If you interested to calculate the sample size. You can refer at this link:

http://www.danielsoper.com/

I am using GPower, and it gives me the information needed to plan for a clinical trial.

Prof. Muayyad,

Try this alternative approach. I attached the calculator in excel here. Your range results of sample size should be between 30 - 35. Link to the article below.

Let me know if it is useful.

Cheers.

https://www.researchgate.net/profile/Paul_Louangrath/contributions

If you interested to calculate the sample size. You can refer at this link:

http://www.danielsoper.com/

Muayyad,

You don't mention your analysis method, which is critical for this.

For simple models that Cohen did simulations for, they're fine. Power depends on sample size, alpha, effect size, and the model. So if you're using a model he didn't, or G-Power doesn't, cover, then you need to look in different directions, up to and including simulation-based power analysis to handle arbitrary models.

Pat

Good points Patrick.

Moreover, in life science we should also consider the biological relevance.

Roberto, agreed. The Cohen definitions for "small, medium, large" effect sizes are much less useful than field- and context-specific values. I'm given to understand Cohen was reluctant to give them at all for fear rules of thumb would become standards.

Prescient.

Dear Patrick,

Thank you for your response, I know that I did not mention the analysis method. Therefore, I continued by saying Cohen book {as a whole} vs. G.pwower, or any other available sources. Unless, based on the type of anlaysis one can prefere a reference above the other!

I am thankful to all scholars who shared their opinions, and I think there are nice ideas and other free avialbel sources which could be helpful for researchers and Graduate students.

Muayyad,

My point was that, depending on the analysis, it might be that neither is useful.

Pat

We are clearly here not talking about non-parametric tests.

To make things more concrete, lets say in runing ANOVA, MANOVA, Regression, Correlations, ttest..etc. can we favour one reference in calculating the sample size?!

I would like to thank Prof. Ahmad for asking this useful question and appreciate all scholar for their functional responses. It is very critical in research indeed.

In my opinion, the problem with G*Power is calculating small sample size that  is not profitable in SEM (by AMOS) - it is often less than 200 or even 100. 

Cohen table (1988) is referred in many papers now, but I do not know how much it is accurate! The sample size it gives us is very near to what is derived from Cochran's sample size table! 

Through searching for sample size calculators, I found STATA as another way to estimate sample size. It has been referred in a few articles. Please kindly read more about it in the following links:

http://www.stata.com/stata13/power-and-sample-size/

http://www.stata.com/manuals13/pss.pdf

Although I watched some related videos in youtube and read some texts about it, I do not have enough knowledge in this regard. Just mentioned as an option.

Hope it is useful.

Best regards,

I'd like to thank Prof. Muayyed for asking this very important question

You can refer to this useful link for free sample size calculator: http://www.danielsoper.com/statcalc3/category.aspx?id=19

and

http://www.surveysystem.com/sscalc.htm

Regards

Dear Dr. Alloubani

Regarding to the first link introduced by you and already by Mr. Wan Mohamad, under the tab of "A-priori Sample Size Calculator for Structural Equation Models", would you please let me know how this calculator should be used in the cases that we have different models? (e.g. I draw 8 separate models in my thesis base on my hypotheses)

Further, would you please let me know whether the observed variables linked to the latent ones should be counted as the number of the observed variables or not? If yes, the recommended minimum sample size will be so high!

I hope I asked my questions clearly!

Thank you for your valuable time,

Regards,

I am here trying join all the links mentioned in this roster for calculating sample size online, hoping to will be easy for all to get access to the site which one feel comfort with. The G. power remains another option, but it needs to download its free software first. Good luck to all.

http://www.danielsoper.com/

http://www.stata.com/stata13/power-and-sample-size/

http://www.stata.com/manuals13/pss.pdf

http://www.danielsoper.com/statcalc3/category.aspx?id=19

http://www.surveysystem.com/sscalc.htm

Thanks a lot.

I would add another useful hub for “Power and Sample Size Programs”:

http://www.epibiostat.ucsf.edu/biostat/sampsize.html

Thank you Dr. Roberto and thanks extend to all colleagues who participated with their valuable opinions

Dear RG colleague,

Determining the sample sizes involve resource and statistical issues. Usually, researchers regard 100 participants as the minimum sample size when the population is large. However, In most studies the sample size is determined effectively by two factors: (1) the nature of data analysis proposed and (2) estimated response rate.

For example, if you plan to use a linear regression a sample size of 50+ 8K is required, where K is the number of predictors. Some researchers believes it is desirable to have at least 10 respondents for each item being tested in a factor analysis, Further, up to 300 responses is not unusual for Likert scale development according to other researchers.

Another method of calculating the required sample size is using the Power and Sample size program (www.power-analysis.com).

Regards,

Sekaran (20013) wrote:

"Roscoe (1975) proposes the following rules of thumb for determining sample size:

1. Sample sizes larger than 30 and less than 500 are appropriate for most research.

2. Where samples are to be broken into sub-samples; (male/females, juniors/seniors, etc.), a minimum sample size of 30 for each category is necessary.

3. In multivariate research (including multiple regression analyses), the sample size should be several times (preferably 10 times or more) as large as the number of variables in the study.

4. For simple experimental research with tight experimental controls (matched pairs, etc.), successful research is possible with samples as small as 10 to 20 in size."

Reference

Sekaran, U., 2003. Research methods for business: A skill building approach. John Wiley & Sons.