The dataset is a subset of data derived from an experiment in social psychology

The dataset is a subset of data derived from an experiment in social psychology titled ‘Shame, Blame and Contamination: Mental Illness Stigma and the Family’, and the example shows how to test whether particular kinds of health conditions induce different levels of family stigma.
The data come from an experiment where the interest is in whether particular kinds of health condition induce different levels of family stigma. We analyse data from an experiment in which participants were randomly assigned to vignettes where a family member is described as having one of three health conditions: schizophrenia, drug dependence and emphysema, and subsequently asked whether they felt ashamed of the family member)s illness.
This example therefore addresses the following research question:
Does shame about family members depend on the type of illness suffered?
Stated in the form of a null hypothesis:
H0 = There is no difference in family shame regardless of illness suffered.
This example uses a subset of data from TESS2004. This extract includes 960 respondents. The two variables we examine are:
A vignette describing a family member suffering from one of three health conditions: schizophrenia, drug dependency and emphysema (condition).
A question tapping the extent to which the respondent would feel ashamed if a family member had the illness (shame).
The first variable, condition, is coded 1 = “Schizophrenia”, 2 = “Drug dependence”, 3 = “Emphysema”. These values denote which health condition was mentioned in a vignette shown to respondents (randomly allocated). There were several versions of each of these three, but these have been pooled in the current dataset. An example is:
“Joan is the wife of Frank, a 30 year-old-man with schizophrenia. Frank lives with his family in an apartment and works as a clerk at a nearby store. Frank has been hospitalized several times because of his illness. The illness has disrupted his life significantly.”
The shame variable invites a response to the statement: “Joan should feel ashamed about Frank)s illness”. Responses are on a 7-point scale: strongly disagree, disagree, somewhat disagree, neither agree nor disagree, somewhat agree, agree, strongly agree.
We treat the shame variable as continuous for the purposes of this example. The condition variable is categorical, representing the three experimental groups, making the one-way ANOVA appropriate for this example.