R15 Models
EK Questions
Are the sleep variables that are on the same row as the non-sleep variables referring to the prior night?
I think the models that might make the most sense here are prior night sleep - > next day outcomes – otherwise, we’d be looking at sleep affecting outcomes 2 days later
Aversion to Suicide/Cognitive Bias on Concurrent & Next Day NSSI/SI
Background: Sleep moderation of Aversion to Suicide/Cognitive Bias on Concurrent & Next Day NSSI/SI (note: if doing both concurrent & next day is “too much” for a single paper, we can focus on concurrent only)
Hypothesis: Disrupted Sleep (on the night preceding/associated with the daily reports) will moderate the effects of subjective aversion to death and cognitive biases toward self-harm on concurrent and next-day NSSI Acts and Suicide Intent
Variables
- Aversion/Fearlessness About Death: Avers
- Cognitive Bias: CogBias
- Sleep Variables to test as moderators: Objective Sleep (SleepTime, Efficnt); subjective sleep problems (SlpProb)
- NSSI Outcome: NSIact (Note: Recoded to be
dichotomous
NSIact_dich) - Suicide Outcome: S_Intent (would be interesting to look at S_Plan but I think there are too few instances)
Aversion/Fearlessness About Death Models with NSSI outcome
Note: NSSI is yes/no
Regression results
Aversion x Sleep Probs looks to be the most promising model here
| Sleep Time | Sleep Eff | Sleep Probs | |||||||
|---|---|---|---|---|---|---|---|---|---|
| Predictors | Odds Ratios | CI | p | Odds Ratios | CI | p | Odds Ratios | CI | p |
| (Intercept) | 0.13 | 0.13 – 0.13 | <0.001 | 0.13 | 0.13 – 0.13 | <0.001 | 0.12 | 0.12 – 0.12 | <0.001 |
|
aversion to self-harm(FAD 1+FAD 2) |
0.86 | 0.86 – 0.86 | <0.001 | 0.86 | 0.86 – 0.86 | <0.001 | 0.87 | 0.87 – 0.87 | <0.001 |
| Minutes | 1.00 | 1.00 – 1.00 | 0.672 | ||||||
| Avers_pcent:SleepTime_pcent | 1.00 | 1.00 – 1.00 | 0.314 | ||||||
| Sleep Efficiency Score(%) | 1.00 | 1.00 – 1.00 | 0.109 | ||||||
| Avers_pcent:Efficnt_pcent | 1.00 | 1.00 – 1.00 | <0.001 | ||||||
|
sleep problems+felt sleepy all day |
1.30 | 1.30 – 1.30 | <0.001 | ||||||
| Avers_pcent:SlpProb_pcent | 0.94 | 0.94 – 0.95 | <0.001 | ||||||
| Random Effects | |||||||||
| σ2 | 3.29 | 3.29 | 3.29 | ||||||
| τ00 | 3.56 IDcode | 3.57 IDcode | 3.98 IDcode | ||||||
| ICC | 0.52 | 0.52 | 0.55 | ||||||
| N | 130 IDcode | 130 IDcode | 130 IDcode | ||||||
| Observations | 2664 | 2664 | 3220 | ||||||
| Marginal R2 / Conditional R2 | 0.010 / 0.525 | 0.010 / 0.525 | 0.019 / 0.557 | ||||||
Plot of Aversion x Sleep Probs
Simple slopes probe of Aversion x Sleep Probs
## JOHNSON-NEYMAN INTERVAL
##
## When SlpProb_pcent is OUTSIDE the interval [-2.55, -2.45], the slope of
## Avers_pcent is p < .05.
##
## Note: The range of observed values of SlpProb_pcent is [-2.43, 3.48]
##
## SIMPLE SLOPES ANALYSIS
##
## Slope of Avers_pcent when SlpProb_pcent = -0.9516498613 (- 1 SD):
##
## Est. S.E. z val. p
## ------- ------ -------- ------
## -0.09 0.04 -2.03 0.04
##
## Slope of Avers_pcent when SlpProb_pcent = -0.0004720145 (Mean):
##
## Est. S.E. z val. p
## ------- ------ --------- ------
## -0.14 0.00 -270.35 0.00
##
## Slope of Avers_pcent when SlpProb_pcent = 0.9507058323 (+ 1 SD):
##
## Est. S.E. z val. p
## ------- ------ -------- ------
## -0.20 0.04 -5.04 0.00Aversion/Fearlessness About Death Models with SI outcome
Regression results
No significant interaction effects
| Sleep Time | Sleep Eff | Sleep Probs | |||||||
|---|---|---|---|---|---|---|---|---|---|
| Predictors | Estimates | CI | p | Estimates | CI | p | Estimates | CI | p |
| (Intercept) | 1.77 | 1.61 – 1.93 | <0.001 | 1.77 | 1.61 – 1.93 | <0.001 | 1.82 | 1.65 – 1.98 | <0.001 |
|
aversion to self-harm(FAD 1+FAD 2) |
-0.09 | -0.11 – -0.07 | <0.001 | -0.09 | -0.11 – -0.07 | <0.001 | -0.09 | -0.11 – -0.08 | <0.001 |
| Minutes | -0.00 | -0.00 – 0.00 | 0.209 | ||||||
| Avers_pcent:SleepTime_pcent | 0.00 | -0.00 – 0.00 | 0.282 | ||||||
| Sleep Efficiency Score(%) | 0.00 | -0.00 – 0.01 | 0.490 | ||||||
| Avers_pcent:Efficnt_pcent | 0.00 | -0.00 – 0.00 | 0.110 | ||||||
|
sleep problems+felt sleepy all day |
0.11 | 0.08 – 0.15 | <0.001 | ||||||
| Avers_pcent:SlpProb_pcent | -0.01 | -0.03 – 0.00 | 0.099 | ||||||
| Random Effects | |||||||||
| σ2 | 0.75 | 0.75 | 0.78 | ||||||
| τ00 | 0.83 IDcode | 0.83 IDcode | 0.86 IDcode | ||||||
| ICC | 0.53 | 0.53 | 0.52 | ||||||
| N | 130 IDcode | 130 IDcode | 130 IDcode | ||||||
| Observations | 2663 | 2663 | 3219 | ||||||
| Marginal R2 / Conditional R2 | 0.015 / 0.533 | 0.015 / 0.533 | 0.025 / 0.536 | ||||||
Cognitive Bias Models with NSSI outcome
Note: NSSI is yes/no
Regression results
Cognitive bias has a decent main effect but no significant interactions
| Sleep Time | Sleep Eff | Sleep Probs | |||||||
|---|---|---|---|---|---|---|---|---|---|
| Predictors | Odds Ratios | CI | p | Odds Ratios | CI | p | Odds Ratios | CI | p |
| (Intercept) | 0.06 | 0.03 – 0.10 | <0.001 | 0.06 | 0.03 – 0.10 | <0.001 | 0.06 | 0.04 – 0.10 | <0.001 |
|
cog bias to NSSI(NSSI Urge+NSSI Ideation Freq) |
1.98 | 1.83 – 2.14 | <0.001 | 1.97 | 1.82 – 2.12 | <0.001 | 1.94 | 1.81 – 2.08 | <0.001 |
| Minutes | 1.00 | 1.00 – 1.00 | 0.683 | ||||||
| CogBias_pcent:SleepTime_pcent | 1.00 | 1.00 – 1.00 | 0.050 | ||||||
| Sleep Efficiency Score(%) | 1.00 | 0.98 – 1.02 | 0.799 | ||||||
| CogBias_pcent:Efficnt_pcent | 1.00 | 0.99 – 1.01 | 0.632 | ||||||
|
sleep problems+felt sleepy all day |
1.11 | 0.95 – 1.30 | 0.195 | ||||||
| CogBias_pcent:SlpProb_pcent | 0.95 | 0.90 – 1.00 | 0.064 | ||||||
| Random Effects | |||||||||
| σ2 | 3.29 | 3.29 | 3.29 | ||||||
| τ00 | 6.96 IDcode | 6.85 IDcode | 6.93 IDcode | ||||||
| ICC | 0.68 | 0.68 | 0.68 | ||||||
| N | 130 IDcode | 130 IDcode | 130 IDcode | ||||||
| Observations | 2670 | 2670 | 3228 | ||||||
| Marginal R2 / Conditional R2 | 0.216 / 0.748 | 0.213 / 0.745 | 0.210 / 0.746 | ||||||
Cognitive Bias Models with SI outcome
Regression results
Sleep time and sleep probs are significant
| Sleep Time | Sleep Eff | Sleep Probs | |||||||
|---|---|---|---|---|---|---|---|---|---|
| Predictors | Estimates | CI | p | Estimates | CI | p | Estimates | CI | p |
| (Intercept) | 1.77 | 1.61 – 1.94 | <0.001 | 1.78 | 1.61 – 1.94 | <0.001 | 1.81 | 1.65 – 1.97 | <0.001 |
|
cog bias to NSSI(NSSI Urge+NSSI Ideation Freq) |
0.18 | 0.16 – 0.19 | <0.001 | 0.18 | 0.16 – 0.19 | <0.001 | 0.17 | 0.16 – 0.18 | <0.001 |
| Minutes | -0.00 | -0.00 – 0.00 | 0.585 | ||||||
| CogBias_pcent:SleepTime_pcent | -0.00 | -0.00 – -0.00 | 0.011 | ||||||
| Sleep Efficiency Score(%) | 0.00 | -0.00 – 0.01 | 0.562 | ||||||
| CogBias_pcent:Efficnt_pcent | -0.00 | -0.00 – 0.00 | 0.217 | ||||||
|
sleep problems+felt sleepy all day |
0.05 | 0.02 – 0.08 | 0.001 | ||||||
| CogBias_pcent:SlpProb_pcent | 0.02 | 0.01 – 0.03 | 0.003 | ||||||
| Random Effects | |||||||||
| σ2 | 0.58 | 0.58 | 0.62 | ||||||
| τ00 | 0.84 IDcode | 0.84 IDcode | 0.87 IDcode | ||||||
| ICC | 0.59 | 0.59 | 0.59 | ||||||
| N | 130 IDcode | 130 IDcode | 130 IDcode | ||||||
| Observations | 2669 | 2669 | 3227 | ||||||
| Marginal R2 / Conditional R2 | 0.115 / 0.640 | 0.114 / 0.639 | 0.115 / 0.633 | ||||||
Plot/probe for Cog Bias x Sleep Time
Plot of Cog Bias x Sleep Time
Simple slopes probe of Cog Bias x Sleep Time
## JOHNSON-NEYMAN INTERVAL
##
## When SleepTime_pcent is OUTSIDE the interval [765.46, 6053.16], the slope
## of CogBias_pcent is p < .05.
##
## Note: The range of observed values of SleepTime_pcent is [-503.91, 875.82]
##
## SIMPLE SLOPES ANALYSIS
##
## Slope of CogBias_pcent when SleepTime_pcent = -121.8009758 (- 1 SD):
##
## Est. S.E. t val. p
## ------ ------ -------- ------
## 0.19 0.01 21.75 0.00
##
## Slope of CogBias_pcent when SleepTime_pcent = -0.9392126 (Mean):
##
## Est. S.E. t val. p
## ------ ------ -------- ------
## 0.18 0.01 28.87 0.00
##
## Slope of CogBias_pcent when SleepTime_pcent = 119.9225506 (+ 1 SD):
##
## Est. S.E. t val. p
## ------ ------ -------- ------
## 0.16 0.01 18.82 0.00Plot/probe for Cog Bias x Sleep Probs
Plot of Cog Bias x Sleep Probs
Simple slopes probe of Cog Bias x Sleep Probs
## JOHNSON-NEYMAN INTERVAL
##
## When SlpProb_pcent is OUTSIDE the interval [-31.79, -6.29], the slope of
## CogBias_pcent is p < .05.
##
## Note: The range of observed values of SlpProb_pcent is [-2.43, 3.48]
##
## SIMPLE SLOPES ANALYSIS
##
## Slope of CogBias_pcent when SlpProb_pcent = -0.954610981 (- 1 SD):
##
## Est. S.E. t val. p
## ------ ------ -------- ------
## 0.16 0.01 19.40 0.00
##
## Slope of CogBias_pcent when SlpProb_pcent = -0.001424105 (Mean):
##
## Est. S.E. t val. p
## ------ ------ -------- ------
## 0.17 0.01 30.13 0.00
##
## Slope of CogBias_pcent when SlpProb_pcent = 0.951762770 (+ 1 SD):
##
## Est. S.E. t val. p
## ------ ------ -------- ------
## 0.19 0.01 25.24 0.00Sleep moderation of Neg Self Perception and Emotional Reactivity on concurrent & next day NSSI/SI
Hypothesis: Disrupted sleep (on the night immediately preceding the daily reports) will moderate (intensify) the effects of self-criticism and emotional reactivity on same and next day NSSI acts (or urges) and Suicide Intent.
Variables
- Emotional Reactivity: EmotReactivity
- Neg Self-Perception: SelfCritical
- Sleep Variables to test as moderators: Objective Sleep (Sleeptime, Efficnt, #Awakenings); subjective sleep problems (SlpProb) -
- NSSI Outcome: NSIact (Note: Recoded to be
dichotomous
NSIact_dich) - Suicide Outcome: S_Intent
Emotional Reactivity Models with NSSI outcome
Note: NSSI is yes/no
Regression results
Moderation with of awakenings is significant but the ES is supeerrrr small
| Sleep Time | Sleep Eff | Number of awakenings | |||||||
|---|---|---|---|---|---|---|---|---|---|
| Predictors | Odds Ratios | CI | p | Odds Ratios | CI | p | Odds Ratios | CI | p |
| (Intercept) | 0.11 | 0.11 – 0.11 | <0.001 | 0.11 | 0.11 – 0.11 | <0.001 | 0.11 | 0.11 – 0.11 | <0.001 |
|
emotional reactivity total |
1.65 | 1.65 – 1.65 | <0.001 | 1.65 | 1.64 – 1.65 | <0.001 | 1.65 | 1.65 – 1.66 | <0.001 |
| Minutes | 1.00 | 1.00 – 1.00 | 0.794 | ||||||
| EmotReactivity_pcent:SleepTime_pcent | 1.00 | 1.00 – 1.00 | 0.258 | ||||||
| Sleep Efficiency Score(%) | 1.00 | 1.00 – 1.00 | 0.023 | ||||||
| EmotReactivity_pcent:Efficnt_pcent | 1.00 | 1.00 – 1.00 | <0.001 | ||||||
| #Awakenings | 1.00 | 1.00 – 1.00 | 0.800 | ||||||
| EmotReactivity_pcent:Awaken_pcent | 1.01 | 1.01 – 1.01 | <0.001 | ||||||
| Random Effects | |||||||||
| σ2 | 3.29 | 3.29 | 3.29 | ||||||
| τ00 | 3.96 IDcode | 3.97 IDcode | 3.97 IDcode | ||||||
| ICC | 0.55 | 0.55 | 0.55 | ||||||
| N | 130 IDcode | 130 IDcode | 130 IDcode | ||||||
| Observations | 2690 | 2690 | 2690 | ||||||
| Marginal R2 / Conditional R2 | 0.058 / 0.572 | 0.057 / 0.573 | 0.059 / 0.574 | ||||||
Emotional Reactivity Models with SI outcome
Regression results
Nothing is significant
| Sleep Time | Sleep Eff | Number of awakenings | |||||||
|---|---|---|---|---|---|---|---|---|---|
| Predictors | Estimates | CI | p | Estimates | CI | p | Estimates | CI | p |
| (Intercept) | 1.77 | 1.61 – 1.93 | <0.001 | 1.77 | 1.61 – 1.93 | <0.001 | 1.77 | 1.61 – 1.93 | <0.001 |
|
emotional reactivity total |
0.21 | 0.19 – 0.24 | <0.001 | 0.22 | 0.19 – 0.24 | <0.001 | 0.22 | 0.19 – 0.24 | <0.001 |
| Minutes | -0.00 | -0.00 – 0.00 | 0.800 | ||||||
| EmotReactivity_pcent:SleepTime_pcent | -0.00 | -0.00 – 0.00 | 0.097 | ||||||
| Sleep Efficiency Score(%) | 0.00 | -0.00 – 0.01 | 0.420 | ||||||
| EmotReactivity_pcent:Efficnt_pcent | -0.00 | -0.00 – 0.00 | 0.957 | ||||||
| #Awakenings | 0.00 | -0.00 – 0.00 | 0.765 | ||||||
| EmotReactivity_pcent:Awaken_pcent | -0.00 | -0.00 – 0.00 | 0.597 | ||||||
| Random Effects | |||||||||
| σ2 | 0.68 | 0.68 | 0.68 | ||||||
| τ00 | 0.83 IDcode | 0.83 IDcode | 0.83 IDcode | ||||||
| ICC | 0.55 | 0.55 | 0.55 | ||||||
| N | 130 IDcode | 130 IDcode | 130 IDcode | ||||||
| Observations | 2688 | 2688 | 2688 | ||||||
| Marginal R2 / Conditional R2 | 0.052 / 0.573 | 0.052 / 0.572 | 0.052 / 0.572 | ||||||
Neg Self Perception Models with NSSI outcome
Note: NSSI is yes/no
Regression results
Nothing significant| Sleep Time | Sleep Eff | Number of awakenings | |||||||
|---|---|---|---|---|---|---|---|---|---|
| Predictors | Odds Ratios | CI | p | Odds Ratios | CI | p | Odds Ratios | CI | p |
| (Intercept) | 0.11 | 0.07 – 0.16 | <0.001 | 0.11 | 0.07 – 0.16 | <0.001 | 0.11 | 0.07 – 0.16 | <0.001 |
|
self-judgement+self-hate+self criticism |
1.99 | 1.76 – 2.24 | <0.001 | 1.99 | 1.76 – 2.24 | <0.001 | 1.98 | 1.76 – 2.24 | <0.001 |
| Minutes | 1.00 | 1.00 – 1.00 | 0.662 | ||||||
| SelfCritical_pcent:SleepTime_pcent | 1.00 | 1.00 – 1.00 | 0.966 | ||||||
| Sleep Efficiency Score(%) | 1.00 | 0.99 – 1.02 | 0.583 | ||||||
| SelfCritical_pcent:Efficnt_pcent | 0.99 | 0.98 – 1.01 | 0.470 | ||||||
| #Awakenings | 1.00 | 0.99 – 1.01 | 0.978 | ||||||
| SelfCritical_pcent:Awaken_pcent | 1.00 | 0.99 – 1.01 | 0.544 | ||||||
| Random Effects | |||||||||
| σ2 | 3.29 | 3.29 | 3.29 | ||||||
| τ00 | 4.04 IDcode | 4.04 IDcode | 4.05 IDcode | ||||||
| ICC | 0.55 | 0.55 | 0.55 | ||||||
| N | 130 IDcode | 130 IDcode | 130 IDcode | ||||||
| Observations | 2664 | 2664 | 2664 | ||||||
| Marginal R2 / Conditional R2 | 0.070 / 0.583 | 0.071 / 0.583 | 0.070 / 0.583 | ||||||
Neg Self Perception Models with NSSI outcome
Regression results
Moderation with of Sleeptime is significant at p=.049 so I don’t think we should pursue
| Sleep Time | Sleep Eff | Number of awakenings | |||||||
|---|---|---|---|---|---|---|---|---|---|
| Predictors | Estimates | CI | p | Estimates | CI | p | Estimates | CI | p |
| (Intercept) | 1.77 | 1.61 – 1.93 | <0.001 | 1.78 | 1.62 – 1.94 | <0.001 | 1.77 | 1.61 – 1.93 | <0.001 |
|
self-judgement+self-hate+self criticism |
0.31 | 0.28 – 0.34 | <0.001 | 0.31 | 0.28 – 0.34 | <0.001 | 0.31 | 0.28 – 0.34 | <0.001 |
| Minutes | -0.00 | -0.00 – 0.00 | 0.935 | ||||||
| SelfCritical_pcent:SleepTime_pcent | -0.00 | -0.00 – -0.00 | 0.049 | ||||||
| Sleep Efficiency Score(%) | 0.00 | -0.00 – 0.01 | 0.340 | ||||||
| SelfCritical_pcent:Efficnt_pcent | 0.00 | -0.00 – 0.00 | 0.573 | ||||||
| #Awakenings | -0.00 | -0.00 – 0.00 | 0.871 | ||||||
| SelfCritical_pcent:Awaken_pcent | -0.00 | -0.00 – 0.00 | 0.494 | ||||||
| Random Effects | |||||||||
| σ2 | 0.65 | 0.65 | 0.65 | ||||||
| τ00 | 0.83 IDcode | 0.83 IDcode | 0.83 IDcode | ||||||
| ICC | 0.56 | 0.56 | 0.56 | ||||||
| N | 130 IDcode | 130 IDcode | 130 IDcode | ||||||
| Observations | 2662 | 2662 | 2662 | ||||||
| Marginal R2 / Conditional R2 | 0.071 / 0.592 | 0.071 / 0.591 | 0.071 / 0.591 | ||||||
Mediation Models
Model Aversion/FAD and cognitive bias mediate the effect of NSSI acts on next day suicide intent.
Notes I think the effect of NSSI is in the wrong direction here– there’s no significant effect when I use NSSI acts as a continuous variable
Aversion/FAD and cognitive bias mediate the effect of NSSI acts on next day suicide intent.
Mediation output
## lavaan 0.6-19 ended normally after 81 iterations
##
## Estimator ML
## Optimization method NLMINB
## Number of model parameters 19
##
## Used Total
## Number of observations 2899 3726
## Number of clusters [IDcode] 130
##
## Model Test User Model:
##
## Test statistic 65.511
## Degrees of freedom 2
## P-value (Chi-square) 0.000
##
## Parameter Estimates:
##
## Standard errors Standard
## Information Observed
## Observed information based on Hessian
##
##
## Level 1 [within]:
##
## Regressions:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## S_Intent_t2 ~
## Avers (b2) -0.019 0.010 -1.983 0.047 -0.019 -0.038
## CogBias (b1) 0.037 0.008 4.774 0.000 0.037 0.104
## NSIct_dch -0.121 0.059 -2.043 0.041 -0.121 -0.044
## CogBias ~
## NSIct_dch (a1) 3.672 0.127 28.816 0.000 3.672 0.480
## Avers ~
## NSIct_dch (a2) -0.499 0.102 -4.873 0.000 -0.499 -0.092
##
## Variances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## .S_Intent_t2 0.814 0.022 37.200 0.000 0.814 0.990
## .CogBias 4.883 0.131 37.218 0.000 4.883 0.769
## .Avers 3.156 0.085 37.211 0.000 3.156 0.991
##
##
## Level 2 [IDcode]:
##
## Regressions:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## S_Intent_t2 ~
## Avers -0.089 0.020 -4.523 0.000 -0.089 -0.307
## CogBias 0.273 0.030 9.146 0.000 0.273 0.709
## NSIact_dich -0.812 0.298 -2.727 0.006 -0.812 -0.214
## CogBias ~
## NSIact_dich 4.664 0.822 5.675 0.000 4.664 0.474
## Avers ~
## NSIact_dich -2.074 1.201 -1.727 0.084 -2.074 -0.158
##
## Intercepts:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## .S_Intent_t2 1.161 0.225 5.164 0.000 1.161 1.273
## .CogBias 4.372 0.253 17.303 0.000 4.372 1.848
## .Avers 7.459 0.370 20.186 0.000 7.459 2.367
##
## Variances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## .S_Intent_t2 0.408 0.058 7.038 0.000 0.408 0.490
## .CogBias 4.336 0.571 7.592 0.000 4.336 0.775
## .Avers 9.676 1.224 7.908 0.000 9.676 0.975
##
## Defined Parameters:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## total_indirect 0.057 0.043 1.347 0.178 0.057 0.026
## cogbias_ind 0.138 0.029 4.709 0.000 0.138 0.050
## aver_ind 0.010 0.005 1.837 0.066 0.010 0.004Mediation Figure
Top line going from one variable to the other is level 1
Moderated mediation: sleep time moderating a and b paths
## lavaan 0.6-19 ended normally after 163 iterations
##
## Estimator ML
## Optimization method NLMINB
## Number of model parameters 33
##
## Used Total
## Number of observations 2411 3726
## Number of clusters [IDcode] 129
##
## Model Test User Model:
##
## Test statistic 8834.327
## Degrees of freedom 12
## P-value (Chi-square) 0.000
##
## Parameter Estimates:
##
## Standard errors Standard
## Information Observed
## Observed information based on Hessian
##
##
## Level 1 [within]:
##
## Regressions:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## S_Intent_t2 ~
## Avers (b2) -0.017 0.021 -0.800 0.424 -0.017 -0.034
## CogBias (b1) 0.061 0.021 2.888 0.004 0.061 0.169
## NSIct_d -0.147 0.063 -2.333 0.020 -0.147 -0.055
## SleepTm (mb0) -0.000 0.000 -0.085 0.933 -0.000 -0.004
## CgBsXST (mb1) -0.000 0.000 -1.304 0.192 -0.000 -0.083
## AvrsXST (mb2) 0.000 0.000 0.830 0.406 0.000 0.047
## CogBias ~
## NSIct_d (a1) 2.804 0.437 6.419 0.000 2.804 0.374
## SleepTm (ma10) -0.001 0.000 -2.371 0.018 -0.001 -0.047
## NSI_XST (ma1) 0.002 0.001 1.717 0.086 0.002 0.101
## Avers ~
## NSIct_d (a2) -0.472 0.349 -1.356 0.175 -0.472 -0.089
## SleepTm (ma20) -0.001 0.000 -1.658 0.097 -0.001 -0.037
## NSI_XST (ma2) 0.000 0.001 0.114 0.909 0.000 0.008
##
## Variances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## .S_Intent_t2 0.779 0.023 33.736 0.000 0.779 0.970
## .CogBias 4.820 0.143 33.782 0.000 4.820 0.778
## .Avers 3.092 0.092 33.778 0.000 3.092 0.992
##
##
## Level 2 [IDcode]:
##
## Regressions:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## S_Intent_t2 ~
## Avers -0.180 0.151 -1.192 0.233 -0.180 -0.529
## CogBias 0.274 0.300 0.913 0.361 0.274 0.604
## NSIact_dich -0.834 0.341 -2.445 0.014 -0.834 -0.181
## SleepTime -0.002 0.002 -0.854 0.393 -0.002 -0.157
## CogBiasXSlepTm -0.000 0.001 -0.029 0.977 -0.000 -0.019
## AversXSleepTim 0.000 0.000 0.599 0.549 0.000 0.270
## CogBias ~
## NSIact_dich -0.050 35.679 -0.001 0.999 -0.050 -0.005
## SleepTime -0.003 0.006 -0.440 0.660 -0.003 -0.092
## NSIct_dchXSlpT 0.011 0.088 0.128 0.898 0.011 0.448
## Avers ~
## NSIact_dich 0.006 69.094 0.000 1.000 0.006 0.000
## SleepTime -0.005 0.012 -0.458 0.647 -0.005 -0.136
## NSIct_dchXSlpT -0.005 0.170 -0.030 0.976 -0.005 -0.154
##
## Intercepts:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## .S_Intent_t2 2.109 1.091 1.932 0.053 2.109 1.951
## .CogBias 5.552 2.559 2.170 0.030 5.552 2.331
## .Avers 9.821 4.804 2.044 0.041 9.821 3.099
##
## Variances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## .S_Intent_t2 0.404 0.073 5.500 0.000 0.404 0.346
## .CogBias 4.458 0.636 7.011 0.000 4.458 0.786
## .Avers 9.669 1.244 7.774 0.000 9.669 0.963
##
## Defined Parameters:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## total_indirect 0.102 0.075 1.361 0.174 0.102 0.039
## cogbias_ind 0.171 0.065 2.628 0.009 0.171 0.063
## aver_ind 0.008 0.012 0.689 0.491 0.008 0.003Moderated mediation: sleep problems moderating a and b paths
## lavaan 0.6-19 ended normally after 156 iterations
##
## Estimator ML
## Optimization method NLMINB
## Number of model parameters 33
##
## Used Total
## Number of observations 2399 3726
## Number of clusters [IDcode] 129
##
## Model Test User Model:
##
## Test statistic 6074.606
## Degrees of freedom 15
## P-value (Chi-square) 0.000
##
## Parameter Estimates:
##
## Standard errors Standard
## Information Observed
## Observed information based on Hessian
##
##
## Level 1 [within]:
##
## Regressions:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## S_Intent_t2 ~
## Avers (b2) 0.010 0.014 0.751 0.452 0.010 0.021
## CogBias (b1) 0.021 0.014 1.551 0.121 0.021 0.060
## NSIct_d -0.151 0.063 -2.391 0.017 -0.151 -0.057
## SleepTm (mb0) -0.000 0.000 -0.867 0.386 -0.000 -0.020
## CgBsXSP (mb1) 0.005 0.004 1.287 0.198 0.005 0.057
## AvrsXSP (mb2) -0.005 0.003 -1.326 0.185 -0.005 -0.044
## CogBias ~
## NSIct_d (a1) 3.655 0.302 12.116 0.000 3.655 0.488
## SlpProb (ma10) 0.337 0.053 6.387 0.000 0.337 0.131
## NSI_XSP (ma1) -0.089 0.101 -0.883 0.377 -0.089 -0.037
## Avers ~
## NSIct_d (a2) -0.182 0.243 -0.747 0.455 -0.182 -0.034
## SlpProb (ma20) -0.073 0.043 -1.706 0.088 -0.073 -0.040
## NSI_XSP (ma2) -0.079 0.082 -0.966 0.334 -0.079 -0.046
##
## Variances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## .S_Intent_t2 0.781 0.023 33.665 0.000 0.781 0.993
## .CogBias 4.738 0.141 33.706 0.000 4.738 0.765
## .Avers 3.082 0.091 33.691 0.000 3.082 0.991
##
##
## Level 2 [IDcode]:
##
## Regressions:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## S_Intent_t2 ~
## Avers -0.030 0.074 -0.402 0.688 -0.030 -0.113
## CogBias 0.103 0.108 0.954 0.340 0.103 0.292
## NSIact_dich -0.803 0.313 -2.562 0.010 -0.803 -0.224
## SlpProb -0.069 0.367 -0.187 0.851 -0.069 -0.054
## CogBiasXSlpPrb 0.060 0.040 1.495 0.135 0.060 0.579
## AversXSlpProb -0.019 0.030 -0.637 0.524 -0.019 -0.198
## CogBias ~
## NSIact_dich 4.016 3.114 1.290 0.197 4.016 0.395
## SlpProb 0.820 0.407 2.015 0.044 0.820 0.229
## NSIct_dchXSlpP 0.175 1.143 0.153 0.878 0.175 0.049
## Avers ~
## NSIact_dich 10.388 4.518 2.299 0.021 10.388 0.767
## SlpProb 0.204 0.588 0.348 0.728 0.204 0.043
## NSIct_dchXSlpP -4.629 1.660 -2.789 0.005 -4.629 -0.975
##
## Intercepts:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## .S_Intent_t2 1.335 0.889 1.502 0.133 1.335 1.596
## .CogBias 2.313 1.036 2.232 0.026 2.313 0.975
## .Avers 6.989 1.497 4.669 0.000 6.989 2.209
##
## Variances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## .S_Intent_t2 0.396 0.059 6.730 0.000 0.396 0.566
## .CogBias 4.150 0.559 7.422 0.000 4.150 0.737
## .Avers 8.716 1.155 7.549 0.000 8.716 0.871
##
## Defined Parameters:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## total_indirect 0.111 0.060 1.833 0.067 0.111 0.037
## cogbias_ind 0.078 0.051 1.538 0.124 0.078 0.029
## aver_ind -0.002 0.004 -0.530 0.596 -0.002 -0.001##Sleep (time/prob) –> SelfCriticism—>SIntent with Emotion reactivity moderating SC –>SI path.
## lavaan 0.6-19 ended normally after 96 iterations
##
## Estimator ML
## Optimization method NLMINB
## Number of model parameters 20
##
## Used Total
## Number of observations 2411 3726
## Number of clusters [IDcode] 129
##
## Model Test User Model:
##
## Test statistic 3773.543
## Degrees of freedom 4
## P-value (Chi-square) 0.000
##
## Parameter Estimates:
##
## Standard errors Standard
## Information Observed
## Observed information based on Hessian
##
##
## Level 1 [within]:
##
## Regressions:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## S_Intent_t2 ~
## SelfCrtcl (b) -0.003 0.037 -0.068 0.946 -0.003 -0.003
## SlpProb -0.026 0.020 -1.332 0.183 -0.026 -0.029
## SleepTime -0.000 0.000 -0.748 0.455 -0.000 -0.016
## EmtRctvty (m0) -0.040 0.035 -1.147 0.252 -0.040 -0.062
## EmoXSC (m1) 0.014 0.008 1.819 0.069 0.014 0.144
## SelfCritical ~
## SlpProb (a1) 0.253 0.023 10.827 0.000 0.253 0.222
## SleepTime (a2) -0.000 0.000 -1.630 0.103 -0.000 -0.033
##
## Variances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## .S_Intent_t2 0.765 0.023 33.766 0.000 0.765 0.991
## .SelfCritical 1.146 0.034 33.790 0.000 1.146 0.948
##
##
## Level 2 [IDcode]:
##
## Regressions:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## S_Intent_t2 ~
## SelfCritical -0.113 0.165 -0.685 0.493 -0.113 -0.123
## SlpProb -0.038 0.124 -0.302 0.763 -0.038 -0.024
## SleepTime -0.002 0.001 -1.636 0.102 -0.002 -0.116
## EmotReactivity -0.497 0.218 -2.281 0.023 -0.497 -0.518
## EmoXSC 0.158 0.043 3.656 0.000 0.158 1.213
## SelfCritical ~
## SlpProb 0.663 0.151 4.404 0.000 0.663 0.394
## SleepTime 0.001 0.001 0.738 0.461 0.001 0.069
##
## Intercepts:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## .S_Intent_t2 2.173 0.785 2.768 0.006 2.173 2.164
## .SelfCritical 2.061 0.711 2.898 0.004 2.061 1.879
##
## Variances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## .S_Intent_t2 0.399 0.064 6.237 0.000 0.399 0.395
## .SelfCritical 1.011 0.137 7.399 0.000 1.011 0.840
##
## Defined Parameters:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## total_indirect -0.001 0.009 -0.068 0.946 -0.001 -0.001
## SlpProb_ind -0.001 0.009 -0.068 0.946 -0.001 -0.001
## SlpTime_ind 0.000 0.000 0.068 0.946 0.000 0.000Sleep –> emotional reactivity / self-critical –> NSSI/SI
Mediation output
## lavaan 0.6-19 ended normally after 114 iterations
##
## Estimator ML
## Optimization method NLMINB
## Number of model parameters 23
##
## Used Total
## Number of observations 2425 3726
## Number of clusters [IDcode] 129
##
## Model Test User Model:
##
## Test statistic 906.300
## Degrees of freedom 4
## P-value (Chi-square) 0.000
##
## Parameter Estimates:
##
## Standard errors Standard
## Information Observed
## Observed information based on Hessian
##
##
## Level 1 [within]:
##
## Regressions:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## S_Intent_t2 ~
## SlfCrtcl (b2) 0.051 0.020 2.539 0.011 0.051 0.064
## EmtRctvt (b1) 0.014 0.016 0.880 0.379 0.014 0.022
## SleepTim -0.000 0.000 -0.610 0.542 -0.000 -0.013
## EmotReactivity ~
## SleepTim (a1a) -0.001 0.000 -2.617 0.009 -0.001 -0.058
## Efficnt (a1b) 0.005 0.004 1.082 0.279 0.005 0.024
## SelfCritical ~
## SleepTim (a2a) -0.000 0.000 -2.410 0.016 -0.000 -0.054
## Efficnt (a2b) -0.000 0.003 -0.003 0.997 -0.000 -0.000
##
## Variances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## .S_Intent_t2 0.766 0.023 33.866 0.000 0.766 0.995
## .EmotReactivity 1.825 0.054 33.916 0.000 1.825 0.997
## .SelfCritical 1.212 0.036 33.899 0.000 1.212 0.997
##
##
## Level 2 [IDcode]:
##
## Regressions:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## S_Intent_t2 ~
## SelfCritical 0.436 0.074 5.922 0.000 0.436 0.546
## EmotReactivity 0.215 0.079 2.721 0.007 0.215 0.256
## SleepTime -0.002 0.001 -1.734 0.083 -0.002 -0.145
## EmotReactivity ~
## SleepTime -0.002 0.001 -1.051 0.293 -0.002 -0.114
## Efficnt 0.013 0.025 0.522 0.602 0.013 0.056
## SelfCritical ~
## SleepTime -0.000 0.001 -0.088 0.930 -0.000 -0.009
## Efficnt 0.039 0.026 1.540 0.124 0.039 0.161
##
## Intercepts:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## .S_Intent_t2 -0.144 0.511 -0.281 0.778 -0.144 -0.164
## .EmotReactivity 3.435 1.947 1.764 0.078 3.435 3.302
## .SelfCritical 0.953 2.008 0.475 0.635 0.953 0.870
##
## Variances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## .S_Intent_t2 0.471 0.068 6.937 0.000 0.471 0.615
## .EmotReactivity 1.069 0.147 7.258 0.000 1.069 0.988
## .SelfCritical 1.170 0.157 7.454 0.000 1.170 0.975
##
## Defined Parameters:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## total_indirect 0.000 0.000 0.632 0.527 0.000 -0.008
## cogbias_ind 0.000 0.000 0.645 0.519 0.000 -0.001
## EmoR_ind -0.000 0.000 -0.147 0.883 -0.000 -0.003Mediation Figure
Top line going from one variable to the other is level 1