What's does "group size" mean when comparing correlations? I have a very naive questions when it comes to comparing correlations. I want to compare two correlations using the cocor package in R.
The two correlations show the relationship of two variables X and Y in two groups of subjects (A and B) who speak the different languages (a and b). Group A had 32 participants who performed a task with 310 items tapping into variable X and Y in language a. Group B had 21 participants who performed a task tapping into variable X and Y with 430 items in language b.
To compare the correlations of X and Y in group A and B I need to tell cocor the size of the group. Now, my question is: For the sake of comparing the correlations, my group size here does not concern the actual number of participants, right? Instead, it should be the number of items that each task contained for the respective group as shown here:
cocor.indep.groups(r1.jk = corA_XY, r2.hm = corB_XY, n1 = 310, n2 = 430)

I applied this logic because it could also be the case that I have a correlation in which I correlate X and Z from two different groups of participants (g1 = 31 people, g2 = 29 people) and want to compare that correlation R1 to the correlation of X and Z from a yet two other groups of participants (g3 = 30 people, g4 = 33 people). In that case, what group size would give to the respective funciton form cocor?
Any thoughts?
 A: Your question is a bit vague. If you have 2 variables X and Y ad two groups A (n1 = 32) and B (n2 = 21), then what are 310 and 430 items? Your data are 2-dimensional, aren't they?  
In fact, n1 and n2 in the cocor::cocor.indep.groups() function are the size of groups as it is written in the package's documentation. 
Here, I make an example for the dataset iris:
iris_virginica <- subset(iris, Species == "virginica")

iris_versicolor <- subset(iris, Species == "versicolor")

r1.jk <- cor(iris_virginica$Sepal.Length, iris_virginica$Sepal.Width)
r1.jk
0.4572278

r2.hm <- cor(iris_versicolor$Sepal.Length, iris_versicolor$Sepal.Width)
r2.hm
0.5259107

cocor::cocor.indep.groups(
  r1.jk = r1.jk, 
  r2.hm = r2.hm, 
  n1 = nrow(iris_virginica),
  n2 = nrow(iris_versicolor),
  conf.level = 0.95
  )

  Results of a comparison of two correlations based on independent groups

Comparison between r1.jk = 0.4572 and r2.hm = 0.5259
Difference: r1.jk - r2.hm = -0.0687
Group sizes: n1 = 50, n2 = 50
Null hypothesis: r1.jk is equal to r2.hm
Alternative hypothesis: r1.jk is not equal to r2.hm (two-sided)
Alpha: 0.05

fisher1925: Fisher's z (1925)
  z = -0.4396, p-value = 0.6603
  Null hypothesis retained

zou2007: Zou's (2007) confidence interval
  95% confidence interval for r1.jk - r2.hm: -0.3761 0.2376
  Null hypothesis retained (Interval includes 0)

If it is not clear, you can make a reproducible example by dput() and ask in StackOverflow.
