Interpretaciones reales, no capturas de pantalla
Cinco outputs reales de SPSS, interpretados por completo con el mismo motor que obtienes al iniciar sesión — en tu idioma, con las tablas, listos para tu tesis.
Group Statistics
Gender N Mean Std. Deviation Std. Error Mean
LS_Mean 1 180 3.4222 .73626 .05488
2 218 3.6908 .70679 .04787
Independent Samples Test
Levene's Test for Equality of t-test for Equality of
Variances Means
[…]
Run MATRIX procedure:
Written by Andrew F. Hayes, Ph.D. www.afhayes.com
Documentation available in Hayes (2022). www.guilford.com/p/hayes3
Model : 4
Y : LS_Mean
X : IQ_Mean
M : TU_Mean
Sample
Size: 400
OUTCOME VARIABLE:
TU_Mean
Model Summary
R R-sq MSE F df1 df2
p
.2603 .0677 .5153 28.9156 1.0000 398.0000 .00
00
Model
coeff se t p LLCI ULCI
constant 2.4815 .1891 13.1212 .0000 2.1097 2.8533
IQ_Mean .2748 .0511 5.3773 .0000 .1743 .3752
Standardized coefficients
coeff
IQ_Mean .2603
OUTCOME VARIABLE:
LS_Mean
Model Summary
Page 26
R R-sq MSE F df1 df2
p
.5519 .3046 .3725 86.9548 2.0000 397.0000 .00
00
Model
coeff se t p LLCI ULCI
constant 1.0707 .1924 5.5638 .0000 .6924 1.4491
IQ_Mean .4201 .0450 9.3370 .0000 .3316 .5085
TU_Mean .2796 .0426 6.5620 .0000 .1959 .3634
Standardized coefficients
coeff
IQ_Mean .4047
TU_Mean .2844
OUTCOME VARIABLE:
LS_Mean
Model Summary
R R-sq MSE F df1 df2
p
.4787 .2292 .4118 118.3430 1.0000 398.0000 .00
00
Model
coeff se t p LLCI ULCI
constant 1.7647 .1691 10.4376 .0000 1.4323 2.0971
IQ_Mean .4969 .0457 10.8786 .0000 .4071 .5867
Standardized coefficients
coeff
IQ_Mean .4787
Total effect of X on Y
Effect se t p LLCI ULCI c_
cs
.4969 .0457 10.8786 .0000 .4071 .5867 .47
87
Direct effect of X on Y
Page 27
Effect se t p LLCI ULCI c'_
cs
.4201 .0450 9.3370 .0000 .3316 .5085 .40
47
Indirect effect(s) of X on Y:
Effect BootSE BootLLCI BootULCI
TU_Mean .0768 .0180 .0442 .1142
Completely standardized indirect effect(s) of X on Y:
Effect BootSE BootLLCI BootULCI
TU_Mean .0740 .0167 .0431 .1081
Level of confidence for all confidence intervals in output:
95.0000
Number of bootstrap samples for percentile bootstrap confidence intervals:
5000
------- H1: Prueba t para muestras independientes (LS_Mean por género)
- H1: Prueba de Levene para igualdad de varianzas
- H8: Análisis de mediación — PROCESS Modelo 4 (IQ_Mean → TU_Mean → LS_Mean; 5000 muestras bootstrap)
H1
Se realizó una prueba t para muestras independientes con el fin de examinar si LS_Mean difería entre el grupo de género 1 y el grupo de género 2. El grupo de género 3 (n = 2) fue excluido mediante especificación de sintaxis, lo que resultó en un N de análisis de 398. Previo a la prueba t, se evaluó la prueba de Levene para igualdad de varianzas, F(1, 396) = 0.31, p = .578, lo que indicó que el supuesto de igualdad de varianzas se cumplió; por lo tanto, se utilizó la fila de varianzas iguales. La prueba reveló una diferencia estadísticamente significativa en LS_Mean entre el grupo de género 1 (M = 3.42, SD = 0.74) y el grupo de género 2 (M = 3.69, SD = 0.71), t(396) = −3.703, p < .001, 95% CI [−.411, −.126]. La diferencia de medias fue de −0.27 a favor del grupo de género 2. La d de Cohen = 0.37, lo que indica un tamaño del efecto pequeño (Cohen, 1988). La hipótesis fue respaldada.
Tabla 1: Estadísticas descriptivas de LS_Mean por género
| Grupo | n | M | SD |
|---|---|---|---|
| Género 1 | 180 | 3.42 | 0.74 |
| Género 2 | 218 | 3.69 | 0.71 |
Nota. N de análisis = 398. El grupo de género 3 (n = 2) fue excluido. Levene F(1, 396) = 0.31, p = .578; se asumieron varianzas iguales. t(396) = −3.703, p < .001, 95% CI [−.411, −.126]. d de Cohen = 0.37.
H8
Se realizó un análisis de mediación con PROCESS Modelo 4 (Hayes, 2022) utilizando 5000 muestras bootstrap con el fin de examinar si TU_Mean mediaba la relación entre IQ_Mean y LS_Mean. La trayectoria a, que representa el efecto de IQ_Mean sobre TU_Mean, fue significativa, B = 0.27, SE = 0.05, t(398) = 5.377, p < .001, 95% CI [.174, .375]. La trayectoria b, que representa el efecto de TU_Mean sobre LS_Mean controlando por IQ_Mean, también fue significativa, B = 0.28, SE = 0.04, t(398) = 6.562, p < .001, 95% CI [.196, .363].
El efecto indirecto de IQ_Mean sobre LS_Mean a través de TU_Mean fue significativo, B = 0.08, BootSE = 0.02, 95% Boot CI [.044, .114], dado que el intervalo de confianza (CI) excluyó el cero. El efecto indirecto completamente estandarizado fue .074. El efecto directo de IQ_Mean sobre LS_Mean, tras controlar por TU_Mean, se mantuvo significativo, B = 0.42, SE = 0.05, t(398) = 9.337, p < .001, 95% CI [.332, .509]. El efecto total de IQ_Mean sobre LS_Mean fue B = 0.50, SE = 0.05, t(398) = 10.879, p < .001, 95% CI [.407, .587]. Dado que el efecto indirecto fue significativo y el efecto directo también se mantuvo significativo tras incluir la variable mediadora, este patrón constituye una mediación parcial. La hipótesis fue respaldada.
Tabla 8: Efectos directo, indirecto y total para la mediación IQ_Mean → TU_Mean → LS_Mean
| Efecto | B | SE | t | p | 95% CI |
|---|---|---|---|---|---|
| Trayectoria a: IQ_Mean → TU_Mean | 0.27 | 0.05 | 5.377 | < .001 | [.174, .375] |
| Trayectoria b: TU_Mean → LS_Mean | 0.28 | 0.04 | 6.562 | < .001 | [.196, .363] |
| Efecto directo: IQ_Mean → LS_Mean | 0.42 | 0.05 | 9.337 | < .001 | [.332, .509] |
| Efecto total: IQ_Mean → LS_Mean | 0.50 | 0.05 | 10.879 | < .001 | [.407, .587] |
| Efecto indirecto (IQ_Mean → TU_Mean → LS_Mean) | 0.08 | 0.02† | — | — | [.044, .114]† |
Nota. N = 400. PROCESS Modelo 4 (Hayes, 2022); 5000 muestras bootstrap, intervalos de confianza (CI) al 95%. †Bootstrap SE e intervalo de confianza (CI). Efecto indirecto completamente estandarizado = .074. Se indicó mediación parcial.
Referencias
Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Lawrence Erlbaum Associates.
Hayes, A. F. (2022). Introduction to mediation, moderation, and conditional process analysis: A regression-based approach (3rd ed.). The Guilford Press.
- H1: Independent Samples t-Test (LS_Mean by Gender)
- H1: Levene's Test for Equality of Variances
- H8: Mediation Analysis — PROCESS Model 4 (IQ_Mean → TU_Mean → LS_Mean; 5,000 bootstrap samples)
H1
An independent samples t-test was conducted to examine whether LS_Mean differed between Gender group 1 and Gender group 2. Gender group 3 (n = 2) was excluded by syntax specification, yielding an analysis N of 398. Prior to the t-test, Levene's test for equality of variances was evaluated, F(1, 396) = 0.31, p = .578, indicating that the assumption of equal variances was met; the equal-variances row was therefore used. The test revealed a statistically significant difference in LS_Mean between Gender group 1 (M = 3.42, SD = 0.74) and Gender group 2 (M = 3.69, SD = 0.71), t(396) = −3.703, p < .001, 95% CI [−.411, −.126]. The mean difference was −0.27 in favour of Gender group 2. Cohen's d = 0.37, indicating a small effect size (Cohen, 1988). The hypothesis was supported.
Table 1: Group Statistics for LS_Mean by Gender
| Group | n | M | SD |
|---|---|---|---|
| Gender 1 | 180 | 3.42 | 0.74 |
| Gender 2 | 218 | 3.69 | 0.71 |
Note. Analysis N = 398. Gender group 3 (n = 2) was excluded. Levene's F(1, 396) = 0.31, p = .578; equal variances assumed. t(396) = −3.703, p < .001, 95% CI [−.411, −.126]. Cohen's d = 0.37.
H8
PROCESS Model 4 mediation analysis (Hayes, 2022) was conducted using 5,000 bootstrap samples to examine whether TU_Mean mediated the relationship between IQ_Mean and LS_Mean. The a-path, representing the effect of IQ_Mean on TU_Mean, was significant, B = 0.27, SE = 0.05, t(398) = 5.377, p < .001, 95% CI [.174, .375]. The b-path, representing the effect of TU_Mean on LS_Mean controlling for IQ_Mean, was also significant, B = 0.28, SE = 0.04, t(398) = 6.562, p < .001, 95% CI [.196, .363].
The indirect effect of IQ_Mean on LS_Mean through TU_Mean was significant, B = 0.08, BootSE = 0.02, 95% Boot CI [.044, .114], as the confidence interval excluded zero. The completely standardized indirect effect was .074. The direct effect of IQ_Mean on LS_Mean, after controlling for TU_Mean, remained significant, B = 0.42, SE = 0.05, t(398) = 9.337, p < .001, 95% CI [.332, .509]. The total effect of IQ_Mean on LS_Mean was B = 0.50, SE = 0.05, t(398) = 10.879, p < .001, 95% CI [.407, .587]. Because the indirect effect was significant and the direct effect also remained significant after including the mediator, this pattern constitutes partial mediation. The hypothesis was supported.
Table 8: Direct, Indirect, and Total Effects for the Mediation of IQ_Mean → TU_Mean → LS_Mean
| Effect | B | SE | t | p | 95% CI |
|---|---|---|---|---|---|
| a-path: IQ_Mean → TU_Mean | 0.27 | 0.05 | 5.377 | < .001 | [.174, .375] |
| b-path: TU_Mean → LS_Mean | 0.28 | 0.04 | 6.562 | < .001 | [.196, .363] |
| Direct effect: IQ_Mean → LS_Mean | 0.42 | 0.05 | 9.337 | < .001 | [.332, .509] |
| Total effect: IQ_Mean → LS_Mean | 0.50 | 0.05 | 10.879 | < .001 | [.407, .587] |
| Indirect effect (IQ_Mean → TU_Mean → LS_Mean) | 0.08 | 0.02† | — | — | [.044, .114]† |
Note. N = 400. PROCESS Model 4 (Hayes, 2022); 5,000 bootstrap samples, 95% confidence intervals. †Bootstrap SE and CI. Completely standardized indirect effect = .074. Partial mediation was indicated.
References
Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Lawrence Erlbaum Associates.
Hayes, A. F. (2022). Introduction to mediation, moderation, and conditional process analysis: A regression-based approach (3rd ed.). The Guilford Press.
¿Listo para tu propio output?
Sube tu PDF o pega tu texto — la misma calidad, listo en minutos.