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

Tuesday, November 20, 2018 edit

Model 3 With Global North vs Global South edit

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
Return to "Reading time/Work log/2018-11-20" page.