Research on life events has also been criticized for the reliability and validity of
measures (that is, do people produce the same scores if asked to complete the
questionnaire after a time interval and do the measures actually measure what they are
intended to measure?). In the 1970s and 1980s researchers considered reliability over
time (usually called test–retest reliability). One study found that, when people repeated
the measure after 1–2 weeks, there was only 70 per cent agreement (Steele, Henderson
and Duncan-Jones, 1980) and other research has indicated that reliability declines
dramatically over longer periods (Dohrenwend, 2006). Related to this, there are
concerns that retrospective reports may not be valid when, as is often the case, they
are reported by people who are already diagnosed with a disease. In these circumstances
people may be inclined to want to find an explanation for their illness, leading them
to report more life events than those who have no illness (Brown, 1974). This factor
may account for some of the positive relationships found in the early studies. Nowadays
limited credence is given to studies unless they use longitudinal approaches, which assess
life events prior to the development (or at least the diagnosis) of disease.
Over the years there have been improvements to life events methodology and many
different measures of stressful life events have been developed. Some have been
established that are relevant to particular subgroups. For example, there are a range of
scales specifically for use with children and adolescents (Grant et al., 2004). Others have
tried to overcome the methodological limitations of the checklist approach by using
the much more time-consuming method of semi-structured interviews where people
52 STRESS AND HEALTH
Look at Table 3.1 and work out your own LCU score for the past year. Do you think
this is a reasonable indication of the stress you have experienced in the last year?
Before you read on, write down any problems you can see with this approach to
measuring stress.
There have been a number of criticisms of the life events approach highlighting
limitations that may account for the large number of positive findings in the early
literature. Some critics have commented on the fact that where two people both
have the same score their subjective experience may actually be very different. In
one study, researchers interviewed people about their responses and found that,
for some people, the death of a close friend involved the death of a childhood
friend who they had not seen for a long time whereas for others it was a much
more significant loss (Dohrenwend et al., 1990). This has been labelled the problem
of ‘intracategory variability’ (Dohrenwend, 2006).
A further criticism, which you may have identified, is that the Social
Readjustment Rating Scale does not discriminate between positive and negative
events (Jones and Kinman, 2001). However, most people would assume that only
negative events would be harmful for health, while positive events might even be
beneficial. Because of this criticism many more recent approaches to life events
measurement also take into account people’s appraisals of each event.ACTIVITY 3.2