CoChat

Exploring How Older Adults Use a Smart Speaker–Based Voice Assistant in Their First Interactions: Qualitative Study

Open access
Kim, Sunyoung · 2021 · JMIR mHealth and uHealth

Published version of a preprint

The DOI matches its Crossref record; 50 of the 50 references it cites were found in a registry or an index; 5 of the 21 sources its claims cite were checked for support.

Abstract

BACKGROUND: Smart speaker-based voice assistants promise support for the aging population, with the advantages of hands-free and eyes-free interaction modalities to handle requests. However, little is known about how older adults perceive the benefits of this type of device. OBJECTIVE: This study investigates how older adults experience and respond to a voice assistant when they first interact with it. Because first impressions act as strong predictors of the overall attitude and acceptability of new technologies, it is important to understand the user experiences of first exposure. METHODS: We conducted semistructured interviews with 18 people 74 years and older who had never used a smart speaker before, investigating the patterns of use, usability issues, and perspectives that older adults have when using a voice assistant for the first time. RESULTS: The overall first response to a voice assistant was positive, thanks to the simplicity of a speech-based interaction. In particular, a positive and polite response to complete the interaction with a voice assistant was prevalent, such as expressing gratitude or giving feedback about the quality of answers. Two predominant topics of commands made in the first interaction include asking health care-related questions and streaming music. However, most of the follow-up reactions were unfavorable because of the difficulty in constructing a structured sentence for a command; misperceptions about how a voice assistant operates; and concerns about privacy, security, and financial burdens. Overall, a speech-based interaction was perceived to be beneficial owing to its efficiency and convenience, but no other benefits were perceived. CONCLUSIONS: On the basis of the findings, we discuss design implications that can positively influence older adults' first experiences with a voice assistant, including helping better understand how a voice assistant works, incorporating mistakes and common interaction patterns into its design, and providing features tailored to the needs of older adults.

Overview

CoChat Summary

## TL;DR

This qualitative study examined how 18 older adults aged 74+ who had never used a smart speaker responded in their first interactions with a voice assistant. First impressions were generally positive because speech felt simple, efficient, and convenient, but users quickly encountered problems with command structure, misconceptions about how the system works, and concerns about privacy, security, and cost.

## Key findings

  • Participants made 234 conversational interactions with the voice assistant. The most common command topics were health care-related questions (91/234, 38.9%), streaming classical/older music (66/234, 28.2%), and asking for directions/locations (30/234, 12.8%).
  • Many participants ended interactions with polite or appreciative responses such as gratitude or feedback, and some remarks served to start follow-up human conversation.
  • The main difficulties were constructing a structured command sentence, misunderstanding how the voice assistant operates, and concerns about privacy, security, and financial burdens.
  • The speech-based modality was viewed as efficient and convenient, but participants reported no other benefits and often concluded that the device was “good, but not for me.”

## Methods

  • Design: Qualitative study using semistructured interviews and thematic analysis (open coding, axial coding, selective coding).
  • Sample/data: 18 participants from a local assisted-living facility in the greater New York area; all were 74 years or older and had no prior experience with a voice assistant. Interviews lasted 1.5 to 2 hours and were audio-recorded and transcribed.
  • Main analysis: Participants first interacted freely with an Amazon Echo voice assistant, then responded to scenario-based card decks (7 categories), and finally discussed their perspectives; the authors analyzed interview transcripts for recurring themes.

## Limitations

  • The paper states that the sample size was small and may not be representative of the general population.
  • It also states that convenience sampling from one assisted-living facility may limit generalizability and introduce selection bias or effects from participant homogeneity.
  • The study only examined first interactions in a controlled interview setting, so the text makes evident that findings may not generalize to long-term, everyday use or to users who already own and regularly use a voice assistant.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.2196/20427

Cites

50 of 50 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Figures

Figure 1 The setting of a semistructured interview with a group of participants.
Figure 1 The setting of a semistructured interview with a group of participants.CC BY · PubMed Central

Verification

Claims the paper makes

4 Valid · 1 Not valid · 9 Not checked · 78 Own results and interpretations

47 more claims

Create a free account to see all 92 claims and every source verdict.

See all claims — free account
  • Probably not a substantial claim. 13
  • It may state two separate claims as one. 31

Create a free account to see these claims.

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

11 more claims about this paper

Create a free account to see every claim other papers make about this one.

See all claims — free account

Citations

Cites

Show all 50 references

Cited by

See all 153

Metadata

Authors
Kim, S.
Affiliations
Rutgers, The State University of New Jersey (US)
Corresponding author
Kim, Sunyoung
Venue
JMIR mHealth and uHealth
Published
2021-01-13
Volume
9
Issue
1
Pages
e20427
ISSN
2291-5222 (ISSN-L)
OpenAlex
W3118958215
Cited by
153
References
38
Citations by year
2026: 30 · 2025: 32 · 2024: 24 · 2023: 22 · 2022: 33 · 2021: 12
FWCI
25.19
Type
journal article
Language
en
Open access
Licence
cc-by
Locations
  • JMIR mhealth and uhealth (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
Topics
  • Health Sciences › Medicine › Geriatrics and Gerontology › Technology Use by Older Adults
  • Social Sciences › Psychology › Social Psychology › Social Robot Interaction and HRI
  • Health Sciences › Health Professions › Occupational Therapy › Assistive Technology in Communication and Mobility
Keywords
voice assistants, smart speakers, older adults, first impressions, user experience, speech-based interaction, usability, privacy concerns, music streaming, technology acceptance, design implications
MeSH terms
Smartphone, Aged, Humans, Speech, User-Computer Interface, Voice, Privacy

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

The claim verdicts this report counts are experimental: their accuracy is not yet measured.

A language model drew the claims from the paper on 2026-09-29, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-09-29 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Cited sources are checked with a language model’s help (gpt-6-sol).

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.