| Type: | Package |
| Title: | Meta-Analytic Effect Size Calculation for Pre-Post Designs with Correlation Imputation |
| Version: | 1.2.0 |
| Description: | Tools for the calculation of effect sizes (standardised mean difference) and mean difference in pre-post controlled studies, including robust imputation of missing variances (standard deviation of changes) and correlations (Pearson correlation coefficient). The main function 'metacor_dual()' implements several methods for imputing missing standard deviation of changes or Pearson correlation coefficient, and generates transparent imputation reports. Designed for meta-analyses with incomplete summary statistics. For details on the methods, see Higgins et al. (2023) and Fu et al. (2013). |
| License: | MIT + file LICENSE |
| URL: | https://github.com/ikerugr/metacor, https://ikerugr.r-universe.dev/metacor |
| BugReports: | https://github.com/ikerugr/metacor/issues |
| Depends: | R (≥ 3.6) |
| Imports: | stats, officer, stringr |
| Suggests: | knitr, rmarkdown, testthat (≥ 3.0.0) |
| Config/testthat/edition: | 3 |
| Encoding: | UTF-8 |
| RoxygenNote: | 7.3.2 |
| VignetteBuilder: | knitr |
| NeedsCompilation: | no |
| Packaged: | 2025-09-25 20:05:05 UTC; Iker |
| Author: | Iker J. Bautista [aut, cre], Saul M. Rodriguez [ctb] |
| Maintainer: | Iker J. Bautista <ikerugr@gmail.com> |
| Repository: | CRAN |
| Date/Publication: | 2025-09-26 07:00:03 UTC |
Effect Sizes and Imputation for Meta-Analysis of Pre-Post Studies and Pre-Post intervention and control groups studies (metacor_dual)
Description
Calculates effect sizes (i.e., SMDpre, SMDchange, ScMDpooled, ScMDpre) and allows for various imputation methods (i.e., none, cv, direct, mean) for missing SDdiff and correlation coefficients in pre-post meta-analyses, with or without a control group. Generates a detailed imputation report in Word format.
Usage
metacor_dual(
df,
digits = NULL,
add_to_df = TRUE,
method = "both",
apply_hedges = TRUE,
SMD_method = "SMDpre",
MeanDifferences = FALSE,
impute_method = "none",
verbose = TRUE,
report_imputations = FALSE,
custom_sd_diff_int = NULL,
custom_sd_diff_con = NULL,
single_group = FALSE
)
Arguments
df |
Data frame with the necessary columns for intervention and (optionally) control groups. |
digits |
Number of decimal places to round results (default: NULL). |
add_to_df |
Logical. If TRUE, results are added to the original data frame. |
method |
Method for SDdiff calculation (i.e., 'p_value', 'ci', 'both'). |
apply_hedges |
Logical. Apply Hedges' g correction? (default: TRUE) |
SMD_method |
Method for effect size (i.e., 'SMDpre', 'SMDchange', 'ScMDpooled', 'ScMDpre'). |
MeanDifferences |
Logical. Calculate mean differences and variances? (default: FALSE) |
impute_method |
Imputation method for missing SDdiff (i.e., 'none', 'direct', 'mean', 'cv'). |
verbose |
Logical. Print messages during processing? (default: TRUE) |
report_imputations |
Logical. Generate Word imputation report? (default: FALSE) |
custom_sd_diff_int |
List with elements 'row' and 'value' for manual sd_diff_int values. |
custom_sd_diff_con |
List with elements 'row' and 'value' for manual sd_diff_con values. |
single_group |
Logical. Is the design single-group only? (default: FALSE) |
Value
Data frame with calculated variables. Optionally, a Word report ('imputation_report.docx') is generated.
References
Higgins, J. P. T., Thomas, J., Chandler, J., Cumpston, M., Li, T., Page, M. J., & Welch, V. A. (Eds.). (2023). Cochrane handbook for systematic reviews of interventions (Version 6.3). Cochrane. https://training.cochrane.org/handbook Fu, R., Vandermeer, B.W., Shamliyan, T.A., ONeil, M.E., Yazdi, F., Fox, S.H., & Morton, S.C. (2013). Handling Continuous Outcomes in Quantitative Synthesis. Methods Guide for Comparative Effectiveness Reviews. AHRQ Publication No. 13-EHC103-EF. https://effectivehealthcare.ahrq.gov/reports/final.cfm
Examples
df <- data.frame(
study_name = c("Study1", "Study2", "Study3", "Study4",
"Study5", "Study6", "Study7", "Study8", "Study9"),
p_value_Int = c(1.038814e-07, NA, NA, NA, NA, 2.100000e-02, NA, NA, NA),
n_Int = c(10, 10, 10, 10, 15, 15, 10, 10, 10),
meanPre_Int = c(8.17, 10.09, 10.18, 9.85, 9.51, 7.70, 10.00, 11.53, 11.20),
meanPost_Int = c(10.12, 12.50, 12.56, 10.41, 10.88, 9.20, 10.80, 13.42, 12.00),
sd_pre_Int = c(1.83, 0.67, 0.66, 0.90, 0.62, 0.90, 0.70, 0.60, 1.90),
sd_post_Int = c(1.85, 0.72, 0.97, 0.67, 0.76, 1.10, 0.70, 0.80, 1.80),
upperCI_Int = c(NA, NA, NA, NA, NA, NA, NA, NA, NA),
lowerCI_Int = c(NA, NA, NA, NA, NA, NA, NA, NA, NA)
)
result <- metacor_dual(df)
print(result)