#Github integration

Enrichment Data Analysis in R

This data has already been analyzed in Excel. Now, we are going to recruit the power of R to do a deeper analysis.

The raw data was loaded up into R by splitting the master data by week and loading it using the import dataset function. We will utilize the pipe function when selecting the specific tests.

Let’s first load the data and run some simple summaries.

Load Data

Calculate the percent error for time

Calculating this to determine if we need to re-run video in Ethovision and also understand the how the percent error might affect any conclusions we make.

##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
## -64.733  -3.789  -1.983  -3.721  -0.875   0.000
##     Min.  1st Qu.   Median     Mean  3rd Qu.     Max. 
## -50.3556  -2.7083  -1.3444  -2.4108  -0.2667   0.0000
##     Min.  1st Qu.   Median     Mean  3rd Qu.     Max. 
## -96.8000  -2.7611  -0.8778  -3.9922  -0.2222   0.0000

Calculate the percent error for time 2022 sample

Percentage error calculations for Pre-enrichment data samples

##      Min.   1st Qu.    Median      Mean   3rd Qu.      Max. 
## -89.38422  -0.40067   0.02523  -0.57817   0.54389  21.92253
##     Min.  1st Qu.   Median     Mean  3rd Qu.     Max. 
## -21.9918  -0.6498  -0.4106  -1.4753  -0.2553   0.2116

Percentage error calculations for Preference Testing

Novel Tank test

Calculate time spent at Novel Tank Zone

Novel Object test

Data wrangling and sorting

Clean up the data based on the necessary values. I’m keeping the Weeks separate so we can compare with the 2022 data easily + one r function away from combining all.

For 2022 samples

Split the data by Retention interval.Cleaned data has all variables as Characters except for Object_Zone_1 & 2. I’ll calculate percent time first for each zone and sort based on object side.

For 2022 data we can use the FO test to measure the object side preference so we won’t split the data based on side. Calculating percent time on each side.

##Preference testing 2022