everyone – not everyone wants to take
part in research and even if they did, the time and cost commitment would
be enormous. So instead we work with a manageable group of research
volunteers. They provide their data and we use this to estimate the values that
would be measured in a population.
Research studies, and particularly randomised controlled trials, often use a
special calculation called a power calculation to decide how many people
we need to recruit to a study. A study must have sufficient power to infer
the correct result. The minimum power level which is
of statistical significance and power calculations. In qualitative research,
arguments are instead made for commonality or typicality (Fossey et al.,
2002). Within the studies included in EQUIP, commonality was inferred by the
fact participants were recruited by virtue of having certain characteristics (e.g.
service users under the care of a community mental health team).
BEE (RESEARCH) PRINT.indd 116
Given the interpretative nature of qualitative analysis, credibility refers to the
interpretations made about the data (Green and Thorogood
signiﬁcant eﬀect of the intervention. For example, this might be a change on
a depression scale that moved a person from severe to moderate depression,
or from moderate to mild depression. Once this has been agreed, researchers
conduct a statistical test called a power calculation. This calculates the
minimal sample size required to detect a significant difference between the
treatment and control groups.
The final key feature of an RCT yet to
be covered is the need for researchers
to conduct an intention-to-treat
analysis. This refers to the need to