Lesson 5 part

.idea/dictionaries/Dusty.xml

@@ -1,6 +1,7 @@
 <component name="ProjectDictionaryState">
   <dictionary name="Dusty">
     <words>
+      <w />
       <w>.</w>
       <w>...elt</w>
       <w>...length</w>
@@ -1114,7 +1115,6 @@
       <w>`.__C__C++Function`</w>
       <w>`.__C__C++Object`</w>
       <w>`.__C__C++OverloadedMethods`</w>
-      <w />
       <w>`.__C__{`</w>
       <w>`.__T__!:base`</w>
       <w>`.__T__$:base`</w>
@@ -1463,7 +1463,6 @@
       <w>`mday&lt;-`</w>
       <w>`minute&lt;-`</w>
       <w>`month&lt;-`</w>
-      <w />
       <w>`on_failure&lt;-`</w>
       <w>`packageSlot&lt;-`</w>
       <w>`polygons&lt;-`</w>

lesson5/lesson5_student.rmd

@@ -15,8 +15,19 @@ Notes:
 Notes:
 
 ```{r Third Qualitative Variable}
-ggplot(aes(x = gender, y = age),
-       data = subset(pf, !is.na(gender))) + geom_histogram()
+library(ggplot2)
+library(dplyr)
+
+pf.fc_by_age_gender <- pf %>%
+  filter(!is.na(gender)) %>%
+  group_by(age, gender) %>%
+  summarize(mean_friend_count = mean(friend_count),
+            median_friend_count = median(friend_count),
+            n = n()) %>%
+  ungroup() %>%
+  arrange(age)
+
+head(pf.fc_by_age_gender)
 ```
 
 ***
@@ -25,7 +36,8 @@ ggplot(aes(x = gender, y = age),
 Notes:
 
 ```{r Plotting Conditional Summaries}
-
+ggplot(aes(x = age, y = median_friend_count), data = pf.fc_by_age_gender) +
+  geom_line(aes(color = gender))
 ```
 
 ***
@@ -33,7 +45,7 @@ Notes:
 ### Thinking in Ratios
 Notes:
 
-***
+What is the ratio of friends for males vs females
 
 ### Wide and Long Format
 Notes:
@@ -44,8 +56,14 @@ Notes:
 Notes:
 
 ```{r}
-install.packages('reshape2')
+#install.packages('reshape2')
 library(reshape2)
+
+pf.fc_by_age_gender.wide <- dcast(pf.fc_by_age_gender,
+                                  age ~ gender,
+                                  value.var = 'median_friend_count')
+
+head(pf.fc_by_age_gender.wide)
 ```
 
 
@@ -55,7 +73,9 @@ library(reshape2)
 Notes:
 
 ```{r Ratio Plot}
-
+ggplot(aes(x = age, y = female/male), data = pf.fc_by_age_gender.wide) +
+  geom_line() +
+  geom_hline(aes(yintercept = 1), alpha=0.3, linetype = 2)
 ```
 
 ***
@@ -64,7 +84,8 @@ Notes:
 Notes:
 
 ```{r Third Quantitative Variable}
-
+pf$year_joined <- floor(2014 - pf$tenure/365)
+head(pf)
 ```
 
 ***
@@ -73,7 +94,8 @@ Notes:
 Notes:
 
 ```{r Cut a Variable}
-
+pf$year_joined.bucket = cut(pf$year_joined, c(2004, 2009, 2011, 2012, 2014))
+table(pf$year_joined.bucket)
 ```
 
 ***
@@ -82,7 +104,8 @@ Notes:
 Notes:
 
 ```{r Plotting it All Together}
-
+ggplot(aes(x = age, y = friend_count), data = subset(pf, !is.na(year_joined.bucket))) +
+  geom_line(aes(color = year_joined.bucket), stat='summary', fun.y = median)
 ```
 
 ***
@@ -91,6 +114,9 @@ Notes:
 Notes:
 
 ```{r Plot the Grand Mean}
+ggplot(aes(x = age, y = friend_count), data = subset(pf, !is.na(year_joined.bucket))) +
+  geom_line(aes(color = year_joined.bucket), stat='summary', fun.y = mean) +
+  geom_line(stat = 'summary', fun.y = mean, linetype = 2)
 
 ```