Empathetic - Measurement

Measurement

Research into the measurement of empathy has sought to answer a number of questions: who should be carrying out the measurement? What should pass for empathy and what should be discounted? What unit of measure (UOM) should be adopted and to what degree should each occurrence precisely match that UOM are also key questions that researchers have sought to investigate.

Researchers have approached the measurement of empathy from a number of perspectives.

Behavioural measures normally involve raters assessing the presence or absence of certain either predetermined or ad-hoc behaviours in the subjects they are monitoring. Both verbal and non-verbal behaviours have been captured on video by experimenters such as Truax (1967b). Other experimenters, including Mehrabian and Epstein (1972), have required subjects to comment upon their own feelings and behaviours, or those of other people involved in the experiment, as indirect ways of signalling their level of empathic functioning to the raters.

Physiological responses tend to be captured by elaborate electronic equipment that has been physically connected to the subject's body. Researchers then draw inferences about that person's empathic reactions from the electronic readings produced (e.g. Levenson and Ruef, 1992; Leslie et al., 2004).

Bodily or "somatic" measures can be looked upon as behavioural measures at a micro level. Their focus is upon measuring empathy through facial and other non-verbally expressed reactions in the empathiser. These changes are presumably underpinned by physiological changes brought about by some form of "emotional contagion" or mirroring (e.g. Levenson and Ruef, 1992*; Leslie et al., 2004*). It should be pointed out that these reactions, whilst appearing to reflect the internal emotional state of the empathiser, could also, if the stimulus incident lasted more than the briefest period, be reflecting the results of emotional reactions that are based upon more pieces of thinking through (cognitions) associated with role-taking ("if I were him I would feel...").

Paper-based indices involve one or more of a variety of methods of responding. In some experiments, subjects are required to watch video scenarios (either staged or authentic) and to make written responses which are then assessed for their levels of empathy (e.g. Geher, Warner and Brown, 2001); scenarios are sometimes also depicted in printed form (e.g. Mehrabian and Epstein, 1972). Measures also frequently require subjects to self-report upon their own ability or capacity for empathy, using Likert-style numerical responses to a printed questionnaire that may have been designed to tap into the emotional, cognitive-affective or largely cognitive substrates of empathic functioning. Some questionnaires claim to have been able to tap into both cognitive and emotional substrates (e.g. Davis, 1980). More recent paper-based tools include The Empathy Quotient (EQ) created by Baron-Cohen and Wheelwright which comprises a self-report questionnaire consisting of 60 items.

For the very young, picture or puppet-story indices for empathy have been adopted to enable even very young, pre-school subjects to respond without needing to read questions and write answers (e.g. Denham and Couchoud, 1990). Dependent variables (variables that are monitored for any change by the experimenter) for younger subjects have included self reporting on a 7-point smiley face scale and filmed facial reactions (Barnett, 1984).

A certain amount of confusion exists about how to measure empathy. These may be rooted in another problem: deciding what is empathy and what is not. In general, researchers have until now been keen to pin down a singular definition of empathy which would allow them to design a measure to assess its presence in an exchange, in someone's repertoire of behaviours or within them as a latent trait. As a result they have been frequently forced to ignore the richness of the empathic process in favour of capturing surface, explicit self-report or third-party data about whether empathy between two people was present or not. In most cases, instruments have unfortunately only yielded information on whether someone had the potential to demonstrate empathy (Geher et al., 2001)*. Gladstein (1987) summarises the position noting that empathy has been measured from the point of view of the empathiser, the recipient for empathy and the third-party observer. He suggests that since the multiple measures used have produced results that bear little relation to one another, researchers should refrain from making comparisons between scales that are in fact measuring different things. He suggests that researchers should instead stipulate what kind of empathy they are setting out to measure rather than simplistically stating that they are setting out to measure the unitary phenomenon "empathy"; a view more recently endorsed by Duan and Hill (1996).

In the field of medicine, a measurement tool for carers is the Jefferson Scale of Physician Empathy, Health Professional Version (JSPE-HP). At least one study using this tool with health sciences' students has found that levels of empathy are greater amongst females than males, and also are greater amongst older students than younger students.