- see [[220323_112352 t1-t0 difference]] model for comparison
model 1: quasipoisson model with time 0 count (`t0LC`) as predictor/covariate
- two rows per user
- `t1`: outcome (bad, good domain count)
- `conditionC`: -0.5, 0.5 (control, treatment indicator)
- `qualityC`: -0.5, 0.5 (bad, good domain indicator)
- `t0LC`: time0 count (bad, good domain count; log-transformed then centered)
```r
> m <- feglm(t1 ~ conditionC * qualityC * t0LC | block, dt1, family = "quasipoisson", vcov = ~ screen_name)
NOTE: 266 fixed-effects (2,910 observations) removed because of only 0 outcomes.
> m
GLM estimation, family = quasipoisson, Dep. Var.: t1
Observations: 62,866
Fixed-effects: block: 5,158
Standard-errors: Clustered (screen_name)
Estimate Std. Error t value Pr(>|t|)
conditionC 0.053382 0.036337 1.469054 0.141828
qualityC 2.415746 0.032736 73.795682 < 2.2e-16 ***
t0LC 0.986016 0.016214 60.813525 < 2.2e-16 ***
conditionC:qualityC -0.194418 0.064484 -3.014978 0.002572 **
conditionC:t0LC 0.016316 0.029032 0.562013 0.574111
qualityC:t0LC -1.005236 0.027283 -36.845423 < 2.2e-16 ***
conditionC:qualityC:t0LC -0.010885 0.054878 -0.198359 0.842766
```
model 2: quasipoisson model with time 0 count (`t0_badLC`, `t0_goodLC`) as separate predictors
- two rows per user
- `t0_badLC`: time0 bad domain count (log-transformed then centered)
- `t0_goodLC`: time0 good domain count (log-transformed then centered)
```r
> m <- feglm(t1 ~ conditionC * qualityC * (t0_badLC + t0_goodLC) | block, dt2, family = "quasipoisson", vcov = ~ screen_name)
NOTE: 266 fixed-effects (2,910 observations) removed because of only 0 outcomes.
> summary(m)
GLM estimation, family = quasipoisson, Dep. Var.: t1
Observations: 62,866
Fixed-effects: block: 5,158
Standard-errors: Clustered (screen_name)
Estimate Std. Error t value Pr(>|t|)
conditionC 0.055417 0.048358 1.145983 0.251811
qualityC 3.612403 0.044214 81.702289 < 2.2e-16 ***
t0_badLC 0.727530 0.017496 41.582303 < 2.2e-16 ***
t0_goodLC 0.236764 0.021528 10.997869 < 2.2e-16 ***
conditionC:qualityC -0.185196 0.088616 -2.089875 0.036637 *
conditionC:t0_badLC 0.040933 0.034325 1.192500 0.233074
conditionC:t0_goodLC -0.013279 0.039269 -0.338141 0.735259
qualityC:t0_badLC -1.521137 0.028291 -53.766779 < 2.2e-16 ***
qualityC:t0_goodLC 0.497291 0.036347 13.681918 < 2.2e-16 ***
conditionC:qualityC:t0_badLC -0.010142 0.057090 -0.177643 0.859004
conditionC:qualityC:t0_goodLC 0.042030 0.072963 0.576050 0.564585
```