You can now create new variables for analysis in Statwing:
|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”
Or use similar tools to clean up dirty data:
|Grouping categories together||f & female & Female → “Female”|
|Filtering out errant data||If Date is before 1/1/1900
then → (delete)
else → (keep)
Let’s walk through some examples to get a feel for it. We’ll use a dataset of eleven years of point of sale data, where each row of data is one purchase.
Creating new variables from a time variable
Each purchase happened on a specific date. So if you Described the variable Purchase Date, you’d see:
What if you wanted to know if more purchases happened on Sundays than on Mondays? Or in January versus February across all years of data?
Select one of Statwing’s time-specific functions to create a new variable, in this case tranforming dates like “2/3/2005” or “6/13/2013” to months like “February” and “June”…
…resulting in this:
With hand-entered data, it’s not uncommon to see messy data like this:
Select the variable, then “Clean”, then drag the categories into the correct positions:
So now your data will look like this:
Sometimes it’s useful to turn numeric data into categorical data by bucketing. So this…
…can now easily be turned into this:
You get the picture. As always, our goal is to make data analysis delightful and efficient.
To try these new tools out, sign in, upload some data (or use a sample dataset) and then use the “Clean” and “Create” buttons on the left sidebar.