Sample size will depend on the size effect you are trying to detect. For example, you need at least 50 people per group to identify an effect size of .5 standard deviations with power of .8, assuming normality, independence, homogeneity of variance and an alpha of .05.
Dr. Paul R. Swank,
Children's Learning Institute
Professor, Department of Pediatrics, Medical School
Adjunct Professor, School of Public Health
University of Texas Health Science Center-Houston
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Jean Providence NZABONIMPA
Sent: Friday, May 11, 2012 6:06 AM
To: [hidden email]
Subject: Guidance on statistical design and data analysis
Dear all,
Sorry for my layman questions in statistics. What is the minimum sample sample size to compare two or three groups? Please give me a statistical advice on the minimum sample size. I am looking into methods effects resulting from a quasi-experimental sequential mixed methods design (qualitative and quantitative) using survey questionnaire, semi-structured interviews, and essays. I will compare data from two groups collected at two different times. One group will be subjected to two methods of data collection and another group will be only subjected to one method of data collection. Also, what statistical tests are more appropriate to detect differences in data collected? I will be quantitizing the qualitative data by presence/absence, mentioned/unmentioned, yes/no and compare with survey data. Both qualitative and quantitative tools are designed to collect data on the same concepts.
Your guidance is highly appreciated.
Thanks in advance
Jean Providence
| Free forum by Nabble | Edit this page |