Alteryx Designer Desktop Discussions

Find answers, ask questions, and share expertise about Alteryx Designer Desktop and Intelligence Suite.
SOLVED

Null value and incorrect value

elephant
7 - Meteor

Hi, I having two column of data one is income and one is spending on food. For my spending on food column I have data that is null and I had data that look incorrect (number in bold) for null value i using imputation tool to input the average value am i doing it correctly? For the incorrect (number in bold) data any advice of what I should do? Here a sample of data

 

Income        Spending on food

10900          890

2400            950

17000          1490

9700             9

15900          1310

10700

11400          38

6000            2000

5609            

4050             1

7 REPLIES 7
binuacs
20 - Arcturus

@elephant something like below? it replaces the NULL to 0

image.png

Qiu
21 - Polaris
21 - Polaris

@elephant 
How do you define "data that look incorrect"?

elephant
7 - Meteor

Like earn 5000 but spend only 8 

caltang
17 - Castor
17 - Castor

Err... how is that incorrect?

Calvin Tang
Alteryx ACE
https://www.linkedin.com/in/calvintangkw/
elephant
7 - Meteor

like it earn 9700 but spend 8 isnt that consider an incorrect or not accurate data?

caltang
17 - Castor
17 - Castor

What... high income people can''t be frugal on food? Haha!

 

Let's circle back to your rawest form of data. Can you provide a sample? Let's have a look.

Calvin Tang
Alteryx ACE
https://www.linkedin.com/in/calvintangkw/
AndrewDMerrill
13 - Pulsar

You have a few options that you can take, all of which require you to define what "looks incorrect" means. As an example, ratio ([food cost]/[Income] < 1%):

  1. You can just drop these row entirely using the Filter Tool
  2. If possible/feasible, you can contact whoever collected the data in an attempt to correct these seemingly incorrect data points, replacing the values with the correct amounts when proper information is provided.
  3. You can use the Formula Tool to set these values to NULL, and treat them like other NULL values replacing with the average.

Which option you choose, depends on how much data you have, what your use case is, and how much data is affected.

Labels