Hi,
I have a set of subjects in a database, with their symptoms logged on a given date in one of the 4 columns. Outlining this below with some simplifications.
Now the first symptom (fever) is by far the most frequent and it is logged pretty much every day - sometimes even twice or more (see last 2 rows), if there has been a manual entry error / duplicate. The rest of the symptoms are hit and miss, they are logged very irregularly, and even if they were occurring on the same day as fever, they may be logged in a wrong row (see "cramps" on 03-Jan-2020 logged in the same row as "fever" on 02-Jan-2020).
Some days are symptom-free, so no entries on that day whatsoever (e.g. for the first subject below there's a week without any symptoms between 04 Jan and 10 Jan 20).
I'm trying to see whether I can build a clustered column or maybe some other chart type, which would have on horizontal axis a custom data range (let's say 01 Jan 20 - 01 Apr 20), and then 0...4 bars on each day, depending on whether some of the symptoms were logged that day. These bars should be colored differently so that it's readily visible whether there were a lot of symptom free days for that subject in a custom data range... is it possible with the dataset in question? I don't need different height for the bars if events were logged multiple times throughout a given day - as long as a symptom has been logged at least once on a given day, I just need a bar...
I have a set of subjects in a database, with their symptoms logged on a given date in one of the 4 columns. Outlining this below with some simplifications.
Now the first symptom (fever) is by far the most frequent and it is logged pretty much every day - sometimes even twice or more (see last 2 rows), if there has been a manual entry error / duplicate. The rest of the symptoms are hit and miss, they are logged very irregularly, and even if they were occurring on the same day as fever, they may be logged in a wrong row (see "cramps" on 03-Jan-2020 logged in the same row as "fever" on 02-Jan-2020).
Some days are symptom-free, so no entries on that day whatsoever (e.g. for the first subject below there's a week without any symptoms between 04 Jan and 10 Jan 20).
I'm trying to see whether I can build a clustered column or maybe some other chart type, which would have on horizontal axis a custom data range (let's say 01 Jan 20 - 01 Apr 20), and then 0...4 bars on each day, depending on whether some of the symptoms were logged that day. These bars should be colored differently so that it's readily visible whether there were a lot of symptom free days for that subject in a custom data range... is it possible with the dataset in question? I don't need different height for the bars if events were logged multiple times throughout a given day - as long as a symptom has been logged at least once on a given day, I just need a bar...
Subject | Fever | Nausea | Cramps | Runny nose |
001 | 01-Jan-2020 | |||
001 | 02-Jan-2020 | 03-Jan-2020 | ||
001 | 03-Jan-2020 | |||
001 | 11-Jan-2020 | 11-Jan-2020 | ||
001 | 01-Feb-2020 | |||
002 | 01-Mar-2020 | 01-Mar-2020 | ||
002 | 02-Mar-2020 | 02-Mar-2020 | ||
002 | 03-Mar-2020 | |||
002 | 03-Mar-2020 | 05-Mar-2020 |