Survey Sample SelectionMunicipal census records for the <strong>Town</strong> <strong>of</strong> H<strong>in</strong>gham were used to identify residents <strong>of</strong>H<strong>in</strong>gham aged 45 and over. A mail<strong>in</strong>g list <strong>of</strong> all town residents, <strong>in</strong>clud<strong>in</strong>g name, address,and date <strong>of</strong> birth, was obta<strong>in</strong>ed through the H<strong>in</strong>gham <strong>Town</strong> Clerk. The names andaddresses on this list were updated by the mail house service used by the Department <strong>of</strong>Elder Services (Ryan Weaver Mail<strong>in</strong>g Services). From the updated list, researchers randomlyselected a sample <strong>of</strong> 3,400 residents, or approximately 30% <strong>of</strong> the residents aged 45 andolder on the corrected address list.RecruitmentApproximately one week prior to the distribution <strong>of</strong> the questionnaire, a personallyaddressed postcard was mailed to selected respondents <strong>in</strong>vit<strong>in</strong>g their participation <strong>in</strong> thedata collection. The message on the postcard was from Barbara Farnsworth, Director <strong>of</strong>Elder Services, and was designed to make participants aware that they would receive amail<strong>in</strong>g <strong>in</strong> the com<strong>in</strong>g week. In October 2012, we distributed the questionnaireaccompanied by a cover letter signed by Barbara Farnsworth. The cover letter outl<strong>in</strong>ed thepurpose <strong>of</strong> the survey and the measures taken to protect the rights and privacy <strong>of</strong> theparticipants. Those who wished to respond to the questionnaire onl<strong>in</strong>e, rather than return<strong>in</strong>gthe questionnaire by mail, were directed to an electronic version <strong>of</strong> the questionnaire hostedon the SurveyMonkey website. UMass Boston was clearly identified <strong>in</strong> the materials as theresearch partner for the study. As an <strong>in</strong>centive for participation, respondents were entered<strong>in</strong> a draw<strong>in</strong>g for one <strong>of</strong> ten H<strong>in</strong>gham Downtown Association $50 gift cards. These gift cardswere donated <strong>in</strong> part by the H<strong>in</strong>gham Downtown Association. Identification numbers were<strong>in</strong>cluded on each questionnaire so that the researchers could monitor who was eligible forthe draw<strong>in</strong>g. The database conta<strong>in</strong><strong>in</strong>g these numbers is securely ma<strong>in</strong>ta<strong>in</strong>ed by theresearchers.Response RateDur<strong>in</strong>g October and November <strong>of</strong> 2012, a total <strong>of</strong> 749 surveys were received, result<strong>in</strong>g <strong>in</strong> anoverall response rate <strong>of</strong> 22% (see Table 1). 1 The return rate was higher for Seniors (28%)than for Boomers (15%), and was especially high for those aged 80 and older (34%). 2 Only2% <strong>of</strong> the responses were returned onl<strong>in</strong>e.1 Time and budget constra<strong>in</strong>ts precluded issu<strong>in</strong>g follow-up mail<strong>in</strong>gs as a means <strong>of</strong> <strong>in</strong>creas<strong>in</strong>g the overallresponse rate. Although our return rate is with<strong>in</strong> acceptable limits for surveys <strong>of</strong> this type, it is lower than wehad hoped for. Surveys were received <strong>in</strong> selected homes on about October 18-21. About this time, HurricaneSandy was be<strong>in</strong>g tracked on its path toward the east coast; it made landfall <strong>in</strong> the New England area onOctober 29 th . Although Sandy had limited impact on H<strong>in</strong>gham, we speculate that preparations and mediasurround<strong>in</strong>g the hurricane may have reduced participation to the survey.2 “Baby Boomers” refer to the cohort <strong>of</strong> <strong>in</strong>dividuals born follow<strong>in</strong>g World War II. In the U.S., this cohort istypically def<strong>in</strong>ed as those born between 1946 and 1964. As a result, <strong>in</strong> 2012 the Baby Boom cohort was aged48 to 66. We use the term “Boomers” <strong>in</strong> this report to refer to residents aged 45 to 59. This draws adist<strong>in</strong>ction between “Boomers” and “Seniors” (def<strong>in</strong>ed as residents age 60 and older, consistent with languageused <strong>in</strong> the Older Americans Act). As well, this approach facilitates comparison <strong>of</strong> the data presented from thecommunity survey with data drawn from the U.S. Census Bureau, which reports data for the 45-59 population,but not for those aged 48-59.14
Table 1: H<strong>in</strong>gham Community Survey Response RatesTotal Boomers SeniorsSurveys mailed 3,400 1,539 1,861Incorrect address, deceased, ormoved out <strong>of</strong> H<strong>in</strong>gham60 27 33Surveys received (basel<strong>in</strong>e) 3,340 1,512 1,828Surveys returned 749 231 518Response rate 22% 15% 28%ApproachData were entered directly <strong>in</strong>to an SPSS database (version 18.0). Most <strong>of</strong> the items on thequestionnaire were closed-ended questions. These data were analyzed us<strong>in</strong>g descriptivestatistics, <strong>in</strong>clud<strong>in</strong>g frequencies and crosstabs. Responses elicited through open-endedquestions were coded by the research staff and tabulated.All participants <strong>in</strong> the study were aged 45 and older; 69% were aged 60 or more and 19%were aged 80 or more. Although data are not available to generate precise estimates on thedemographic characteristics <strong>of</strong> all potential participants, <strong>in</strong>clud<strong>in</strong>g those who did notrespond to the survey, data from the 2010 Census suggest that 53% <strong>of</strong> the H<strong>in</strong>ghamresidents aged 45+ are aged 60 and over and 16% are aged 80+. Respondents to thecommunity survey are therefore more likely to be Seniors (as compared to Boomers),relative to the community at large.Key <strong>in</strong>formant <strong>in</strong>terviewsOne-on-one <strong>in</strong>terviews were conducted with six <strong>in</strong>dividuals who play leadership roles <strong>in</strong> thecommunity: the <strong>Town</strong> Adm<strong>in</strong>istrator; a member <strong>of</strong> the H<strong>in</strong>gham Board <strong>of</strong> Selectmen;representatives <strong>of</strong> the Police Department; the Fire Chief; the <strong>Town</strong> nurse; and arepresentative from the H<strong>in</strong>gham Interfaith Food Pantry. These <strong>in</strong>terviews focused on ways<strong>in</strong> which H<strong>in</strong>gham’s ag<strong>in</strong>g population is affect<strong>in</strong>g the community at large and other <strong>Town</strong><strong>of</strong>fices and organizations, and to provide <strong>in</strong>put on the perceived effectiveness <strong>of</strong> thecommunity <strong>in</strong> respond<strong>in</strong>g to elders’ needs.Focus groupsTwo focus groups were held <strong>in</strong> H<strong>in</strong>gham to preview the draft results from the H<strong>in</strong>gham dataanalysis and to share <strong>in</strong>sights. A goal <strong>of</strong> hold<strong>in</strong>g these focus groups was to engage <strong>in</strong> <strong>in</strong>depthdiscussion on topics relat<strong>in</strong>g to the report, and to <strong>in</strong>volve community members and<strong>Town</strong> <strong>of</strong>fices <strong>in</strong> the process <strong>of</strong> design<strong>in</strong>g the Recommendations section <strong>of</strong> the report. Onefocus group (N= 10) was composed <strong>of</strong> representatives <strong>of</strong> <strong>Town</strong> <strong>of</strong>fices and nonpr<strong>of</strong>itorganizations. The other focus group (N= 5) was composed <strong>of</strong> community members aged15