Researching Loneliness: The Relevance of Mixed-Methods Approaches (Social Wellbeing UOM)

Reading Time: 7 minutes

by Jamie Bonnici, Research Officer, Faculty for Social Wellbeing


The Faculty for Social Wellbeing recently embarked on a number of initiatives to explore the phenomenon of loneliness in Malta. Following a conference where the Faculty’s documentary, “The Wound of Loneliness” was presented, a nationally representative prevalence study was carried out in order to formally investigate the phenomenon. Adopting a mixed-methods approach to researching loneliness is critical, in order to understand both the magnitude of the issue, as well as the personal and contextual factors that are implicated in the experience of loneliness. However, researching such a complex and subjective theme is not without its difficulties.

Loneliness has been defined by prominent researchers in the area as “a distressing feeling that accompanies the perception that one’s social needs are not being met by the quantity or especially the quality of one’s social relationships” (Hawkley & Cacciopo, 2010, p. 218). The phenomenon was largely neglected by social science researchers until the mid-20th century, yet has recently seen a rise in interest due, in part, to studies demonstrating the many negative implications – both physical and psychological – that are associated with prolonged or high levels of loneliness in people’s lives.

The impact of loneliness

Longitudinal research has shown that loneliness predicts health problems, most notably being implicated in cardiovascular diseases, high systolic blood pressure, and mortality rates, among others. We also know that loneliness affects people from all ages and walks of life; studies show that up to 80% of young people under the age of 18 report that they are lonely at least some of the time, with rates gradually decreasing for middle-aged adults and then increasing again for those aged 70 and over (Berguno, Leroux, McAinsh, & Shaikh, 2004; Pinquart & Sorensen, 2001; Weeks, 1994). In addition to contributing to poor physical health outcomes, prolonged or elevated levels of loneliness can also contribute to psychological disorders, such as depression.

Several valid and reliable measures have been developed for measuring loneliness for the purposes of academic research, such as the de Jong Gierveld Scale for Emotional and Social Loneliness (de Jong Gierveld & Kamphuis, 1985; de Jong Gierveld & Van Tilburg, 1999), and the UCLA Loneliness Scale (Russel, Peplau, & Ferguson, 1987). Yet, it is unclear whether such measures are equally suited for use in a practical context; for instance, in order to predict which individuals might be at a higher risk of experiencing adverse effects of loneliness, or to measure the success of a targeted intervention aiming to reduce loneliness.

Being a social phenomenon, loneliness is inherently difficult to measure in a purely objective way – there are no blood tests, genetic markers, or outward physical signs that a physician could use to assess their patient’s predicament. For this reason, researchers have predominantly made use of self-report measures – although these are also vulnerable to inaccurate measurement, since loneliness is a deeply personal experience that individuals may find difficult to describe or quantify as discrete responses in a survey.

The standardised tools that are designed to measure loneliness have attempted to overcome some of the potential uncertainty by breaking the phenomenon of loneliness down into various elements, so that individual variations in conceptualising the issue are less likely to lead to incorrect measurement. This is done by asking questions related to the experience of loneliness, such as whether a person has someone to talk to about their day to day problems, rather than directly asking them to rate their level of loneliness on a scale from low to very high.

Self-report measures of loneliness could also be susceptible to reporting bias, where an individual rates their social connections more negatively if they are in a particularly low-mood state when completing the survey. Researchers (e.g. Penning, Liu, & Chou, 2014) have nonetheless been able to demonstrate strong measurement invariance for established loneliness scales, such as the de Jong Gierveld Loneliness Scale, meaning that the tool effectively captures the construct across different age groups and successive measurements.

Furthermore, even if a formal screening test for loneliness were to be developed, it might be difficult in practical terms to roll out at a large-scale level. Perhaps this is the reason that countries such as the United Kingdom have seen the adoption of innovative methods to informally screen for loneliness in the general population; One such method, the “High Intensity User Service”, was founded by a paramedic who recognised that loneliness was the “number one reason” for repeated ambulance calls in her town (Monteith, in Orton, 2019).

At risk of loneliness

The service identifies individuals who may be at risk of loneliness due to a high number of ambulance or doctor visits, so that representatives can check up on them and offer to visit them for a chat and a cup of tea. The project has been massively successful, being scaled out on a national level and helping over 4,000 people to date. Those who have used the service report that it has helped them to deal with their issues, and ultimately the service can reduce admissions to Emergency services – thus minimising pressures on the hospital system.

Another consideration when researching loneliness is to be aware of other factors that may influence a person’s experience of loneliness, whether by increasing its effects, acting as a protective element, or by outright causing the phenomenon to occur. It has been noted that when we study loneliness, we are not simply measuring compromised social relations between individuals; rather, discussing loneliness may also be serving as a covert way of discussing other issues – such as poverty and social inequality – that may not be as easy to talk about at a societal level (Victor, 2010).

In fact, recent research by Niedzwiedz and colleagues (2016) has demonstrated that, among the older population, social inequality acts as a significant determinant of loneliness; those individuals at the poorest levels of the population were found to be 10% more likely to experience loneliness, when compared to the wealthiest in the population. However, this relationship between social inequality and loneliness has also been found to be mediated by participation in social activities. Conducting a thorough review of existing literature, prior to designing data collection tools, is thus central to researching loneliness since this ensures that researchers capture as many of the variables that may be implicated in the experience as possible.


It is helpful to make use of both quantitative and qualitative approaches when researching a topic such as loneliness, especially when limited empirical data exist with regards to the local context. Qualitative research methods allow us to gain a comprehensive picture of the phenomenon of loneliness, by providing a sense of context in which to better understand results from quantitative measures such as standardised questionnaires. From a social policy and public health perspective, it is equally as important to gain empirical evidence of how many people are experiencing significant levels of loneliness as it is to understand common threads that emerge from studying the lived experience of such individuals.

Furthermore, we need to be cognizant of including all members of the population wherever possible; This is why the Faculty is building upon its findings from the first national prevalence study with a qualitative investigation on the topic of loneliness, which will focus on the in-depth experience of loneliness amongst individuals residing in institutionalised settings, who did not form part of the initial study’s sample.

As researchers, it is imperative that we remain mindful of these methodological and practical issues when attempting to study a topic such as loneliness. Whilst much progress has been made in understanding loneliness in an academic sense, we also need to consider how applicable our findings are to the real-world context of identifying at-risk individuals and delivering interventions.


Berguno, G., Leroux, P., McAinsh, K., & Shaikh, S. (2004). Children’s experience of loneliness at school and its relation to bullying and the quality of teacher interventions. The qualitative report9(3), 483-499.

De Jong Gierveld, J., & Kamphuls, F. (1985). The development of a Rasch-type loneliness scale. Applied psychological measurement9(3), 289-299.

De Jong Gierveld, J., & Van Tilburg, T. G. (1999). Manual of the loneliness scale. VU University, Amsterdam.

Hawkley, L. C., & Cacioppo, J. T. (2010). Loneliness matters: a theoretical and empirical review of consequences and mechanisms. Annals of behavioral medicine40(2), 218-227.

Monteith, N. in Howard, R., Hall, S., Williams, S., & Pritchard-McLean, K. (Writers) & O’Brien, R. (Director). (2019). 2.10. [Television series episode] In Orton, P. (Producer), The Russell Howard Hour. London: Sky One.

Niedzwiedz, C. L., Richardson, E. A., Tunstall, H., Shortt, N. K., Mitchell, R. J., & Pearce, J. R. (2016). The relationship between wealth and loneliness among older people across Europe: Is social participation protective?. Preventive medicine91, 24-31.

Penning, M. J., Liu, G., & Chou, P. H. B. (2014). Measuring loneliness among middle-aged and older adults: the UCLA and de Jong Gierveld Loneliness Scales. Social Indicators Research118(3), 1147-1166.

Pinquart, M., & Sorensen, S. (2001). Influences on loneliness in older adults: A meta-analysis. Basic and applied social psychology23(4), 245-266.

Russell, D., Peplau, L. A., & Ferguson, M. L. (1978). Developing a measure of loneliness. Journal of personality assessment42(3), 290-294.

Victor, C. (2015). Measurement and meaning in loneliness research. Presentation, Brunel University: United Kingdom.

Weeks, D. J. (1994). A review of loneliness concepts, with particular reference to old age. International Journal of Geriatric Psychiatry9(5), 345-355.


%d bloggers like this: