GPower is good for basic applications. The r-package pwr is also pretty useful.
You can do Monte Carlo Simulations, it should work with the r-packages simsem or lavaan.
Here is some material:
Prajapati, B., Dunne, M., & Armstrong, R. (2010). Sample size estimation and statistical power analyses. Optometry Today, July 2010.
Thoemmes, F., MacKinnon, D.P., & Reiser, M.R. (2010). Power Analysis for Complex Mediational Designs Using Monte Carlo Methods. Structural Equation Modeling, 17(3), 510-534.
Muthen, L.K. & Muthen B.O. (2002). How to use a Monte Carlo study to decide on sample size and determine power. Structural Equation Modeling, 4, 599{620.
2 Recommendations
1st Jan, 2015
Robert Milewski
Medical University of Bialystok
Power analysis is implemented in almost every statistical package, e.g. Stata, Statistica.
1st Jan, 2015
Ronán Michael Conroy
Royal College of Surgeons in Ireland
What kind of sample size problem do you have? We could be more specific if you were more specific…
1st Jan, 2015
Sharon E Phillips
Vanderbilt University
This is a link to a program for computing sample size:
Minitab has a complete set of tools for statistical analysis.
1st Jan, 2015
Georgios Kotsakis
University of Texas Health Science Center at San Antonio
Hi Vaidya,
Almost any statistical package you purchase will have functions for sample size estimation. Someone already recommended R statistical package. The nice thing with R is that it comes for free and coding is very straight-forward. All the packages come for free as well. The "pwr" is the easiest one to use for rudimentary power analysis in R, although there are a couple of others available.
G*power is really nice if you want to add graphics. It gives you a nice representation of how your 1-b increases with the changes in your sample size.
As with all things in statistics, it's a good thing to know what you are doing and to make sure that it is meaningful. It is really easy to get a false sense of security with your power analysis, even for really small sample sizes, if you "goof" the effect size estimation. Russ Lenth has a really good page about the basics of sample size estimation and links to online calculators that you may find useful:
i hope you find the following attached program file (Gpower.rar) useful.
n.b. this program has a reference included within to be used in your article.
GPower.rar9.54 MB
4 Recommendations
7th Jul, 2018
Abolfazl Ghoodjani
McGill University
A long time has come to answer this question. However, the best software to estimate the sample size is Minitab.
Minitab 17 or 18 / Stat / Power and Sample Size
1 Recommendation
8th Aug, 2018
Eimear Keyes
University College Cork
I am a statistician at Statsols.
We make nQuery, a straightforward sample size and power calculation software, used by 90% of organisations with FDA approved clinical trials. The software covers hundreds of test designs.
https://www.statsols.com/nquery
We also hosted a webinar recently where you can see sample size calculation in nQuery in action - https://www.statsols.com/webinar/the-5-essential-steps-for-sample-size-determination
12th Dec, 2018
Hrishikesh Kulkarni
Cytel
Hi Vaidya,
I am a statistician at Cytel. We develop East software for power and sample size calculation for simple fixed and group sequential trials. Then we have an advanced module (ADAPT) for sample size re-estimation. Refer to this link to understand the modular structure of East https://www.cytel.com/software/east#section_modularformat
You can pick and choose modules as per your needs. Just click on these boxes to learn more.
Here is our YouTube channels with some useful videos for you- https://www.youtube.com/user/CytelVideos/playlists
Hope this helps!
Cheers,
Hrishi
12th Dec, 2018
Miland Joshi
University of Central Lancashire
If you want to avoid spending many thousands of rupees or equivalent, I would extract formulae from downloadable internet articles or papers or library books, put them in Excel spreadsheets, and there's your software!
The book by David Machin (ISBN 9781118874943) is supposed to have associated software. However I would make sure you can access it before spending a large amount of money (or try and contact the publisher or author).
The book by Steven Julious (ISBN 9781584887393) has formulae which, if you can borrow the book from a library, can be entered into Excel.