The Online Survey
To supplement the results from the Statistics Bureau’s STULA, an online survey titled “Survey on Social Connections and Daily Life” was conducted by the author in February 2013 with the assistance of INTAGE Research Inc. The online survey covered registered monitors who were between the ages of 20 and 59, not in school, not married, and not engaged in a salary-paying job. In hopes of receiving generated data that remained as consistent as possible with the composition of the population of UMNEP used by the STULA, the respondent population was designed to have a ratio of 14:10:7:4 among people in their 20s, 30s, 40s, and 50s, respectively.
The survey gathered 3622 valid responses, 458 of which were from divorced or widowed respondents. Thus the remaining 3164 responses were used to create the population of UMNEP for the discussion in this chapter. The scope included 1420 (44.9%) responses from people in their 20s, 915 (28.9%) responses from people in their 30s, 570 (18.0%) responses from people in their 40s, and 259 (8.2%) from people in their 50s, which means that three quarters of the survey population was in the 20–39 age group. There were 1571 (49.65%) responses from men and 1593 (50.35%) from women.
In the survey, the respondents were asked to indicate whether they had “seen and talked face to face with a family member, friend, or acquaintance (not including indirect forms of contact such as telephone conversations and e-mails or conversations with store employees or other strangers)” over the preceding 2 days. The SNEP here consisted of respondents who indicated that they had not conversed with any “friends/acquaintances,” “significant others/partners,” or “others” on either of the preceding days.
The Determinants of SNEP
In Chap. 2, the data from the Statistics Bureau’s STULA were used to delineate the attributes that make someone prone to becoming SNEP. Although the survey contains a wide array of highly detailed items, it does not address some factors that may predispose someone to become SNEP.
Thus, the online survey was designed to include these factors, specifically concentrating on past experiences such as work experience and personal childhood relationships and on a broader definition of health that extended beyond medical treatment or care.
Incorporating these new independent variables into the mix of variables already present in the dataset—such as age, gender, and educational background—a probit analysis was performed on the determinants of SNEP. The results appear in Table 4.1.
Determinants of solitary non-employment
In the previous analysis, the data showed that people in their 20s were less likely to become SNEP than the elderly. As shown in Table 4.1, however, the 20–29 age group had a significantly higher probability of becoming SNEP, albeit still at a 10% significance level. On the other end of the age spectrum, the 50–59 age group had a significantly low probability of becoming SNEP, at just 1% significance level. The differences between the results of the STULA analysis and the results of the online survey likely stem from lower participation in the online survey of SNEP in their 50s and older. That is why readers need to remember that the results presented in Chap. 4 are, generally, skewed toward younger SNEP.
Chapter 2 explained that people who never continued school after junior high school (including high school dropouts) were likely to become SNEP in 2011. On the other hand, the analysis of the online survey results showed almost no significant correlation between educational background and solitary non-employment. People with higher levels of educational achievement tend to respond to Internet-based surveys more often. Thus, the scope of SNEP formed from the online survey data probably omitted not only many elderly people but also many high school dropouts and other similar segments.
Even considering these remarks, the results of the analysis shown in Table 4.1 provide statistical support for the facts presented in the figures from Chap. 4. For example, people who had good friends in junior high school had a significantly lower probability of becoming SNEP compared with people who did not have good friends during childhood and adolescence. Likewise, people who remember having regular contact with adults from outside their family or school during junior high school days were also less likely to become SNEP. Those who were bullied in junior high school, however, did not demonstrate a statistically significant tendency toward becoming SNEP compared with those who were not bullied. The online survey data also revealed that people who, after graduating from school or completing their educational courses, never worked had a significantly higher probability of becoming SNEP currently.
While Chap. 2 explained that people who spent no time receiving medical treatment or care were more likely to become SNEP, the analysis on the online survey results indicated that people who “were not healthy but were not visiting any medical facilities for treatment” had a significantly high probability of becoming SNEP. These results demonstrate the possibility that the facts described in Chap. 2 were influenced by a significant number of people who did not seek medical treatment or care despite having health problems, which isolated them from society and weakened their interpersonal communication.
The Determinants of Concerns for the Future
Table 4.2 shows the estimated results of the probit analysis to determine concerns for the future, based on several independent variables.
Determinants of worry about later-life years
First, as a dependent variable, the tendency of people who had strong worries is measured by using the “very” worried response as 1 and all other responses as 0. The results of this investigation showed that both individual- and family-type SNEP had significantly higher rates of strong worries than non-solitary non-employed persons did. Another group of people who felt very worried about their later years was the segment of people who currently had health concerns: those who “were not healthy and were visiting medical facilities for treatment” or “were not healthy but were not visiting any medical facilities for treatment.”
By contrast, those at either end of the age spectrum—young people in their 20s and borderline elderly people in their 50s—did not appear to feel apprehensive about their senior lives. Regarding the 20–29 age group, later-life problems are probably too far into the future for them to be felt as immediate concerns. Meanwhile, people in their 50s probably feel less confused and distraught than other groups because many of them are already preparing for their lives as elderly citizens and see the potential problems as imminent.
Going a step further, another probit analysis was conducted using all responses that indicated concerns for the future (those who were “very” or “somewhat” worried) to be 1 and those who did not feel any such concerns to be 0 as the dependent variable.
Being SNEP, regardless of the type, had no significant effect on whether respondents felt concerns for the future in this broader sense. The results also show that, just as in the refined analysis, unhealthy people were apprehensive about their senior lives, while men and people in their 20s did not tend to be worried.
The probit analysis was performed on the determinants of awareness of welfare, using the “would go on welfare if possible” and “already receive welfare” responses to be 1 and the “would only go on welfare in certain circumstances” and “never want to receive welfare, whatever the circumstances may be” responses to be 0 as the dependent variable. Table 4.3 presents the estimated results.
Determinants of feeling about receiving welfare
While the family-type SNEP coefficient was insignificant compared with those of other groups, including that of the non-solitary non-employed population, the individual-type SNEP showed a significantly high probability of either going on welfare if possible or already receiving welfare. The data show that both men and people who worried about their health tended to be more interested in welfare or were already receiving welfare, whereas highly educated people (graduates of universities or graduate schools) appeared to be more resistant to the idea.