Introduction
Peer Relationships, Internalizing Symptoms, and Gender in Adolescence: Theoretical Perspectives
Bullying Victimization and Internalizing Symptoms
The Promotive Benefits of Friendship and Social Support
Current Study
Hypothesis and underpinning theory/model | Prediction(s) (path directions in Fig. 1 where applicable) | Possible outcomes | Interpretation | Empirical support | |
---|---|---|---|---|---|
H1: Bullying victimization has adverse consequences for later peer relationships and mental health. Interpersonal risk model (Coyle et al., 2021) | H1A: Bullying victimization negatively predicts later friendship and social support | \({\beta }_{\mathrm{5,1}} < 0\), \({\beta }_{\mathrm{8,4}} < 0\) | \({\beta }_{\mathrm{5,1}} < 0\), \({\beta }_{\mathrm{8,4}} < 0\) \({\beta }_{\mathrm{5,1}} < 0,\) \({\beta }_{\mathrm{8,4}}\ge 0\) (or \({\beta }_{\mathrm{5,1}}\ge 0\), \({\beta }_{\mathrm{8,4}} < 0\)) \({\beta }_{\mathrm{5,1}}\ge 0,\) \({\beta }_{\mathrm{8,4}}\ge 0\) | H1A supported H1A partially supported H1A rejected | Siennick and Turanovic (2023) |
H1B: Bullying victimization positively predicts later Internalizing symptoms | \({\beta }_{\mathrm{6,1}} > 0\), \({\beta }_{\mathrm{9,4}} > 0\) | \({\beta }_{\mathrm{6,1}} > 0\), \({\beta }_{\mathrm{9,4}} > 0\) \({\beta }_{\mathrm{6,1}} > 0\), \({\beta }_{\mathrm{9,4}}\le 0\) (or \({\beta }_{\mathrm{6,1}}\le 0\), \({\beta }_{\mathrm{9,4}} > 0\)) \({\beta }_{\mathrm{6,1}}\le 0\), \({\beta }_{\mathrm{9,4}}\le 0\) | H1B supported H1B partially supported H1B rejected | Sentse et al. (2017), Bartlett et al. (2023) | |
H2: Positive peer relationships confer protection against later bullying victimization and Internalizing symptoms. Interpersonal risk model (Coyle et al., 2021) | H2A: Friendship and social support negatively predicts later bullying victimization | \({\beta }_{\mathrm{4,2}} < 0\), \({\beta }_{\mathrm{7,5}} < 0\) | \({\beta }_{\mathrm{4,2}} < 0\), \({\beta }_{\mathrm{7,5}} < 0\) \({\beta }_{\mathrm{4,2}} < 0\), \({\beta }_{\mathrm{7,5}}\ge 0\)(or \({\beta }_{\mathrm{4,2}}\ge 0\), \({\beta }_{\mathrm{7,5}} < 0\)) \({\beta }_{\mathrm{4,2}}\ge 0\), \({\beta }_{\mathrm{7,5}}\ge 0\) | H2A supported H2A partially supported H2A rejected | Siennick and Turanovic (2023) |
H2B: Friendship and social support negatively predicts later Internalizing symptoms | \({\beta }_{\mathrm{6,2}} < 0\), \({\beta }_{\mathrm{9,5}} < 0\) | \({\beta }_{\mathrm{6,2}} < 0\), \({\beta }_{\mathrm{9,5}} < 0\) \({\beta }_{\mathrm{6,2}} < 0\), \({\beta }_{\mathrm{9,5}}\ge 0\)(or \({\beta }_{\mathrm{6,2}}\ge 0\), \({\beta }_{\mathrm{9,5}} < 0\)) \({\beta }_{\mathrm{6,2}}\ge 0\), \({\beta }_{\mathrm{9,5}}\ge 0\) | H2B supported H2B partially supported H2B rejected | Murray et al. (2021), Bernasco et al. (2022), Van Droogenbroeck et al. (2018) | |
H3: Mental health difficulties precede relational difficulties with peers. Symptoms-driven model (Kochel et al., 2012) | H3A: Internalizing symptoms positively predict later bullying victimization | \({\beta }_{\mathrm{4,3}} > 0\), \({\beta }_{\mathrm{7,6}} > 0\) | \({\beta }_{\mathrm{4,3}} > 0\), \({\beta }_{\mathrm{7,6}} > 0\) \({\beta }_{\mathrm{4,3}} > 0\), \({\beta }_{\mathrm{7,6}}\le 0\) (or \({\beta }_{\mathrm{4,3}}\le 0\), \({\beta }_{\mathrm{7,6}} > 0\)) \({\beta }_{\mathrm{4,3}}\le 0\), \({\beta }_{\mathrm{7,6}}\le 0\) | H3A supported H3A partially supported H3A rejected | LoParo et al. (2023), Sentse et al. (2017) |
H3B: Internalizing symptoms negatively predict later friendships and social support | \({\beta }_{\mathrm{5,3}} < 0\), \({\beta }_{\mathrm{8,6}} < 0\) | \({\beta }_{\mathrm{5,3}} < 0\), \({\beta }_{\mathrm{8,6}} < 0\) \({\beta }_{\mathrm{5,3}} < 0\), \({\beta }_{\mathrm{8,6}}\ge 0\) (or \({\beta }_{\mathrm{5,3}}\ge 0\), \({\beta }_{\mathrm{8,6}} < 0\)) \({\beta }_{\mathrm{5,3}}\ge 0\), \({\beta }_{\mathrm{8,6}}\ge 0\) | H3B supported H3B partially supported H3B rejected | Beeson et al. (2020) | |
H4: Peer relationships and mental health are stable within-persons over time. Developmental cascades model (Masten & Cicchetti, 2010) | H4A: Bullying victimization positively predicts later bullying victimization | \({\beta }_{\mathrm{4,1}} > 0\), \({\beta }_{\mathrm{7,4}} > 0\) | \({\beta }_{\mathrm{4,1}} > 0\), \({\beta }_{\mathrm{7,4}} > 0\) \({\beta }_{\mathrm{4,1}} > 0\), \({\beta }_{\mathrm{7,4}}\le 0\) (or \({\beta }_{\mathrm{4,1}}\le 0\), \({\beta }_{\mathrm{7,4}} > 0\)) \({\beta }_{\mathrm{4,1}}\le 0\), \({\beta }_{\mathrm{7,4}}\le 0\) | H4A supported H4A partially supported H4A rejected | Siennick and Turanovic (2023), Fredrick et al. (2022), Bartlett et al. (2023) |
H4B: Friendship and social support positively predicts later friendship and social support | \({\beta }_{\mathrm{5,2}} > 0\), \({\beta }_{\mathrm{8,5}} > 0\) | \({\beta }_{\mathrm{5,2}} > 0\), \({\beta }_{\mathrm{8,5}} > 0\) \({\beta }_{\mathrm{5,2}} > 0\), \({\beta }_{\mathrm{8,5}}\le 0\) (or \({\beta }_{\mathrm{5,2}}\le 0\), \({\beta }_{\mathrm{8,5}} > 0\)) \({\beta }_{\mathrm{5,2}}\le 0\), \({\beta }_{\mathrm{8,5}}\le 0\) | H4B supported H4B partially supported H4B rejected | Siennick and Turanovic (2023) | |
H4C: Internalizing symptoms positively predict later Internalizing symptoms | \({\beta }_{\mathrm{6,3}} > 0\), \({\beta }_{\mathrm{9,6}} > 0\) | \({\beta }_{\mathrm{6,3}} > 0\), \({\beta }_{\mathrm{9,6}} > 0\) \({\beta }_{\mathrm{6,3}} > 0\), \({\beta }_{\mathrm{9,6}}\le 0\) (or \({\beta }_{\mathrm{6,3}}\le 0\), \({\beta }_{\mathrm{9,6}} > 0\)) \({\beta }_{\mathrm{6,3}}\le 0\), \({\beta }_{\mathrm{9,6}}\le 0\) | H4C supported H4C partially supported H4C rejected | Siennick and Turanovic (2023), Fredrick et al. (2022), Bartlett et al. (2023) | |
H5: Peer relationships and mental health share a reciprocal relationship over time. | H1 or H2, and H3, supported or partially supported Any other outcome | H5 supported H5 rejected | LoParo et al. (2023), Siennick and Turanovic (2023), Fredrick et al. (2022), Bartlett et al. (2023), Boyes et al. (2014) | ||
H6: Developmental cascades between peer relationships and mental health vary by gender. Gender socialization and intensification theory (Sravanti & Sagar Kommu, 2020) | Multi-group RI-CLPM analyses (chi-square difference tests and the test of small difference in fit) will indicate structural variance by gender | Multi-group RI-CLPM analyses indicate structural variance by gender Multi-group RI-CLPM analyses indicate structural invariance by gender | H6 supported, RI-CLPM conducted separately for boys and girls H6 rejected, single RI-CLPM conducted | Fredrick et al. (2022), Sentse et al. (2017) |
Methods
Design and Sample
Characteristic | Percentage | Number of observations |
---|---|---|
Gender | ||
Girls | 51% | 13,395 |
Boys | 49% | 13,063 |
SEN | ||
Yes | 16% | 4333 |
No | 84% | 22,125 |
FSM | ||
Yes | 28% | 7522 |
No | 72% | 18,936 |
Ethnicity | ||
White | 65% | 17,246 |
Minority | 35% | 9212 |
Measures
Bullying victimization
Friendship and social support
Internalizing symptoms
Socio-demographic variables
Analytic Strategy
Results
Construct | BVT1 | BVT2 | BVT3 | FST1 | FST2 | FST3 | IST1 | IST2 | IST3 |
---|---|---|---|---|---|---|---|---|---|
Girls | |||||||||
BVT1 | 1.00 | ||||||||
BVT2 | 0.63 | 1.00 | |||||||
BVT3 | 0.47 | 0.60 | 1.00 | ||||||
FST1 | −0.47 | −0.33 | −0.22 | 1.00 | |||||
FST2 | −0.35 | −0.46 | −0.31 | 0.51 | 1.00 | ||||
FST3 | −0.29 | −0.34 | −0.45 | 0.39 | 0.52 | 1.00 | |||
IST1 | 0.56 | 0.38 | 0.28 | −0.63 | −0.45 | −0.36 | 1.00 | ||
IST2 | 0.39 | 0.52 | 0.34 | −0.43 | −0.61 | −0.42 | 0.63 | 1.00 | |
IST3 | 0.33 | 0.39 | 0.51 | −0.31 | −0.41 | −0.57 | 0.49 | 0.57 | 1.00 |
Latent Means | 1.42 | 1.43 | 1.36 | 3.85 | 3.77 | 3.82 | 0.80 | 0.82 | 0.78 |
Latent Std. Dev. | 0.51 | 0.54 | 0.51 | 0.79 | 0.81 | 0.79 | 0.41 | 0.42 | 0.43 |
Boys | |||||||||
BVT1 | 1.00 | ||||||||
BVT2 | 0.53 | 1.00 | |||||||
BVT3 | 0.36 | 0.47 | 1.00 | ||||||
FST1 | −0.48 | −0.32 | −0.24 | 1.00 | |||||
FST2 | −0.29 | −0.51 | −0.29 | 0.50 | 1.00 | ||||
FST3 | −0.26 | −0.33 | −0.48 | 0.40 | 0.52 | 1.00 | |||
IST1 | 0.56 | 0.38 | 0.28 | −0.60 | −0.42 | −0.35 | 1.00 | ||
IST2 | 0.41 | 0.57 | 0.36 | −0.41 | −0.57 | −0.42 | 0.57 | 1.00 | |
IST3 | 0.28 | 0.34 | 0.55 | −0.29 | −0.34 | −0.54 | 0.42 | 0.49 | 1.00 |
Latent Means | 1.35 | 1.38 | 1.34 | 3.85 | 3.80 | 3.77 | 0.53 | 0.50 | 0.48 |
Latent Std. Dev. | 0.52 | 0.57 | 0.59 | 0.79 | 0.83 | 0.85 | 0.36 | 0.39 | 0.42 |
Structural (In)variance by Gender
Model | χ2 | df | RMSEA (90% CI) | CFI | TLI | SRMR | ∆χ2 | ∆df | p |
---|---|---|---|---|---|---|---|---|---|
Model fits | Comparison with Model 1 | ||||||||
Model 1 | 5036.796 | 726 | 0.030(0.029, 0.031) | 0.975 | 0.970 | 0.030 | – | – | – |
Model 2 | 5058.334 | 753 | 0.029(0.028, 0.030) | 0.974 | 0.971 | 0.031 | 51.426 | 27 | 0.003 |
Model 3 | 5016.063 | 740 | 0.029(0.029, 0.030) | 0.975 | 0.970 | 0.031 | 10.202 | 14 | 0.747 |
Model Fit
Girls | Boys | ||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|
Path | Time | b | p | CI | Beta | Hypotheses | b | p | CI | Beta | Hypotheses |
Cross-lagged pathways | |||||||||||
BV → FS | T1 → T2 | −0.08* | 0.04 | [−0.15, −0.01] | −0.07 | H1A supported | 0.07 | 0.13 | [−0.02, 0.15] | 0.06 | H1A rejected |
BV → FS | T2 → T3 | −0.08* | 0.04 | [−0.15, −0.01] | −0.08 | H1A supported | −0.03 | 0.45 | [−0.10, 0.05] | −0.03 | H1A rejected |
BV → IS | T1 → T2 | 0.05 | 0.30 | [−0.04, 0.13] | 0.04 | H1B partially supported | 0.09** | 0.007 | [0.02, 0.16] | 0.09 | H1B supported |
BV → IS | T2 → T3 | 0.14*** | <0.001 | [0.06, 0.22] | 0.14 | H1B partially supported | 0.09** | 0.007 | [0.02, 0.16] | 0.09 | H1B supported |
FS → BV | T1 → T2 | −0.05 | 0.17 | [−0.13, 0.02] | −0.05 | H2A rejected | 0.00 | 1.00 | [−0.06, 0.06] | 0.00 | H2A rejected |
FS → BV | T2 → T3 | −0.05 | 0.17 | [−0.13, 0.02] | −0.05 | H2A rejected | 0.00 | 1.00 | [−0.06, 0.06] | 0.00 | H2A rejected |
FS → IS | T1 → T2 | −0.07 | 0.10 | [−0.15, 0.01] | −0.06 | H2B partially supported | −0.04 | 0.27 | [−0.10, 0.03] | −0.03 | H2B rejected |
FS → IS | T2 → T3 | −0.10* | 0.01 | [−0.18, −0.02] | −0.10 | H2B partially supported | −0.04 | 0.27 | [−0.10, 0.03] | −0.04 | H2B rejected |
IS → BV | T1 → T2 | 0.00 | 0.92 | [−0.08, 0.07] | 0.00 | H3A rejected | 0.08* | 0.02 | [0.01, 0.16] | 0.08 | H3A supported |
IS → BV | T2 → T3 | 0.00 | 0.92 | [−0.08, 0.07] | 0.00 | H3A rejected | 0.08* | 0.02 | [0.01, 0.16] | 0.08 | H3A supported |
IS → FS | T1 → T2 | −0.16** | 0.002 | [−0.26, −0.06] | −0.15 | H3B partially supported | −0.17*** | <0.001 | [−0.28, −0.07] | −0.17 | H3B supported |
IS → FS | T2 → T3 | −0.08 | 0.09 | [−0.16, 0.01] | −0.08 | H3B partially supported | −0.12** | 0.003 | [−0.20, −0.04] | −0.12 | H3B supported |
Autoregressive effects | |||||||||||
BV → BV | T1 → T2 | 0.37*** | <0.001 | [0.25, 0.48] | 0.34 | H4A supported | 0.24*** | <0.001 | [0.16, 0.33] | 0.23 | H4A supported |
BV → BV | T2 → T3 | 0.37*** | <0.001 | [0.25, 0.48] | 0.37 | H4A supported | 0.24*** | <0.001 | [0.16, 0.33] | 0.24 | H4A supported |
FS → FS | T1 → T2 | 0.17** | 0.003 | [0.06, 0.28] | 0.16 | H4B supported | 0.15** | 0.007 | [0.04, 0.26] | 0.15 | H4B supported |
FS → FS | T2 → T3 | 0.23*** | <0.001 | [0.13, 0.33] | 0.23 | H4B supported | 0.21*** | <0.001 | [0.11, 0.30] | 0.20 | H4B supported |
IS → IS | T1 → T2 | 0.31*** | <0.001 | [0.20, 0.42] | 0.29 | H4C supported | 0.24*** | <0.001 | [0.14, 0.34] | 0.23 | H4C supported |
IS → IS | T2 → T3 | 0.15** | 0.002 | [0.05, 0.24] | 0.15 | H4C supported | 0.16*** | <0.001 | [0.07, 0.24] | 0.16 | H4C supported |
Girls | Boys | ||||||
---|---|---|---|---|---|---|---|
Path | Time | Beta | p | CI | Beta | p | CI |
Concurrent relationships between within-person components | |||||||
FS ~ BV | T1 | −0.41*** | <0.001 | [−0.48, −0.34] | −0.35*** | <0.001 | [−0.42, −0.27] |
FS ~ BV | T2 | −0.35*** | <0.001 | [−0.39, −0.31] | −0.40*** | <0.001 | [−0.44, −0.36] |
FS ~ BV | T3 | −0.35*** | <0.001 | [−0.39, −0.31] | −0.40*** | <0.001 | [−0.44, −0.36] |
IS ~ BV | T1 | 0.52*** | <0.001 | [0.46, 0.59] | 0.45*** | <0.001 | [0.38, 0.52] |
IS ~ BV | T2 | 0.40*** | <0.001 | [0.37, 0.44] | 0.47*** | <0.001 | [0.43, 0.51] |
IS ~ BV | T3 | 0.40*** | <0.001 | [0.37, 0.44] | 0.47*** | <0.001 | [0.43, 0.51] |
IS ~ FS | T1 | −0.58*** | <0.001 | [−0.63, −0.53] | −0.48*** | <0.001 | [−0.54, −0.42] |
IS ~ FS | T2 | −0.49*** | <0.001 | [−0.53, −0.44] | −0.43*** | <0.001 | [−0.46, −0.39] |
IS ~ FS | T3 | −0.43*** | <0.001 | [−0.47, −0.38] | −0.43*** | <0.001 | [−0.46, −0.39] |
Relationships between between-person components | |||||||
FS ~ BV | T1–T3 | −0.55*** | <0.001 | [−0.66, −0.44] | −0.67*** | <0.001 | [−0.78, −0.57] |
IS ~ BV | T1–T3 | 0.60*** | <0.001 | [0.52, 0.68] | 0.70*** | <0.001 | [0.62, 0.79] |
IS ~ FS | T1–T3 | −0.73*** | <0.001 | [−0.80, −0.66] | −0.74*** | <0.001 | [−0.81, −0.67] |