### Purpose of this repository This is a record of the raw and processed data, which is analysed in our manuscript "[Gender differences in individual variation in academic grades fail to fit expected patterns for STEM](http://dx.doi.org/10.1038/s41467-018-06292-0)". Here, we provide pre-processed and processed data as .csv files, and imputed data as .Rdata files (using m = 100 imputations). The imputed data are student age and study year, which were not always reported in the paper (but correlate strongly with source (i.e. school level) and publication year). These continuous variables are presented either raw or Z-scaled (i.e. standardised so that they have a mean of 0 and a standard deviation of 1). ### Source of data **Supplementary Table 1.xlsx** contains the list of studies included in our meta-analysis ### Meta-data **Data Description.xlsx** contains descriptions for the column names present in .csv data files, in alphabetical order ### Data files in this repository #### pre-processed File name | Description ---------|--------- datAll_20180226_sorted.csv | all data before effect size calculations datAll_20180226_preprocessed.csv | all data after effect size calculations. Please note that the Vhg column needs to be squared before analysis, as there was an error in our function (see Code repository) #### processed - data with average imputation File name | Description ---------|--------- datall_processed_avMI.csv | main dataframe with average imputed values added datall_school_processed_avMI.csv | school data with average imputed values added datall_uni_processed_avMI.csv | uni data with average imputed values added datallfm_stacked_processed_avMI.csv | stacked main data frame (females and males in one colum) with average imputed values added datallfm_school_stacked_processed_avMI.csv | stacked school data with average imputed values added datallfm_uni_stacked_processed_avMI.csv | uni data with average imputed values added compare_STEM_humanities_df.csv | data for Table S12 (comparing STEM and non-STEM together) #### imputations - data with 100 imputations File name | Description ---------------|--------- imputation_m100_corr.Rdata | main dataframe with imputed mean_age and study_year imputation_school_m100_corr.Rdata | school data with imputed mean_age and study_year imputation_uni_m100_corr.Rdata | uni data with imputed mean_age and study_year imputationfm_m100_corr.Rdata | stacked main dataframe (females and males in one column) with imputed mean_age and study_year imputationfm_school_m100_corr.Rdata | stacked school data with imputed mean_age and study_year imputationfm_uni_m100_corr.Rdata | stacked uni data with imputed mean_age and study_year imputation_scaled_m100_corr.Rdata | main dataframe with imputed z-scaled mean_age and study_year imputation_school_scaled_m100_corr.Rdata | school data with imputed z-scaled mean_age and study_year imputation_uni_scaled_m100_corr.Rdata | uni data with imputed z-scaled mean_age and study_year imputationfm_scaled_m100_corr.Rdata | stacked main dataframe (females and males in one column) with imputed z-scaled mean_age and study_year imputationfm_school_scaled_m100_corr.Rdata | stacked school data with imputed z-scaled mean_age and study_year imputationfm_uni_scaled_m100_corr.Rdata | stacked uni data with imputed z-scaled mean_age and study_year ### Location of other information #### Code Processing of the pre-processed data file, and subsequent analyses, requires a series of R markdown scripts (.Rmd files), which are presented in another component of this project. #### List of included studies Supplementary information, **Table S1**