Skip to main content
Skip header Skip navigation

AR Methodology

Qualitative Reports banner blue and orange


The purpose of the ethnographic/qualitative study was to provide detailed information and stories about the experiences of participants receiving the “cash option” in Arkansas’ IndependentChoices program. The study focused on how “care units,” composed of consumers and/or representatives, paid workers, and counselors interacted around issues of consumer directed care. The study addresses several broad research questions: (1) has IndependentChoices made a difference in the lives of consumers, representatives, and workers, and if so, how; (2) how does participation in the IndependentChoices compare with previous arrangements; (3) in what manner are services provided, and (4) how does the program work?

Back to the Top


Following the approach taken by Keigher and Luz (1997), the focus was on both consumers (IndependentChoices participants and their families) and their paid workers (both family and non-family member workers). Additionally, interviews were conducted with counselors assigned to provide oversight for those consumers included as part of the care unit interviews.  Qualitative data collection (involving interviews, visits to consumers homes, and travel throughout the state) took place in Arkansas from March through June 2000. Face-to-face, semi-structured interviews lasting about one hour each were conducted with the members of 27 care units. The tape-recorded interviews were conducted by two trained interviewers and took place in the consumers’ homes. Each consumer interviewed had participated in the Arkansas IndependentChoices Program for at least eight months.  Consumers of different ages, races, sexes, monthly cash benefit amounts, and type of care workers were interviewed.Three semi-structured interview protocols (consumer/representative, paid caregiver, and counselor) were developed through an iterative process of consultation with the Cash & Counseling Program management team (included representatives from the program management office, Mathematica Policy Research, IndependentChoices Program in Arkansas, ASPE-HHS, RWJ). Interviews included open-ended questions organized under three headings: background, services, and operations of the program. In March of 2000, two interviewers conducted pilot interviews in Arkansas with nine care units. Following the pilot interviews the interviewers were carefully debriefed, further discussion occurred with the program management team, and the protocols were finalized (Appendix A: Interview Protocols).

Kiegher, Sharon M. and Luz, Clare. (1997). A Pilot Study of Milwaukee’s Gray Market in Independent Care: Common Stakes in Homecare of the Elderly. School of Social Welfare, University of Wisconsin-Milwaukee.

Back to the Top


Selection criteria for the care units (triads including consumer/representative, paid worker, and counselor) were developed in close consultation with the Management team. Four main selection criteria were identified: age 18-64/65+, white/African American, type of worker (family/non-family), and representative/no representative. Other variable taken into consideration to ensure maximum variability were rural/urban, recently discharged from a nursing home, drop-out from the program, variations in use of the cash benefit, disability level (determined by level of Medicaid benefit, but above $200 per month), gender, new users (i.e., OMB’s definition of new, meaning no prior program experience), high counselor intervention, age extremes, consumer without fiscal intermediary, and one worker with two program participants. A further criterion in selecting the sample was to include consumers from the two regions of the state, Little Rock and Helena, covered by the two counseling/fiscal intermediary agencies.The names of eligible consumers (i.e., had participated in the program for at least eight months and fitting the selection criteria) to contact were derived from lists provided by Mathematica Policy Research and Arkansas program office. Research staff then telephoned potential subjects to enlist their participation and schedule home visits to conduct interviews.

Back to the Top

Data Collection/Interviews

A total of 59 interviews were conducted (27 care units composed of 27 consumers/representatives, 27 paid workers, and 5 counselors). Two interviewers, one male Caucasian and one female Caucasian made three visits to Arkansas (including the visit for pilot interviews) between March and June, 2000 to conduct the interviews. All interviews with consumers and paid workers were face-to-face and conducted in the home of the consumer. In most cases consumers and paid workers were interviewed during the same visit. Interviews averaged about one hour and were tape-recorded.Five counselors (3 from the Little Rock agency and 2 from the Helena agency) were interviewed face-to-face at a later date. Counselors were interviewed separately and asked about their activities as counselors in the IndependentChoices program generally, and specifically, about the consumers in the study for whom they were responsible.

Back to the Top


Individual tape recordings were transcribed verbatim as Word documents and then uploaded into software designed for qualitative data management and analysis (NUD*IST 4). The software was used to manage the large quantity of textual data, facilitate coding, and discover crosscutting topic/themes. Stories were written from the interview transcripts.

Back to the Top

How We Wrote the Stories

We wrote each of the twenty-seven stories based on related interviews with a consumer, personal care worker, and counselor. In writing the stories, we tried to incorporate all three perspectives on a topic or story whenever possible. Each story relies heavily on direct quotations with a limited amount of our writing to connect the narrative. We also used the interviewers’ field notes and comments to add detail to the stories.

Occasionally, the facts or chronologies were not clear or were in conflict with the information in other interviews within a unit. In some cases the discrepancies came about because of a time lag between interviews, differences of viewpoint about the topic, or a lack of knowledge on the part of a member of the unit. In cases where facts were unclear or contradictory, we checked the interviewer’s notes and reconciled evidence within the transcripts themselves.

Each of the author’s wrote the original story drafts for a third of the cases. We discussed story content and interpretations as a group and read each other’s drafts and the interview transcripts to check for completeness and accuracy.We added topic headings to facilitate reading and comparison between stories. Names are all pseudonyms. Places, companies, and agency names have been eliminated or fictitious names used in their stead. Within quotations, three periods represent a short break within an answer such as an interviewer breaking in to ask a question. Four periods are used to indicate a considerable break in the conversation. For example, sometimes a consumer returned to a topic or expanded on an answer much later in an interview. Brackets within a quote include our clarification of a reference to a person or event. The terms caregiver, care worker, and care provider were used interchangeably as were consumer and participant.Quotations were rendered as close to verbatim as possible in order to preserve the rhythm and regional flavor of the interviewees’ speech patterns. Above all, the participants speak for themselves about what life is like for them in the IndependentChoices program.

Back to the Top

Cash and Counseling Banner of Participants' Smiling Faces