Notes.Rmd

---
title: "Notes"
output:
  html_document:
    toc: true
    toc_float: true
    collapsed: false
    number_sections: false
    toc_depth: 1
    #code_folding: hide
---

```{r setup, include=FALSE}
knitr::opts_chunk$set(message=FALSE,warning=FALSE, cache=TRUE)
```

# going over a lesson on 2/4/2019
```{r}
a<- 1
b<-"Hello World"
d<-"3"
print(a)

class(d)

vector<- c(1,2,3,4,5,6,7,8)
vector[2:4]
vector[c(2,2,2,2,2)]
vector[vector <6] # show the numbers less than 7
vector[vector>=6] # greater or equal to 6
vector[vector!=4] # show numbers that are not 5

vector[8]<-3000 # put 3000 in the place of 8th number
vector

vector[1:2] <- 10

class(vector)
vector[7]<- "j" # put a number in the 7th place in the vector

vector

a<-c("1","2","3")
class(a)
b<- a
class(b)
as.numeric(b) # store charecters 123 as a number

a<- "ljjasjdflsjasf;lasdjflk;asdjfl;asdjfl"
strsplit(a,split="") # if you had a paragraph and you wanted to split the paragraph or put a period in R

a[1]

b[(1)][5]

example_list <- list(a=11, b=4, d=c(1,2,3,4))
example_list


data.frame() # particular way of structuring data


f<- matrix(0,nrow = 5, ncol = 3) # creating 5 rows and 3 columns

first_names <- c("asdf", "Asdfasd","asdfasdf")
ages<-c(3343,23122,67456)
grades<-c(89,56,99)

everybody <- data.frame(first_names,ages,grades)

everybody$ages[2] <- 12
```

```{r}
data1 <- c(1,2,3)
data1[1]
data1[-1]

```

# 2/11/2019 3rd lecture
Downloading R repository

on February 25th we will talk about data analysis 
assignment 3 (R Basics) - R package assignment - how R code works, how it is shared
I will be submiting a link to a new R repository
R functions are prepackaged - in Packages - collect functions you have worked on

click instal tab in packages - repository cran allows y ou to save on cloud - choose your package and install
next step load the package - click the package button


# creating a simple function of adding 1 to whatever number you indicated
'''
my_function<- function(input){
  #body of code
return(output)
}
'''
```{r}
add_one <- function(x){
  return(x+1)
}
  add_one(9)


```

'''
{r, eval=FALSE}
add_one <- function(x){
# add one tox, and save the output
save_result <- x+1


another way to do it
'''
{r}
add_one <- function(x) return(x+1)
'''
```{r}
add_one <- function(x) return(x+1)

  add_one(7)


```

# how can we roll a die and get a number
'''
{r}
sample(1:6,1)
roll_dice <- function(){
return(sample(1:6,1))
}
roll_dice() # make sure you put something in the parenthesis
'''
```{r}
roll_dice <- function(x){
  return(sample(1:6,1))

}
  roll_dice()

```

creating a sum function 
```{r}
x <- c(1,2,3)
my_sum <- function(x){  # this is declaring a funtion
  get_sum<-sum(x)
  return(get_sum)
}
# to declare a variable
  a <- c(1,2,3)
```
# find a mean of restricted range of numbers
```{r}
ranged_mean  <- function(x, min_val, max_val){
  restricted_values <- x[x>min_val & x < max_val]
  return(mean(restricted_values))

}
  some_numbers <-c(3,4,3,2,3,4,5,6,7,8,8,8,9,8)
  ranged_mean(some_numbers,3,9)

```

```{r,eval=F}
ranged_mean  <- function(x, min_val, max_val){
  restricted_values <- x[x>min_val & x < max_val]
  outputs <- list(original_values = x, 
                  restricted_values=restricted_values
                  restricted_mean = mean(restricted_values)

}
  some_numbers <-c(3,4,3,2,3,4,5,6,7,8,8,8,9,8)
  ranged_mean(some_numbers,2,4)

```


# how to make new R project

steps
1. 
2.
3.

# 2/25/19February 25th data visualization today
APA paper with papaja on March 25th

making graphs 

a <- data.fram()
simple reaction time - how fast can you respond to the light

first column DV - simple reaction time
SRT   IV - how bright the light is  AGE   Subject   Address
150   Dim
200   bright
210   bright 

```{r}
Names <- c("Peter", "Paul", "Mary",NA)
Ages <- c(100,120,90, 80)
Sex <- c("M","M","F",NA)

my_dataframe <- data.frame(Names,Ages,Sex) # start adding things in the perenthesis
```

when you see a graph you will see lines corresponding to the numbers 
you will have to create a table with  all those numbers 

DV  Gender    XY
1   B         Y
2   A         Y
3   A         x
can have 100s of numbers

other things I should know 
my_dataframe$Names  # illustrates a default 
my_dataframe$Ages
li
if u didn't want it to be a factor you have to convert it to a diffrent thing
my_dataframe$Names <- as.character(my_dataframe$Names)

```{r,}
library(ggplot2)
a <- c(1,2,3,2,3,4,5,4, 4, 2)
b <- c(4,3,4,3,2,1,2,3, 7, 1)
plot_df <- data.frame(a,b)
ggplot(plot_df, aes(x=a,y=b))+
  geom_point() # connect the data to diff parts of the graph
  
# basic scatterplot
#ggplot(plot_df, aes(x=a,y=b))+
 # geom_point()  

```

here we are plotting the same graph but addit custom settings 

```{r}
ggplot(NULL, aes(x=a,y=b))+
  geom_point(size=2)+
  geom_smooth(method=lm)+ # this is linear regression line
  coord_cartesian(xlim=c(0,7),ylim=c(0,10))+ # if you want to increase a range
  xlab("x-axis label")+ # change a name of x axis
  ylab("y-axis label")+
  ggtitle("I made a scatterplot")+ # add a title
  theme_classic(base_size=12)+ # classic background; base size is the size of the font from 12 to whatever
  theme(plot.title = element_text(hjust = 0.5))
```

what if u wanted to make a graph but you want it to take only a prortion of the page, like in the corner of the page
base size will make sure your smaller graph has the same font

3 bars graph

```{r}
factor_one <- as.factor(c("A","B","C"))
dv_means <- c(20,30,40)
dv_SEs   <- c(4,3.4,4)
plot_df <- data.frame(factor_one,
                      dv_means,
                      dv_SEs)
ggplot(plot_df, aes(x=factor_one,y=dv_means))+ # what do you want to go into x and y axis
  geom_bar(stat="identity")

```

bar graph with 2 factors 
```{r}
factor_one <- rep(as.factor(c("A","B","C")),2) # this 2 means repeat ABC twice
factor_two <- rep(as.factor(c("IIA","IIB")),3)
dv_means <- c(20,30,40,20,40,40)
dv_SEs   <- c(4,3.4,4,3,2,4)
plot_df <- data.frame(factor_one,
                      factor_two,
                      dv_means,
                      dv_SEs)

# basic bar graph

ggplot(plot_df, aes(x=factor_one,y=dv_means, # aestetics definition has gotten 
                    group=factor_two, 
                   # color=factor_two))+  # color here means the border collor
                    fill=factor_two))+
  geom_bar(stat="identity", 
           position="dodge",
           color="black")
```

```{r}
ggplot(plot_df, aes(x=factor_one,y=dv_means,
                    group=factor_two,
                    color=factor_two,
                    fill=factor_two))+
  geom_bar(stat="identity", position="dodge")+
  geom_errorbar(aes(ymin=dv_means-dv_SEs,
                    ymax=dv_means+dv_SEs),
                position=position_dodge(width=0.9),
                width=.2,
                color="black")+
  coord_cartesian(ylim=c(0,100))+
  xlab("x-axis label")+
  ylab("y-axis label")+
  ggtitle("Bar graph 2 factors")+
  theme_classic(base_size=12)+
  theme(plot.title = element_text(hjust = 0.5))

```

recreating the first figure 

we have 8 names (Dora, Rita, Lisa)

# 1st promblem in the 4th assignment - we will have names column - each name will appear twice in each condition
measurement - Mean freaquncy

Names   MF    Condition
Dara    60    social cond
Dara    60    none social

```{r}
Names <-rep(c("Dara", "Azalea", "Barbi", "Rowena", "Fiona"),each=2)
MF <- rnorm(10,45,25) # 10 is the number of numbers you want, and 45 to 25 is the range
Condition <- rep(c("Social","NonSocial"),5)
Aversity <- rep(c("A","N_A" ),times=c(4,6))
  plot_df <- data.frame(Names, MF, Condition, Aversity)
  
  ggplot(plot_df, aes(x=Condition, y=MF, group=Names, linetype=Aversity))+
    geom_line()+
    geom_text(label=Names)+
    theme_classic()


```


# 3/11/2019

```{r}
library(dplyr)
df <- starwars


```

# some of the moves you can manipulate the data without a dplyr package
df$name # dolloar sign gives yoiu the names in the file

df[1,] # row 1
df[1:4,] # rows 1 to 4 and all of the columns

```{r}
df[df$name=="Luke Skywalker",] # find luke's row and give me all of the columns of the skywalker
df[df$height >180,]
```
```{r}
dim(df) #c(number of rows, number of columns)

df <-cbind(df, random_number=runif(dim(df)[1],0,1))
```


df$species <- as.factor(df$species)
levels(df$species)

how many characters have tatooine 
```{r, eval=FALSE}
df[df$homeworld=="Tatooine",]
dim(df[df$homeworld=="Tatppome",)[1]

```

# group by certain things

Dataframe %>%
```{r} 
library(dplyr)
new_df <- df %>% # adding things on
  filter(height > 100) %>%
  group_by(homeworld) %>% 
  summarise(mean_birth_year = mean (birth_year, na.rm = TRUE))
  
```

```{r}
View (new_df)
new_df <- df %>%
  filter(height >120,
         height <180,
         birth_year >20)
  
```


```{r} 
new_df <- df %>%
  filter(gender =="male")
  
```

```{r} 
new_df <- df %>%
  group_by(hair_color) %>%
  summarise(counts=length(name))
  
```

```{r,eval=FALSE} 
new_df <- df %>%
  group_by(hair_color, eye_color) %>%
  summarise(mean_years = mean(birth_year,na.rm = TRUE)),
      sd_years = sd(birth_year,na.rm = TRUE),
      counts = length(names)
  
```
how many characterers in each movie


how to load data
```{r, eval=FALSE} 
hsq <- read.csv(data.table("data/hsq.csv"))

```

# getwd() function will tell you where the working directory is 

3/18

# Statistics

## t test
```{r}
x <- c(1,2,3,4,5,2,4,7,4,9)
t.test(x, mu=2)
```

SE of the mean is sd/sqrt of N
T = x-mu / SEM
T dist = find t value of 10 numbers 10000 times - the t scores make up T distribution

```{r}
x <- c(1,2,3,4,5,2,4,7,4,9)
t.test(x, mu=2, alternative = "two.sided")
```


```{r}
x <- c(1,2,3,4,5,2,4,7,4,9)
t.test(x, mu=2, alternative = "greater")
```

Paired T-test
```{r}
x <- c(1,2,3,4,5,2,4,7,4,9)
y <- c(3,2,5,4,3,2,5,6,7,8)
t.test(x, y, paired=TRUE)
```
welch two sample t-test

# simulation of 10000 t-tests
```{r}
a <- replicate(10000,t.test(rnorm(10,0,1),rnorm(10,0,1))$p.value)
hist(a)
```

# getting t values instead of p values

```{r}
a <- replicate(10000,t.test(rnorm(10,0,1),rnorm(10,0,1))$statistic)
hist(a)
```

report_t(t_results)'

in a t-test = testing difference between the means

in an ANOVA = one omnibus tes tells us whether there is a difference

```{r, eval=F}
Factor1 <- rep(c("A","B","C"),each=5)
  DV <- rnorm(n=15,mean=0,sd=1)
  all_data1 <- data.frame(DV,Factor1)
    View(all_data1)

```

# load a function into package without loading a package - knitr:: kable()

# Randomization Test
```{r}
A <- c(1,2,3,4,5,6,7,8,9,10)
B <- c(2,4,6,8,10,12,14,16,18,20)

all <- c(A,B) 

mean_difference <- c()
for(i in 1:10000){
  shuffle <- sample(all)
  newA <- shuffle[1:10] # first ten num put in a new A
  newB <- shuffle[11:20] # the next ten nums
  mean_difference[i] <- mean(newB)-mean(newA) # sizes of mean diff we could have done if we assigned people randomly
}

observed <- mean(B)-mean(A)
length(mean_difference[mean_difference >= observed])/10000 # how many means fall within

```


```{r}
RandomNumbers <- runif(100, min = 0, max = 100)
hist(RandomNumbers)
```

```{r}
RandomNumbers <- runif(100, min = 0, max = 100)
hist(RandomNumbers)
```

## 3/25/19

```{r}
a<- rnorm(10,0,1)
b<- rnorm(10,0,1)
mean_diff <- mean(a) - mean(b)
mean_diff <- replicate(10000,
                       mean(rnorm(10,0,1))-mean(rnorm(10,0,1)))
sort(mean_diff)[9500]


```
```{r}
hist(mean_diff)

```

show that the properties of the t-dist are the same as the properties of the analytic dist


```{r}
pt(q=c(.5,1,1.5,2,2.5),df=9) 
```


```{r}
pt(q=c(.5,1,1.5,2,2.5),df=9) # looking the sampling distribution of t
t_s <- replicate(10000,t.test(rnorm(10,0,1),mu=0)$statistic)
length(t_s[t_s > .5])
```

```{r}
hist(t_s)

```

```{r}
t_s <- replicate(10000,t.test(rnorm(10,0,1),mu=0)$statistic)
length(t_s[t_s < .5])/10000
```

```{r}
t_s <- replicate(10000,t.test(rnorm(10,0,1),mu=0)$statistic)
length(t_s[t_s < 1])/10000
```

```{r,eval=F}
t_s <- replicate(10000,t.test(rnorm(10,0,1),mu=0)$statistic)
length(t_s[t_s < .5])/10000
sim_ps <- c(length(t_s[t_s <.5])/1000
ral_ps
sim_ps
sum(abs)
```


sampling distribution of correlations


```{r}
cor(rnorm(10,0,1), rnorm(10,0,1))
```

replicate 10000 times
```{r}
hist(replicate(10000,cor(rnorm(10,0,1), rnorm(10,0,1))))
```

converge all r values to absolute values and sort them from smallest 0 to largest

```{r}
sim_rs<-replicate(10000,cor(rnorm(100,0,1),rnorm(100,0,1)))
sort(abs(sim_rs))[9500]
```

F distribution

calculate f value for the main effect of group

```{r}
run_anova <- function(){
A <- rnorm(4,0,1)
B <- rnorm(4,0,1)
C <- rnorm(4,0,1)
conds<- rep(c("A","B","C"),each=4)
DV<-c(A,B,C)
df<-data.frame(conds,DV)
sum_out <-summary(aov(DV~conds,df))
return(sum_out[[1]]$'F value'[1])
}
save_fs <-replicate(10000,run_anova())
hist(save_fs)
sort(save_fs)[9500]
qf(.95,2,9)
```