- see [[220331_174150 absolute count badness|badness absolute count]], [[220331_183210 fraction bad|fraction badness]] # parameters - bad/goodness measures: fc, mbfc (factual, bias, mean(factual, bias), min(factual_bias)), afm (reliable, bias, mean(reliable, bias), min(reliable, bias)) (see [[different measures of domain quality]]) - goodness thresholds: $[5, 10, 15...90, 95]$ - values **below** threshold are set to $100$ (maximum badness) - values **above** threshold are scaled $[99...0]$ (decreasing badness) - outcome: **summed** badness (range: $[0, \infty]$) - winsorize user summed values (both at t1 and t0): $[0.85, 0.90, 0.95, 0.99, 1.00]$ (1.00 = no winsorizing) - winsorizing applied **before** *filling in missing values with 0* ## bad/goodness threshold examples ![[dv_mbfc_badness_threshold70.png|600]] ![[dv_fc_badness_threshold15.png|600]] # results models - sum badness, **quasipoisson** model: `feglm(t1 ~ conditionC * t0LC | block, dt1, family = "quasipoisson", vcov = "HC1")` - `conditionC` $[-0.5, 0.5]$ - `t0` time0/pre-campaign badness (not shown because all estimates are highly significant) - sum badness: `log(t0 + 1)` then mean-centered - `block` as fixed effect - red: statistically significant effect (p < .05) ![[winsorize-1.00.png]] ![[winsorize-0.99.png]] ![[winsorize-0.95.png]] ![[winsorize-0.90.png]] ![[winsorize-0.85.png]]