> # ============================================================

> # Pre-task_Grammar_Instruction_LREs_Young_Learners – Table 7

> # Pearson correlations between MLAT-ES scores and week-to-week LRE gains

> # ============================================================

> library(readxl)

> library(dplyr)

> # ---- 1. Import all sheets ----

> d1 <- read_excel("C:/Users/dmard/Documents/PTGI_LRE_repro_package/data/PTGI_LRE_data_1.xlsx",
+                  sheet = "Dict. 1")

> d2 <- read_excel("C:/Users/dmard/Documents/PTGI_LRE_repro_package/data/PTGI_LRE_data_1.xlsx",
+                  sheet = "Dict. 2")

> datos <- read_excel("C:/Users/dmard/Documents/PTGI_LRE_repro_package/data/PTGI_LRE_data_1.xlsx",
+                     sheet = "D1+D2+IDs")

> # ---- 2. Combine relevant columns ----

> combined <- d1 %>%
+   select(Code, Group, `HE/SHE_D1_SUM`, `3ps_D1_SUM`) %>%
+   left_join(
+     d2 %>% select(Code, `HE/SHE_D2_SUM`, `3ps_D2_SUM`),
+     by = "Code"
+   ) %>%
+   left_join(
+     datos %>% select(Code, MLAT_Q) %>%
+       mutate(MLAT_Q = as.numeric(gsub(",", ".", MLAT_Q))),
+     by = "Code"
+   )

> # ---- 3. Compute gains ----

> combined <- combined %>%
+   mutate(
+     HE_SHE_Gain = `HE/SHE_D2_SUM` - `HE/SHE_D1_SUM`,
+     X3ps_Gain   = `3ps_D2_SUM` - `3ps_D1_SUM`
+   )

> # ---- 4. Split by group ----

> ctrl <- subset(combined, Group == "Control")

> exp  <- subset(combined, Group == "Experimental")

> # ---- 5. Pearson correlations ----

> cat("\n--- CONTROL GROUP ---\n")

--- CONTROL GROUP ---

> print(cor.test(ctrl$MLAT_Q, ctrl$HE_SHE_Gain, method = "pearson"))

	Pearson's product-moment correlation

data:  ctrl$MLAT_Q and ctrl$HE_SHE_Gain
t = -1.1567, df = 3, p-value = 0.3311
alternative hypothesis: true correlation is not equal to 0
95 percent confidence interval:
 -0.9648653  0.6409667
sample estimates:
       cor 
-0.5553547 


> print(cor.test(ctrl$MLAT_Q, ctrl$X3ps_Gain,   method = "pearson"))

	Pearson's product-moment correlation

data:  ctrl$MLAT_Q and ctrl$X3ps_Gain
t = 0.04, df = 3, p-value = 0.9706
alternative hypothesis: true correlation is not equal to 0
95 percent confidence interval:
 -0.8770436  0.8872807
sample estimates:
       cor 
0.02308785 


> cat("\n--- EXPERIMENTAL GROUP ---\n")

--- EXPERIMENTAL GROUP ---

> print(cor.test(exp$MLAT_Q, exp$HE_SHE_Gain, method = "pearson"))

	Pearson's product-moment correlation

data:  exp$MLAT_Q and exp$HE_SHE_Gain
t = 1.236, df = 3, p-value = 0.3044
alternative hypothesis: true correlation is not equal to 0
95 percent confidence interval:
 -0.6182274  0.9673743
sample estimates:
      cor 
0.5808689 


> print(cor.test(exp$MLAT_Q, exp$X3ps_Gain,   method = "pearson"))

	Pearson's product-moment correlation

data:  exp$MLAT_Q and exp$X3ps_Gain
t = -0.3329, df = 3, p-value = 0.7611
alternative hypothesis: true correlation is not equal to 0
95 percent confidence interval:
 -0.9181222  0.8320833
sample estimates:
       cor 
-0.1887431