Keywords

1 Background

Due to demographic change, the proportion of people over 65 is rising [1] and most older adults wish to live at home as long as possible [2,3,4] which is beneficial for their quality of life [5]. This will lead to an increase in the number of people in need of support in the future, which may exacerbate the existing shortage of nursing staff in the out- and inpatient sector and increase health care costs [6,7,8,9,10]. To address the problems in the out- and inpatient sector, more and more technical innovations are being developed and used to support older adults at home and caregivers in institutions. It is to be expected that the use of robots in older adults’ homes or institutions will also play an increasingly important role, in maintaining the independence and well-being of older adults [11,12,13] and to relieve caregivers [14].

To achieve accepted robotic solutions, the opinions and concerns of the population need to be considered. Despite the actuality (popularity) of the topic, ambivalent results exist indicating low acceptance [15] as well as positive attitudes toward robots [16] and there is currently little detailed knowledge about the attitudes and concerns of the Swiss population regarding the use of robots for older adults. Previous surveys focus on specific robots such as special humanoid robots [e.g. 17] or industrial robots [18] but do not address robots for older adults.

In the present study, attitudes, wishes and concerns of the Swiss population towards robot use for older adults are surveyed in order to support further research and development of successful and accepted robots for older adults. Different findings, cultural sensitivity [17], rapid technical development, and a universal approach were considered in developing an innovative questionnaire that addresses robot use for older adults specifically, the situation in Switzerland, and robots in general. The study is ongoing, with the aim of collecting national representative data for later in-depth analyses of the acceptance of robots.

2 Method

As part of a project funded by the University of Applied Sciences St.Gallen in 2018, a questionnaire for the survey study was developed. Based on a literature search, various models of technology/robot acceptance and attitude research were compiled and evaluated for their suitability for the objective of the questionnaire to be prepared. Factors that were present in theories, models and scales concerning robot acceptance often cited in literature were selected by two researchers of the study group and included the following: Multi-dimensional robot attitude scale [19], Almere Model [20, 21], Usability, social acceptance, user experience and societal impact (USUS) [22], Acceptance model for industrial context [18], Technology acceptance model 3 (TAM 3) [23], Model of domestic social robot acceptance [24], Theory of Reasoned Action (TRA) [25], Technology Acceptance Model (TAM) [26], and the Unified Theory of Acceptance and Use of Technology Model (UTAUT) [27]. The selected factors were discussed in an internal scientific colloquium.

Based on this, 13 factors were identified as variables that can influence robot acceptance. Factors related to the person/individual are general attitude toward robots, negative emotions, positive emotions, self-efficacy and trust. Factors related to the robot and its functions are usefulness, perceived barriers (later renamed to effort), joy/fun, safety, control and usability. Factors related to the context are the attitude of the social network and social support. The factors were each collected with one to two items. Further aspects considered relevant for acceptance in literature are included: ethical concerns (seven items), effects on society (one item), preferred appearance of the robot (one item), conceivable functions of the robot (eight items) and possible application areas of robots (two items) [28,29,30,31,32,33]. A Likert scale with seven categories was chosen as answer format for the variables. Sociodemographic variables were recorded according to the recommendations of Flandorfer [34]. Experience with robots [35], technical experience, interest in technology, and the area of professional activity (self-developed) were included, one question each. This resulted in a three-page questionnaire with 48 items that can be completed in 10–15 min.

The questionnaire was pretested by 61 participants. After the pretest, linguistic adjustments were made, and the response scale was reduced to six categories (from 1 “totally disagree” to 6 “totally agree”). Internal consistency resulted in a Cronbach’s Alpha of >.9. The online version of the final questionnaire can be viewed on the website of the Institute (http://www.fhsg.ch/alter). The questionnaire was then distributed via online-link and paper-pencil version at events for relatives and older adults in nursing homes, at municipal events, by e-mail to persons of the institute’s database, and to persons of the institute’s broad network organized in various associations and societies, and by posting the link to the questionnaire on the institute’s homepage. All statistics were calculated using the IBM SPSS 24 software package. The survey started in January 2019 and data are collected continuously. Descriptive statistics are shown for variables described above, based on the data available on March 2020. Answer categories 1, 2, and 3 are taken together to calculate disagreement, and answer categories 4, 5, and 6 to calculate agreement.

3 Results

3.1 Participants

So far, 189 people filled out the questionnaire. The respondents were on average 65.4 years old (SD 16.6 years, range 17 to 96 years). 57.6% were women and 64.5% lived with a partner. 9.2% completed mandatory school education and 65.8% have a tertiary level qualification. The majority (96.8%) lived in a private household and 50.3% rated their residential area as rather rural. 29.4% considered themselves to be very interested in technology, 52.4% as interested, 14.4% as less interested and 3.7% as not at all interested in technology. 59.7% valued themselves as experienced with technology. 54.6% had no previous experience with a robot.

3.2 General Attitudes Towards Robots

The majority of respondents said they were interested in robots, stated having a positive view of robots and would themselves use a robot: 72.3% agreed with the statement that they were interested in robot use and robotic solutions for older adults, 67.5% said they had a generally positive view of robots, and 69.7% said they would personally use a robot. Most of the respondents stated that they could imagine the support of older adults by a robot mainly in the private home (78.6%) or also in a retirement or nursing home (74.3%). 59.9% could imagine the support in a hospital. 68.4% said they could imagine using a robot for themselves, 20.9% for their parents or grandparents, and 61.5% for other people.

3.3 Assessment of Ability to Handle Robots

Most respondents rated themselves as competent in handling robots: 86.0% agreed that they would be able to use a robot after instruction, and 83.1% agreed with the statement that they had all the skills to handle a robot.

3.4 Desired Functions and Application Areas of Robots

Of the eight functions asked, five were conceivable for most respondents. 85.0% could imagine the robot calling for help if something happened (e.g. a fall), 84.0% a reminder function (e.g. taking medication), 72.7% support in everyday activities (such as picking up things), 69.0% monitoring and reporting vital signs (e.g. pulse or blood pressure), and 68.4% the joint performance of movement exercises. For most of the respondents talking and conversing with a robot was not imaginable (72.2%), for 69.0% nursing activities (e.g. washing) by a robot were not imaginable, and the robot as a communication aid to communicate with friends and relatives were not imaginable for 58.3%.

3.5 Preferred Appearance of Robots

In terms of appearance, a robot was preferred that looks more like a machine (25.3%), like a human being (22.5%), 21.9% would not care about appearance, 14.6% preferred a look like an object of utility, 11.2% preferred a look like a fantasy creature, and 4.5% preferred an animal-like appearance. The evaluation separated by gender showed that male participants preferred a machine-like appearance (28.8%) or they would not care about appearance (26.0%), and female participants preferred the human appearance (27.2%) or the machine-like appearance (22.3%).

3.6 Ethical Concerns Regarding the Use of Robots for Older Adults

Participants indicated their agreement to various ethical concerns, which must be considered when using robots with older adults. Concerns mainly related to the lack of interpersonal contact and problems with handling sensitive data. 76.5% and 60.4% respectively agreed with these statements. 49.3% expressed concern that workers will lose their jobs and 45.6% expressed the fear that privacy is not guaranteed. Concerns that the older adults could be deceived (42.4%), that the self-determination of older adults could be restricted (40.0%), and that the dignity of older adults is being violated (38.1%) were expressed.

4 Discussion

To collect attitudes, wishes and concerns about robot use for older adults, a survey study was conducted in Switzerland. The participants up to now were highly educated and highly interested in technology. Therefore, the actual sample is not yet representative of Switzerland. However, it is known from other studies, that well integrated, educated and technology-oriented people are more likely to take part in studies [36,37,38,39]. Due to the recruitment procedure, it was not possible to collect any information about return rates.

Most respondents were generally interested in robots and had positive thoughts about robots in general. That means that not all user-groups and functions of robots seem accepted. Most respondents would use robots for themselves or others but not for their (grand)parents, although many respondents indicated that the use in a retirement or nursing home would be quite conceivable. Possible explanations that should be further investigated are whether the respondents consider their (grand)parents not to be able to handle a robot, or whether respondents feel guilty considering a robot caring for their (grand)parents. The most problematic ethical concerns stated were a lack of interpersonal contact and data security. Developers must address these findings when creating or adapting a robot.

To be able to consider as many factors as possible for a comprehensive opinion, each aspect of robot acceptance was realized with 1-2 items. In the original scales, the respective factors often contained several items. Direct comparability with specific individual scales is therefore not always possible.

The participants were not shown a specific picture of a robot, which could have influenced the answers: Some relied on experience, others only on an internal picture [see 33]. This must be considered for interpretation but is handled like this in other studies too [16, 35].

So far, the data was published for the first time and evaluated descriptively. The aim is to continue the data collection, to achieve representativeness and to provide knowledge for research and development, and to adapt robots to the wishes and needs of the population. There is a need for further research on attitudes and concerns about robots and in-depth, multivariate analyses. Whether the desired or imaginable functions of robots are influenced by other moderating aspects should be assessed. It could make a difference whether the use of a robot is compared to a subjectively less attractive alternative (e.g. entering a nursing home), or whether the robot is used alone or together with a caregiver. Currently, the different age-groups are too small for in-depth analysis. Sub-group analyses (age, gender, educational level) will be carried out with a larger sample. In addition, a theoretical model of robot acceptance will be tested.