diff --git a/.gitignore b/.gitignore
index e3c9989..fa89596 100644
--- a/.gitignore
+++ b/.gitignore
@@ -67,3 +67,4 @@ ML/outputs
 
 
 Final Paper/
+test.Rda
diff --git a/ML/2-modeling.R b/ML/2-modeling.R
index b9735d8..a3cf251 100644
--- a/ML/2-modeling.R
+++ b/ML/2-modeling.R
@@ -47,9 +47,12 @@ ds_train <- rsample$training(model_data_split)
 ds_test  <- rsample$testing(model_data_split)
 
 # verify distribution of data
-table(ds_train$ft4_dia) %>% prop.table()
-table(ds_test$ft4_dia) %>% prop.table()
+strata1 <- table(ds_train$ft4_dia) %>% prop.table() %>% tibble::enframe() %>% dplyr::rename(Train = value)
+strata2 <- table(ds_test$ft4_dia) %>% prop.table() %>% tibble::enframe() %>% dplyr::rename(Test = value)
 
+strata_table <- strata1 %>%
+  dplyr::left_join(strata2) %>%
+  dplyr::rename(Class = name)
 
 # random forest classification -----------------------------------------------------------
 
diff --git a/_quarto.yml b/_quarto.yml
index a5a5489..71a5344 100644
--- a/_quarto.yml
+++ b/_quarto.yml
@@ -10,6 +10,8 @@ book:
     - chapter1.qmd
     - chapter2.qmd
     - chapter3.qmd
+    - chapter4.qmd
+    - chapter5.qmd
     - references.qmd
   abstract: "This is a test to see what happens with this"
 
diff --git a/chapter4.qmd b/chapter4.qmd
new file mode 100644
index 0000000..980ab16
--- /dev/null
+++ b/chapter4.qmd
@@ -0,0 +1,45 @@
+# Results
+
+```{r}
+#| include: false
+#| cache: true
+
+library(magrittr)
+load("test.Rda")
+
+```
+
+The final data set used for this analysis consisted of 11,340
+observations. All observations contained a TSH and Free T4 result and
+less than three missing results from all other analytes selected for the
+study. The dataset was then randomly split into a training set
+containing 9071 observations and a testing set containing 2269
+observations. The data was split using stratification of the Free T4
+laboratory diagnostic value. @tbl-strata shows the split percentages.
+
+```{r}
+#| label: tbl-strata
+#| tbl-cap: Data Stratification 
+#| echo: false
+
+strata_table %>% knitr::kable()
+
+```
+
+First, the report shows the ability of classification algorithms to
+predict whether Free T4 will be diagnostic, with the prediction quality
+measured by Area Under Curve (AUC) and accuracy. Data regarding the
+univariate association between each predictor analyte and the Free T4
+Diagnostic value is then presented. Finally, data is presented with the
+extent to which FT4 can be predicted by examining the correlation
+statistics denoting the relationship between measured and predicted Free
+T4 values.
+
+## Predictability of Free T4 Classifications
+
+In clinical decision-making, a key consideration in interpreting
+numerical laboratory results is often just whether the results fall
+within the normal reference range [@luo2016]. In the case of Free T4
+reflex testing, the results will either fall within the normal range
+indicating the Free T4 is not diagnostic of Hyper or Hypo Throydism, or
+they will fall outside those ranges indicating they are diagnostic.
diff --git a/chapter5.qmd b/chapter5.qmd
new file mode 100644
index 0000000..8b13789
--- /dev/null
+++ b/chapter5.qmd
@@ -0,0 +1 @@
+
diff --git a/references.bib b/references.bib
index 60cb331..526df30 100644
--- a/references.bib
+++ b/references.bib
@@ -335,3 +335,18 @@ DOI: 10.13026/S6N6-XD98}
 	url = {https://dl.acm.org/doi/10.1145/2939672.2939785},
 	address = {New York, NY, USA}
 }
+
+@article{luo2016,
+	title = {Using Machine Learning to Predict Laboratory Test Results},
+	author = {Luo, Yuan and Szolovits, Peter and Dighe, Anand S. and Baron, Jason M.},
+	year = {2016},
+	month = {06},
+	date = {2016-06},
+	journal = {American Journal of Clinical Pathology},
+	pages = {778--788},
+	volume = {145},
+	number = {6},
+	doi = {10.1093/ajcp/aqw064},
+	note = {PMID: 27329638},
+	langid = {eng}
+}