Reading in CSVs with headers on multiple rows

data.csv

multirow-headers.rmd

# Dealing with multi-row headers

To load data that looks like this:

![CSV image](https://pbs.twimg.com/media/CmIg6rVUYAADNQQ.jpg)

Load the zoo and dplyr packages

```r
library("zoo")
library("dplyr")
```

Then create headers that have all the info from both header rows.

```r
## Read in the first header row and replace blank cells with the most recent
## non-blank value using na.locf()
head1 <- read.csv("data.csv", nrow = 1, header = FALSE,
                  stringsAsFactors = FALSE, colClasses=rep("character", 9),
                  na.strings = "") %>%
  as.character() %>%
  na.locf()

## Read in second header row
head2 <- read.csv("data.csv", skip = 1, nrow = 1, header = FALSE,
         stringsAsFactors = FALSE)

## Paste headers together
header <- paste(head1, head2, sep = "-")

header
## [1] "B0S2-abundance"     "B0S2-percent cover" "B0S2-height"       
## [4] "B0S3-abundance"     "B0S3-percent cover" "B0S3-height"       
## [7] "B0S4-abundance"     "B0S4-percent cover" "B0S4-height"
```

Then load the data without headers and add the combined headers back on.

```
dat <- read.csv("data.csv", skip = 2, stringsAsFactors = FALSE,
                header = FALSE) %>%
  setNames(header)
```