starting_with_data.Rmd

---
title: "Starting with data"
author: "Najko Jahn"
date: "5. Oktober 2015"
output:
  html_document:
    keep_md: true
---

Adapted and re-used from : [Data carpentry -- Starting with R for data analysis](https://github.com/datacarpentry/R-ecology)

# Skills

* Load spreadsheets into R
* Explore the nature of the datasets
* Store data on your local disk

In many cases, you will work with data. R supports several ways of loading datasets. One is to fetch spreadsheets from the web.

## Spreadsheet data

Let's download journal metadata from the Directory of Open Access Jorunals into our data folder by using `curl` methods.

```{r}
download.file("https://doaj.org/csv", "examplesdoaj.csv", method = "curl")
```

Load the dataset into R console with `read.csv`

```{r}
doaj <- read.csv(file = "examples/doaj.csv", sep = ",", header = TRUE)
```

Show the first six rows

```{r}
head(doaj)
```

How many rows and columns has the dataset?

```{r}
dim(doaj)
```

Show the column names

```{r}
names(doaj)
```

To compactly display the structure of arbitrary R objects use `str`

```{r}
str(doaj)
```

**Task 1**: 
Load in the spreadsheet examples/neuro-scimago.csv.

Based on the output of `str(neuro)`, can you answer the following questions?**

* What is the class of the object `neuro`?
* How many different publishers does the Directory of Open Access Journals spreadsheet contain?


## Working with empty cells

Empty cells often affect your analysis. R discriminate missing values with `NA`

```{r}
x <- c(2:20, NA)
mean(x)
mean(x, na.rm = TRUE)
```

As you have may noticed, the `doaj` dataset has empty cells not designated as `NA`

```{r}
head(doaj)
```

To take care that empty cells are corectly loaded into R, use the `na.strings` option

```{r}
doaj <- read.csv(file = "examples/doaj.csv", sep = ",", header = TRUE, na.strings = "")
head(doaj)
```

## Saving datasets

To print your data to a `csv` file, use the `write.csv` command.

```{r}
write.csv(doaj, file = "examples/backup-doaj.csv")
```