Hi all,
I have a model goes:
formula_final2_mlm ← bf(ztheta_P ~ zthetaB_G * mo(Attain_Lvl) + zthetaB_A + zthetaB_B + zthetaB_C + District + Gender + YearG * zlog_hp + z_resource + (1|Sch_ID))
test_final2_mlm ← brm(
formula = formula_final2_mlm,
data = final_imputed_datasets[[1]],
prior = my_priors,
chains = 4,
cores = 4,
iter = 1000,
warmup = 500,
control = list(adapt_delta = 0.99, max_treedepth = 15),
file = “models/test_final2_mlm”
)
and in the output I have
zthetaB_G -0.27 0.31 -1.03 0.15 1.00 1073 942
moAttain_Lvl 0.06 0.09 -0.14 0.24 1.00 1211 1005
moAttain_Lvl:zthetaB_G 0.28 0.13 0.06 0.55 1.00 1169 1004
I’m a bit confused of how to interpret mo() effect here, so the context is that thetaB_G is teacher’s general belief and attain_lvl is teacher’s perceived classroom attainment level (from 1-5). I was wondering the 0.28 represents that for each threshold averagely the beta increase 0.28 more when controlling teacher’s belief, or it means from 1-5, the cumulative beta increase is 0.28 averagely?
Also I have the conditional effect here (bottom left), seems like it should be each threshold 0.28 as for attainment = 5 the slope is quite steep. I just want to confirm I understand it correct?
Many thanks,
Chi