---
title: "R Notebook"
output: html_notebook
---

Let us load data on a phonological research (Clarett, 2017):

```{r}
df <- read.csv("http://math-info.hse.ru/f/2018-19/ling-data/icelandic.csv")
```

The research is dedicated to the relationship between vowel duration in Icelandic language and phonological features of following consonants. Five native speakers were asked to read some texts aloud, their speech was recorded and then the duration of different sounds was measured.

**Key variables:**

* ``
* ``
* ``
* ``
* ``
* ``
* ``
* ``

Why Icelandic language? In this language we can observe the aspiration effect: vowel duration is different for cases when 

```{r}
# general boxplot
boxplot(df$vowel.dur)
```

```{r}
# number of outliers
length(boxplot(df$vowel.dur)$out)
```

```{r}
table(df$apriration)
```

```{r}
# choose two subsamples
asp <- df[df$aspiration == 'yes',]
nasp <- df[df$apriration == 'no',]
```

```{r}
# summary for aspirated and non-aspirated cases
summary(asp$vowel.dur)
summary(nasp$vowel.dur)
```

```{r}
# boxplot for groups
boxplot(df$vowel.dur ~ df$aspriration)
```

```{r}
# more interesting - boxplot for all groups
boxplot(df$vowel.dur ~ df$cons1)
```

```{r}
# correct - make sure we work with the same type of a consonant
asp <- df[df$aspiration == 'yes' & df$height == 'mid', ]
nasp <- df[df$aspiration == 'no' & df$height == 'mid', ]
```

```{r}
# again summary for a corrected case
summary(asp$vowel.dur)
summary(nasp$vowel.dur)

nrow(asp)
nrow(nasp)
```

```{r}
# formal testing
t.test(asp$vowel.dur, nasp$vowel.dur)
```

```{r}
# H1: mu_aps < mu_nasp
t.test(asp$vowel.dur, nasp$vowel.dur, alternative = "less")
```