(p.251) Appendix D Full Description of the Study Purpose, Design, and Methods
(p.251) Appendix D Full Description of the Study Purpose, Design, and Methods
The purpose of the mixed methods study presented in chapters 3 through 5 was to determine the relationship between senior citizens’ music participation and their perceptions of quality of life. Quality of life was based on Flanagan’s (1978, 1982) model revised by Burckhardt and Anderson (2003), which included material and physical well-being; relationships with other people; social, community, and civic activities; personal development and fulfillment; recreation; and independence. Music participation activities were defined as organized musical groups that rehearse regularly, including choral and instrumental ensembles, in Evergreen Town, a large retirement community. We were specifically interested in the following questions: (a) To what extent did senior citizens participate in music activities in the community? (b) What were senior citizens’ perceptions of their quality of life? (c) Did senior citizens’ music participation relate to their perceptions of quality of life? (d) Was there a difference between music participants’ and music nonparticipants’ perceptions of quality of life? (e) What were senior citizens’ perceived benefits of music participation? (f) For what reasons did senior citizens choose not to participate in music? (g) What activities (musical or no-musical) contributed to senior citizens’ quality of life?
Mixed Methods Design
To answer the stated questions, we used a mixed methods design to acquire both breadth and depth of understanding on the topic. Initial quantitative survey results were followed by qualitative focus group interviews. Findings were mixed at the interpretation stage. Mixed methods research, formalized only since 1959 to incorporate multiple methods in research, has been characterized by well-documented variations in philosophies, definitions, techniques, and developments in the last few decades (Creswell, 2009; Creswell & Plano Clark, 2011). However, there has also been consensus that in mixed methods research, researchers would combine “elements of qualitative and quantitative research approaches (e.g., use of qualitative and quantitative viewpoints, data collection analysis, inference techniques) for the (p.252) broad purposes of breadth and depth of understanding and corroboration” (R. B. Johnson, Onwuegbuzie, & Turner, 2007, p. 123). From viewpoints to data collection and from data analysis to interpretation, elements of both qualitative and quantitative approaches could be incorporated. The combined effect of qualitative and quantitative methods should be greater than the sum of the individual elements (Creswell & Plano Clark, 2011; Hall & Howard, 2008). On the one hand, the integration of both types of data has been deemed advantageous in (a) offering the strengths of both quantitative and qualitative research, (b) answering a broader and more complete range of research questions, (c) adding the strengths of one method to overcome the weaknesses in another method, (d) providing stronger evidence for a conclusion through convergence and corroboration of findings, (e) adding insights and understanding that might be missed when only a single method was used, (f) increasing the generalizability of the results, and (g) producing more complete knowledge necessary to inform theory and practice (R. B. Johnson & Onwuegbuzie, 2004). On the other hand, weaknesses in the mixed methods design have been identified, such as the degree of difficulty for a single researcher to carry it out, the need for the researcher to learn about multiple approaches and to understand how to mix them appropriately, the increase in expense, the increase in time involved, and the need to work out the details about mixing the paradigms (R. B. Johnson & Onwuegbuzie, 2004). In the current inquiry, the advantages of the mixed methods design outweighed the known weaknesses. We believed that the strengths of both quantitative and qualitative methods would provide the best understanding of the relationships in question. We recognized the challenges of the need for more extensive data collection and more time-intensive data analyses. We were comfortable and familiar with both quantitative and qualitative research, and we were extremely grateful for the support of the NAMM Foundation and the Sounds of Learning initiative; therefore, we were ready for this challenge.
In this study, we triangulated results of different methods to seek convergence, corroboration, and correspondence. We adopted the sequential explanatory design (Creswell & Plano Clark, 2011; Creswell, Plano Clark, Gutmann, & Hanson, 2003), a type of mixed methods design in which quantitative data collection was followed by qualitative data collection. It is sequential, because the quantitative elements are followed by the qualitative elements. The “mixing” did not occur until after the results of both methods were revealed, at the interpretation phase. The quantitative elements were based on an established instrument and a demographic questionnaire, combined in the form of a survey. The qualitative elements were based on focus group interviews of a subsample. Qualitative data were used to elaborate, enhance, illustrate, and clarify findings from the quantitative data. Figure D.1 represents the various phases of such a design.
There were 308 participants in the study. This sample represented a response rate of 35%. Studies elsewhere that included senior citizens’ research response rates indicated that those who participated in research studies tended to have fewer health problems, were better educated (Herzog, Rodgers, & Kulka, 1983; Launer, Wind, & Deeg, 1994; Lorant, Demarest, Miermans, & Oyen, 2007), and did better on a cognitive test (Launer, Wind, & Deeg, 1994). These characteristics supported the “nonhomebound” requirement to participate in the study. (p.253)
The survey was our tool of choice for the collection of quantitative data. It consisted of two sections. The first section was the Quality of Life Scale (Burckhardt & Anderson, 2003), which had been validated for use in diverse groups (Burckhardt & Anderson, 2003; Burckhardt, Anderson, Archenholtz, & Hägg, 2003). The version used in this study was downloaded from http://www.myalgia.com based on the recommendation of Burckhardt (personal communication, June 9, 2009) (see appendix E). It contained 16 items in the following conceptual categories: (a) material and physical well-being; (b) relationships with other people; (c) social, community, and civic activities; (d) personal development and fulfillment; (e) recreation; and (f) independence. Study participants were to rate each item on a seven-point Likert-type scale (1 = terrible, 7 = delighted). It was based on a model developed by Flanagan (1978, 1982), which included “listening to music” as a component under “recreation” and had been referred to in music participation studies of senior citizens (p.254) (Coffman, 2002; Coffman & Adamek, 1999). For reliability information of the Quality of Life Scale, see appendix H.
The second section of the survey, appendix F, was a researcher-designed demographic questionnaire. We used it to collect data on age, sex, length of residency in the community, prior home state and occupation, music participation (type, frequency, and length), and reason(s) for participating (or not participating) in music. At the end of the survey, we asked if the study participant was willing to participate in a 60- to 75-minute focus group interview.
We provided information about the study to senior citizens in advance of the data collection dates, via brief information sessions held during regular meetings of various musical and nonmusical groups. We wanted to acquaint the senior citizens with our purpose and process with the hope that they would become more willing to participate in the study. We believed that they would be more likely to participate if they were familiarized with the study ahead of time, and that the information session would build interest and momentum within the community for the study. We distributed and collected the survey forms during and following subsequent meetings of these groups. Participants took 15 to 30 minutes to complete the forms. Survey administrators were available to answer questions in each data collection session. We chose to administer the survey in person, rather than via phone or mail, to aim at a better response rate and more valid and better-quality data. It has been documented that lower response rate and missing responses increased with age beyond 60 years in mailed surveys (Kaldenberg, Koenig, & Becker, 1994). Due to the length and complexity of the survey for this age group, we also avoided conducting the survey over the phone (Craik, 1999). Regardless, we still encountered resistance from those who were unwilling to sign anything despite our explanations of the study. Signature on an informed consent form was a prerequisite to participate in the study.
The Focus Group Interviews
An interview guide was developed ahead of time for use with a subsample of the survey to collect more in-depth qualitative data. A guided approach was used in the focus group interviews.
The interview guide provides topics . . . within which the interviewer is free to explore, probe, and ask questions that will elucidate and illuminate that particular subject. Thus, the interviewer remains free to build a conversation within a particular subject area, to word questions spontaneously, and to establish a conversational style but with the focus on a particular subject that has been predetermined. (Patton, 2002, p. 343)
This approach allowed for an ideal balance between spontaneity and structure.
Among those who returned the survey with an agreement to be interviewed (n = 131; 47.8%), we interviewed a subsample (eight music nonparticipants and 54 music participants) in 16 focus group interviews. Each focus group interview included two to six participants and was video and audio recorded for use in transcription. Two of these focus group interviews were dedicated to music nonparticipants. Focus group interview was chosen over individual interview because senior citizens learn better, participate more freely, and (p.255) recall more about their lived experiences in collaborative, familiar environments with friends (Johansson, Andersson, & Rönnberg, 2000; Powers & Love, 2000).
For music participants, questions were asked about their reasons for music participation and about their perceived benefits of music participation. For music nonparticipants, questions were asked about their reasons for not participating in musical groups. For both music participants and nonparticipants, we asked about the activities that they believed to have contributed to their quality of life and the musical activities in which they had participated in the past. See appendix G for the interview guide. The focus group interviews led to a total of 645 minutes of recording for transcription. We then coded the transcriptions for further analyses. Details about the coding are included in appendix O. Findings of the focus interviews and their relations with the survey are presented in chapters 4 and 5. (p.256) (p.257)