Often times, people collect data and put it into an excel spreadsheet because it’s a free software that is open to the public. There are different types of information that gets incorporated in the file. Character variable represent the columns that contain only letters and symbols. Actually anything having to do with symbols could potentially be considered a character or string variable. You also have columns that contain numbers only – numeric variables. Then you have variables which look like character fields but you plan on analyzing it. You meant to make it numeric but you just didn’t have to time to do so. Or you figure ‘well I will do it later’. And that’s fine, you can do it later in a program. But in the interest of time, if you analyze it – make it numeric. Most statistical software programs will not process character variables as a statistic with one exception. The frequency command works for most string variables. It will give you a count of all the possible values in your data file. But aside from percentages and frequencies, you need to tell the computer what the text represents in the file. For instance, suppose you have a field for Intensity with entries ‘Mild’, ‘Moderate’, and ‘Severe’. Your computer does not know that Mild is a lower category than the others. Nor does your software understand that you mean to quantify it. Recommendation: Please convert categorical string variables from the start of your study. It saves everyone time with entry and verification. So for instance, you could assign a 1=Mild; 2=Moderate; and 3=Severe. It’s a heck of a lot easier to enter a 1,2, or 3 then spell everything out for each patient entry. And it also helps the programmer in that you don’t have to worry about some entries being in different formats. I recommend just keep track of what the values represent in your documentation.
If you enjoyed this blog there are more to follow-Amy