--- tags: analysis, macmanes, respiomatry --- # Do some chambers have higher SD?? This is done on "ovenight.csv" which was 2 days recodring of empty cages, without water, baseline calc every 3rd cage. The code I used ``` waterrandomsampling %>% group_by(Animal) %>% summarise(mean = mean(SD_VO2)) ``` ### These are the average SDs of the cages ##### Water: there does not seem to be any specific cages with higher variance A0=0.01819597 A1=0.01754496 A2=0.01879907 A3=0.01838103 A4=0.01769739 A5=0.01773998 A6=0.01806375 ##### CO2: there does not seem to be any specific cages with higher variance A0=0.009228428 A1=0.0090518 A2=0.009335356 A3=0.010005 A4=0.009171164 A5=0.009080128 A6=0.009827624 ##### O2: there does not seem to be any specific cages with higher variance A0=0.02458157 A1=0.02286483 A2=0.02571785 A3=0.02597231 A4=0.02727174 A5=0.02393258 A6=0.02677925 ### Temporal Patterns: these are plots of the SD over time. ##### Water: none ![](https://i.imgur.com/WrHLRvm.png) ##### CO2: none ![](https://i.imgur.com/VZSzWwS.png) ##### O2: there does seem to be higher variability early on in recording?? ![](https://i.imgur.com/CLDUGFj.png)