It is not possible to directly calculate the unreported sample size from the given means and standard deviations of two study groups. This is because the sample size is one of the variables used to calculate means and standard deviations, along with the actual data values.
However, it may be possible to estimate the sample size based on the reported means and standard deviations, as well as some additional assumptions. Here is one possible approach:
Assume that the sample sizes for both groups are equal (i.e., N1 = N2).
Calculate the pooled standard deviation (s_pooled) using the formula: s_pooled = sqrt(((N1-1)*s1^2 + (N2-1)*s2^2) / (N1+N2-2)) where s1 and s2 are the standard deviations of the two groups, and N1 and N2 are the assumed equal sample sizes.
Calculate the effect size (d) using the formula: d = (x1 - x2) / s_pooled where x1 and x2 are the means of the two groups.
Use a table or calculator to look up the corresponding sample size for the calculated effect size and desired level of power and significance.
Keep in mind that this approach involves several assumptions, and the estimated sample size may not be accurate in practice. It's always best to consult with a statistician or use a reliable statistical software package to ensure accurate sample size calculations.
I want to compare different data sets but cannot access raw data. Only the averages and standard deviations are available. How can I do tests such as Duncan or Tukey?
Amir Heydari I think, if you want to compare different data (two groups) that have mean value and standard deviation, it is possible to use Meta-Analysis.
If you have the means and standard deviations of two study groups and want to estimate the unreported sample size, you can use the following formula:
n = (tα/2 + tβ)^2 * [(s1^2 + s2^2)/(μ1 - μ2)^2]
where:
n is the estimated sample size for each group
tα/2 is the critical value of the t-distribution for a two-tailed test with α/2 significance level and (n1 + n2 - 2) degrees of freedom, where α is the level of significance (e.g. 0.05 for a 95% confidence interval)
tβ is the critical value of the t-distribution for the power of 1-β (e.g. 0.8 for 80% power) and (n1 + n2 - 2) degrees of freedom
s1 and s2 are the standard deviations of the two groups
μ1 and μ2 are the means of the two groups
Note that this formula assumes that the sample sizes of the two groups are equal. If they are not equal, you can use the harmonic mean of the sample sizes instead of (n1 + n2)/2.
It's worth noting that this formula provides an estimate of the required sample size based on the assumption that the means and standard deviations accurately represent the population. In reality, there may be other factors that affect the sample size calculation, such as the desired level of precision or the complexity of the study design.