Figuring out an effect size with a P value for a t-test, sample means, and sample sizes An author did not report an effect size so I am hoping you might be able to help me calculate it.
I know the Means are:
Condition 1: 326
Condition 2: 558

Sample size:
Cond1: 11
Cond2: 12

 
T-test p value: .04

 A: First, recall the form of the statistic
http://en.wikipedia.org/wiki/Student%27s_t-test#Unequal_sample_sizes.2C_equal_variance
Then, see here:
http://en.wikipedia.org/wiki/Effect_size#t-test_for_mean_difference_between_two_independent_groups
Now you need an estimate of $\sigma$. You have that $p \approx 0.04$, so we can approximate the $t$-value using t-tables (or the relevant function):
 qt(1-0.04/2,df=11+12-2)
[1] 2.189427

However, the true value might actually correspond to some rather different value of $t$ -- between 2.13 and 2.25 if the p-value was rounded (true $p$ is between 0.035 and 0.045) -- or between 2.19 and 2.33 if a conservative (rounding up) approach was used (i.e. if say the p-value was calculated as 0.0367 and then reported as being no more than 0.04).
Rearranging the t-statistic formula in terms of $Sp$ (equivalently, $S_{X_1X_2}$ at that first link) of 
$$
\frac{\bar {X}_1 - \bar{X}_2}{t\cdot \sqrt{\frac{1}{n_1}+\frac{1}{n_2}}}   
$$
Taking the $|t|$ value at 2.19 for the purpose of illustration, this implies an $Sp$ of 
$$
\frac{|326 - 558|}{2.19 \cdot \sqrt{\frac{1}{11}+\frac{1}{12}}}  \approx 254
$$
From here you should be able to use the formulas at the second link above.
