--- title: "Data Analysis and Visualization <br> Paper 2" author: Gagan Atreya date: today format: html: toc: true toc-location: left toc-depth: 4 theme: lumenR fontsize: large code-fold: true code-tools: true code-summary: "Display code" code-overflow: wrap editor: markdown: wrap: 72 --- # **Section 1. Data importation and cleaning** ```{r, error = F, message=F, warning = F} ## clear environment: rm (list= ls ())options (digits = 2 )## Correct asterisk pattern for stargazer tables: <- "<em>*p<0.05; **p<0.01; ***p<0.001</em>" ## Install "pacman" package if not installed # (remove the # symbol from the line below): # install.packages("pacman") ## Load R packages: :: p_load (data.table, tidyverse, haven, labelled, vtable, ## Import datasets. These are cleaned datasets with variables created and named in the individual country R scripts: # Gambia: <- fread ("~/Dropbox/2023_Gagan/Paper02/data/dsgmb.csv" )# Pakistan: <- fread ("~/Dropbox/2023_Gagan/Paper02/data/dspak.csv" )# Tanzania: <- fread ("~/Dropbox/2023_Gagan/Paper02/data/dstza.csv" )# Uganda: <- fread ("~/Dropbox/2023_Gagan/Paper02/data/dsuga.csv" )# Combine into one dataset: <- rbind (dsgmb, dspak, dstza, dsuga)## Clean demographic variables: ## gender: $ gender <- ifelse (ds$ gender == "Male" , "Male" ,ifelse (ds$ gender == "Female" , "Female" , NA ))## ses: $ ses <- ifelse (ds$ ses == "" , NA , ds$ ses)## Religion: $ religion <- ifelse (ds$ religion == "" , NA , ds$ religion)$ religion <- ifelse (ds$ religion == "Christian (Catholic)" , "Christian: Catholic" , ifelse (ds$ religion == "Christian (Protestant)" , "Christian: Protestant" ,ifelse (ds$ religion == "Muslim (Shia)" , "Muslim: Shia" ,ifelse (ds$ religion == "Muslim (Sunni)" , "Muslim: Sunni" , ds$ religion))))## Marital status $ married <- ifelse (ds$ married == "Not married" , "Unmarried" , ds$ married)$ married <- ifelse (ds$ married == "" , NA , ds$ married)## Separate into individual country datasets: <- ds[ds$ country == "Gambia" , ]<- ds[ds$ country == "Pakistan" , ]<- ds[ds$ country == "Tanzania" , ]<- ds[ds$ country == "Uganda" , ]``` ### Missing data: Gambia ```{r, error = F, message=F, warning = F} <- map (dsgmb, ~ mean (is.na (.))* 100 ) <- as.data.frame (missing_ds)<- as_tibble (cbind (variable = names (missing_ds), t (missing_ds)))$ percent_missing <- missing_ds$ V2$ percent_missing <- as.numeric (missing_ds$ percent_missing)<- select (missing_ds,- c (2 ))<- as.data.frame (missing_ds)$ percent_missing <- as.numeric (missing_ds$ percent_missing)<- as.data.table (missing_ds)<- missing_ds[1 : 31 ,]<- missing_ds[32 : 61 ,]<- missing_ds_01 %>% lollipop_chart (x = variable,y = percent_missing,line_color = "black" ,point_color = "black" )+ labs (y = "% missing data" ,x = "" ,title = "Gambia" )+ theme_bw ()<- missing_ds_02 %>% lollipop_chart (x = variable,y = percent_missing,line_color = "black" ,point_color = "black" )+ labs (y = "% missing data" ,x = "" ,title = "" )+ theme_bw ()| lp02``` ### Missing data: Pakistan ```{r, error = F, message=F, warning = F} <- map (dspak, ~ mean (is.na (.))* 100 ) <- as.data.frame (missing_ds)<- as_tibble (cbind (variable = names (missing_ds), t (missing_ds)))$ percent_missing <- missing_ds$ V2$ percent_missing <- as.numeric (missing_ds$ percent_missing)<- select (missing_ds,- c (2 ))<- as.data.frame (missing_ds)$ percent_missing <- as.numeric (missing_ds$ percent_missing)<- as.data.table (missing_ds)<- missing_ds[1 : 31 ,]<- missing_ds[32 : 61 ,]<- missing_ds_01 %>% lollipop_chart (x = variable,y = percent_missing,line_color = "black" ,point_color = "black" )+ labs (y = "% missing data" ,x = "" ,title = "Pakistan" )+ theme_bw ()<- missing_ds_02 %>% lollipop_chart (x = variable,y = percent_missing,line_color = "black" ,point_color = "black" )+ labs (y = "% missing data" ,x = "" ,title = "" )+ theme_bw ()| lp02``` ### Missing data: Tanzania ```{r, error = F, message=F, warning = F} <- map (dstza, ~ mean (is.na (.))* 100 ) <- as.data.frame (missing_ds)<- as_tibble (cbind (variable = names (missing_ds), t (missing_ds)))$ percent_missing <- missing_ds$ V2$ percent_missing <- as.numeric (missing_ds$ percent_missing)<- select (missing_ds,- c (2 ))<- as.data.frame (missing_ds)$ percent_missing <- as.numeric (missing_ds$ percent_missing)<- as.data.table (missing_ds)<- missing_ds[1 : 31 ,]<- missing_ds[32 : 61 ,]<- missing_ds_01 %>% lollipop_chart (x = variable,y = percent_missing,line_color = "black" ,point_color = "black" )+ labs (y = "% missing data" ,x = "" ,title = "Tanzania" )+ theme_bw ()<- missing_ds_02 %>% lollipop_chart (x = variable,y = percent_missing,line_color = "black" ,point_color = "black" )+ labs (y = "% missing data" ,x = "" ,title = "" )+ theme_bw ()| lp02``` ### Missing data: Uganda ```{r, error = F, message=F, warning = F} <- map (dsuga, ~ mean (is.na (.))* 100 ) <- as.data.frame (missing_ds)<- as_tibble (cbind (variable = names (missing_ds), t (missing_ds)))$ percent_missing <- missing_ds$ V2$ percent_missing <- as.numeric (missing_ds$ percent_missing)<- select (missing_ds,- c (2 ))<- as.data.frame (missing_ds)$ percent_missing <- as.numeric (missing_ds$ percent_missing)<- as.data.table (missing_ds)<- missing_ds[1 : 31 ,]<- missing_ds[32 : 61 ,]<- missing_ds_01 %>% lollipop_chart (x = variable,y = percent_missing,line_color = "black" ,point_color = "black" )+ labs (y = "% missing data" ,x = "" ,title = "Uganda" )+ theme_bw ()<- missing_ds_02 %>% lollipop_chart (x = variable,y = percent_missing,line_color = "black" ,point_color = "black" )+ labs (y = "% missing data" ,x = "" ,title = "" )+ theme_bw ()| lp02``` ### Missing data: full dataset ```{r, error = F, message=F, warning = F} <- map (ds, ~ mean (is.na (.))* 100 ) <- as.data.frame (missing_ds)<- as_tibble (cbind (variable = names (missing_ds), t (missing_ds)))$ percent_missing <- missing_ds$ V2$ percent_missing <- as.numeric (missing_ds$ percent_missing)<- select (missing_ds,- c (2 ))<- as.data.frame (missing_ds)$ percent_missing <- as.numeric (missing_ds$ percent_missing)<- as.data.table (missing_ds)<- missing_ds[1 : 31 ,]<- missing_ds[32 : 61 ,]<- missing_ds_01 %>% lollipop_chart (x = variable,y = percent_missing,line_color = "black" ,point_color = "black" )+ labs (y = "% missing data" ,x = "" ,title = "Full dataset" )+ theme_bw ()<- missing_ds_02 %>% lollipop_chart (x = variable,y = percent_missing,line_color = "black" ,point_color = "black" )+ labs (y = "% missing data" ,x = "" ,title = "" )+ theme_bw ()| lp02``` # **Section 2: Demographics** - Age- Gender- Socio-economic status- Religious affiliation- Education **(NEEDS MAJOR WORK)**- Marital status- Ethnicity## Variable: Sample size by Country ```{r, error = F, message = F, warning = F} <- table (ds$ country)## Table of user language by country: ## Sample size by country: <- ds %>% # drop_na(Country) %>% lollipop_chart (x = country,line_color = "black" ,point_color = "black" )+ labs (y = "Frequency" ,x = "" ,title = "Sample size by country" )+ theme_bw ()``` ## Variable: Age ```{r, error = F, message = F, warning = F} summary (ds$ age)%>% drop_na (age)%>% ggplot (aes (x = age))+ geom_histogram (color = "black" ,fill = "gray" ,bins = 50 )+ geom_textvline (label = "Mean = 37.00" , xintercept = 37.00 , vjust = 1.1 , lwd = 1.05 , linetype = 2 )+ labs (x = "Age" , y = "Frequency" , title = "Age distribution (full sample)" )+ theme_bw ()%>% drop_na (age)%>% ggplot (aes (x = age))+ geom_histogram (color = "black" ,fill = "gray" ,bins = 50 )+ labs (x = "Age" , y = "Frequency" , title = "Age distribution by country" )+ facet_wrap (~ country, nrow = 2 )+ theme_bw ()``` ## Variable: Gender ```{r, error = F, message = F, warning = F} <- ds %>% drop_na (gender, age) %>% lollipop_chart (x = gender,line_color = "black" ,point_color = "black" )+ labs (y = "Frequency" ,x = "" ,title = "Gender distribution (full sample)" )+ # facet_wrap(~Country, nrow = 2)+ theme_bw ()<- ds %>% drop_na (gender, age) %>% ggplot (aes (y = age, x = gender))+ geom_boxplot (fill = "grey" )+ labs (y = "Age" ,x = "" ,title = "Age and gender distribution by country" )+ facet_wrap (~ country, nrow = 2 )+ coord_flip ()+ theme_bw ()``` ## Variable: Socio-economic status ```{r, error = F, message = F, warning = F} %>% drop_na (ses) %>% lollipop_chart (x = ses,line_color = "black" ,point_color = "black" )+ labs (y = "Frequency" ,x = "" ,title = "Socioeconomic status (full sample)" )+ theme_bw ()<- dsgmb %>% drop_na (ses) %>% lollipop_chart (x = ses,line_color = "black" ,point_color = "black" )+ labs (y = "Frequency" ,x = "" ,title = "Gambia" )+ theme_bw ()<- dspak %>% drop_na (ses) %>% lollipop_chart (x = ses,line_color = "black" ,point_color = "black" )+ labs (y = "Frequency" ,x = "" ,title = "Pakistan" )+ theme_bw ()<- dstza %>% drop_na (ses) %>% lollipop_chart (x = ses,line_color = "black" ,point_color = "black" )+ labs (y = "Frequency" ,x = "" ,title = "Tanzania" )+ theme_bw ()<- dsuga %>% drop_na (ses) %>% lollipop_chart (x = ses,line_color = "black" ,point_color = "black" )+ labs (y = "Frequency" ,x = "" ,title = "Uganda" )+ theme_bw ()## All four plots together: <- (sesgmb | sespak) / (sestza | sesuga) + plot_annotation ("Socio-economic status by country" )``` ## Variable: Marital status ```{r, error = F, message = F, warning = F} %>% drop_na (married) %>% lollipop_chart (x = married,line_color = "black" ,point_color = "black" )+ labs (y = "Frequency" ,x = "" ,title = "Marital status (full sample)" )+ theme_bw ()<- dsgmb %>% drop_na (married) %>% lollipop_chart (x = married,line_color = "black" ,point_color = "black" )+ labs (y = "Frequency" ,x = "" ,title = "Gambia" )+ theme_bw ()<- dspak %>% drop_na (married) %>% lollipop_chart (x = married,line_color = "black" ,point_color = "black" )+ labs (y = "Frequency" ,x = "" ,title = "Pakistan" )+ theme_bw ()<- dstza %>% drop_na (married) %>% lollipop_chart (x = married,line_color = "black" ,point_color = "black" )+ labs (y = "Frequency" ,x = "" ,title = "Tanzania" )+ theme_bw ()<- dsuga %>% drop_na (married) %>% lollipop_chart (x = married,line_color = "black" ,point_color = "black" )+ labs (y = "Frequency" ,x = "" ,title = "Uganda" )+ theme_bw ()## All four plots together: <- (maritalgmb | maritalpak) / (maritaltza | maritaluga) + plot_annotation ("Marital status by country" )``` ## Variable: Religious affiliation ```{r, error = F, message = F, warning = F} <- ds %>% drop_na (religion) %>% lollipop_chart (x = religion,line_color = "black" ,point_color = "black" )+ labs (y = "Frequency" ,x = "" ,title = "Religious distribution (full sample)" )+ theme_bw ()<- dsgmb %>% drop_na (religion) %>% lollipop_chart (x = religion,line_color = "black" ,point_color = "black" )+ labs (y = "Frequency" ,x = "" ,title = "Gambia" )+ theme_bw ()<- dspak %>% drop_na (religion) %>% lollipop_chart (x = religion,line_color = "black" ,point_color = "black" )+ labs (y = "Frequency" ,x = "" ,title = "Pakistan" )+ theme_bw ()<- dstza %>% drop_na (religion) %>% lollipop_chart (x = religion,line_color = "black" ,point_color = "black" )+ labs (y = "Frequency" ,x = "" ,title = "Tanzania" )+ theme_bw ()<- dsuga %>% drop_na (religion) %>% lollipop_chart (x = religion,line_color = "black" ,point_color = "black" )+ labs (y = "Frequency" ,x = "" ,title = "Uganda" )+ theme_bw ()## All four plots together: <- (religiongmb | religionpak) / (religiontza | religionuga) + plot_annotation ("Religious affiliation by country" )``` ## Variable: Education ```{r, error = F, message = F, warning = F} # table(dsgmb$education) # table(dspak$education) # table(dstza$education) # table(dsuga$education) ``` ## Variable: Ethnicity ```{r, error = F, message = F, warning = F} ## Gambia: <- as.data.frame (table (dsgmb$ ethnicity))$ Var1 <- as.character (eth$ Var1)$ ethnicity <- ifelse (eth$ Freq < 2 , "Other" , eth$ Var1)<- as.list (eth$ ethnicity)<- dsgmb[, c ("ethnicity" )]$ ethnicity <- ifelse (dsgmb2$ ethnicity %in% l1, dsgmb2$ ethnicity, "Other" )$ ethnicity <- ifelse (dsgmb2$ ethnicity == "Serere" , "Serer" ,ifelse (dsgmb2$ ethnicity == "" , NA , dsgmb2$ ethnicity))<- dsgmb2 %>% drop_na (ethnicity) %>% lollipop_chart (x = ethnicity,line_color = "black" ,point_color = "black" )+ labs (y = "Frequency" ,x = "" ,title = "Gambia" )+ theme_bw ()## Pakistan: <- as.data.frame (table (dspak$ ethnicity))$ Var1 <- as.character (eth$ Var1)$ ethnicity <- ifelse (eth$ Freq < 7 , "Other" , eth$ Var1)<- as.list (eth$ ethnicity)<- dspak[, c ("ethnicity" )]$ ethnicity <- ifelse (dspak2$ ethnicity %in% l1, dspak2$ ethnicity, "Other" )$ ethnicity <- ifelse (dspak2$ ethnicity == "Urduspeaking" , "Urdu speaking" ,ifelse (dspak2$ ethnicity == "" , NA , dspak2$ ethnicity))<- dspak2 %>% drop_na (ethnicity) %>% lollipop_chart (x = ethnicity,line_color = "black" ,point_color = "black" )+ labs (y = "Frequency" ,x = "" ,title = "Pakistan" )+ theme_bw ()## Tanzania: <- as.data.frame (table (dstza$ ethnicity))$ Var1 <- as.character (eth$ Var1)$ ethnicity <- ifelse (eth$ Freq < 7 , "Other" , eth$ Var1)<- as.list (eth$ ethnicity)<- dstza[, c ("ethnicity" )]$ ethnicity <- ifelse (dstza2$ ethnicity %in% l1, dstza2$ ethnicity, "Other" )$ ethnicity <- ifelse (dstza2$ ethnicity == "Serere" , "Serer" ,ifelse (dstza2$ ethnicity == "" , NA , dstza2$ ethnicity))<- dstza2 %>% drop_na (ethnicity) %>% lollipop_chart (x = ethnicity,line_color = "black" ,point_color = "black" )+ labs (y = "Frequency" ,x = "" ,title = "Tanzania" )+ theme_bw ()## Uganda: <- as.data.frame (table (dsuga$ ethnicity))$ Var1 <- as.character (eth$ Var1)$ ethnicity <- ifelse (eth$ Freq < 7 , "Other" , eth$ Var1)<- as.list (eth$ ethnicity)<- dsuga[, c ("ethnicity" )]$ ethnicity <- ifelse (dsuga2$ ethnicity %in% l1, dsuga2$ ethnicity, "Other" )$ ethnicity <- ifelse (dsuga2$ ethnicity == "japhadollah" , "Japhadollah" ,ifelse (dsuga2$ ethnicity == "Atesot" , "Iteso" ,ifelse (dsuga2$ ethnicity == "Etesot" , "Iteso" ,ifelse (dsuga2$ ethnicity == "Itesot" , "Iteso" ,ifelse (dsuga2$ ethnicity == "" , NA , dsuga2$ ethnicity)))))<- dsuga2 %>% drop_na (ethnicity) %>% lollipop_chart (x = ethnicity,line_color = "black" ,point_color = "black" )+ labs (y = "Frequency" ,x = "" ,title = "Uganda" )+ theme_bw ()## All four plots together: <- (ethgmb | ethpak) / (ethtza | ethuga) + plot_annotation ("Ethnic distribution by country" )``` # **Section 3. Measures of interest** - Endorsement of Barrier Bound Leadership (BBL)- Endorsement of Barrier Crossing Leadership (BCL)- Imagistic items: - Event: Positive affect - Event: Negative affect - Event: Episodic recall - Event: Shared perception - Event: Reflection - Event: Transformative for individual - Event: Transformative for group- Ingroup fusion- Ingroup identification- Outgroup fusion- Outgroup identification- Perception of religious freedom- Experience of religious freedom- Empathic concern- Perspective taking- Perceived history of discrimination- Frequency of positive contact with OG- Frequency of negative contact with OG- Outgroup cooperation- Outgroup hostility- Willingness to fight outgroup- Outgroup affect (made up of five sub-items)# **Section 3b: Outgroup attitudinal measures ** ### Perceived history of discrimination ```{r, error = F, message = F, warning = F} # single item measure <- ds %>% drop_na (history_discrimination) %>% ggplot (aes (y = history_discrimination, x = country))+ geom_boxplot (fill = "grey" )+ labs (y = "Perceived discrimination" ,x = "" ,title = "Perceived history of discrimination by country" )+ #facet_wrap(~country, nrow = 2)+ coord_flip ()+ theme_bw ()``` ### Positive contact with outgroup: ```{r, error = F, message = F, warning = F} # single item measure <- ds %>% drop_na (freq_positive_contact) %>% ggplot (aes (y = freq_positive_contact, x = country))+ geom_boxplot (fill = "grey" )+ labs (y = "Frequency" ,x = "" ,title = "Positive contact with outgroup by country" )+ #facet_wrap(~country, nrow = 2)+ coord_flip ()+ theme_bw ()``` ### Negative contact with outgroup: ```{r, error = F, message = F, warning = F} # single item measure <- ds %>% drop_na (freq_negative_contact) %>% ggplot (aes (y = freq_negative_contact, x = country))+ geom_boxplot (fill = "grey" )+ labs (y = "Frequency" ,x = "" ,title = "Negative contact with outgroup by country" )+ #facet_wrap(~country, nrow = 2)+ coord_flip ()+ theme_bw ()``` ### Willingness to fight outgroup: ```{r, error = F, message = F, warning = F} # single item measure <- ds %>% drop_na (fight_outgroup) %>% ggplot (aes (y = fight_outgroup, x = country))+ geom_boxplot (fill = "grey" )+ labs (y = "Willingness" ,x = "" ,title = "Willingness to fight outgroup by country" )+ #facet_wrap(~country, nrow = 2)+ coord_flip ()+ theme_bw ()``` ### Outgroup hostility: ```{r, error = F, message = F, warning = F} # 2 items make up full scale: # og_host_01 # og_host_02 <- ds %>% drop_na (og_hostility) %>% ggplot (aes (y = og_hostility, x = country))+ geom_boxplot (fill = "grey" )+ labs (y = "Hostility" ,x = "" ,title = "Outgroup hostility by country" )+ #facet_wrap(~country, nrow = 2)+ coord_flip ()+ theme_bw ()``` ### Outgroup cooperation: ```{r, error = F, message = F, warning = F} # 2 items make up full scale: # og_coop_01 # og_coop_02 <- ds %>% drop_na (og_cooperation) %>% ggplot (aes (y = og_cooperation, x = country))+ geom_boxplot (fill = "grey" )+ labs (y = "Cooperation" ,x = "" ,title = "Outgroup cooperation by country" )+ #facet_wrap(~country, nrow = 2)+ coord_flip ()+ theme_bw ()``` ### Outgroup affect: ```{r, error = F, message = F, warning = F} # 5 items make up full scale: #NegativeToPositive #HostileToFriendly #SuspiciousToTrusting #ContemptToRespect #ThreatenedToRelaxed <- ds %>% drop_na (og_affect) %>% ggplot (aes (y = og_affect, x = country))+ geom_boxplot (fill = "grey" )+ labs (y = "Affect" ,x = "" ,title = "Outgroup affect by country" )+ #facet_wrap(~country, nrow = 2)+ coord_flip ()+ theme_bw ()``` # **Section 4: TO DO** - Replicate Table 2 from paper 1 for all of the above variables - This is quite easy. Copy R code from one of the individual country Rpubs files.- Replicate Figure 6 from Paper 1 - This will take a bit of work.- Replicate Table 3 from paper 1 (CFA for BCL and BBL) - Done. See section 5 below- Replicate Table 4 from paper 1 (CFA for imagistic items)- Replicate Table 5 from paper 1 (Standardized Loadings from Three Factor EFA)- Replicate Table 6 from paper 1 (CFA for Fusion/Identification measures) - Done. See sections 6 and 7 below- Everything else from Table 6 onwards... # **Section 5: EFA and CFA: BCL and BBL** ## EFA: ```{r, error = F, message = F, warning = F} <- cbind.data.frame (ds$ ENDBCL01, ds$ ENDBCL02, ds$ ENDBCL03, $ ENDBBL01, ds$ ENDBBL02, ds$ ENDBBL03)<- na.omit (leadership)scree (leadership)<- factanal (leadership, 2 , rotation= "promax" )``` ## CFA: ```{r, error = F, message = F, warning = F} ## Remove 'ds$ END ' from variable names: names (leadership) <- sub (".* \\ END" , "" , names (leadership))#correlated two factor solution, marker method <- 'BCL =~ BCL01+BCL02+BCL03 BBL =~ BBL01+BBL02+BBL03' <- cfa (twofacs, data= leadership, std.lv= TRUE ) summary (cfa02,fit.measures= TRUE ,standardized= TRUE )``` # **Section 6: EFA and CFA: Ingroup Fusion and Identification** ## EFA ```{r, error = F, message = F, warning = F} ## IG IGbonds dataframe: <- cbind.data.frame (ds$ IGF01, ds$ IGF02, ds$ IGF03,$ IGI01, ds$ IGI02, ds$ IGI03)names (IGbonds) <- sub ('ds \\ $' , '' , names (IGbonds))<- na.omit (IGbonds)scree (IGbonds)<- factanal (IGbonds, 2 , rotation= "promax" )``` ## CFA ```{r, error = F, message = F, warning = F} #correlated two factor solution, marker method <- 'IGFusion =~ IGF01+IGF02+IGF03 IGIdentification =~ IGI01+IGI02+IGI03' <- cfa (twofacs, data= IGbonds, std.lv= TRUE ) summary (cfa02,fit.measures= TRUE ,standardized= TRUE )``` # **Section 7: EFA and CFA: Outgroup Fusion and Identification** ## EFA ```{r, error = F, message = F, warning = F} ## OG OGbonds dataframe: <- cbind.data.frame (ds$ OGF01, ds$ OGF02, ds$ OGF03,$ OGI01, ds$ OGI02, ds$ OGI03)names (OGbonds) <- sub ('ds \\ $' , '' , names (OGbonds))<- na.omit (OGbonds)scree (OGbonds)<- factanal (OGbonds, 2 , rotation= "promax" )``` ## CFA ```{r, error = F, message = F, warning = F} #correlated two factor solution, marker method <- 'OGFusion =~ OGF01+OGF02+OGF03 OGIdentification =~ OGI01+OGI02+OGI03' <- cfa (twofacs, data= OGbonds, std.lv= TRUE ) summary (cfa02,fit.measures= TRUE ,standardized= TRUE )``` # **Section 8: EFA and CFA: Imagistic items (after removing certain items)** ## EFA ```{r, error = F, message = F, warning = F} <- cbind.data.frame (ds$ event_episodic_recall_01, $ event_episodic_recall_02,$ event_shared_perception_01, $ event_shared_perception_02,$ event_reflection_01, $ event_reflection_02,$ event_transformative_indiv_01,#ds$event_transformative_indiv_02, $ event_transformative_group_01#ds$event_transformative_group_02 ``` # **Section 8: EFA: Imagistic items** ```{r, error = F, message = F, warning = F} <- cbind.data.frame (ds$ event_episodic_recall_01, $ event_episodic_recall_02,$ event_shared_perception_01, $ event_shared_perception_02,$ event_reflection_01, $ event_reflection_02,#ds$event_transformative_indiv_01, $ event_transformative_indiv_02,#ds$event_transformative_group_01, $ event_transformative_group_02names (imagistic) <- sub ('ds \\ $event \\ _' , '' , names (imagistic))<- na.omit (imagistic)<- cor (imagistic[, c (1 : 8 )])``` ### Correlation plot: ```{r, error = F, message = F, warning = F} corrplot (mtx, method = "number" , number.cex = 0.7 ,col= c ("darkred" , "red" , "red" ,"darkgrey" , "blue" , "darkblue" ))``` ### Scree plot: ```{r, error = F, message = F, warning = F} scree (imagistic)``` ### Four factor model: ```{r, error = F, message = F, warning = F} <- factanal (imagistic, 4 , rotation= "promax" )print (fit04$ loadings, cutoff = 0.35 )``` ### Three factor model: ```{r, error = F, message = F, warning = F} <- factanal (imagistic, 3 , rotation= "promax" )print (fit03$ loadings, cutoff = 0.35 )``` # **Section 9: CFA: Imagistic items** ```{r, error = F, message = F, warning = F} ## Remove 'ds$ END ' from variable names: names (imagistic) <- sub (".* \\ $" , "" , names (imagistic))names (imagistic)#correlated two factor solution, marker method <- 'Reflection =~ reflection_01+reflection_02 Vivid_recall =~ episodic_recall_01+episodic_recall_02 Perceived shareness =~ shared_perception_01+shared_perception_02 Defining_experience =~ transformative_indiv_02+transformative_group_02' <- cfa (fourfacs, data= imagistic, std.lv= TRUE ) summary (cfa04,fit.measures= TRUE ,standardized= TRUE )``` 
