Suppose that you have data from two groups – heights of male and female students, salaries of CEOs from public and private businesses, or, what I will use here, the number of books owned by science and humanities majors. Further suppose that the data look like this
science -- 7,1,3,2 humanities -- 17,12,9,6,15,13
It is very common for students to want to enter these data in the following format:
| science | humanities |
|---|---|
| 7 | 17 |
| 1 | 12 |
| 3 | 9 |
| 2 | 6 |
| 15 | |
| 13 |
However, this violates the principle that each row should contain information about only one individual. For example, the first row shows the number of books for two individuals, one a science and the other a humanities student. In addition, these data will result in an error about unequal column lengths when loaded into R.
These data should be entered in what is called stacked format. In stacked format, the data look like the following.
| type | books |
|---|---|
| science | 7 |
| science | 1 |
| science | 3 |
| science | 2 |
| humanities | 17 |
| humanities | 12 |
| humanities | 9 |
| humanities | 6 |
| humanities | 15 |
| humanities | 13 |
In this way, each row corresponds to a single individual. For examle, the first row represents a single student, who is in the sciences and has seven books.
See this FAQ for how to get (i.e., load) these data into R.