S1 Main analysis

EPDS

# predict epds by time and group
mod_epds1 <- mmrm(epds ~ time * group + epds_pre + us(time|group/id), 
                  method = "Kenward-Roger", data = data)

# switch time factor
data$time2 <- factor(data$time, levels = c("T2","T1"))
mod_epds1.1 <- mmrm(epds ~ time2 * group + epds_pre + us(time2|group/id), 
                  method = "Kenward-Roger", data = data)

kable(tidy(mod_epds1, conf.int = T), caption = "Model 1: EPDS (Time Ref = T1)")
Model 1: EPDS (Time Ref = T1)
term estimate std.error df statistic p.value conf.low conf.high
(Intercept) 4.8345099 1.0829927 73.28262 4.464028 0.0000286 2.6762489 6.9927709
epds_pre 0.6113849 0.0803578 70.23853 7.608279 0.0000000 0.4511258 0.7716440
groupExp -1.6862586 1.0636283 73.26451 -1.585383 0.1171864 -3.8059378 0.4334206
timeT2 -1.5333247 0.5716504 42.42001 -2.682277 0.0103776 -2.6866237 -0.3800257
timeT2:groupExp 2.9063478 0.9969495 58.68121 2.915241 0.0050254 0.9112297 4.9014659
kable(tidy(mod_epds1.1, conf.int = T), caption = "Model 2: EPDS (Time Ref = T2)")
Model 2: EPDS (Time Ref = T2)
term estimate std.error df statistic p.value conf.low conf.high
(Intercept) 3.3012441 1.1585227 77.61833 2.849529 0.0056041 0.9946239 5.6078642
epds_pre 0.6113803 0.0804032 70.23617 7.603927 0.0000000 0.4510306 0.7717301
groupExp 1.2200826 1.0957489 73.74985 1.113469 0.2691229 -0.9633677 3.4035330
time2T1 1.5333059 0.5716188 42.41970 2.682392 0.0103746 0.3800705 2.6865414
time2T1:groupExp -2.9063275 0.9960972 58.67930 -2.917715 0.0049908 -4.8997413 -0.9129138



Post-hoc power (as requested during peer-review)
vc_list <- VarCorr(mod_epds1)
Sigma_a <- as.matrix(vc_list[["Ctrl"]])
Sigma_b <- as.matrix(vc_list[["Exp"]])
Ra <- cov2cor(Sigma_a)
Rb <- cov2cor(Sigma_b)

sigmaa <- sqrt(Sigma_a[2, 2])
sigmab <- sqrt(Sigma_b[2, 2])

beta <- coef(mod_epds1)
epds_pre_mean <- mean(data$epds_pre, na.rm = TRUE)
new_a <- data.frame(time = factor("T2", levels = levels(data$time)),
                    group = factor("Ctrl", levels = levels(data$group)),
                    epds_pre = epds_pre_mean)
new_b <- data.frame(time = factor("T2", levels = levels(data$time)),
                    group = factor("Exp", levels = levels(data$group)),
                    epds_pre = epds_pre_mean)
X_a <- model.matrix(~ time * group + epds_pre, new_a)
X_b <- model.matrix(~ time * group + epds_pre, new_b)
X_a_use <- X_a[, names(beta), drop = FALSE]
X_b_use <- X_b[, names(beta), drop = FALSE]
mu_a <- as.numeric(X_a_use %*% beta)
mu_b <- as.numeric(X_b_use %*% beta)
delta <- abs(mu_b - mu_a)       

power.mmrm(
  N        = length(unique(data$id)),
  Ra       = Ra,
  ra       = rep(1,2),
  sigmaa   = sigmaa,
  Rb       = Rb,
  rb       = rep(1,2),
  sigmab   = sigmab,
  lambda   = 0.7924,
  delta    = delta,
  sig.level = 0.05
)
## 
##      Power for Mixed Model of Repeated Measures (Lu, Luo, & Chen, 2008) 
## 
##              n1 = 41.99844
##              n2 = 53.00156
##      retention1 = 1, 1
##      retention2 = 1, 1
##           delta = 1.220089
##       sig.level = 0.05
##           power = 0.2244424
##     alternative = two.sided





PSI

# predict psi by time and group
mod_psi1 <- mmrm(psi ~ time * group + psi_pre + us(time|group/id), 
                  method = "Kenward-Roger", data = data)

# switch time factor
data$time2 <- factor(data$time, levels = c("T2","T1"))
mod_psi1.1 <- mmrm(psi ~ time2 * group + psi_pre + us(time2|group/id), 
                  method = "Kenward-Roger", data = data)

kable(tidy(mod_psi1, conf.int = TRUE), caption = "Model 1: PSI (Time Ref = T1)")
Model 1: PSI (Time Ref = T1)
term estimate std.error df statistic p.value conf.low conf.high
(Intercept) 17.0569094 5.5292976 69.05806 3.0848239 0.0029285 6.0264279 28.0873910
groupExp -4.4497214 3.2142914 66.68982 -1.3843553 0.1708646 -10.8660225 1.9665797
psi_pre 0.8221064 0.0718433 64.48998 11.4430572 0.0000000 0.6786041 0.9656088
timeT2 -1.5399587 2.2536148 37.97122 -0.6833283 0.4985481 -6.1022768 3.0223594
timeT2:groupExp 4.1741587 3.4110506 62.27455 1.2237164 0.2256692 -2.6438364 10.9921537
kable(tidy(mod_psi1.1, conf.int = TRUE), caption = "Model 2: PSI (Time Ref = T2)")
Model 2: PSI (Time Ref = T2)
term estimate std.error df statistic p.value conf.low conf.high
(Intercept) 15.5170180 5.6879908 78.81181 2.7280315 0.0078525 4.1949351 26.8391009
groupExp -0.2755245 3.3187190 70.37184 -0.0830214 0.9340703 -6.8938861 6.3428370
psi_pre 0.8221055 0.0718005 64.49372 11.4498630 0.0000000 0.6786888 0.9655223
time2T1 1.5399608 2.2599124 37.97311 0.6814250 0.4997376 -3.0350992 6.1150207
time2T1:groupExp -4.1742015 3.4119625 62.27506 -1.2234019 0.2257870 -10.9940182 2.6456152





S2 Moderation of treatment effects

EPDS

Child age
mod_epds2 <- mmrm(epds ~ time * group * age_c + epds_pre + us(time|group/id), 
                  method = "Kenward-Roger", data = data)

mod_epds2.1 <- mmrm(epds ~ time2 * group * age_c + epds_pre + us(time2|group/id), 
                  method = "Kenward-Roger", data = data)

kable(tidy(mod_epds2, conf.int = T), caption = "Model 1: EPDS (Time Ref = T1)")
Model 1: EPDS (Time Ref = T1)
term estimate std.error df statistic p.value conf.low conf.high
(Intercept) 6.3088895 1.6035977 50.33955 3.9342097 0.0002563 3.0885070 9.5292720
age_c -0.1799192 0.2039350 41.52183 -0.8822379 0.3827273 -0.5916171 0.2317788
epds_pre 0.5927851 0.0856536 70.48612 6.9207299 0.0000000 0.4219752 0.7635951
groupExp -2.7597139 2.1837358 67.70717 -1.2637581 0.2106507 -7.1176322 1.5982044
groupExp:age_c 0.1592669 0.3367016 63.95716 0.4730210 0.6378086 -0.5133805 0.8319144
timeT2 -0.9439212 1.2047680 40.05222 -0.7834880 0.4379494 -3.3787493 1.4909068
timeT2:age_c -0.1030553 0.1725217 39.96322 -0.5973468 0.5536456 -0.4517446 0.2456341
timeT2:groupExp 0.9811894 2.0219665 56.03339 0.4852649 0.6293804 -3.0692430 5.0316218
timeT2:groupExp:age_c 0.3279749 0.3107376 51.79220 1.0554722 0.2961082 -0.2956253 0.9515751
kable(tidy(mod_epds2.1, conf.int = T), caption = "Model 2: EPDS (Time Ref = T2)")
Model 2: EPDS (Time Ref = T2)
term estimate std.error df statistic p.value conf.low conf.high
(Intercept) 5.3650116 1.8046190 49.04252 2.9729331 0.0045629 1.7385735 8.9914498
age_c -0.2829736 0.2353758 40.04442 -1.2022201 0.2363394 -0.7586695 0.1927223
epds_pre 0.5927803 0.0856856 70.49276 6.9180896 0.0000000 0.4219069 0.7636538
groupExp -1.7784749 2.2515573 68.46367 -0.7898866 0.4323213 -6.2708367 2.7138869
groupExp:age_c 0.4872435 0.3409095 66.74499 1.4292458 0.1575974 -0.1932625 1.1677494
time2T1 0.9439205 1.2050273 40.05680 0.7833188 0.4380470 -1.4914229 3.3792640
time2T1:age_c 0.1030561 0.1725601 39.96776 0.5972186 0.5537300 -0.2457096 0.4518217
time2T1:groupExp -0.9812475 2.0202915 56.02585 -0.4856960 0.6290769 -5.0283365 3.0658416
time2T1:groupExp:age_c -0.3279733 0.3103930 51.78443 -1.0566389 0.2955810 -0.9508842 0.2949376





App usage
mod_epds4 <- mmrm(epds ~ time * appuse + epds_pre + us(time|group/id), 
                  method = "Kenward-Roger", data = data[data$group == "Exp",])

mod_epds4.1 <- mmrm(epds ~ time2 * appuse + epds_pre + us(time2|group/id), 
                  method = "Kenward-Roger", data = data[data$group == "Exp",])

kable(tidy(mod_epds4, conf.int = T), caption = "Model 1: EPDS (Time Ref = T1)")
Model 1: EPDS (Time Ref = T1)
term estimate std.error df statistic p.value conf.low conf.high
(Intercept) 1.6085632 1.4805763 37.70829 1.0864440 0.2841791 -1.3894684 4.6065948
appuse 0.0051391 0.0054722 34.43259 0.9391212 0.3542101 -0.0059767 0.0162549
epds_pre 0.7008865 0.1071593 30.88264 6.5406030 0.0000003 0.4823000 0.9194731
timeT2 1.9413789 0.9828136 30.39130 1.9753276 0.0573738 -0.0647112 3.9474690
timeT2:appuse -0.0059432 0.0053298 28.48546 -1.1150803 0.2741355 -0.0168524 0.0049661
kable(tidy(mod_epds4.1, conf.int = T), caption = "Model 2: EPDS (Time Ref = T2)")
Model 2: EPDS (Time Ref = T2)
term estimate std.error df statistic p.value conf.low conf.high
(Intercept) 3.5499495 1.4850485 35.83999 2.3904603 0.0222068 0.5376645 6.5622344
appuse -0.0008041 0.0051387 29.62659 -0.1564764 0.8767190 -0.0113044 0.0096962
epds_pre 0.7008867 0.1072132 30.88351 6.5373152 0.0000003 0.4821905 0.9195830
time2T1 -1.9413871 0.9822597 30.39146 -1.9764500 0.0572404 -3.9463461 0.0635718
time2T1:appuse 0.0059432 0.0053226 28.48558 1.1165848 0.2735022 -0.0049514 0.0168378





PSI

Child age
mod_psi2 <- mmrm(psi ~ time * group * age_c + psi_pre + us(time|group/id), 
                  method = "Kenward-Roger", data = data)

mod_psi2.1 <- mmrm(psi ~ time2 * group * age_c + psi_pre + us(time2|group/id), 
                  method = "Kenward-Roger", data = data)

kable(tidy(mod_psi2, conf.int = T), caption = "Model 1: PSI (Time Ref = T1)")
Model 1: PSI (Time Ref = T1)
term estimate std.error df statistic p.value conf.low conf.high
(Intercept) 27.3439401 7.1725131 61.01591 3.812324 0.0003230 13.0016895 41.6861908
age_c -1.1873323 0.5717774 36.67133 -2.076564 0.0449058 -2.3462139 -0.0284506
groupExp -18.0961908 6.7332102 65.79453 -2.687602 0.0091051 -31.5402622 -4.6521193
groupExp:age_c 2.3791818 1.0499076 64.42569 2.266087 0.0268130 0.2820172 4.4763465
psi_pre 0.7838814 0.0807940 65.38819 9.702220 0.0000000 0.6225427 0.9452200
timeT2 -8.3219749 4.5294544 35.30249 -1.837302 0.0745949 -17.5144406 0.8704909
timeT2:age_c 1.0507986 0.6488716 35.18834 1.619425 0.1142852 -0.2662287 2.3678260
timeT2:groupExp 14.5906683 6.8406384 58.13177 2.132940 0.0371633 0.8983037 28.2830329
timeT2:groupExp:age_c -1.7747768 1.0404203 55.15658 -1.705827 0.0936662 -3.8596927 0.3101391
kable(tidy(mod_psi2.1, conf.int = T), caption = "Model 2: PSI (Time Ref = T2)")
Model 2: PSI (Time Ref = T2)
term estimate std.error df statistic p.value conf.low conf.high
(Intercept) 19.0216982 7.7435764 75.88649 2.4564487 0.0163173 3.5986560 34.4447403
age_c -0.1365327 0.7121332 41.88816 -0.1917236 0.8488851 -1.5737895 1.3007240
groupExp -3.5053655 7.2695999 71.69781 -0.4821951 0.6311362 -17.9980916 10.9873606
groupExp:age_c 0.6043814 1.1112102 69.27240 0.5438947 0.5882602 -1.6122669 2.8210297
psi_pre 0.7838850 0.0806435 65.39032 9.7203729 0.0000000 0.6228470 0.9449230
time2T1 8.3219651 4.5446022 35.30288 1.8311757 0.0755247 -0.9012391 17.5451693
time2T1:age_c -1.0507967 0.6510979 35.18873 -1.6138844 0.1154856 -2.3723423 0.2707489
time2T1:groupExp -14.5907276 6.8478089 58.13400 -2.1307148 0.0373541 -28.2974337 -0.8840215
time2T1:groupExp:age_c 1.7747819 1.0412266 55.15886 1.7045107 0.0939131 -0.3117478 3.8613115





App usage
mod_psi4 <- mmrm(psi ~ time * appuse + psi_pre + us(time|group/id), 
                  method = "Kenward-Roger", data = data[data$group == "Exp",])

mod_psi4.1 <- mmrm(psi ~ time2 * appuse + psi_pre + us(time2|group/id), 
                  method = "Kenward-Roger", data = data[data$group == "Exp",])

kable(tidy(mod_psi4, conf.int = T), caption = "Model 1: PSI (Time Ref = T1)")
Model 1: PSI (Time Ref = T1)
term estimate std.error df statistic p.value conf.low conf.high
(Intercept) 6.9417312 7.5332336 32.64179 0.9214809 0.3635603 -8.3911398 22.2746021
appuse 0.0137551 0.0172156 33.05016 0.7989925 0.4299996 -0.0212683 0.0487785
psi_pre 0.8814760 0.0970092 28.08793 9.0865194 0.0000000 0.6827897 1.0801623
timeT2 3.5617274 3.1238880 28.10896 1.1401585 0.2638426 -2.8361498 9.9596045
timeT2:appuse -0.0085993 0.0168811 26.52025 -0.5094062 0.6146804 -0.0432659 0.0260672
kable(tidy(mod_psi4.1, conf.int = T), caption = "Model 2: PSI (Time Ref = T2)")
Model 2: PSI (Time Ref = T2)
term estimate std.error df statistic p.value conf.low conf.high
(Intercept) 10.5034691 7.3033213 28.23920 1.4381771 0.1613748 -4.4509995 25.4579376
appuse 0.0051558 0.0156809 27.62205 0.3287967 0.7447893 -0.0269848 0.0372965
psi_pre 0.8814757 0.0971517 28.08765 9.0731923 0.0000000 0.6824975 1.0804539
time2T1 -3.5617175 3.1186119 28.10892 -1.1420843 0.2630547 -9.9487893 2.8253543
time2T1:appuse 0.0085993 0.0168418 26.52022 0.5105930 0.6138602 -0.0259866 0.0431852





S3 Sensitivity analysis

EPDS

excluding participants with no app usage

mod_epds5 <- mmrm(epds ~ time * group + epds_pre + us(time|group/id), 
                  method = "Kenward-Roger", data = data[data$appuse > 0 | is.na(data$appuse),])
mod_epds5.1 <- mmrm(epds ~ time2 * group + epds_pre + us(time2|group/id), 
                  method = "Kenward-Roger", data = data[data$appuse > 0 | is.na(data$appuse),])

kable(tidy(mod_epds5, conf.int = T), caption = "Model 1: EPDS (Time Ref = T1)")
Model 1: EPDS (Time Ref = T1)
term estimate std.error df statistic p.value conf.low conf.high
(Intercept) 5.1288242 1.1216805 71.15612 4.572447 0.0000198 2.8923418 7.3653066
epds_pre 0.5830902 0.0852514 70.60164 6.839658 0.0000000 0.4130872 0.7530932
groupExp -1.3728961 1.1180550 65.86028 -1.227933 0.2238428 -3.6052534 0.8594612
timeT2 -1.5271374 0.5715379 42.43927 -2.671979 0.0106517 -2.6801941 -0.3740808
timeT2:groupExp 2.7934358 1.0545753 50.49229 2.648873 0.0107517 0.6757705 4.9111011
kable(tidy(mod_epds5.1, conf.int = T), caption = "Model 2: EPDS (Time Ref = T2)")
Model 2: EPDS (Time Ref = T2)
term estimate std.error df statistic p.value conf.low conf.high
(Intercept) 3.6017395 1.1935536 76.14739 3.017660 0.0034631 1.2246459 5.9788331
epds_pre 0.5830853 0.0852829 70.60088 6.837070 0.0000000 0.4130193 0.7531513
groupExp 1.4205311 1.1452330 68.00654 1.240386 0.2190951 -0.8647418 3.7058040
time2T1 1.5271350 0.5715243 42.44020 2.672039 0.0106500 0.3741066 2.6801635
time2T1:groupExp -2.7934198 1.0536159 50.49139 -2.651270 0.0106851 -4.9091595 -0.6776801



adding covariates

mod_epds6 <- mmrm(epds ~ time * group + epds_pre + SP + edu + citizen +
                    finaid + cts + preg_planned + us(time|group/id), 
                  method = "Kenward-Roger", data = data)
mod_epds6.1 <- mmrm(epds ~ time2 * group + epds_pre + SP + edu + citizen +
                    finaid + cts + preg_planned + us(time2|group/id), 
                  method = "Kenward-Roger", data = data)

kable(tidy(mod_epds6, conf.int = T), caption = "Model 1: EPDS (Time Ref = T1)")
Model 1: EPDS (Time Ref = T1)
term estimate std.error df statistic p.value conf.low conf.high
(Intercept) 6.9384114 3.0685558 66.00092 2.2611325 0.0270529 0.8118437 13.0649791
citizen 1.3797309 1.4811082 61.69634 0.9315531 0.3551986 -1.5812520 4.3407139
cts -0.0804287 0.1347181 58.25491 -0.5970148 0.5528109 -0.3500712 0.1892138
edu -0.1045481 0.4134496 66.37514 -0.2528679 0.8011518 -0.9299399 0.7208436
epds_pre 0.6025155 0.0868145 58.37140 6.9402607 0.0000000 0.4287609 0.7762702
finaid -3.6168370 1.3787807 55.56593 -2.6232141 0.0112219 -6.3793430 -0.8543311
groupExp -1.6097248 1.1463420 60.16748 -1.4042274 0.1653923 -3.9026189 0.6831692
preg_planned -1.8099873 1.2480103 58.16271 -1.4502984 0.1523470 -4.3080026 0.6880280
SP 0.4173183 1.4306401 59.98280 0.2917004 0.7715221 -2.4444049 3.2790416
timeT2 -1.2950867 0.6006221 37.71710 -2.1562421 0.0375076 -2.5112822 -0.0788912
timeT2:groupExp 2.1430896 1.0192843 52.23605 2.1025435 0.0403428 0.0979660 4.1882133
kable(tidy(mod_epds6.1, conf.int = T), caption = "Model 2: EPDS (Time Ref = T2)")
Model 2: EPDS (Time Ref = T2)
term estimate std.error df statistic p.value conf.low conf.high
(Intercept) 5.6432324 3.1052187 68.88105 1.8173382 0.0735149 -0.5517002 11.8381651
citizen 1.3797675 1.4832251 61.69485 0.9302483 0.3558683 -1.5854488 4.3449839
cts -0.0804268 0.1349415 58.25363 -0.5960121 0.5534761 -0.3505166 0.1896630
edu -0.1045375 0.4137413 66.37356 -0.2526639 0.8013088 -0.9305119 0.7214369
epds_pre 0.6025167 0.0869728 58.37014 6.9276444 0.0000000 0.4284452 0.7765882
finaid -3.6168144 1.3811067 55.56476 -2.6187799 0.0113523 -6.3839818 -0.8496469
groupExp 0.5333510 1.1948908 63.65568 0.4463596 0.6568522 -1.8539665 2.9206686
preg_planned -1.8100088 1.2504620 58.16139 -1.4474720 0.1531337 -4.3129327 0.6929151
SP 0.4173179 1.4326767 59.98145 0.2912855 0.7718379 -2.4484803 3.2831161
time2T1 1.2950872 0.6002718 37.71755 2.1575015 0.0374025 0.0796017 2.5105728
time2T1:groupExp -2.1430813 1.0166760 52.23608 -2.1079295 0.0398534 -4.1829714 -0.1031911





PSI

excluding participants with no app usage

mod_psi5 <- mmrm(psi ~ time * group + psi_pre + us(time|group/id), 
                  method = "Kenward-Roger", data = data[data$appuse > 0 | is.na(data$appuse),])
mod_psi5.1 <- mmrm(psi ~ time2 * group + psi_pre + us(time2|group/id), 
                  method = "Kenward-Roger", data = data[data$appuse > 0 | is.na(data$appuse),])

kable(tidy(mod_psi5, conf.int = T), caption = "Model 1: PSI (Time Ref = T1)")
Model 1: PSI (Time Ref = T1)
term estimate std.error df statistic p.value conf.low conf.high
(Intercept) 16.7647316 5.7188826 65.83608 2.9314698 0.0046349 5.3460835 28.183380
groupExp -3.6622098 3.3598121 59.92865 -1.0900043 0.2800748 -10.3829992 3.058580
psi_pre 0.8261623 0.0746454 61.80107 11.0678188 0.0000000 0.6769386 0.975386
timeT2 -1.5408675 2.2530992 37.98850 -0.6838880 0.4981966 -6.1020738 3.020339
timeT2:groupExp 2.7177914 3.3334364 58.57302 0.8153122 0.4181966 -3.9534170 9.389000
kable(tidy(mod_psi5.1, conf.int = T), caption = "Model 2: PSI (Time Ref = T2)")
Model 2: PSI (Time Ref = T2)
term estimate std.error df statistic p.value conf.low conf.high
(Intercept) 15.2241275 5.8741850 74.91084 2.5917004 0.0114755 3.521923 26.9263324
groupExp -0.9444507 3.3891420 62.53071 -0.2786696 0.7814179 -7.718106 5.8292041
psi_pre 0.8261587 0.0746044 61.80139 11.0738582 0.0000000 0.677017 0.9753003
time2T1 1.5408679 2.2594539 37.98799 0.6819648 0.4993984 -3.033205 6.1149407
time2T1:groupExp -2.7177534 3.3342209 58.57338 -0.8151090 0.4183120 -9.390531 3.9550241



adding covariates

mod_psi6 <- mmrm(psi ~ time * group + psi_pre + SP + edu + citizen +
                    finaid + cts + preg_planned + us(time|group/id), 
                  method = "Kenward-Roger", data = data)
mod_psi6.1 <- mmrm(psi ~ time2 * group + psi_pre + SP + edu + citizen +
                    finaid + cts + preg_planned + us(time2|group/id), 
                  method = "Kenward-Roger", data = data)

kable(tidy(mod_psi6, conf.int = T), caption = "Model 1: PSI (Time Ref = T1)")
Model 1: PSI (Time Ref = T1)
term estimate std.error df statistic p.value conf.low conf.high
(Intercept) 4.3776108 10.0145075 63.43621 0.4371269 0.6635038 -15.6320927 24.3873143
citizen 2.6635208 4.4126006 58.79621 0.6036170 0.5484201 -6.1667130 11.4937546
cts 0.5285038 0.4071192 56.45022 1.2981548 0.1995131 -0.2869105 1.3439181
edu 2.0102048 1.2909308 61.97358 1.5571747 0.1245215 -0.5703509 4.5907605
finaid 0.3088264 4.2378485 55.87983 0.0728734 0.9421671 -8.1810071 8.7986600
groupExp -4.4939552 3.6226185 59.42535 -1.2405268 0.2196562 -11.7417127 2.7538023
preg_planned -0.3701674 3.8805960 57.27791 -0.0953893 0.9243385 -8.1401099 7.3997750
psi_pre 0.7986455 0.0789524 50.77292 10.1155248 0.0000000 0.6401246 0.9571663
SP -0.4783158 4.3927445 58.45252 -0.1088877 0.9136645 -9.2699001 8.3132685
timeT2 -0.4010091 1.9488872 36.50934 -0.2057631 0.8381190 -4.3516218 3.5496036
timeT2:groupExp 2.5889151 3.3684611 50.76934 0.7685750 0.4457062 -4.1742988 9.3521290
kable(tidy(mod_psi6.1, conf.int = T), caption = "Model 2: PSI (Time Ref = T2)")
Model 2: PSI (Time Ref = T2)
term estimate std.error df statistic p.value conf.low conf.high
(Intercept) 3.9766851 10.0203626 63.72797 0.3968604 0.6927963 -16.0429362 23.9963064
citizen 2.6635400 4.4168253 58.79551 0.6030440 0.5487986 -6.1751502 11.5022301
cts 0.5285147 0.4075543 56.44907 1.2967958 0.1999773 -0.2877713 1.3448007
edu 2.0101722 1.2912911 61.97283 1.5567149 0.1246307 -0.5711046 4.5914489
finaid 0.3087952 4.2403818 55.87862 0.0728225 0.9422074 -8.1861174 8.8037079
groupExp -1.9050284 3.5024199 59.68580 -0.5439178 0.5885242 -8.9116700 5.1016131
preg_planned -0.3701916 3.8849710 57.27699 -0.0952881 0.9244186 -8.1488966 7.4085134
psi_pre 0.7986440 0.0790606 50.77166 10.1016674 0.0000000 0.6399059 0.9573821
SP -0.4781988 4.3950021 58.45139 -0.1088051 0.9137297 -9.2743050 8.3179074
time2T1 0.4009994 1.9498435 36.51118 0.2056572 0.8382011 -3.5515450 4.3535438
time2T1:groupExp -2.5889276 3.3625485 50.76396 -0.7699302 0.4449095 -9.3402876 4.1624324