Statistics Playground

import os
files = []
for root, _, fnames in os.walk("No Videos"):
    files.extend([os.path.join(root, filename) for filename in fnames if filename.endswith('.csv')])

%%capture
import pandas as pd
df = pd.DataFrame()
list_ = []
for f in files:
    fpd = pd.read_csv(f,index_col=None, header=0)
    fpd.participantId = os.path.dirname(f).split('/')[-1]
    list_.append(fpd)
frame = pd.concat(list_)

frame.head()

	participantId	condition	timeSinceStartup	correctNodeHit	keypressed	calibrationData	bubbleSize	numberNodes	targetNode	currentSelectedNode	currentState	correctedEyeX	correctedEyeY	rawEyeX	rawEyeY
0	1	NaN	3.825378	NaN	HAPRING_TIP	noCalibrationDataSet	NaN	0.0	NaN	-1.0	Introduction	0.0	0.0	960.0	540.0
1	1	NaN	7.018346	NaN	HAPRING_TIP	noCalibrationDataSet	NaN	0.0	NaN	-1.0	Introduction	0.0	0.0	960.0	540.0
2	1	NaN	7.033599	NaN	NaN	noCalibrationDataSet	NaN	0.0	NaN	-1.0	Introduction	0.0	0.0	960.0	540.0
3	1	NaN	7.034281	NaN	NaN	noCalibrationDataSet	NaN	0.0	NaN	-1.0	Introduction	0.0	0.0	960.0	540.0
4	1	NaN	7.036498	NaN	NaN	noCalibrationDataSet	NaN	0.0	NaN	-1.0	Introduction	0.0	0.0	960.0	540.0

frame.condition.unique()

array([nan, 'WITHCUSTOMCALIB', 'MOUSE', 'EYE', 'noCalibrationDataSet'], dtype=object)

import numpy as np
df  = pd.DataFrame(frame, columns=['participantId', 'condition','timeSinceStartup']).groupby(['participantId', 'condition'])

total_time_df = df.agg({'condition': 'first','timeSinceStartup': lambda x: max(x) - min(x)})
total_time_df = total_time_df[total_time_df.condition.isin(['WITHCUSTOMCALIB', 'MOUSE', 'EYE'])]

import matplotlib.pyplot as plt
plt.style.use('ggplot')

pd.DataFrame.boxplot(total_time_df, by='condition')
plt.show()