Evaluating measures to assess loneliness in autistic adults

Abstract

Several measures have been developed to assess loneliness in the general population. Here, we examined if, and how accurately, widely-used measures capture the experience of loneliness in autistic adults. A total of 203 autistic participants from the UK completed an online survey, which included two widely-used loneliness questionnaires: the University of California, Los Angeles (UCLA) Loneliness Scale Version 3 and the Social and Emotional Loneliness Scale for Adults (SELSA). Participants completed the measures and provided their views on how well they captured their experiences of loneliness. A direct measure, asking participants how often they felt lonely, was also completed. Results demonstrated that scores on the UCLA and SELSA questionnaires positively correlated with the direct measure of loneliness, suggesting that these tools accurately index autistic experiences of loneliness. However, qualitative responses suggested that participants found both scales difficult to complete. Using reflexive thematic analysis, we identified three themes that encapsulated participants’ views: (1) failing to distinguish between the characteristics/experiences of loneliness and being autistic, (2) loneliness is not a static trait and (3) inappropriate wording. Given that our sample of autistic participants tended to prefer the UCLA scale to the SELSA, we provide initial recommendations around how the UCLA scale could be adapted for autistic adults. Lay abstract: There has been increasing interest in research on loneliness in autistic adults. Much of this research has involved giving autistic adults widely-used questionnaires that are thought to measure how lonely people are. However, these questionnaires have been developed for the general public. We do not know whether these questionnaires accurately measure how lonely autistic adults are. We asked 203 autistic adults to complete an online survey that included two widely-used loneliness questionnaires: (1) the University of California, Los Angeles (UCLA) Loneliness Scale Version 3 and (2) the Social and Emotional Loneliness Scale for Adults (SELSA). We also asked participants to rate how lonely they were from 1 (often/always) to 5 (never). Participants were then asked to give their views on the questionnaires (e.g. what they thought was good, and what they thought was not so good about them). We found that the scores on the UCLA scale and the SELSA aligned with participants’ ratings of how lonely they were, which suggests that these two questionnaires accurately measure loneliness in autistic people. However, our participants also identified several ways to improve the questionnaires. This included (1) better distinguishing the characteristics/experiences of loneliness from those of being autistic; (2) better reflecting how loneliness may change at different times and in different contexts and (3) making the phrasing of the questions clearer. Overall, our autistic participants tended to prefer the UCLA scale to the SELSA. Therefore, we present some recommendations about how the UCLA scale could be changed to be more suitable for autistic people.