Delve is committed to training and teaching. This commitment extends beyond psychotherapy. Because we have professionals with research and statistics training and experience, we have the opportunity to share some resources with others who may benefit!
On this page, we share several worksheets, excel calculators, code snippets, and links to documents that may help students in their research endeavors. If you spot any mistakes, please let me know directly at adam@delvepsych.com. I am not the author of all of these documents, but I did author many of them.
Writing
General tools, calculators, how-tos
Links
Calculate confidence interval for indirect effects
Calculation for the Sobel Test
Multiple Imputation using Stochastic Regression
You will need to install the ‘mice’ package.
Example code:
library(mice)
data1 = read.csv('[filepath]')
tempdata = mice(data = data1,m=5, maxit=5,seed=500, methods='mice.impute.pmm')
completedData <- complete(tempdata,5)
write.csv(completedData, file = '[destinationfilepath]')
Citations supporting this approach
Buuren, S. V., & Groothuis-Oudshoorn, K. (2010). mice: Multivariate imputation by chained equations in R. Journal of statistical software, 1-68.
Brockmeier, L. L., Kromrey, J. D., & Hines, C. V. (1998). Systematically missing data and multiple regression analysis: an empirical comparison of deletion and imputation techniques. Multiple Linear Regression Viewpoints, 25, 20-39.
Dealing with Alpha-Inflation
Structural Equation Modeling
Meta-Analyses
A macro for calculating d, v, w, and Meanes
*' ---------------------------------------------------------------------------------------------------------------------------------------------------
*' Adjust D based on small sample sizes. ES refers to raw sample sizes. D refers to edjusted sample sizes.
*' ---------------------------------------------------------------------------------------------------------------------------------------------------
COMPUTE D = ES* (1- (3/ ( (4* (NT+NC) ) -9 ) ) ) .
COMPUTE V = ( (NT+NC) / (NT*NC) ) + ( (D**2) / (2* (NT+NC) ) ) .
COMPUTE w = 1/V .
EXECUTE .
*' ---------------------------------------------------------------------------------------------------------------------------------------------------
*' Run the Meanes macro below based on D and W (Don't use ES).
*' ---------------------------------------------------------------------------------------------------------------------------------------------------
INCLUDE "C:\Users\Main\Documents\Stats\MEANES.SPS" .
MEANES ES= D /W= W /PRINT=c .
*' E.g., MEANES ES = D /W = IVWEIGHT .
*' ---------------------------------------------------------------------------------------------------------------------------------------------------
*' If Homogenity test is significant, use the random variance component to adjust V and W using the syntax below
*' ---------------------------------------------------------------------------------------------------------------------------------------------------
COMPUTE VAdjusted = V+0.224 .
EXECUTE .
COMPUTE wadjusted = 1/VAdjusted .
EXECUTE .
*' ---------------------------------------------------------------------------------------------------------------------------------------------------
*' Rerun the Meanes macro using D and wadjusted
*' ---------------------------------------------------------------------------------------------------------------------------------------------------
INCLUDE "C:\Users\Main\Documents\Stats\MEANES.SPS" .
MEANES ES= D /W= wadjusted /PRINT=c .
*' E.g., MEANES ES = D /W = IVWEIGHT .