Classifying your columns.
Once your data has been uploaded, you will arrive at the Column classification page.
In the Classify Column area, you are prompted to identify whether each data column in your uploaded project consists of a variable, or a response. You are also able to exclude specific columns of data from your analysis.
Variable. Selecting this option indicates that the data in the column should be considered as a variable. If conducting an experiment, these are your independent variables or conditions. If conducting correlational research, these are your predictors. If conducting a survey, then these are your demographics, such as gender, age or political affiliation, or any other characteristic you care to collect. Further information on working with variables can be found in the Comparing your variables page of these Practicals, as well as during discussion on geists in the Get a sense of the whole page of the Technicals.
Response. Selecting this option includes all of the column's text in the pool of natural language to be analyzed. If you selected to combine qualitative columns on the Settings page, then classifying more than one column as a response will combine all of the data in each selected column together into one pool for analysis. If, however, you chose to keep qualitative columns separate, then each column classified as a response will be treated as separate pools of natural language data during analysis.
Exclude. Selecting this option excludes all of the data in the column from the analysis.
After you have classified your columns, you will want to review the Column Name area, and if needed, remove all symbols and punctuation.
Once you have classified your columns and removed symbols and punctuation from the Column Name area, you can analyze your data by clicking the Commit Classification button, located at the bottom of the page.