Is there a sample calculator for quantitative dissertations? I am, planning to do my dissertation on the role of critical thinking instructional strategies on ESL students' academic achievement, but I am not sure how big my sample should be.
What you would like to do is conduct a power analysis. If I understand correctly, you're planning to randomly assign ESL students to one of two critical thinking instructional strategy conditions (or one experimental and one control). Below is a link to an online stats calculator for power analysis. You'll be conducting an independent samples t-test (so choose two groups > t-test on group means). Based on previous studies on interventions like your, make up numbers for the group means and standard deviation. What really matters is the difference in means and the standard deviation. This calculator lets you specify many things, but I'd stick with the defaults because they're the conventions in social science. Stick with alpha of .05 (p-value for significance), power of .80 (80% chance of finding an effect if one exists), and have an equal number of participants in each group. Hope this helps Ayad! ~ Kevin
http://biomath.info/power/
The attached paper by Cohen 1992 gives guidelines on sample sizing in relation to effect and power sizes
In line with what Kevin and Silbrun suggested, you could use the freely available G*Power program to determine this. I'm attaching a PowerPoint tutorial for getting started with it.
- Badges
- Science topic
- Similar topics
- Medical Science Nursing
- Dissertations