What's the hypothesis that you want to test?
On 09/06/07, Davies, Nicola <[log in to unmask]> wrote:
> Hi All,
> I am using a programme called gpower in order to try and determine the sample size needed in my research. My research will involve one group of patients (breast cancer patients having chemotherapy) and I will be measuring each patient's quality of life at two points in time. I will therefore have two scores for each patients. Now in order to work out the power on this programme, do I choose 'means: difference between two dependent means (matched pairs)' or 'means: difference from constant (one sample case)'. Alternatively, there is also an option for 'correlation: point biserial model'. I am so confused!!!
> Kind Regards,
> Nicola Davies,
> BSc; MSc Comm.; PhD Candidate
> Liasion Officer for the DHP Postgraduate Subcommittee
Learning statistics blog: www.jeremymiles.co.uk/learningstats
Psychology Research Methods Wiki: www.researchmethodsinpsychology.com