CluelessFellowExcelor
New Member
- Joined
- Jun 15, 2023
- Messages
- 3
- Office Version
- 365
- 2021
- 2019
- Platform
- Windows
Dear all,
I have a huge problem and a huge excel file of thousands of rows of information.
There are many "participants" with test results taken on different dates. Each participant can have multiple test done on different dates and these are all listed in multiple rows. Each participant is separated by a thick line.
I need each participant and their results all in 1 row instead of multiple rows. This is also to say, if i had 1000 participants, i wish to have 1000 rows, each row is a unique patient, unlike what it is now.
An example of my nightmare is attached in 1 image, and the ideal end result in the 2nd image.
Question:
-Is there a way or do i need to do it manually? I don't think the transpose functions will work well.
-For those well versed with statistical analysis, whether R or SPSS, is it better to have 1 unique patient's result all in 1 row instead of what it is now?
Thank you
EDIT:
The results from the least recent from left to the most recent to the right
I have a huge problem and a huge excel file of thousands of rows of information.
There are many "participants" with test results taken on different dates. Each participant can have multiple test done on different dates and these are all listed in multiple rows. Each participant is separated by a thick line.
I need each participant and their results all in 1 row instead of multiple rows. This is also to say, if i had 1000 participants, i wish to have 1000 rows, each row is a unique patient, unlike what it is now.
An example of my nightmare is attached in 1 image, and the ideal end result in the 2nd image.
Question:
-Is there a way or do i need to do it manually? I don't think the transpose functions will work well.
-For those well versed with statistical analysis, whether R or SPSS, is it better to have 1 unique patient's result all in 1 row instead of what it is now?
Thank you
EDIT:
The results from the least recent from left to the most recent to the right
Attachments
Last edited by a moderator: