Language Log has a great post looking at differences in empathy between males and females, and highlights a new study showing race differences as well.
The punchline is that it’s actually really hard to say whether either of these results reflect true differences because the samples tend to be unrepresentative of the population, and measures of empathy tend to be influenced by the social situation in which they’re taken.
They grab this paragraph from a review article on empathy measurement:
In general, sex differences in empathy were a function of the methods used to assess empathy. There was a large sex difference favoring women when the measure of empathy was self-report scales; moderate differences (favoring females) were found for reflexive crying and self-report measures in laboratory situations; and no sex differences were evident when the measure of empathy was either physiological or unobtrusive observations of nonverbal reactions to another’s emotional state.
This article is from way back in ’83, but more recent studies have tended to support the main idea that the overall difference between men and women in empathy is fairly negligible when behaviour, rather than self-report, is examined.
These sorts of social influences on experimental findings are known as ‘demand characteristics‘.
The classic example is an attractive female researcher asking men about penis size, but the effects can be quite subtle and only come to light in subsequent replications of the study (if at all!).
One of my favourite studies in this area looked at the supposed tendency for people who experience ‘sensory deprivation’ to have hallucinations and suffer severe emotional and cognitive impairment.
In 1964 psychologists Martin Orne and Karl Scheibe compared two groups of participants in a sensory deprivation experiment.
One group of participants was greeted by white coated researchers standing next to emergency equipment, were asked for their medical history and given serious looking tests, were told to report any strange sensory distortions and were informed that if they wanted to stop the experiment, they had to press a panic button.
The other group was greeted informally by researchers in casual clothes, weren’t given any medical checks, and were told to report their experiences freely as they occurred. To stop the experiment, they just had to knock on the window.
The actual sensory deprivation procedure was the same for both groups, but the participants given the formal medical introduction reported greater emotional disturbance, unusual experiences and mental distress. Furthermore, they tended to do much worse on the cognitive tests given afterwards.
While this didn’t ‘disprove’ any of the unpleasant effects of sensory deprivation, it did show that they are heavily mediated by expectation which is implicitly inferred from the testing situation.
Needless to say, this can affect any type of study, so scientists are always on the look out to see if it might be responsible for new findings.