Development of Japanese version of the Friendship Scale and its reliability and validity:A new measurement scale for evaluating social isolation. (English translated by DeepL))

This paper is an English translation of the following Japanese paper translated into English using DeepL. Please understand that the English may not be accurate in some parts.

http://hdl.handle.net/10232/00032397

--------------------------------------------------------------------------------------

Development of Japanese version of the Friendship Scale and its reliability and validity: A new measurement scale for evaluating social isolation.

Cultural Science Reports of Kagoshima University, 90, 1-12, 2023.

Koji Abe, Ph.D.

Kagoshima University

INTRODUCTION

    Since the spread of the COVID-19 in January 2020 and the declaration of a state of emergency in April 2020, our lives have undergone extremely significant changes. In particular, face-to-face interactions have been severely restricted, and opportunities for interpersonal communication have decreased for all age groups. However, the impact of several years of restrictions on interpersonal exchanges on our lives and health is expected to be significant. One of the most important effects of these limitations on interpersonal interaction is social isolation(Banerjee & Rai, 2020)。Donovan & Blazer (2020) reported a 24% increase in socially isolated older adults in the United States compared to pre-pandemic levels of COVID-19. The increase in social isolation is a major cause of various diseases such as depression(Koizumi et al., 2005; Krendl & Perry, 2021), suicide(Awata et al., 2005; Sher, 2020), dementia(Fratiglioni et al., 2000), heart disease, and stroke(Colantonio et al, 1993; Orth-Gomer et al., 1993), and is a major factor in shortening life expectancy(Holt-Lunstad et al., 2010).

    Loades et al. (2020) revealed that the limitation of interpersonal interaction due to COVID-19 has a significant impact on social isolation and loneliness not only among the older people but also among children and adolescents. In addition, Yamamoto et al. (2022) conducted a longitudinal study in Japan from 2020 to 2021 and reported increased loneliness and reduced social networks, with particularly large effects on mental health among young people.

    In order to examine the relationship between COVID-19 epidemic and social activities and quality of life, it will be extremely important to accumulate studies on social isolation in a wide range of age groups. However, although several methods and scales have been developed to measure social isolation, only a limited number of methods have been reported in Japan. 

Measuring Social Isolation

    There are two methods of measuring social isolation: one is that combines objective indicators such as the number of interactions with others, household composition, and number of friends, in line with Townsend's (1963) definition of "the objective state of having little contact with one's family or community," and the other is that measures the state of poor social interaction as social isolation (Kawai, 2009; Saito et al., 2015).

    One representative measure is the Lubben Social Network Scale (Lubben, 1988: a Japanese version has been developed by Kurimoto et al.(2011)), which measures subjective feelings of loneliness and desolation caused by unfulfilled interpersonal needs for intimacy (Kudo & Nishikawa, 1983). In studies conducted on adolescents in particular, it has been reported that this subjective sense of loneliness has not only a negative meaning but also a positive meaning, such as the view that, from the viewpoint of personality development, enduring loneliness and staying in solitude leads to self-reliance (Okonogi, 1979). Hawthorne (2006), after reviewing the studies on these numerous measures of social isolation, found that the objective situation of lack of interaction with others, social support, and social connections, as well as feelings of alienation, loneliness, and social isolation, were associated with the development of social and emotional relationships (Ochiai, 1988). Social isolation is defined as the combination of two aspects: the objective state of not having interaction with others, social support, or social connections, and the subjective state of feeling alienated, alone, and lonely (Hawthorne, 2006). If we follow this definition, it seems necessary to measure social isolation in terms of items related to both objective situations and subjective states, but there are not many scales that satisfy this condition. For example, the Inventory of Socially Supportive Behaviours by Henderson et al. (1980) is a scale consisting of 6 factors and 52 items and containing the above two aspects, but the reliability of each factor is not sufficient. Rokach (2000) also created a scale consisting of 82 items with 5 factors, including interpersonal isolation and alienation. This scale is intended to capture loneliness over the life cycle and can be used widely from adolescents to the older people. However, the large number of items makes it burdensome to answer, and it is difficult to regard it as a suitable scale for tracking changes in social isolation over time. 

    To solve these problems, Hawthorne (2006) developed the Friendship Scale as a scale that can simultaneously measure objective conditions such as the existence of social ties and interactions with others and subjective conditions such as feelings of alienation and loneliness. The Friendship Scale is a six-item scale with relatively simple items that can be applied not only to the older people but also to a wide range of generations, and all items can be answered within about three minutes. Exploratory factor analysis and validation factor analysis were conducted during the development process, allowing comparison of the factor structure with other sample data.

    In recent years, the usefulness of this scale has been internationally recognized and it has been translated into different languages and used in Malaysia (Nikmat et al., 2014), Singapore (Poon, et al., 2020), Pakistan (Khan & Adil, 2020), and Brazil (Alves et al., 2022). Hawthorne (2006) developed a scale for the older people, and a recent study has found a scale that can quickly measure social isolation in a wide range of ages during the COVID-19 epidemic (Nagarajan et al., 2020). For example, Alves et al. (2022) studied 160 adults with an average age of 30 years and reported the reliability and validity of the Brazilian-Portuguese version of the scale. In this study, we developed a Japanese version of the Friendship Scale by Hawthorne (2006) and conducted a survey of young adults whose lives may have been greatly affected by limited interpersonal interaction, in order to examine the reliability and validity of the scale in Japan.

METHODS

Participants

    A self-administered questionnaire survey was conducted among students at University A and Junior College B in 2019 and 2020. Specifically, data from surveys conducted at University A and Junior College B in July 2019 and at Junior College B in June 2020 were used for factor analysis. Of these, only the data from the survey conducted at Junior College B in June 2020 included items on satisfaction with campus life and evaluation of campus life in order to examine their validity. Data from a follow-up survey conducted in July 2020 among students at Junior College B were also used in the analysis of test-retest reliability. Questionnaires were distributed before and after psychology-related classes at the respective universities and asked to be completed by the subjects. The cover page of the questionnaire clearly states that participation in the study is voluntary and that the respondent may stop answering during the course of the survey, and only those who agree to participate in the study were asked to complete the form.

    In a survey conducted at Junior College B in 2020, participants were asked to save a photo of random alphanumeric characters printed on a questionnaire using a smartphone during the initial survey in order to match the data.

    In other words, an attempt was made to enable matching by filling in the questionnaire with its alphanumeric characters stored on the smartphone at the time of follow-up. Questionnaires were distributed to 314 people in 2019 and 121 people in 2020, for a total of 435 people. Data from 408 people were used for factor analysis, excluding data from one person who did not agree to answer the questionnaire and data from 26 people with missing values in the item on social isolation. Data from 121 people surveyed in 2020 were also used for validation. For the follow-up survey, data from 111 people who participated in both the initial and follow-up surveys and whose consent was obtained were used in the analysis.

Ethical considerations

    A self-administered questionnaire was used in this study. On the front cover of the distributed questionnaires, it was clearly stated that consent to the survey was voluntary and that respondents could stop answering during the survey, and in addition, a box-type graphic was shown on the front cover with the text 'If you agree to participate in this study, please tick the box in the square'. In the present study, only data from those who indicated explicit consent in these instructions were used in the analysis.

Japanese version of the Friendship Scale

    In this study, we first attempted to obtain permission from Dr Graeme Hawthorne, the developer of the Friendship scale, for translation and use in our research. However, Dr. Graeme Hawthorne passed away in 2016, and it was found that he was unable to obtain permission directly. We therefore contacted his spouse, Dr Lesleyanne Hawthorne, who is believed to have inherited the copyright. As a result, permission was obtained from Dr Lesleyanne Hawthorne for translation and use in this study.

    For the translation procedure of the Friendship scale, a double back-translation procedure was used. First, based on the items of the English version of the scale, two translators made Japanese translations of the items individually, and based on these two translations, a Japanese translation of the Friendship scale was made by adjusting the contents of the scale. Next, the items in the Japanese translation were translated into English by another translator. The expressions and contents of the original English version of the scale and the English translation were examined closely, and it was confirmed that the original scale and the English translation were almost the same. Finally, the Japanese version of the Friendship scale was created by the researcher by making minor modifications to the wording to make it more suitable for the survey.

    As in the English version, this study used a five-point scale, ranging from "0. always" to "4. never. In interpreting the scores, a score close to 0 means isolation, while a higher score indicates a richer social connection (Hawthorne, 2006). Items 1, 3, and 4 were treated as reversed items when calculating scale scores.

Depression

    For depression, the K-6 by Kessler et al. (2002) was used; the Japanese version of the K-6 was developed by Furukawa et al. (2008), and its reliability and validity were confirmed. This scale is a five-point scale ranging from "1. always" to "5. not at all" for six items indicating depressive symptoms. In this study, the total score of the six items was calculated and used for analysis. The reliability of the data in this study was α = .886.

Satisfaction and Evaluation of Campus Life

　To test the validity of the Friendship scale, we measured satisfaction with and evaluation of college life, which may be related to isolation and loneliness among college students in the COVID-19 epidemic. Satisfaction with campus life was measured using a 7-point scale from 1. extremely satisfied to 7. extremely dissatisfied, using the item "Overall, how satisfied are you with your campus life?

Evaluation of campus life was measured using a 5-point scale from 1. very good to 5. very bad, using the item "How much better is campus life compared to what you thought it was before you entered the university?

Analysis

    In this study, exploratory and confirmatory factor analyses were first conducted to verify whether the Japanese version of the Friendship scale has the same factor structure as the original scale. If a factor structure different from that of the original scale was extracted, it was considered to have the possibility of having a unique factor structure, and a model different from that of the original scale was constructed for comparative analysis. Next, internal consistency and retest reliability were calculated to examine reliability. Finally, correlation analyses were conducted between the friendship scale and satisfaction with college life, evaluation of college life, and depression (K-6) to examine criterion-related validity.

RESULTS

    Table 1 shows the descriptive statistics results of the Japanese version of the Friendship Scale. The mean of item 6 was slightly high at 4.32, but neither the skewness nor the kurtosis, which indicate the skewness of the distribution, were significantly high.

    Next, an exploratory factor analysis using the maximum likelihood method was conducted, assuming a one-factor solution as in the original scale (Table 2). As a result, it was confirmed that the factor loadings of all items were greater than ±.35. The alpha coefficient indicating reliability was .805. Thus, the Japanese version of the Friendship scale, like the original scale, can be regarded as a one-factor structure. However, low communality was found in some items, suggesting that, unlike the original scale, the Japanese version of the Friendship scale may have multiple factors.

    Therefore, we conducted an exploratory factor analysis with maximum likelihood Promax rotation without fixing the number of factors a priori (Table 3). As a result, unlike the original scale, two factors were extracted, judging from the decrease in eigenvalues and scree plot. The first factor was named the "loneliness" factor (α=.836), because it was composed of items expressing feelings of personal loneliness and isolation, such as "I felt isolated from other people" and "When with other people I felt separate from them”. The second factor was named the "feeling of isolation" factor (α=.813), because it was composed of items that indicated whether or not and to what extent respondents felt connected to others, such as "I found it easy to get in touch with others when I needed to" and "It had been easy to relate to others”.

    The inter-factor correlation between the two factors was -.444, suggesting that although they were extracted as different factors, they had a strong inter-factor correlation.

    Since a confirmatory factor analysis was conducted in Hawthorne's (2006) study to examine factorial validity, the same analysis was conducted in this study. First, assuming a simple one-factor structure, the analysis yielded a goodness of fit of χ2 (9) = 314.299, p < .001, GFI = .765, AGFI = .453, CFI = .678, RMSEA = .289, AIC = 338.299, which were not interpretable. Therefore, as in Hawthorne (2006), we attempted to construct and analyze a model that introduced error correlation. In this study, as in Hawthorne (2006), a model was constructed by introducing error correlations between item 1 and item 3, and between item 3 and item 4 (Figure 1). The results showed that the goodness of fit was χ2(7) = 162.413, p < .001, GFI = .897, AGFI = .690, CFI = .836, RMSEA = .234, and AIC = 190.413, a slight improvement over the initial model.

    A confirmatory factor analysis was also performed for the two-factor solution model shown in Table 3 (Figure 2). The results showed that in the simple two-factor structural model, without assuming error correlation, the goodness of fit was χ2 (8) = 4.654, p = .794, GFI = .996, AGFI = .990, CFI = 1.000, RMSEA = .000, AIC = 30.654.

    Finally, to verify the retest reliability and criterion-related validity of the Japanese version of the Friendship scale, we conducted a correlation analysis between the scores of the Japanese version of the Friendship scale in a follow-up survey of 111 subjects one month later, variables related to satisfaction and evaluation of campus life conducted at the junior college B, and depression ( Table 4). There were strong significant correlations between the corresponding factors and total scores between the scores of the initial and follow-up surveys (loneliness factor r = .524, p < .001; isolation factor r = .611, p < .001; total score r = .628, p < .001).

    Regarding the relationship between satisfaction with college life and the Friendship Scale, a negative correlation was found with the total score (r = -.328, p < .001) and the loneliness factor (r = -.331, p < .001) and a positive correlation with the isolation factor (r = .233, p = .01). Similarly, the ratings of college life were negatively correlated with total scores (r = -.346, p < .001) and loneliness (r = -.271, p = .003) and positively correlated with isolation (r = .318, p < .001).

    A strong positive correlation was found between depression and the Friendship scale with total score (r = .454, p < .001) and loneliness (r = .557, p < .001) and a negative correlation with isolation (r = -.192, p < .01).

DISCUSSION

　In this study, we developed a Japanese version of the Friendship scale, a new scale to measure social isolation, and examined its reliability and validity. First, an exploratory factor analysis was conducted on the factor structure of the Friendship scale, assuming a one-factor structure similar to that of the original scale. This suggests that this scale can be used in applied research in the Japanese environment as well as in the English environment. On the other hand, in the one-factor model, even with the introduction of error correlation as in Hawthorne (2006), the goodness of fit in the validation factor analysis was not satisfactory. The model's adjusted indices also revealed an unobservable correlation between items 1, 3, and 4. Reister et al. (1986) reported that when items related to social life were used in a survey, the responses differed depending on the wording of the items. In the present study, however, the positive expressions of the items suggest that the subjects may have perceived them as different concepts from the items with negative contents. On the other hand, it is known that the inclusion of positive and negative items in a scale such as this scale can prevent responses that include bias (Furnham & Henderson, 1982), and further quantitative psychological examination should be conducted while maintaining the uniqueness of the scale as a whole. Hawthorne (2006) also points this out and speculates that a model consisting of multiple factors may be a better fit to the data.

    In this study, an exploratory factor analysis was conducted without fixing the number of factors to be extracted, and results were obtained that, unlike the original scale, could be judged to be a two-factor structure. In addition, when a validation factor analysis was conducted based on the extracted two-factor structure, values that could be judged to be sufficiently high in terms of the degree of fit to the data were obtained. Therefore, from a quantitative psychological point of view, it is more appropriate to treat this scale as a scale with a two-factor structure; if it is treated as a scale with a two-factor structure, it will be necessary to examine the factors related to each of the extracted "loneliness" and "isolation" factors. On the other hand, since a strong negative correlation was observed between these two factors, it is clear that they measure very similar concepts. Although it is not possible to make a judgment based solely on the results of this study, depending on the purpose of the research, it may be acceptable to treat all items as one concept, or to treat the two extracted factors as individual concepts.

　Hawthorne (2006) pointed out the low reliability of the multidimensional scale measuring social isolation developed by Henderson et al. (1980) and Rokach (2000). However, all values for internal consistency and retest reliability of the Friendship scale in the present study exceeded the normative values. Therefore, the reliability of the Friendship scale can be judged to be sufficiently high. In the field of public health, where health-related indicators are routinely measured in intervention subjects, it is said to be important to reduce the burden of measurement by avoiding the use of indicators with a large number of items and questionnaires as much as possible (Yammarino et al., 1991). The Japanese version of the Friendship Scale developed in this study has only six items, which can be considered reliable enough to be used for measurement over time while reducing the burden on respondents.

    Ueda & Matsuura (2022) reported a positive correlation between loneliness and the frequency of contact with close friends and faculty members among college students in the COVID-19 epidemic. In the present study, significant correlations were also found between Friendship Scale scores and satisfaction with college life and evaluation of college life. Similar correlation coefficient values were found for both satisfaction with campus life and evaluation of campus life for all items of the Friendship Scale. The results of factor-by-factor analysis suggested that satisfaction with campus life was more strongly related to loneliness than to isolation, while evaluation of campus life was more strongly related to isolation than to loneliness. In other words, the fact that the students were not able to lead the kind of college life they had imagined before entering the university and that they were not able to build an interpersonal network may have contributed to their lower evaluation of college life. In addition, the sense of isolation and loneliness caused by the large difference from their imagination is thought to be strongly related to their satisfaction with their college life. Although this study did not examine gender and grade in detail, Kudo & Nishikawa (1983) reported that there were differences in loneliness among college students by gender and grade, and that loneliness was particularly strong among freshmen boys. Since the timing and duration of behavioral restrictions due to COVID-19 epidemic vary considerably from year to year, it is assumed that there are differences in scores on this scale depending on the school year and survey period. It will be necessary in the future to conduct a longitudinal survey of various grades.

    Many measures of social isolation use depression scores for criterion-related validation. In the present study, the K-6 depression scale was also used to test criterion-relevant validity. As a result, a strong correlation was found between the scores of loneliness and all items. On the other hand, a significant but weak correlation was found with the isolation factor. This indicates that the possibility of depression is not so great without active involvement with others, and that subjective feelings of loneliness in this state are more likely to impair mental health. Since it is clear that there is a strong relationship between loneliness and isolation, it will be necessary to carefully monitor the state of isolation and subjective loneliness and, if necessary, to intervene to reduce the isolation (Dickens et al., 2011).

    The Japanese version of the Friendship scale has been translated and used in other languages as well as in English, and its high reliability and validity have been confirmed in each language. Future applications of this scale in international comparative studies and long-term longitudinal studies are expected.

    However, two major issues remain in this study. The first is that the data used in this study were obtained only from university students, and it is unclear whether the scale can be applied to studies of other age groups, especially the older people. The Japanese version of the Friendship Scale developed in this study seems to have high applicability because each item is short and can be uniquely interpreted. However, its relationship to cognitive function in the older people is unknown, and we need to wait for further studies to accumulate. Another major issue is that the university students who were the subjects of this study were only those who attended classes, and students who did not attend classes at the university were not included in the data. This is one of the limitations of questionnaire-based research, but it is also possible that because those who are isolated are not included in the data, only the tendency of those who are not isolated is measured, and only the reliability and validity of the data are examined in this context. In future surveys targeting the older people, it will be necessary to increase the reliability and validity of this scale as well as its applicability by screening those who are isolated and lonely, and by conducting interviews.

    Many psychological studies have focused on loneliness, and only a few studies have focused on social isolation. In addition, factors that prevent loneliness and isolation have not been adequately examined. For example, Lee et al. (2019) investigated loneliness among adults aged 27 to 101 and cited the presence of wisdom as a psychological factor that prevents loneliness, but few studies have clarified the relationship between wisdom and loneliness or social isolation from the perspective of life-span development psychology.  However, few studies have clarified the relationship between wisdom and loneliness or social isolation from the perspective of life-long development It is expected that future studies using this scale will focus on social isolation that occurs in various generations and clarify the related factors.

Acknowledgements

    We would like to express our deepest gratitude to Dr. Lesleyanne Hawthorne for her permission to create and use the Japanese version of the Friendship Scale. We also thank Ms. Ai Shinohara, Ms. Aya Tanabe, and Ms. Chihiro Nakashima, graduates of the Faculty of Law, Economics and Humanities in 2020, for their cooperation in this research.

REFERENCES

Alves, N.C.M., Almeida, P.A., Ferreira, L.C., Teixeira-Salmela, L.F., Leite, H.R., & Oliveira, V.C. (2022). Brazilian-Portuguese version of the Friendship Scale to assess social isolation: cross-cultural adaptation and psychometric properties. Research, Society and Development, 11, e181111234356.

Awata, S., Seki, T., Koizumi, Y., Sato. S., Hozawa, A., Omori, K., (2005). Factors associated with suicidal ideation in an elderly urban Japanese population: a community-based cross-sectional study. Psychiatry and Clinical Neurosciences, 59, 563-569.

Banerjee, D., & Rai, M. (2020). Social isolation in Covid-19: the impact of loneliness. International Journal of Social Psychiatry, 66, 525-527.

Colantonio, A., Kasl, S.V., Ostfeld, A.M., & Berkman, L.F. (1993). Psychosocial predictors of stroke outcomes in an elderly population. Journal of Gerontology, 48, S261-268.

Dickens, A.P., Richards, S.H., Greaves, C.J., & Campbell, J.L. (2011). Interventions targeting social isolation in older people: a systematic review. BMC Public Health, 11, 647.

Donovan, N.J., & Blazer, D.(2020). Social isolation and loneliness in older adults: review and commentary of a National Academies report. American Journal of Geriatric Psychiatry, 28, 1233-1244.

Fratiglioni, L., Wang, H., Ericsson, K., Maytan, M., & Winblad, B (2000). Influence of social network on occurrence of dementia: a community-based longitudinal study. Lancet, 355, 1315-1319.

Furnham, A. & Henderson, M. (1982). The good, the bad and the mad: response bias in self-report measures. Personality and Individual Differences, 3, 311-320.

Furukawa, T.A, Kawakami, N., Saitoh, M., Ono, Y., Nakane, Y., Nakamura, Y., Tachimori, H., Iwata, N., Uda, H., Nakane, H., Watanabe, M., Naganuma, Y., Hata, Y., Kobayashi, M., Miyake, Y., Takeshima, T., & Kikkawa, T. (2008). The performance of the Japanese Version of the K6 and K10 in the World Mental Health Survey Japan．International Journal of Methods in psychiatric Research，17，152-158.

Hawthorne, G. (2006). Measuring social isolation in older adults: development and initial validation of the Friendship Scale. Social Indicators Research, 77, 521-548.

Henderson, S., Duncan-Jones, P., Byrne, D., & Scott, R. (1980). Measuring social relationships: the Interview Schedule for Social Interaction. Psychological Medicine, 10, 723-734.

Holt-Lunstad, J., Smith, T., & Layton, J.B. (2010). Social relationships and mortality risk: a meta-analytic review. PLoS Medicine, 7, e1000316.

Kawai, K.(2009).　The social isolation of the single dwelling elderly in Japan’s major cities. Horitsu Bunka Sha.

Kessler, R. C., Andrews, G., Colpe, L. J., Hiripi, E., Mroczek, D. K., Normand, S. L. T., Walters, E. E., & Zaslavsky, A. M. (2002). Short screening scales to monitor population prevalence and trends in non-specific psychological distress．Psychological Medicine，32，6，959-976．

Khan, A. & Adil, A. (2020). Urdu translation of Friendship Scale: Evidence for the validity and measurement invariance across gender. The Spanish Journal of Psychology, 23, e11, 1-12.

Koizumi,Y., Awata,S., Kuriyama, S., Ohmori, K., Hozawa, A., Seki, T., Matsuoka, H., & Tsuji, I.. (2005). Association between social support and depression status in the elderly: results of a 1-year community-based prospective cohort study in Japan. Psychiatry and Clinical Neurosciences, 59, 563-569.

Krendl., A.C., & Perry, B.L. (2021). The impact of sheltering in place during the COVID-19 pandemic on older adults’ social and mental well-being. The Journals of Gerontology Series B, 76, e53-e58.

Kudoh, T. & Nishikawa, M.(1983).　A study of the feeling of loneliness (Ⅰ): The reliability and validity of the revised UCLA loneliness scale. The Japanese Journal of Experimental Social Psychology, 22，99-108.

Kurimoto, A., Awata, S., Ohkubo, T., Tsubota-Utsugi, M., Asayama K., Takahashi, K., Suenaga, K., Satoh, H., & Imai, Y.(2011).　Reliability and validity of the Japaneseversion of the abbreviated Lubben Social Network Scale, Japanese Journal of Geriatrics, 48，149-157.

Lee, E.E., Depp, C., Palmer, B.W., Glorioso, D., Daly, R., Liu, J., Tu, X.M., Kim, H.C., Tarr, P., Yamada, Y., & Jeste, D.V. (2019). High prevalence and adverse health effects of loneliness in community-dwelling adults across the lifespan: Role of wisdom as a protective factor. International Psychogeriatrics, 31, 1447-1462.

Loades, M.E., Chatburn, E., Higson-Sweeney, N., Reynolds, S., Shafran, R., Brigden, A., & Crawley, E. (2020). Rapid systematic review: the impact of social isolation and loneliness on the mental health of children and adolescents in the context of COVID-19. Journal of the American Academy of Child and Adolescent Psychiatry, S0890-8567(0820)30337-30333.

Lubben, J.E. (1988). Assessing social networks among elderly populations. Family & Community Health, 11, 42-52.

Nagarajan, D., Lee, D.C.A., Robins, L.M., & Haines, T.P. (2020). Risk factor for social isolation in post-hospitalized older adults. Archives of Gerontology and Geriatrics, 88, 104036.

Nikmat, A.W., Al-Mashoor, S.H., & Hashim, N.A. (2015). Quality of life in people with cognitive impairment: nursing homes versus home care. International Psychogeriatrics, 27, 815-824.

Ochiai, Y. (1988).　Chapter20. Friendship and loneliness in adolescence.　Nishihira, N. & Kuze, T.　Handbook of Adolescent Psychology, Fukumura Shuppan.

Okonogi, K. (1979).　Loneliness in adolescence.　Adoclescence，12，16-28.

Orth-Gomer, K., Rosengren, A., & Wilhelmsen, A. (1993). Lack of social support and incidence of coronary heart disease in middle-aged Swedish men. Psychosomatic Medicine, 55, 37-43.

Poon, A., Abdul Wahab, N., Salim, R., & Ow, R. (2020). Well-being and needs of Malay carers of people with mental illness in Singapore. Health and Social Care in the Community, 29, 164-174.

Reister, M., Wallace, M., & Schuessler, K. (1986). Direction-of-wording effects in dichotomous social life feeling items. Sociological Methodology, 16, 1-25.

Rokach, A. (2000). Loneliness and the life cycle. Psychological Reports, 86, 629-642.

Russell, D., Peplau, L., & Cutrona, C. (1980). The revised UCLA Loneliness Scale: concurrent and discriminant validity evidence. Journal of Personality and Social Psychology, 39, 472-480.

Saito, M., Kondo, K., Ojima, T., Hirai, H., & the JAGES group (2015).　Criteria for social isolation based on associations with health indicators among older people: A 10-year follow up of the Aichi Gerontological Evaluation Study. Japanese Journal of Public Health, 62, 95-105.

Sher, L. (2020). The impact of the covid-19 pandemic on suicide rates. QJM: An International Journal of Medicine, 113, 707-712.

Townsend, P. (1963). The Family Life of Old People: An Inquiry in East London. Penguin.

Ueda, J. & Matuura, H. (2022).　What is correlated with the social support in Covid-19 in college students? A survey November in 2020. Japanese Journal of Applied Psychology, 48, 36-37.

Yamamoto, T., Uchiumi, C., Suzuki, N., Sugaya, N., Murillo-Rodriguez, E., Machado, S., Imperatori, C., & Budde, H. (2022). Mental health and social isolation under repeated mild lockdowns in Japan. Scientific Reports, 19, 8452.

Yammarino, F., Skinner, S., & Childers, T. (1991). Understanding mail survey response behavior: a meta-analysis. Public Opinion Quarterly, 55, 613-639.