issues presented when analysing data 117-118
Within-case analysis involves looking at the processes within a case that led up to, or are closely associated with, an outcome that is to be explained. For example, if your concern is to explain ethnic inequalities in educational achievement, you might study what happens in a particular school that could generate such inequalities. These could include: comparing the different ethnic backgrounds of children coming to the school; differential treatment by teachers of children from different ethnic groups; the effects of organisational forms such as streaming, banding and setting, and so on (for an example, see Troyna, 1991 and 1992; and Gomm, 1993).
Cross-case analysis requires the looking across of cases for patterns of association between the outcome to be explained and factors that it is believed might affect it. For example, if you are interested in explaining why there are social class differences in young people’s applications to high-status universities in the UK, you might look at a large number of applicants from different social classes and try to identify what features seem to have been associated with their decisions about which universities to apply to. You may find that there are significant differences between social classes in the decision-making processes involved, perhaps determining whether high-status or lower-status universities are selected (for example, differences in how much account is taken of financial cost, of where one would feel ‘comfortable’, etc). These patterns could be used to develop and test an explanation for social class differences in university applications, and beyond this in who tends to go to which type of university (see Reay et al., 2005; and Cooper et al., 2012:ch3).