Knowledge base | Statwing

Creating a new variable or segment

Statwing offers several specific ways to create new variables from your existing data.

TypeExample
Bucketing numbers into groups
  • 6-7 → “Satisfied”
  • 3-5 → “Neutral”
  • 1-2 → “Unsatisfied”
Grouping categories together
  • USA & Mexico & Canada → “North America”
  • Colombia & Venezuela & (etc.) → “South America”
Mathematical functions and formulas
  • =median(Score1, Score2, Score3) → median of the scores
Logic
  • If Satisfaction < 3 and country = Canada
  • then → “Unsatisfied Canadians”
  • else → “Other”
Time functions
  • 11/2/2012 → “Friday”
  • 1/7/2014 → “Tuesday”
  • 5/12/2014 → “Monday”

 

Start by selecting the “Create” button on the left side of the interface:

create-variable-screenshot

 

You’ll be presented with five different tabs, each of which is a different way to create a variable:

create-variable-tabs

 

***Update: "Bucket a Category" and "Bucket a Number" have been collapsed into one tab since the writing of this help document. They still work as described below, but now that are accessed from the same tab.

 

 

Bucket a Category

Bucket a category when you have a category with many groups and you want to combine them into a simpler, aggregated variable. For example:

Screen Shot 2014-12-18 at 5.40.12 PM

 

 

 

Formula

Use a spreadsheet-like formula to create a new variable.

Screen Shot 2014-12-18 at 4.05.30 PM

 

Statwing has a variety of functions. If you need one that we don’t have, contact us and we can add it quickly, sometimes within 24 hours.

Screen Shot 2014-12-18 at 4.06.55 PMScreen Shot 2014-12-18 at 4.07.19 PM Screen Shot 2014-12-18 at 4.07.04 PMScreen Shot 2014-12-18 at 4.07.14 PM

 

 

 

Time

Time-functions, also accessible via the “Formula” tab, make it easy to analyze cycles by, for example, pulling out the month from a column of dates. So if you’re variable looks like this…

Screen Shot 2014-09-25 at 9.43.17 PM

 

…you can pull out the month of the year…

new-date-var

…resulting in this:

Screen Shot 2014-09-25 at 10.22.49 PM

 

Bucket a Number

Sometimes it’s useful to turn numeric data into categorical data by bucketing. So this…

Screen Shot 2014-09-28 at 5.52.34 PM

…can be turned into this…

Screen Shot 2014-09-28 at 6.00.11 PM

 

…by using the “Bucket a Number” tab:

Screen Shot 2014-12-18 at 4.41.57 PM

 

 

 

Variable by Filters

Use the familiar filter interface from Statwing to identify a specific group and assign it either (1) a specific value like “True” or (2) a value grabbed from another variable.

For example, this set of filters will classify men with a Bachelor’s or Graduate degree as “Highly educated men”, similarly educated women as “Highly educated women”, and any respondent to whom neither of those applies will be classified as “Other”:

Screen Shot 2014-12-18 at 4.46.08 PM

Instead of assigning new categories like “Highly educated men”, we could have pulled in the value from a variable, here pulling in whatever value was in the “Religious preference” variable for that respondent:

Screen Shot 2014-12-18 at 4.48.09 PM