Setting and research participants
We used a cross-sectional design in this study. An anonymous, self-administered questionnaire was mailed directly to all active HPVs in Koka and Konan in September 2005. Participants were all registered in 2005 in the HPV association of either Koka or Konan, two cities in Shiga Prefecture in Japan. Nine HPVs were inactive (six in Koka, three in Konan). There were 604 active HPVs (512 in Koka, 92 in Konan). The HPV organizations of Koka and Konan have often worked in cooperation with each other. In Koka and Konan, the ratio of the population to the number of HPVs was 183.1 and 600.0 (population/HPV), respectively, and the ratio of the area of the city to the number of HPVs was 0.94 and 0.77 (km2/ HPV), respectively.
Koka and Konan are adjoining cities in the same county in central Japan, with the community residents of the two cities having culturally similar characteristics. As of September 2005, Koka and Konan had populations of 93,734 and 55,204, respectively. These cities are located within 100 kilometers of Osaka and Nagoya, major cities easily accessible via expressways. Koka and Konan boast farming and traditional ceramics industries. As with many cities, their industrialization began in the 1960s, and, as a result, secondary and tertiary industries began to expand from this period. In 2005, 5.2% of workers in the Koka area were engaged in primary industries, 41.7% in secondary industries, and 53.1% in tertiary industries. In the Konan area, 1.5% of workers were engaged in primary industries, 46.1% in secondary industries, and 52.4% in tertiary industries.
Some people become HPVs on their own initiative, while others are invited to perform this role. To become an HPV, an individual must be nominated by the head of a neighborhood association and attend a municipality-sponsored Health Promotion Volunteer Training Course provided over approximately one year. In this training course, classes provide information concerning health and characteristics of health problems in the locality are described by healthcare professionals (e.g., public health nurses, physicians, and dietitians). Veteran HPVs discuss their experiences in carrying out HPV activities. After completing the course, HPVs are formally commissioned by the mayor. There is no set term of service, and each volunteer usually records her daily activities. The activities of HPVs are funded by the local municipality. Funds are used for the cost of running the activities, and, as mentioned, HPVs are not paid for their labor. They receive advice and support from public health nurses or dietitians who advise them about customary HPV activities and attend their regular meetings.
Measures
According to the Volunteer Process Model [18–22], we organized the variables into three sequential stages.
Antecedent variables
We asked about the process and reasons for the decision to become an HPV, about motivation for engaging in volunteer activities at the end of the HPV Training Course, and about where the volunteers conduct their activities (environment). The decision to become an HPV was considered either passive or active. Motivations for engaging in volunteer activities at the end of the training course were assessed by one item: “I felt highly motivated to engage in HPV activity at the end of the HPV Training Course.” Respondents answered on a four-point Likert scale (1 = strongly disagree, 2 = disagree, 3 = agree, 4 = strongly agree). In the analysis, motivation was classified into two categories: high (3 and 4) and low (1 and 2).
The activity environment included climate of the volunteers’ HPV organization, social support from HPV colleagues and community members, and the relationship between HPVs and the head of the neighborhood association. Climate of the HPV organization was assessed by six items: “HPVs are always friendly to each other in my organization,” “HPVs have a sense of connectedness in my organization,” “HPVs can easily express their opinion in my organization,” “HPVs can ask anything they do not know about the activities without hesitation,” “HPVs are willing to help each other in my organization,” and “HPVs can easily take a break from their activities when they have a personal matter or a job.” Respondents answered on a four-point Likert scale (1 = strongly disagree, 4 = strongly agree). Scores were added, with a higher score indicating a better organizational climate. Cronbach’s alpha was 0.90. In the analysis, we classified data into two categories (good and bad) by median scores.
We assessed two potential sources of social support for HPVs: HPV colleagues (10 items) and community members (6 items). In accordance with the functional classification presented by House [23], items about HPV colleagues included emotional support (e.g., share my satisfaction with and worries about HPV activities), instrumental support (e.g., help me when I find HPV activities burdensome), informational support (e.g., give me information and knowledge about HPV activities), and appraisal support (e.g., provide appreciation and encouragement for my HPV activities). Items about community members were limited to emotional, informational, and appraisal support. Respondents answered on a four-point Likert scale (1 = strongly disagree, 4 = strongly agree). Scores were added up separately for each source of support. The higher the score, the greater was the recognition of availability of social support from each source. Cronbach’s alpha was 0.95 for colleagues and 0.93 for community members. In the analysis, we classified into two categories (high and low) by median scores.
The relationship between HPVs and the head of the neighborhood association was assessed using one item: “In my district, the head of the neighborhood association is uncooperative regarding HPV activities.” Respondents answered on a four-point Likert scale (1 = strongly disagree, 4 = strongly agree). In the analysis, we classified data into two categories: agree (3 and 4) and disagree (1 and 2).
Experience variables
We included overall activity satisfaction and burden, as assessed by two items. Participants responded to the statement “Overall, I am satisfied with my HPV activity,” using a four-point Likert scale (1 = strongly disagree, 4 = strongly agree). The other item was “Overall, I feel burdened by my HPV activity” and was scored using the same four-point Likert scale. Satisfaction and burden were dichotomized: high (3 and 4) and low (1 and 2).
Consequence variables
Activity involvement and persistence were included, with proxies of these being the number of activities an HPV performed over a 3-month period and years of HPV experience, respectively. The 3-month period we selected was considered usual in that no special event related to HPV activities occurred during that time. In this study, in addition to consequence factors regarding HPV activity, we also included quality of life as a consequence of HPVs’ daily lives. This was assessed by one item for life satisfaction: “I am satisfied with my daily life,” rated on a four-point Likert scale (1 = strongly disagree, 4 = strongly agree). In the analysis, responses were classified into two categories: high (3 and 4) and low (1 and 2).
Demographics
We inquired about gender, age, socioeconomic status, neighborly ties, interest in being a volunteer, and self-rated health. Socioeconomic status included employment and educational level. Employment was classified according to two categories: employed (full-time, self-employed/agriculture, or part-time) and unemployed. Educational level was also classified into two categories: higher education (completed junior college, vocational school, college, or graduate school) and lower education (completed junior high or high school). Neighborly ties were determined by one item, and respondents answered on a four-point Likert scale (1 = almost no ties, 2 = weak ties, 3 = somewhat strong ties, 4 = strong ties). Responses were classified into two categories: strong (3 and 4) and weak (1 and 2).
Statistical analysis
First, according to their answers regarding activity satisfaction and burden, we categorized the participants into four groups: high satisfaction and low burden (hereafter group A), high satisfaction and high burden (hereafter group B), low satisfaction and low burden (hereafter group C), and low satisfaction and high burden (hereafter group D). Moreover, we compared demographics and antecedent variables among the four groups, organized by satisfaction and burden, using a one-way analysis of variance for the continuous scales, the Kruskal-Wallis test for the ordinal scale, and the chi-square test for the nominal scale.
Second, we compared consequence variables among the four groups. A generalized linear model was used to examine the differences, setting consequence variables as the response variables. The total number of activities and years of experience were used as continuous scales, and life satisfaction was a nominal scale. We used age, initial motivation, and affiliation (city) as covariates. Although we analyzed the combined data for Koka and Konan, we adjusted for the difference in city area between the two using the variable of affiliation in the model because the ratio of population to number of HPVs was different between the two cities.
Finally, to investigate which factors were associated with differences among HPV groups when divided by levels of satisfaction and burden (experience variables), a multinomial logistic regression analysis was performed. Demographics and antecedent variables that were significantly different among the four groups were selected as independent variables and included in the model. In addition, affiliation was adjusted for in the model. Results were calculated as adjusted odds ratios (ORs) with 95% confidence intervals (95% CIs). A two-tailed p < 0.05 was considered significant, with p < 0.10 considered marginally significant. We used IBM SPSS 20.