists (Bandura, 1982; Schunk, 1981) focused on the determinants of self-efficacy. For example, Bandura (1982) demonstrated that perceived self-efficacy is gained from performance accomplishments, modeling or observing others, verbal persuasion, and physiological arousal, such as anxiety. In their work, Gist and Mitchell (1992) identified determinants in terms of the locus of self-efficacy and categorized them into external and internal cues. According to the authors, external determinants are primarily under the external properties' control, while internal determinants are primarily under personal control. Properties including task attributes, persuasion, and modeling are affiliated with external determinants, and those including personal abilities, personalities, and arousal are associated with internal determinants. Gist and Mitchell (1992) then reclassified the two types of determinants based on variability. In their work, properties such as task attributes, personal abilities, and personalities comprise the low variability of determinants, whereas attributes such as persuasion, modeling, and arousal comprise the high variability of determinants.

Determinants of self-efficacy, identified by Bandura (1982) and Gist and Mitchell (1992), can be incorporated into a holistic picture, as shown in Figure 1. Two properties such as observing others and verbal persuasion can be classified as a high variability external determinant, whereas two sources including personal abilities and personalities are categorized as a low variability internal determinant. Physiological arousal comprises a high variability internal determinant, whereas performance accomplishments include a low variability internal determinant. All these determinants, from a theoretical perspective, respectively or collectively affect self-efficacy including a sense of competence and a sense of social self-efficacy.

Figure 1.

Theoretical framework of determinants of self-efficacy.

Researchers started to investigate the potency of these sources by examining possible situational and instructional factors within the psychological, educational, and organizational contexts affecting ones' self-efficacy based on its determinants. For example, an empirical study demonstrated positive correlations between therapeutic and self-directed changes in behavior and changes in self-efficacy (Sherer & Maddux, 1982). Psychotherapies and self-directed experiences were then suggested to reinforce a sense of personal efficacy. Through one or several forms of these approaches, people with low self-efficacy could regain confidence and engage more successfully in social activities (Bandura, 2006). Self-efficacy was also found to be influenced by demographic and socioeconomic factors (Callander & Schofield, 2017; Clark, 1996; Perkins, Multhaup, Perkins, & Barton, 2008). These studies contributed to the development of self-efficacy-enhancing interventions. In the same context, self-efficacy can be restored and enhanced by therapeutic treatment, mastery experiences, and public policies aimed at the improvement of socioeconomic status.

Various aging studies on self-efficacy have been performed in the areas of relationships between older adults' occupational self-efficacy and task performance (Fletcher, Hansson, & Bailey, 1992; Peggi & Jopp, 2015), older people's self-efficacy on independent living (Burnett et al., 2011), and enhancement of older adults' computer self-efficacy (Laganá, Olivor, Ainsworth, & Edwards, 2011). However, few empirical studies exist on older adults' self-efficacy determinants. In addition, an in-depth analysis has not yet been conducted to identify the demographic and socioeconomic factors associated with older adults' self-efficacy. This lack of empirical studies and detailed analyses has become a major barrier to the development of self-efficacy-enhancing interventions.

Researchers in Korea began to investigate older adults' self-efficacy in the mid-1990s. Several forms of measurements have been used to explore Korean older adults' self-efficacy. The works of Bandura (2006), Larson et al. (1992), Lent, Hill, and Hoffman (2003), and Sherer and Maddux (1982) were most frequently cited in the development and validation of self-efficacy scales in Korea. A sense of competence was also found to be the core component of general self-efficacy in the Korean literature (Ha & Kim, 2009; Kim, 1997; Kim & Yoo, 2010). Two factors, namely, self-regulatory efficacy and task difficulty preference efficacy, were frequently extracted as a component of self-efficacy (Kim, 1997; Kim & Yoo, 2010; Yang & Kim, 2011). However, social self-efficacy as a dimension of self-efficacy was rarely found in extant research. In addition, not many studies focused on the determinants of older adults' self-efficacy. Nevertheless, family support, peer relationships with friends (Yang & Park, 2010), social activities (Lee, Cho, & Lee, 2010), and work and economic status (Ha & Kim, 2009; Lee, 2012) were reported to be associated with their self-efficacy. However, the determinants of older adults' self-efficacy were barely explored from a comprehensive and multidimensional perspective.

Research Methods

Data Collection

The present study used data collected from the panel survey conducted by Korea's NPRI in 2014 and released to the public near the end of 2015. The survey, entitled the Korean Retirement and Income Study, has been conducted biennially since 2005. The survey was designed to collect information on older people's public pension coverage, economic status, employment, retirement, health status, and family relationships, as well as their perceived self-efficacy. NPRI used the sampling frame from the Korean Population and Housing Census in 2010 in the initial stage and determined the survey households. A group of trained surveyors visited each individual household throughout the nation and conducted a paper and pencil interview to collect data. To improve the sample's representativeness and to reduce selection bias, calibration of weights was performed with benchmarking variables, such as geographical regions, age, and gender. The sample size of the 2014 survey was 6,022.

The survey adopted the successful aging scale for Korean older adults that was developed and validated by Kim and Shin (2005). The scale comprised four parts, including self-efficacy, self-control, satisfaction with children's success, and couple lifelong companion. For the development of the successful aging scale, the authors first developed a theoretical framework based on in-depth reviews on the works of Baltes and Baltes (1990), Fisher (1995), Park and Yi (2002), Rowe and Kahn (1997), Sung and Yoo (2002), Torres (2003), and some other studies. The authors then conducted in-depth interviews with a group of Korean gerontologists. On the basis of the results of the literature reviews and interviews, the authors developed a preliminary successful aging scale. In their study, Kim and Shin (2005) suggested a set of 13 items for the perceived self-efficacy scale for older adults in Korea. However, a confirmatory factor analysis with empirical data has not been conducted on the scale.

Analytic Models and Dimensions of Self-Efficacy as Dependent Variables

For this study, a set of multiple regression models was employed to explore determinants of older people's self-efficacy. The dependent variables were three respective dimensions of self-efficacy explored in a confirmatory factor analysis. With regard to the survey on self-efficacy, older people were asked to answer 13 questions related to their competence and social self-efficacy. Out of the 13 questions, 1 question, namely, “I can live in abundance without my children's financial support” was not associated with the remaining 12 items and was deleted from the factor analysis. A set of 12 items, listed in Table 1, was adopted to explore older adults' perceived self-efficacy.

Table 1.

Dimensions of Self-Efficacy as a Result of Factor Analysis.

Items	Competence	Social self-efficacy
Items	Competence	Inward social self-efficacy	Interactive social self-efficacy
I am confident that:	α = .811	α = .802	α = .793
I have no problem to move my body.	.676	.026	.297
I set important goals for the rest of my life.	.557	.355	.351
I have tasks to achieve every day.	.845	.186	.150
I am recognized as a person who can do what other people need me to do.	.672	.240	.391
I learn something new for my myself from others.	.244	.792	.095
I put myself in a situation to get my job done before I die.	.313	.694	.231
I volunteer to help people around me.	.099	.753	.246
I continuously put myself in social situations.	.300	.549	.355
I have a good relationship with someone I do not know very well.	.370	.038	.806
I have someone with whom I can talk freely.	.338	.352	.595
I get along well with my friends and relatives.	.161	.368	.596
I do some good things, philanthropically, for people.	.263	.393	.638

Note. Numbers in bold represent factor loadings.

As Bandura (2006) suggested, the items were phrased in terms of can do rather than will do. In this context, a 5-point Likert scale—1 for no confidence at all, 2 for little confidence, 3 for moderate confidence, 4 for much confidence, and 5 for complete confidence—was adopted to determine the level of perceived self-efficacy.

The 12 items were analyzed by means of exploratory factor analysis with maximum likelihood extraction, varimax rotation, and eigenvalues greater than 1. The analysis extracted three factors. All expected factor loadings were greater than .5, and none of the remaining factor loadings were greater than .4. The next step was to conduct a confirmatory factor analysis testing three different models. All three models were based on the resulting 12 items in the survey. According to the fit indexes in the model testing (χ² = 2428.98, df = 51, Tucker–Lewis index = .89, comparative fit index = .93, incremental fit index = .93, root mean square error of approximation = .088), the minimum was achieved. None of the correlations between error terms were set free. The correlations among the three factors ranged between .50 and .60, which showed that the model had acceptable fit to the data. The model testing indicated that older people's perceived self-efficacy is a multidimensional construct.

The factor scores of the three categories were used for the measurements of the dependent variables. As seen in Table 1, Factor 1 was composed of four questions, including a sense of competence in physical function, work, hope, and recognition achievement, and was therefore characterized as the competence factor. The reliability coefficient (alpha) of Factor 1 was .811, showing strong intercorrelations among the four items.

The remaining eight items were expected to be condensed into one factor, representing social self-efficacy. However, the eight questions were divided into Factor 2 and Factor 3. Factor 2 comprised four items and represented a set of social self-confidence gained from self-enhancement activities. This factor could be characterized as a set of inward-looking social capabilities, or the domain of social confidence that older adults could approach through the activities of learning from others, putting themselves in social situations, and volunteering to help people around them. Careful observation indicated that this subscale was closely related to psychological preparation and adjustment for progress toward interactive social self-efficacy or Factor 3. Therefore, it could be characterized as inward social self-efficacy, which is more likely to be self-oriented than other-oriented. The reliability coefficient of Factor 2 was .802, demonstrating strong intercorrelations among the four variables.

Factor 3 was composed of the remaining four questions and could be characterized as interactive social self-efficacy. The reliability coefficient of Factor 3 was .793, showing strong intercorrelations among the four variables. All four items in this factor constituted interactive relationship-oriented activities, representing self-confidence in having a good relationship with strangers, talking freely to anyone, and getting along well with friends and relatives. Therefore, a set of 12 questions for self-efficacy was condensed into three factors that individually represent the dimensions of self-efficacy: competence, inward social self-efficacy, and interactive social self-efficacy.

Independent Variables

Six independent variables were selected from the panel survey. As seen in Table 2, three variables, including age, gender, and spouse, were chosen to explore the relationship of older people's self-efficacy with demographic factors. The other three variables, including education, employment, and pension, were selected to examine the relationship between older adults' socioeconomic factors and their self-efficacy. Two variables, including age and gender, could be characterized as potential determinants of self-efficacy primarily under personal control and so were classified as low variability internal variables. The existence of a spouse could be considered as a variable of interpersonal environment, with the time horizon being changeable only after a long period of time. Therefore, this source was categorized as a low variability external variable. Education was labeled as a low variability external information cue that can influence self-efficacy because it could be characterized as a setting for gaining knowledge and skills. The employment variable could be classified as a high variability external variable because employment status can be changed relatively quickly for older adults after retirement in Korea and because it is not primarily under personal control but determined by socioeconomic environment. The pension variable was categorized as a low variability external source because the types of pension are usually given on the basis of their lifetime experiences of employment, and it is primarily under the organization's control.

Table 2.

Independent Variables Categorized Into Demographic and Socioeconomic Factors.

Variables	Compositions	Variability	Locus	Categories
Age	Average = 67.8; Minimum = 50; Maximum = 107	Low	Internal	Demographic factors
Gender	Male = 2,431 (40.4%); Female = 3,591 (59.6%)	Low	Internal
Spouse	Yes = 4,827 (82.4%); No = 1,030 (17.6%); Missing = 165	Low	External
Education	No schooling = 989 (16.4%); Elementary school = 2,196 (36.5%); Middle school: 1,137 (18.9%); High school = 1,235 (20.5%); Community college = 97 (1.6%); College = 312 (5.2%); Graduate school = 48 (0.8%); Missing = 8	Low	External	Socioeconomic factors
Employment	Employment = 2,435 (40.5%); Unemployment = 3,580 (59.5%)	High	External
Types of pension	No pension = 5,575 (92.6%); National pension = 423 (7.0%); Special public pension^a = 24 (0.4%)	Low	External

Note. Total cases = 6,022.

Special public pensions indicate government employee pension, teacher pension, and military pension.

Table 3.

Determinants of Self-Efficacy.

Variables	Factor 1 Competence					Factor 2 Inward social self-efficacy					Factor 3 Interaction-directed social self-efficacy
Variables	B	SE	β	t	Sig.	B	SE	β	t	Sig.	B	SE	β	t	Sig.
Constant	1.407	0.155		9.094	.000	0.035	0.170		0.206	.837	0.357	0.169		2.110	.035
Age	−0.027	0.002	−.243	−16.870	.000	−0.008	0.002	−.072	−4.532	.000	−0.010	0.002	−.091	−5.688	.000
Gender	−0.007	0.027	−.003	−0.262	.793	0.090	0.030	.044	3.040	.002	0.030	0.030	.015	1.029	.304
Spouse	0.123	0.033	.047	3.757	.000	0.086	0.036	.033	2.371	.018	0.124	0.036	.048	3.446	.001
Education	0.059	0.010	.079	5.759	.000	0.123	0.011	.165	11.016	.000	0.059	0.011	.080	5.335	.000
Employment	0.484	0.027	.238	18.172	.000	−0.051	0.029	−.025	−1.729	.084	0.067	0.029	.033	2.312	.021
Type of pension	0.109	0.046	.031	2.384	.017	−0.133	0.050	−.037	−2.662	.008	−0.092	0.050	−.062	−1.836	.066
Statistics	R = .450, R² = .203, F = 246.653 (df = 6, 5837), Sig. = .000					R = .196, R² = .038, F = 38.663 (df = 6, 5837), Sig. = .000					R = .162, R² = .026, F = 26.245 (df = 6, 5387), Sig = .000

Note. No multicollinearity problems were found in the regression analyses. According to the multicollinearity diagnostics, all variance inflation factors in the regression models were located between 1.143 and 1.524.

The data set comprised 6,022 cases, including 2,431 males and 3,591 females. For the gender variable, males were designated as 1, and females as 0. The percentage of females (59.6%) was larger than that of males (40.4%), reflecting the fact that females' life expectancy is generally longer than that of males. Their average age was 67.8. Their ages varied, with the youngest being 50 years old, and the oldest being 107 years old. After omitting 165 cases that were missing from the system, there were 5,857 respondents. Of these, 4,827 respondents or 82.4% had spouses, and the remaining 17.6% were spouseless. For the spouse variable, older adults with a spouse were designated as 1, and the others as 0.

Level of education was divided into seven categories, with 1 for no schooling, 2 for elementary school graduate, 3 for middle school graduate, 4 for high school graduate, 5 for community college graduate, 6 for college graduate, and 7 for graduate school graduate. Older people's level of education in the data set was not high, constituting an average of 2.73, or slightly lower than the middle school level.

The employment variable was designated as 1 for older adults engaged in part-time or full-time work, and 0 for those not engaged in any type of job. Approximately 40% of the respondents held a part-time or full-time job. In this study, special attention was paid to employment status because many older Koreans, in spite of retirement, are forced to continue working due to a lack of financial preparation for retirement and lower levels of pension benefits. In Korea, the average effective age of retirement, indicating the age at which individuals fully leave the job market and stop engaging in income-earning activities, is 71 years old (Organisation for Economic Co-operation and Development, 2014).

The pension variable was divided into three categories: 0 for no pension subscribers, 1 for national pension subscribers, and 2 for special public pension subscribers, including government employee pension, teacher pension, and military pension. Special pension subscribers are expected to receive more benefits than national pension subscribers. The size of the no national pension population was very large whereas that of pension subscribers was very small. Of 6,022 cases, 92.6% did not have a pension. Seven percent constituted national pension holders, and 0.4% were members of special public pensions, including government employee pension, teacher pension, and military pension.

Results

Each of the three subscales—a sense of competence, inward social self-efficacy, and interactive social self-efficacy—was regressed on age, gender, spouse, education, employment, and type of pension. Each regression model uncovered its own set of variables that played a decisive role in older people's perceived self-efficacy, as seen in Table 3. The employment variable was a positive predictor of older adults' perceived competence and interactive social self-efficacy, whereas, surprisingly, the variable was a negative predictor of inward social self-efficacy. The pension variable was positively associated with older people's competence, whereas it was negatively associated with their social self-efficacy, including inward and interactive self-efficacy. Two demographic variables, age and spouse, were predictors of all three dimensions of self-efficacy, in which age was a negative predictor and spouse was a positive one. In addition, the influence of demographic factors on older adults' self-efficacy was more likely to be greater than that of socioeconomic factors.

A Sense of Competence and Its Determinants

A sense of perceived competence as a subscale of self-efficacy was significantly associated with five variables, including age, spouse, education, employment, and type of pension. Out of three demographic variables, gender was not found to be a significant predictor of this subscale. In the regression, the results revealed that older adults' perceived competence decreased with age. Moreover, older people with spouses exhibited higher levels of competence than those without a spouse.

All three socioeconomic variables showed a significant relationship with older adults' perceived sense of competence. Expectedly, higher socioeconomic status was related to higher levels of competence in this study. In addition, older adults with a higher level of education exhibited a higher level of competence than those with a lower level of education. Employment status was significantly correlated with older people's competence. Older people who were engaged in part-time or full-time work showed a higher level of competence than those who were not employed. The pension variable was a critical predictor of older people's perceived competence. Older adults with a national pension exhibited a higher level of competence than those with no pension, and older people with one of the special pensions showed a higher level of competence than those with a national pension.

In terms of older adults' perceived competence, the result was expected. Older people in Korea with a more advantageous condition in demographics and higher socioeconomic status were more likely to exhibit a higher level of competence in this regression analysis.

Inward Social Self-Efficacy and Its Determinants

Variables that played a decisive role in older people's inward social self-efficacy differed from those associated with their sense of competence. Specifically, inward social self-efficacy was significantly associated with all three demographic variables, including age, gender, and spouse. Age was also a negative predictor of older people's perceived sense of inward social self-efficacy. Relatively younger older people showed a higher level of inward social self-efficacy than relatively older ones. Although the gender variable was not a predictor of competence and interactive social self-efficacy, it was a critical predictor of inward social self-efficacy. Male older adults exhibited a higher level of inward social self-efficacy than females in this scale, indicating that males were psychologically better prepared for social self-efficacy than females. As seen in the regression model with competence as a dependent variable, older adults with spouses showed a higher level of self-directed social efficacy than those without a spouse.

All three socioeconomic variables were associated with inward social self-efficacy. Education constituted a significant predictor of inward social efficacy. Relatively well-educated older adults exhibited a higher level of self-efficacy than poorly educated ones. In addition, both employment and pension variables were negative predictors of inward social efficacy. Older adults who were engaged in part-time or full-time employment showed lower inward social self-efficacy or psychological preparedness for social self-efficacy. However, older adults who were not employed could participate in the psychologically preparing activities of learning from others, putting themselves in social situations, and volunteering to help people around them.

Expectedly, the pension variable was a predictor of older people's inward social self-efficacy. However, it was not a positive predictor, but a negative one. Older adults with a national pension exhibited a lower level of self-efficacy than those with no pension, whereas older people with one of the special pensions showed a lower level of self-efficacy than those with a national pension. It was unexpected that better pension benefits led older adults in Korea to possess lower inward social self-efficacy.

Interactive Social Self-Efficacy and Its Determinants

A sense of interactive social self-efficacy was significantly associated with five variables. Out of the three demographic factors, gender was not identified as a significant predictor of this subscale. The same as in the case of the competence subscale, relatively younger older adults with spouses exhibited a higher level of interactive social self-efficacy than relatively older ones without a spouse. Older adults' perceived sense of interactive social self-efficacy decreased with age. Moreover, older people with a spouse tended to have a higher level of interactive social self-efficacy than those without a spouse.

Out of the three socioeconomic variables, the education and employment variables showed a positive association with older adults' sense of interactive social self-efficacy, whereas the pension variable was a negative predictor of this subscale. Older adults with a higher education level exhibited a higher level of interactive social self-efficacy than those with a lower education level. In addition, older people who were engaged in part-time or full-time employment showed a higher level of interactive social self-efficacy than those who were not employed.

Although the pension variable was expected to exhibit a positive relationship with older people's self-efficacy, this was not the case in this study. In fact, the pension variable was found to be a negative predictor of older people's interactive social self-efficacy. Older adults with a national pension showed a lower level of interactive social self-efficacy than those with no pension, whereas older people with one of the special pensions exhibited a lower level of interactive social self-efficacy than those with a national pension. Pensions could be a critical financial factor for older people after retirement. Nevertheless, better pension benefits had a negative effect on their interactive social self-efficacy in this analysis. Further research is needed to elucidate this issue.

Discussion

One of the main purposes of the present study was to conceptualize older people's perceived self-efficacy. The analysis supported the conceptualization of older people's self-efficacy as a multidimensional construct and found strong support for the three subscales. For older adults in Korea, general self-efficacy was condensed into a sense of competence, as asserted in various studies (Chen et al., 2001; Luszczynska et al., 2005; Sherer & Maddux, 1982). This subscale could be conceptualized as one's beliefs in personal capabilities to overcome physical challenges, to accomplish tasks and goals, and to be recognized as a person who can do what other people need them to do.

However, older adults' social self-efficacy was delineated into two subscales: inward social self-efficacy and interactive social self-efficacy. This result differed from previous studies. For example, Sherer and Maddux (1982) extracted two subscales from a set of 23 self-efficacy items: a general self-efficacy subscale and a social self-efficacy subscale. In their study, social self-efficacy pertained to efficacy beliefs regarding social interaction and consisted of six items related to making friends and participating in social gatherings. Smith and Betz (2000) found that social self-efficacy was related to both social confidence and enterprising confidence. On the basis of findings in our current study, inward social self-efficacy could be characterized as a set of inward-looking social capabilities, whereas interactive self-efficacy could be considered as a set of relation-oriented capabilities with various groups of individuals. Inward social self-efficacy was more likely to be self-oriented than other-oriented social self-efficacy. Older adults could approach the domain of inward social self-efficacy through the activities of learning from others, putting themselves in social situations and volunteering to help people around them. They could also reach the domain of interactive social self-efficacy through various relationship-oriented activities.

A sense of competence was found to be a universal construct, which means that it characterizes a basic belief that is inherent in almost all Korean older adults. However, whether or not the two subscales of inward social self-efficacy and interactive social self-efficacy found in this study have a cross-cultural commonality in multicultural contexts remains to be determined. The two social self-efficacy subscales could only be applied to Korean older adults at this point. Further studies in cross-cultural settings are required for universal applicability.

Of the three regression models representing dimensions of older people's self-efficacy, each model uncovered its own set of variables that played a decisive role in perceived self-efficacy. Demographic variables were more effective determinants of older adults' self-efficacy than socioeconomic factors. Out of the three demographic variables, both age and spouse were critical predictors of all three self-efficacy dimensions. Older adults' sense of competence and inward and interactive social self-efficacy decreased with age but increased with the existence of a spouse. This result supports the work of Yang and Park (2010), which showed a positive association between self-efficacy and family relationship. Gender was not a predictor of older adults' sense of competence and interactive social self-efficacy, but it exhibited a significant effect on inward social self-efficacy. In addition, male older people tended to possess a higher level of inward social self-efficacy than females.

Out of the three socioeconomic variables, only the education variable was a positive predictor of all three self-efficacy dimensions, including competence, and inward and interactive social self-efficacy. Therefore, it can be concluded that older people's beliefs in their capabilities to accomplish new tasks and carry these capabilities into new social settings was a function of education and could be reinforced with education. Education could be categorized as a low variability external determinant for the formation of self-efficacy.

Employment status contributed to the improvement of older people's competence and interactive social self-efficacy, which agrees with Korean literature (Ha & Kim, 2009; Lee, 2012). However, the employment variable contributed to decreasing inward social self-efficacy. Older people's capabilities in task-oriented and socially interactive activities could be augmented with part-time or full-time employment. However, social self-confidence gained from self-enhancement activities decreased for older adults having part-time or full-time jobs. In Korea, part-time or full-time jobs in older age do not mean economic stability for better life, but rather economic hardship, because jobs for older adults are usually low-income earning for survival (Organisation for Economic Co-operation and Development, 2014). Older adults with jobs may have a lack of psychological preparedness for social self-efficacy, especially inward social self-efficacy.

The pension variable was positively associated with older people's competence, whereas it was negatively correlated with social self-efficacy, including inward and interactive self-efficacy subscales. Older people's competence as a subscale of self-efficacy increased with pensions with better benefits, but inward and interactive social self-efficacy decreased with pensions with better benefits. Moreover, older adults with a national pension exhibited a higher level of competence than those with no pension, and older people with one of the special pensions showed a higher level of competence than those with a national pension. However, older adults with a national pension exhibited a lower level of social self-efficacy than those with no pension, whereas older people with one of the special pensions showed a lower level of self-efficacy than those with a national pension. Older people with pensions with better benefits exhibited a lower level of social self-efficacy. Further research is needed to identify the precise mechanisms underlying these results.

Conclusion

Self-efficacious older adults view their worlds in ways that are more likely to result in successful adjustment because such older adults hold higher expectations for themselves. Specifically, they trust in their capabilities to overcome physical challenges, to accomplish tasks and goals, and to be recognized as a person who can do what other people need them to do. They see the world, and their ability to respond to it, by approaching the domain of social self-efficacy through self-enhancement activities and relation-oriented activities.

Older people's self-efficacy is more likely to be substantially affected by demographic factors, such as age and spouse, than socioeconomic factors. For older people in Korea, the role of a spouse is decisive in their self-efficacy, and effective spousal relationships reinforce their self-efficacy. In addition, older adults' self-efficacy decreases with age, and thus self-efficacy intervention techniques that are effective for younger adults may not be effective for older adults. Older people's public pension coverage status and benefits as a socioeconomic factor was expected to be a critical predictor of perceived self-efficacy. However, a consistent relationship was not found between the type of pension held by older people and perceived self-efficacy. Pensions may constitute a policy instrument for the improvement of financial status, but it does not always help them to approach the domain of social self-efficacy.

From a theoretical perspective, older adults' social self-efficacy can be condensed into inward social self-efficacy and interactive social self-efficacy. Inward social self-efficacy means capabilities of self-oriented psychological preparedness for social activities, whereas interactive social self-efficacy indicates capabilities of other-oriented social activities. This can be considered as one of the major findings of this study. However, at this point, the two social self-efficacy subscales can be applied to Korean older adults only. Further studies are necessary for universal applicability. In addition, the Korean Retirement and Income Study, a longitudinal survey biennially conducted by Korea's NPRI since 2005, includes demographic aspects, economic status, employment, retirement, health status, and family relationships among older adults. This data set may provide a fruitful opportunity to explore issues distinctive to aging in Korea, as well as aging in a cross-cultural context. Researchers are encouraged to use this data set for their studies.

Footnotes

Declaration of Conflicting Interests

The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author received no financial support for the research, authorship, and/or publication of this article.

Author Biography

Mann Hyung Hur is a professor of Welfare Policy in the Department of Public Administration, Chung-Ang University, Seoul, South Korea. His research interests include aging and the aged, pension policy, information society, internet addiction, and empowerment theories. His work has appeared in the Journal of Community Psychology, Cyberpsychology, Behavior, and Social Networking, and International Journal of Social Welfare.

References

Albert

M. S.

Savage

C. R.

Jones

Berkman

Seeman

Blazer

Rowe

J. W.

(1995) Predictors of cognitive change in older persons: MacArthur studies of successful aging. Psychology and Aging 10(4): 578–589.

Baltes

P. B.

Baltes

M. M.

(1990) Psychological perspectives on successful aging: The model of selective optimization with compensation. In: Baltes

P. B.

(ed.) Successful aging: Perspective from the behavioral sciences, Cambridge, England: Cambridge University Press, pp. 1–34.

Bandura

(1977) Self-efficacy: Toward a unifying theory of behavior change. Psychological Review 84: 191–215.

Bandura

(1982) Self-efficacy mechanism. American Psychologist 37(2): 122–147.

Bandura

(1993) Perceived self-efficacy in cognitive development and functioning. Educational Psychologist 28(2): 117–148.

Bandura

(1997) Self-efficacy: The exercise of control, New York, NY: Freeman.

Bandura

(2002) Social cognitive theory in cultural context. Applied Psychology: An International Review 51(2): 269–290.

Bandura

(2006) Guide for constructing self-efficacy scales. In: Pajares

Urdan

(eds) Self-efficacy beliefs of adolescents, Charlotte, NC: Information Age Publishing, pp. 307–337.

Burnett

Cully

J. A.

Achenbaum

W. A.

Dyer

C. B.

Naik

A. D.

(2011) Assessing self-efficacy for safe and independent living: A cross-sectional study on vulnerable older adults. Journal of Applied Gerontology 30(3): 380–402. doi:10.1177/0733464810362898.

10.

Callander

Schofield

(2017) Impact of multidimensional poverty on the self-efficacy of older people: Results from an Australian longitudinal study. Geriatrics and Gerontology International 17(2): 308–331.

11.

Chen

Gully

Eden

(2001) Validation of a new general efficacy scale. Organizational Research Methods 4: 62–83.

12.

Clark

D. O.

(1996) Age, socioeconomic status, and exercise self-efficacy. Gerontologist 36(2): 157–164.

13.

Connolly

(1989) Social self-efficacy in adolescence: Relations with self-concept, social adjustment, and mental health. Canadian Journal of Behavioural Science 21(3): 258–269.

14.

Constantine

M. G.

Okazaki

Utsey

S. O.

(2004) Self-concealment, social self-efficacy, acculturative stress, and depression in African, Asian, and Latin American international college students. American Journal of Orthopsychiatry 74(3): 230–241.

15.

Elozkan

(2013) The effect of communication skills and interpersonal problem solving skills on social self-efficacy. Educational Sciences: Theory and Practice 13(2): 739–745.

16.

Fisher

(1995) Successful aging, life satisfaction, and generativity in later life. International Journal of Aging and Human Development 41(3): 239–250.

17.

Fletcher

W. L.

Hansson

R. O.

Bailey

(1992) Assessing occupational self-efficacy among middle-aged and older adults. Journal of Applied Gerontology 11(4): 489–501.

18.

Gatti

F. M.

Brivio

Galimberti

(2017) “The future is ours too”: A training process to enable the learning perception and increase self-efficacy in the use of tablets in the elderly. Journal of Educational Gerontology 43(4): 209–224.

19.

Gist

M. E.

Mitchell

T. R.

(1992) Self-efficacy: A theoretical analysis of its determinants and malleability. Academy of Management Review 17(2): 183–211.

20.

C.-K.

Kim

(2009) The effect of elderly job projects satisfaction on life satisfaction: Focusing on mediating effect of self-efficacy. Social Welfare Policy 36(4): 393–414.

21.

Holahan

C. K.

Holahan

C. J.

(1987) Self-efficacy, social support, and depression in aging: A longitudinal analysis. Journal of Gerontology 42(1): 65–68.

22.

Johnson

B. D.

Stone

G. L.

Altmaier

E. M.

Berdahl

D. L.

(1998) The relationship of demographic factors, locus of control and self-efficacy to successful nursing home adjustment. The Gerontologist 38(2): 209–216.

23.

Kim

(1997) A study on the academic failure-tolerance and its correlates. Korean Journal of Educational Psychology 11(2): 1–19.

24.

Kim

Yoo

(2010) The influence of self-efficacy, psychological well-being, and stress management on subjective well-being. Journal of Brain Education 6: 19–52.

25.

Kim

M. H.

Shin

K. R.

(2005) The study of the development of the ‘successful aging’ scale for Korean. Journal of the Korea Gerontological Society 25(2): 35–52.

26.

Laganá

Olivor

Ainsworth

Edwards

(2011) Enhancing computer self-efficacy and attitudes in multi-ethnic older adults: A randomised controlled study. Ageing & Society 31(6): 911–933.

27.

Larson

Suzuki

Gillespie

Potenza

Bechtel

Toulouse

(1992) Development and validation of the counseling self-estimates inventory. Journal of Counseling Psychology 39(1): 105–120.

28.

Lee

(2012) A study on the mediating effect of self-efficacy on the relationship between economic status and successful aging among the elderly. Journal of Welfare for the Aged 57: 289–314.

29.

Lee

S.-H.

Cho

S.-K.

Lee

J.-D.

(2010) Effects of daily exercise program on self-efficacy and social adjustment for senior citizens. Journal of Sport and Leisure Studies 42(1): 639–651.

30.

Lee

(2006) Health promotion behavior and quality of life among community-dwelling elderly in Korea: A cross-sectional survey. International Journal of Nursing Studies 43: 293–300.

31.

Lent

R. W.

Hill

C. E.

Hoffman

M. A.

(2003) The development and validation of the counselor activity self-efficacy scales. Journal of Counseling Psychology 50(1): 97–108.

32.

Lim

E. Z.

Thompson

C. L.

(2016) Measuring active aging among older adults in Singapore. Ageing & Society 36(9): 1853–1869.

33.

Luszczynska

Scholz

Schwarzer

(2005) The general self-efficacy scale: Multicultural validation studies. Journal of Psychology 139(5): 439–457.

34.

McFarlane

A. H.

Bellissimo

Norman

G. R.

(1995) The role of family and peers in social self-efficacy: Links to depression in adolescence. American Journal of Orthopsychiatry 65(3): 402–410.

35.

Oettingen

(1995) Cross-cultural perspectives on self-efficacy. In: Bandura

(ed.) Self-efficacy in changing societies, New York, NY: Cambridge University Press, pp. 149–176.

36.

Organisation for Economic Co-operation and Development (2014) Ageing and employment policies – Statistics on average effective age of retirement, Paris, France: Author.

37.

Park

K. R.

Y. S.

(2002) A study on perception of successful aging in later life. Journal of the Korea Gerontological Study 22(3): 53–66.

38.

Peggi

M. E.

Jopp

D. S.

(2015) Outcomes of occupational self-efficacy in older workers. International Journal of Aging and Human Development 80(4): 357–378.

39.

Perkins

J. M.

Multhaup

K. S.

Perkins

H. W.

Barton

(2008) Self-efficacy and participation in physical and social activity among older adults in Spain and the United States. Gerontologist 48(1): 51–58.

40.

Plouffe

Kalache

(2010) Towards global age-friendly cities: Determining urban features that promote active aging. Journal of Urban Health: Bulletin of the New York Academy of Medicine 87(5): 733–739.

41.

Raskauskas

Rubiano

Offen

Wayland

A. K.

(2015) Do social self-efficacy and self-esteem moderate the relationship between peer victimization and academic performance. Social Psychology of Education 18(2): 297–314.

42.

Rowe

J. W.

Kahn

R. L.

(1997) Successful aging. The Gerontologist 37(4): 433–440.

43.

Schunk

D. H.

(1981) Modeling and attributional effects on children's achievement: A self-efficacy analysis. Journal of Educational Psychology 73: 93–105.

44.

Sherer

Adams

C. H.

(1983) Construct validation of the self-efficacy scale. Psychological Reports 53: 899–902.

45.

Sherer

Maddux

J. E.

(1982) Self-efficacy scale: Construction and validation. Psychological Reports 51(2): 663–671.

46.

Smith

Betz

(2000) Development and validation of a scale of perceived social self-efficacy. Journal of Career Assessment 8(3): 283–301.

47.

Sung

Yoo

(2002) A study on perception of the successful aging concept: Q methodology approach. Journal of the Korea Gerontological Society 22(2): 75–93.

48.

Torres

(2003) A preliminary empirical test of a culturally-relevant theoretical framework for the study of successful aging. Journal of Cross-Cultural Gerontology 18: 79–100.

49.

Tovel

Carmel

(2016) Function self-efficacy scale – FSES: Development, evaluation, and contribution to well-being. Research on Aging 38(6): 643–664.

50.

Wei

Russell

D. W.

Zakalik

R. A.

(2005) Adult attachment, social self-efficacy, self-disclosure, loneliness, and subsequent depression for freshman college students. Journal of Counseling Psychology 52(4): 602–614.

51.

Yang

J.-D.

Park

Y.-K.

(2010) Impact of family and peer support on self-efficacy and juvenile delinquency. Journal of Family Welfare 14(4): 211–235.

52.

Yang

M.-H.

Kim

S.-H.

(2011) Development and validation of the counselor efficacy scales. Korean Journal of Counseling 12(5): 1629–1644.

53.

Zimmerman

B. J.

(2000) Self-efficacy: An essential motive to learn. Contemporary Educational Psychology 25: 82–91.