Research talk:Reading time/Work log/2018-11-20

Tuesday, November 20, 2018

Model 3 With Global North vs Global South

TODO: Look at the distribution of visiblelength / totallength in last in session only.

TODO: Compare total length to visible length.

TODO: Make the ME plots with weights.

TODO: Add firstpaint/total_length as proxy for connection speed. There is a bit of a problem with this because the model matrix is not full rank when I add this ratio. Adding the interaction doesn't make

Statistical models
	model 2	model 3
Intercept	9.6734 (0.0077)^***	9.6587 (0.0077)^***
mobile	0.1172 (0.0015)^***	0.1549 (0.0016)^***
Revision length (bytes)	0.1690 (0.0005)^***	0.1685 (0.0005)^***
time to first paint	-0.0167 (0.0006)^***	-0.0169 (0.0006)^***
time to dom interactive	0.0031 (0.0009)^***	0.0031 (0.0009)^***
sessionlength	-0.0001 (0.0000)^***	-0.0001 (0.0000)^***
lastinsessionTRUE	0.9441 (0.0015)^***	0.9385 (0.0015)^***
nthinsession	0.0002 (0.0000)^***	0.0002 (0.0000)^***
dayofweekMon	0.0878 (0.0020)^***	0.0877 (0.0020)^***
dayofweekSat	0.0154 (0.0020)^***	0.0134 (0.0020)^***
dayofweekSun	0.0328 (0.0020)^***	0.0312 (0.0020)^***
dayofweekThu	0.0497 (0.0020)^***	0.0495 (0.0020)^***
dayofweekTue	0.0291 (0.0020)^***	0.0290 (0.0020)^***
dayofweekWed	0.0697 (0.0019)^***	0.0694 (0.0019)^***
usermonth4	0.0086 (0.0096)	0.0085 (0.0096)
usermonth5	0.0101 (0.0096)	0.0098 (0.0096)
usermonth6	-0.0098 (0.0098)	-0.0100 (0.0098)
usermonth7	-0.0482 (0.0098)^***	-0.0487 (0.0098)^***
usermonth8	-0.0131 (0.0097)	-0.0138 (0.0097)
usermonth9	0.0375 (0.0076)^***	0.0378 (0.0076)^***
usermonth10	0.0047 (0.0075)	0.0030 (0.0075)
GlobalSouthTRUE	0.1305 (0.0012)^***	0.2092 (0.0017)^***
mobileTRUE:lastinsessionTRUE	-0.6616 (0.0021)^***	-0.6541 (0.0021)^***
R²	0.0616	0.0620
Adj. R²	0.0616	0.0620
Num. obs.	9873641	9873641
RMSE	14.3130	14.3102
^*p < 0.001, ^p < 0.01, ^*p < 0.05

Add topic