Romantic Relationships and Sexual Experiences Study 1
Abby Person
2023-08-31
This data analysis was conducted using the Abby_T1T2T3 complete 3.26.12.sav data file.
pacman::p_load(dplyr, tidyr, knitr, ggplot2, kableExtra,
BSDA, car, psych, readr, magrittr,
MASS, gridExtra, Rfit, pwr, pscl, InteractionPoweR, robustHD)
pacman::p_load(tidyverse, MASS,
magrittr, robustHD,
tibble, psych,
kableExtra, e1071,
knitr, tidyr,
lavaan, semPlot,
jtools, car,
lmtest, ggpubr,
FSA, rstatix,
writexl, readxl,
rcompanion, coin, lm.beta, Hmisc)Basic Data Cleaning
Variable creation
Subscales were already created for original datafile such that if a participant answered 80% of the items in a scale or more, the average of their completed items was entered as their missing value.
Scale reliabilities
repdata.1<- subset(repdata, Sex==2)
repdata.2<- subset(repdata.1, Romantic.relationship.T1==1)
#Attachment Avoidance
repdata.2 %>%
select(AAQ.1r, AAQ.2, AAQ.3r, AAQ.5, AAQ.6, AAQ.7, AAQ.8, AAQ.9) %>%
psych::alpha( ,check.keys = F, na.rm = TRUE)##
## Reliability analysis
## Call: psych::alpha(x = ., na.rm = TRUE, check.keys = F)
##
## raw_alpha std.alpha G6(smc) average_r S/N ase mean sd median_r
## 0.84 0.84 0.84 0.4 5.2 0.011 3.3 1.1 0.41
##
## 95% confidence boundaries
## lower alpha upper
## Feldt 0.82 0.84 0.86
## Duhachek 0.82 0.84 0.86
##
## Reliability if an item is dropped:
## raw_alpha std.alpha G6(smc) average_r S/N alpha se var.r med.r
## AAQ.1r 0.82 0.82 0.83 0.40 4.6 0.012 0.040 0.42
## AAQ.2 0.83 0.83 0.83 0.41 4.9 0.012 0.036 0.46
## AAQ.3r 0.86 0.87 0.86 0.48 6.4 0.010 0.012 0.46
## AAQ.5 0.81 0.80 0.80 0.37 4.1 0.014 0.030 0.40
## AAQ.6 0.80 0.80 0.80 0.36 4.0 0.014 0.028 0.40
## AAQ.7 0.82 0.82 0.82 0.40 4.6 0.013 0.033 0.42
## AAQ.8 0.80 0.80 0.79 0.36 3.9 0.014 0.026 0.40
## AAQ.9 0.82 0.82 0.82 0.39 4.5 0.013 0.038 0.42
##
## Item statistics
## n raw.r std.r r.cor r.drop mean sd
## AAQ.1r 432 0.68 0.69 0.61 0.57 3.5 1.6
## AAQ.2 430 0.63 0.62 0.53 0.49 4.2 1.8
## AAQ.3r 430 0.35 0.37 0.22 0.19 3.0 1.5
## AAQ.5 429 0.78 0.79 0.77 0.70 2.8 1.6
## AAQ.6 431 0.80 0.80 0.79 0.72 3.0 1.7
## AAQ.7 428 0.70 0.68 0.62 0.57 4.0 1.8
## AAQ.8 428 0.83 0.83 0.83 0.76 2.9 1.6
## AAQ.9 432 0.71 0.71 0.65 0.60 2.8 1.7
##
## Non missing response frequency for each item
## 1 2 3 4 5 6 7 miss
## AAQ.1r 0.10 0.21 0.24 0.18 0.13 0.10 0.04 0.01
## AAQ.2 0.05 0.18 0.11 0.19 0.19 0.17 0.11 0.02
## AAQ.3r 0.15 0.27 0.26 0.15 0.10 0.05 0.01 0.02
## AAQ.5 0.21 0.32 0.16 0.16 0.09 0.04 0.03 0.02
## AAQ.6 0.23 0.27 0.12 0.16 0.12 0.07 0.02 0.01
## AAQ.7 0.08 0.18 0.14 0.17 0.18 0.14 0.10 0.02
## AAQ.8 0.21 0.31 0.15 0.13 0.11 0.06 0.03 0.02
## AAQ.9 0.27 0.26 0.16 0.14 0.09 0.05 0.03 0.01#Attachment Anxiety
repdata.2 %>%
select(AAQ.4r, AAQ.10, AAQ.11, AAQ.12r, AAQ.13, AAQ.14r, AAQ.15, AAQ.16r, AAQ.17r) %>%
psych::alpha( ,check.keys = F, na.rm = TRUE)##
## Reliability analysis
## Call: psych::alpha(x = ., na.rm = TRUE, check.keys = F)
##
## raw_alpha std.alpha G6(smc) average_r S/N ase mean sd median_r
## 0.81 0.81 0.83 0.32 4.2 0.013 3.3 1.1 0.37
##
## 95% confidence boundaries
## lower alpha upper
## Feldt 0.78 0.81 0.84
## Duhachek 0.79 0.81 0.84
##
## Reliability if an item is dropped:
## raw_alpha std.alpha G6(smc) average_r S/N alpha se var.r med.r
## AAQ.4r 0.79 0.79 0.81 0.32 3.7 0.015 0.027 0.35
## AAQ.10 0.81 0.81 0.82 0.35 4.2 0.013 0.026 0.38
## AAQ.11 0.77 0.76 0.78 0.29 3.2 0.017 0.026 0.21
## AAQ.12r 0.80 0.80 0.82 0.33 3.9 0.014 0.027 0.36
## AAQ.13 0.81 0.80 0.81 0.33 4.0 0.014 0.027 0.38
## AAQ.14r 0.78 0.78 0.80 0.31 3.6 0.015 0.024 0.35
## AAQ.15 0.81 0.80 0.81 0.34 4.1 0.013 0.024 0.38
## AAQ.16r 0.78 0.78 0.80 0.31 3.5 0.015 0.027 0.36
## AAQ.17r 0.78 0.78 0.79 0.30 3.5 0.016 0.024 0.35
##
## Item statistics
## n raw.r std.r r.cor r.drop mean sd
## AAQ.4r 431 0.64 0.62 0.56 0.52 4.0 1.8
## AAQ.10 431 0.45 0.50 0.40 0.33 2.6 1.4
## AAQ.11 430 0.79 0.78 0.77 0.70 2.9 1.9
## AAQ.12r 430 0.62 0.58 0.50 0.46 3.7 2.0
## AAQ.13 428 0.52 0.56 0.49 0.40 2.4 1.4
## AAQ.14r 431 0.70 0.68 0.64 0.59 4.1 1.9
## AAQ.15 430 0.49 0.53 0.46 0.35 3.2 1.6
## AAQ.16r 429 0.70 0.69 0.64 0.59 4.3 1.8
## AAQ.17r 431 0.73 0.71 0.69 0.62 2.9 1.8
##
## Non missing response frequency for each item
## 1 2 3 4 5 6 7 miss
## AAQ.4r 0.08 0.18 0.14 0.14 0.21 0.15 0.08 0.01
## AAQ.10 0.24 0.36 0.15 0.16 0.07 0.03 0.00 0.01
## AAQ.11 0.31 0.25 0.09 0.11 0.11 0.07 0.05 0.02
## AAQ.12r 0.18 0.21 0.09 0.10 0.17 0.15 0.10 0.02
## AAQ.13 0.31 0.32 0.14 0.12 0.07 0.03 0.01 0.02
## AAQ.14r 0.08 0.19 0.13 0.11 0.18 0.19 0.10 0.01
## AAQ.15 0.16 0.26 0.17 0.19 0.13 0.06 0.03 0.02
## AAQ.16r 0.07 0.14 0.12 0.14 0.22 0.21 0.09 0.02
## AAQ.17r 0.30 0.21 0.14 0.12 0.12 0.08 0.03 0.01# Partner Trust
repdata.2 %>%
select(PTrust.2, PTrust.3, PTrust.9, PTrust.10, PTrust.11, PTrust.12, PTrust.16,
PTrust.1, PTrust.7, PTrust.13, PTrust.15, PTrust.17, PTrust.4r, PTrust.5r, PTrust.6r, PTrust.8, PTrust.14r) %>%
psych::alpha( ,check.keys = F, na.rm = TRUE)##
## Reliability analysis
## Call: psych::alpha(x = ., na.rm = TRUE, check.keys = F)
##
## raw_alpha std.alpha G6(smc) average_r S/N ase mean sd median_r
## 0.9 0.91 0.92 0.36 9.7 0.0072 5.5 0.94 0.35
##
## 95% confidence boundaries
## lower alpha upper
## Feldt 0.88 0.9 0.91
## Duhachek 0.88 0.9 0.91
##
## Reliability if an item is dropped:
## raw_alpha std.alpha G6(smc) average_r S/N alpha se var.r med.r
## PTrust.2 0.89 0.90 0.92 0.36 9.0 0.0077 0.023 0.35
## PTrust.3 0.89 0.90 0.91 0.35 8.6 0.0080 0.021 0.34
## PTrust.9 0.89 0.90 0.91 0.35 8.8 0.0079 0.023 0.34
## PTrust.10 0.89 0.90 0.91 0.36 8.8 0.0079 0.022 0.35
## PTrust.11 0.88 0.89 0.91 0.35 8.5 0.0081 0.020 0.34
## PTrust.12 0.89 0.90 0.91 0.35 8.7 0.0080 0.021 0.34
## PTrust.16 0.89 0.90 0.92 0.36 8.9 0.0078 0.023 0.35
## PTrust.1 0.89 0.90 0.92 0.37 9.5 0.0074 0.024 0.36
## PTrust.7 0.90 0.91 0.92 0.38 9.9 0.0070 0.022 0.36
## PTrust.13 0.89 0.90 0.92 0.36 9.1 0.0077 0.024 0.35
## PTrust.15 0.89 0.90 0.91 0.35 8.7 0.0080 0.023 0.34
## PTrust.17 0.89 0.90 0.91 0.36 9.0 0.0078 0.024 0.35
## PTrust.4r 0.90 0.91 0.92 0.39 10.3 0.0068 0.019 0.36
## PTrust.5r 0.89 0.90 0.92 0.37 9.2 0.0076 0.025 0.35
## PTrust.6r 0.90 0.91 0.92 0.38 9.7 0.0072 0.023 0.36
## PTrust.8 0.89 0.90 0.92 0.37 9.2 0.0076 0.024 0.36
## PTrust.14r 0.89 0.90 0.92 0.37 9.5 0.0074 0.024 0.36
##
## Item statistics
## n raw.r std.r r.cor r.drop mean sd
## PTrust.2 429 0.65 0.66 0.64 0.60 5.6 1.5
## PTrust.3 424 0.76 0.77 0.77 0.72 5.8 1.4
## PTrust.9 425 0.72 0.74 0.72 0.68 6.0 1.2
## PTrust.10 426 0.71 0.72 0.70 0.66 5.7 1.4
## PTrust.11 427 0.81 0.82 0.84 0.78 5.7 1.3
## PTrust.12 427 0.76 0.77 0.78 0.72 5.6 1.4
## PTrust.16 423 0.68 0.70 0.68 0.64 6.0 1.3
## PTrust.1 425 0.56 0.54 0.50 0.47 5.1 1.7
## PTrust.7 424 0.45 0.43 0.37 0.35 4.9 1.9
## PTrust.13 425 0.64 0.63 0.61 0.57 5.6 1.7
## PTrust.15 425 0.76 0.77 0.76 0.72 5.7 1.3
## PTrust.17 425 0.66 0.67 0.65 0.61 5.3 1.5
## PTrust.4r 425 0.36 0.33 0.26 0.24 4.7 1.8
## PTrust.5r 427 0.61 0.60 0.58 0.54 5.8 1.4
## PTrust.6r 427 0.48 0.47 0.41 0.39 5.3 1.7
## PTrust.8 425 0.60 0.61 0.58 0.54 5.4 1.4
## PTrust.14r 428 0.55 0.54 0.49 0.46 5.9 1.5
##
## Non missing response frequency for each item
## 1 2 3 4 5 6 7 miss
## PTrust.2 0.01 0.04 0.05 0.09 0.18 0.30 0.32 0.02
## PTrust.3 0.01 0.04 0.04 0.08 0.16 0.26 0.42 0.03
## PTrust.9 0.01 0.01 0.02 0.06 0.14 0.32 0.44 0.03
## PTrust.10 0.02 0.02 0.04 0.08 0.18 0.32 0.33 0.03
## PTrust.11 0.01 0.02 0.04 0.08 0.20 0.32 0.33 0.02
## PTrust.12 0.01 0.02 0.04 0.10 0.20 0.28 0.34 0.02
## PTrust.16 0.01 0.02 0.03 0.06 0.13 0.31 0.43 0.03
## PTrust.1 0.06 0.05 0.06 0.13 0.21 0.23 0.26 0.03
## PTrust.7 0.07 0.08 0.07 0.12 0.18 0.25 0.22 0.03
## PTrust.13 0.04 0.05 0.04 0.08 0.12 0.26 0.41 0.03
## PTrust.15 0.01 0.03 0.03 0.10 0.17 0.34 0.32 0.03
## PTrust.17 0.03 0.03 0.07 0.14 0.20 0.31 0.22 0.03
## PTrust.4r 0.06 0.09 0.13 0.16 0.12 0.25 0.18 0.03
## PTrust.5r 0.02 0.01 0.07 0.08 0.12 0.32 0.39 0.02
## PTrust.6r 0.03 0.06 0.12 0.08 0.13 0.28 0.30 0.02
## PTrust.8 0.02 0.03 0.06 0.09 0.25 0.32 0.23 0.03
## PTrust.14r 0.02 0.02 0.06 0.07 0.07 0.29 0.47 0.02# Responsiveness
repdata.2 %>%
select(REIS.1, REIS.2, REIS.3, REIS.4, REIS.5, REIS.6, REIS.7, REIS.8, REIS.9,
REIS.10, REIS.11, REIS.12, REIS.13, REIS.14, REIS.15, REIS.16, REIS.17, REIS.18) %>%
psych::alpha( ,check.keys = F, na.rm = TRUE)##
## Reliability analysis
## Call: psych::alpha(x = ., na.rm = TRUE, check.keys = F)
##
## raw_alpha std.alpha G6(smc) average_r S/N ase mean sd median_r
## 0.96 0.97 0.97 0.62 29 0.0025 7.3 1.4 0.64
##
## 95% confidence boundaries
## lower alpha upper
## Feldt 0.96 0.96 0.97
## Duhachek 0.96 0.96 0.97
##
## Reliability if an item is dropped:
## raw_alpha std.alpha G6(smc) average_r S/N alpha se var.r med.r
## REIS.1 0.96 0.97 0.97 0.62 28 0.0026 0.0159 0.64
## REIS.2 0.96 0.96 0.97 0.62 27 0.0026 0.0160 0.64
## REIS.3 0.97 0.97 0.98 0.65 31 0.0022 0.0082 0.65
## REIS.4 0.96 0.96 0.97 0.61 27 0.0027 0.0165 0.64
## REIS.5 0.96 0.96 0.97 0.61 27 0.0027 0.0160 0.62
## REIS.6 0.96 0.96 0.97 0.62 27 0.0026 0.0158 0.64
## REIS.7 0.96 0.96 0.97 0.61 27 0.0027 0.0156 0.62
## REIS.8 0.96 0.97 0.97 0.63 29 0.0025 0.0150 0.65
## REIS.9 0.96 0.96 0.97 0.62 27 0.0027 0.0163 0.64
## REIS.10 0.96 0.96 0.97 0.61 27 0.0027 0.0157 0.63
## REIS.11 0.96 0.96 0.97 0.61 27 0.0027 0.0155 0.63
## REIS.12 0.96 0.96 0.97 0.61 27 0.0027 0.0149 0.63
## REIS.13 0.96 0.96 0.97 0.61 27 0.0027 0.0145 0.63
## REIS.14 0.96 0.96 0.97 0.62 27 0.0026 0.0150 0.64
## REIS.15 0.96 0.96 0.97 0.61 26 0.0027 0.0144 0.63
## REIS.16 0.96 0.96 0.97 0.61 27 0.0027 0.0158 0.63
## REIS.17 0.96 0.96 0.97 0.62 27 0.0027 0.0149 0.64
## REIS.18 0.96 0.96 0.97 0.61 27 0.0027 0.0153 0.63
##
## Item statistics
## n raw.r std.r r.cor r.drop mean sd
## REIS.1 428 0.75 0.74 0.73 0.71 7.0 1.8
## REIS.2 422 0.79 0.79 0.78 0.76 7.5 1.6
## REIS.3 424 0.53 0.51 0.47 0.46 6.2 2.2
## REIS.4 425 0.82 0.82 0.81 0.80 7.0 1.6
## REIS.5 425 0.83 0.84 0.83 0.81 7.3 1.7
## REIS.6 422 0.79 0.79 0.78 0.76 7.7 1.6
## REIS.7 424 0.86 0.86 0.86 0.84 7.6 1.7
## REIS.8 424 0.69 0.68 0.65 0.64 7.0 2.0
## REIS.9 424 0.79 0.79 0.77 0.76 6.8 1.9
## REIS.10 427 0.86 0.86 0.85 0.84 7.1 1.9
## REIS.11 423 0.83 0.83 0.82 0.80 7.1 1.8
## REIS.12 424 0.85 0.85 0.85 0.83 7.7 1.6
## REIS.13 426 0.86 0.86 0.86 0.84 7.5 1.7
## REIS.14 424 0.79 0.80 0.79 0.77 7.8 1.5
## REIS.15 424 0.87 0.87 0.87 0.85 7.7 1.5
## REIS.16 425 0.85 0.85 0.84 0.82 7.2 1.9
## REIS.17 425 0.79 0.81 0.80 0.77 8.0 1.4
## REIS.18 423 0.83 0.83 0.83 0.81 7.5 1.7
##
## Non missing response frequency for each item
## 1 2 3 4 5 6 7 8 9 miss
## REIS.1 0.00 0.01 0.04 0.02 0.11 0.09 0.29 0.17 0.26 0.02
## REIS.2 0.00 0.01 0.04 0.02 0.05 0.08 0.24 0.19 0.37 0.03
## REIS.3 0.03 0.04 0.08 0.06 0.14 0.12 0.22 0.15 0.17 0.03
## REIS.4 0.00 0.01 0.03 0.04 0.09 0.13 0.28 0.21 0.21 0.03
## REIS.5 0.00 0.01 0.03 0.03 0.08 0.09 0.23 0.21 0.32 0.03
## REIS.6 0.00 0.01 0.02 0.03 0.06 0.06 0.18 0.21 0.43 0.03
## REIS.7 0.00 0.01 0.02 0.03 0.04 0.08 0.17 0.23 0.42 0.03
## REIS.8 0.00 0.02 0.06 0.04 0.09 0.10 0.21 0.18 0.29 0.03
## REIS.9 0.00 0.02 0.06 0.04 0.11 0.13 0.22 0.17 0.25 0.03
## REIS.10 0.00 0.02 0.05 0.03 0.11 0.11 0.20 0.19 0.29 0.02
## REIS.11 0.01 0.02 0.02 0.04 0.08 0.15 0.20 0.20 0.29 0.03
## REIS.12 0.00 0.00 0.03 0.03 0.05 0.07 0.20 0.20 0.43 0.03
## REIS.13 0.00 0.01 0.03 0.03 0.06 0.10 0.20 0.24 0.34 0.03
## REIS.14 0.00 0.01 0.01 0.02 0.04 0.08 0.18 0.23 0.43 0.03
## REIS.15 0.00 0.00 0.03 0.02 0.04 0.09 0.17 0.27 0.38 0.03
## REIS.16 0.01 0.03 0.03 0.03 0.09 0.12 0.16 0.23 0.32 0.03
## REIS.17 0.00 0.00 0.01 0.02 0.04 0.05 0.14 0.22 0.51 0.03
## REIS.18 0.00 0.01 0.03 0.03 0.06 0.11 0.17 0.22 0.37 0.03# Felt Security
repdata.2 %>%
select(Security.1, Security.2, Security.3, Security.4, Security.5, Security.6, Security.7, Security.8r, Security.9r, Security.10r, Security.11r, Security.12r, Security.13r, Security.14r, Security.15r, Security.16r, Security.17r, Security.18r) %>%
psych::alpha( ,check.keys = F, na.rm = TRUE)##
## Reliability analysis
## Call: psych::alpha(x = ., na.rm = TRUE, check.keys = F)
##
## raw_alpha std.alpha G6(smc) average_r S/N ase mean sd median_r
## 0.94 0.95 0.96 0.49 17 0.0042 6.1 0.87 0.5
##
## 95% confidence boundaries
## lower alpha upper
## Feldt 0.93 0.94 0.95
## Duhachek 0.93 0.94 0.95
##
## Reliability if an item is dropped:
## raw_alpha std.alpha G6(smc) average_r S/N alpha se var.r med.r
## Security.1 0.94 0.95 0.96 0.50 17 0.0043 0.019 0.50
## Security.2 0.94 0.94 0.96 0.49 16 0.0044 0.021 0.50
## Security.3 0.94 0.94 0.96 0.49 16 0.0045 0.020 0.49
## Security.4 0.94 0.94 0.96 0.50 17 0.0044 0.021 0.50
## Security.5 0.94 0.94 0.96 0.49 16 0.0045 0.020 0.49
## Security.6 0.93 0.94 0.96 0.49 16 0.0045 0.021 0.49
## Security.7 0.93 0.94 0.96 0.49 16 0.0045 0.020 0.49
## Security.8r 0.94 0.94 0.96 0.50 17 0.0044 0.022 0.50
## Security.9r 0.94 0.94 0.96 0.49 16 0.0045 0.022 0.49
## Security.10r 0.94 0.94 0.96 0.50 17 0.0044 0.021 0.50
## Security.11r 0.94 0.94 0.96 0.50 17 0.0044 0.021 0.50
## Security.12r 0.94 0.94 0.96 0.49 16 0.0045 0.022 0.49
## Security.13r 0.94 0.94 0.96 0.49 16 0.0045 0.021 0.49
## Security.14r 0.95 0.95 0.96 0.51 18 0.0038 0.019 0.51
## Security.15r 0.93 0.94 0.96 0.48 16 0.0047 0.022 0.48
## Security.16r 0.93 0.94 0.96 0.48 16 0.0046 0.021 0.48
## Security.17r 0.93 0.94 0.96 0.48 16 0.0046 0.021 0.48
## Security.18r 0.94 0.94 0.96 0.50 17 0.0044 0.022 0.50
##
## Item statistics
## n raw.r std.r r.cor r.drop mean sd
## Security.1 429 0.58 0.59 0.56 0.53 6.3 1.13
## Security.2 426 0.72 0.73 0.72 0.68 6.3 1.02
## Security.3 425 0.76 0.77 0.78 0.73 6.3 1.00
## Security.4 426 0.67 0.68 0.66 0.63 5.9 1.26
## Security.5 426 0.77 0.78 0.78 0.74 6.3 1.03
## Security.6 426 0.77 0.78 0.78 0.74 6.1 1.21
## Security.7 423 0.78 0.79 0.79 0.75 6.2 1.17
## Security.8r 428 0.69 0.68 0.66 0.63 6.0 1.26
## Security.9r 426 0.75 0.75 0.74 0.71 6.3 1.24
## Security.10r 426 0.66 0.66 0.64 0.61 6.1 1.10
## Security.11r 424 0.64 0.65 0.63 0.60 6.4 0.92
## Security.12r 425 0.73 0.73 0.72 0.69 6.3 1.17
## Security.13r 425 0.76 0.76 0.76 0.72 6.5 1.01
## Security.14r 427 0.54 0.49 0.45 0.44 4.9 1.92
## Security.15r 425 0.83 0.81 0.80 0.79 6.0 1.53
## Security.16r 424 0.84 0.83 0.83 0.81 6.3 1.18
## Security.17r 425 0.81 0.80 0.80 0.78 6.4 1.13
## Security.18r 428 0.70 0.68 0.66 0.64 5.9 1.57
##
## Non missing response frequency for each item
## 1 2 3 4 5 6 7 miss
## Security.1 0.00 0.00 0.02 0.07 0.09 0.18 0.62 0.02
## Security.2 0.00 0.00 0.02 0.05 0.10 0.26 0.58 0.03
## Security.3 0.00 0.00 0.02 0.05 0.10 0.27 0.55 0.03
## Security.4 0.01 0.02 0.04 0.06 0.18 0.30 0.41 0.03
## Security.5 0.00 0.01 0.02 0.05 0.11 0.27 0.55 0.03
## Security.6 0.00 0.01 0.03 0.06 0.14 0.24 0.52 0.03
## Security.7 0.00 0.01 0.02 0.05 0.13 0.24 0.54 0.03
## Security.8r 0.01 0.02 0.03 0.08 0.08 0.33 0.46 0.02
## Security.9r 0.01 0.02 0.02 0.05 0.06 0.23 0.61 0.03
## Security.10r 0.00 0.01 0.03 0.05 0.09 0.40 0.42 0.03
## Security.11r 0.00 0.00 0.02 0.01 0.06 0.29 0.62 0.03
## Security.12r 0.01 0.01 0.03 0.04 0.06 0.25 0.59 0.03
## Security.13r 0.00 0.01 0.02 0.03 0.05 0.21 0.68 0.03
## Security.14r 0.06 0.09 0.11 0.14 0.11 0.20 0.30 0.02
## Security.15r 0.02 0.04 0.04 0.08 0.08 0.21 0.54 0.03
## Security.16r 0.01 0.01 0.03 0.04 0.07 0.21 0.63 0.03
## Security.17r 0.00 0.01 0.03 0.04 0.05 0.23 0.64 0.03
## Security.18r 0.03 0.02 0.05 0.11 0.07 0.18 0.54 0.02# PRQC
repdata.2 %>%
select(PRQC.1, PRQC.2, PRQC.3, PRQC.4,
PRQC.5, PRQC.6, PRQC.7, PRQC.8, PRQC.9, PRQC.10,
PRQC.11, PRQC.12, PRQC.13, PRQC.14, PRQC.15, PRQC.16, PRQC.17, PRQC.18) %>%
psych::alpha( ,check.keys = F, na.rm = TRUE)##
## Reliability analysis
## Call: psych::alpha(x = ., na.rm = TRUE, check.keys = F)
##
## raw_alpha std.alpha G6(smc) average_r S/N ase mean sd median_r
## 0.95 0.96 0.98 0.56 22 0.0034 6 0.9 0.58
##
## 95% confidence boundaries
## lower alpha upper
## Feldt 0.95 0.95 0.96
## Duhachek 0.95 0.95 0.96
##
## Reliability if an item is dropped:
## raw_alpha std.alpha G6(smc) average_r S/N alpha se var.r med.r
## PRQC.1 0.95 0.95 0.98 0.55 20 0.0038 0.026 0.57
## PRQC.2 0.95 0.95 0.97 0.55 20 0.0038 0.026 0.56
## PRQC.3 0.95 0.95 0.98 0.55 20 0.0038 0.026 0.57
## PRQC.4 0.95 0.95 0.98 0.55 21 0.0037 0.026 0.58
## PRQC.5 0.95 0.95 0.97 0.55 21 0.0037 0.026 0.57
## PRQC.6 0.95 0.95 0.98 0.55 21 0.0037 0.026 0.57
## PRQC.7 0.95 0.96 0.98 0.56 21 0.0037 0.028 0.59
## PRQC.8 0.95 0.95 0.98 0.55 20 0.0038 0.027 0.56
## PRQC.9 0.95 0.95 0.98 0.55 21 0.0038 0.027 0.56
## PRQC.10 0.95 0.96 0.98 0.57 22 0.0035 0.025 0.59
## PRQC.11 0.95 0.96 0.98 0.56 22 0.0036 0.027 0.58
## PRQC.12 0.95 0.96 0.98 0.56 22 0.0036 0.027 0.59
## PRQC.13 0.95 0.95 0.98 0.55 21 0.0037 0.029 0.58
## PRQC.14 0.96 0.96 0.98 0.58 23 0.0032 0.021 0.59
## PRQC.15 0.95 0.96 0.98 0.58 23 0.0032 0.022 0.59
## PRQC.16 0.95 0.96 0.98 0.56 21 0.0036 0.027 0.58
## PRQC.17 0.95 0.95 0.98 0.55 21 0.0037 0.028 0.57
## PRQC.18 0.95 0.95 0.98 0.55 21 0.0037 0.027 0.58
##
## Item statistics
## n raw.r std.r r.cor r.drop mean sd
## PRQC.1 428 0.87 0.86 0.86 0.83 5.8 1.24
## PRQC.2 426 0.85 0.85 0.85 0.82 5.9 1.21
## PRQC.3 425 0.86 0.86 0.86 0.83 5.9 1.21
## PRQC.4 425 0.76 0.78 0.78 0.73 6.3 1.01
## PRQC.5 425 0.81 0.83 0.83 0.78 6.3 1.00
## PRQC.6 425 0.81 0.83 0.83 0.79 6.3 1.02
## PRQC.7 426 0.77 0.76 0.74 0.74 6.0 1.20
## PRQC.8 425 0.84 0.85 0.84 0.82 6.2 1.08
## PRQC.9 428 0.86 0.84 0.84 0.82 6.1 1.17
## PRQC.10 424 0.63 0.64 0.61 0.58 6.0 1.29
## PRQC.11 426 0.73 0.72 0.71 0.68 6.0 1.18
## PRQC.12 426 0.71 0.71 0.70 0.67 6.1 1.14
## PRQC.13 423 0.80 0.78 0.77 0.77 6.0 1.23
## PRQC.14 424 0.56 0.52 0.50 0.48 5.3 1.64
## PRQC.15 420 0.58 0.54 0.52 0.50 5.4 1.59
## PRQC.16 425 0.75 0.76 0.75 0.72 6.4 1.02
## PRQC.17 427 0.78 0.80 0.80 0.77 6.3 1.07
## PRQC.18 428 0.75 0.78 0.77 0.73 6.4 0.97
##
## Non missing response frequency for each item
## 1 2 3 4 5 6 7 miss
## PRQC.1 0.00 0.01 0.06 0.06 0.18 0.32 0.37 0.02
## PRQC.2 0.00 0.01 0.05 0.06 0.19 0.30 0.39 0.03
## PRQC.3 0.00 0.02 0.04 0.08 0.14 0.31 0.42 0.03
## PRQC.4 0.00 0.00 0.02 0.05 0.09 0.21 0.62 0.03
## PRQC.5 0.00 0.00 0.01 0.06 0.11 0.21 0.60 0.03
## PRQC.6 0.00 0.00 0.02 0.06 0.09 0.23 0.60 0.03
## PRQC.7 0.00 0.02 0.03 0.08 0.14 0.29 0.44 0.03
## PRQC.8 0.00 0.01 0.02 0.05 0.13 0.28 0.52 0.03
## PRQC.9 0.00 0.01 0.03 0.06 0.14 0.29 0.47 0.02
## PRQC.10 0.01 0.02 0.04 0.06 0.11 0.33 0.44 0.03
## PRQC.11 0.00 0.01 0.04 0.06 0.10 0.35 0.44 0.03
## PRQC.12 0.00 0.02 0.02 0.07 0.13 0.32 0.45 0.03
## PRQC.13 0.00 0.02 0.03 0.06 0.15 0.30 0.44 0.03
## PRQC.14 0.03 0.05 0.07 0.12 0.18 0.25 0.29 0.03
## PRQC.15 0.03 0.04 0.06 0.10 0.21 0.27 0.29 0.04
## PRQC.16 0.00 0.01 0.01 0.05 0.08 0.22 0.62 0.03
## PRQC.17 0.00 0.01 0.02 0.06 0.09 0.23 0.59 0.02
## PRQC.18 0.00 0.01 0.00 0.06 0.06 0.24 0.62 0.02#Neuroticism
repdata.2 %>%
select(MM.3r, MM.6r, MM.12r, MM.14r, MM.16r, MM.23, MM.24, MM.34r) %>%
psych::alpha( ,check.keys = F, na.rm = TRUE)##
## Reliability analysis
## Call: psych::alpha(x = ., na.rm = TRUE, check.keys = F)
##
## raw_alpha std.alpha G6(smc) average_r S/N ase mean sd median_r
## 0.79 0.79 0.81 0.32 3.8 0.015 5.4 1.3 0.29
##
## 95% confidence boundaries
## lower alpha upper
## Feldt 0.76 0.79 0.82
## Duhachek 0.76 0.79 0.82
##
## Reliability if an item is dropped:
## raw_alpha std.alpha G6(smc) average_r S/N alpha se var.r med.r
## MM.3r 0.76 0.76 0.77 0.31 3.1 0.018 0.022 0.27
## MM.6r 0.76 0.76 0.76 0.31 3.1 0.018 0.015 0.29
## MM.12r 0.76 0.76 0.77 0.31 3.2 0.018 0.021 0.29
## MM.14r 0.78 0.78 0.80 0.33 3.5 0.016 0.026 0.29
## MM.16r 0.79 0.79 0.81 0.36 3.9 0.015 0.022 0.33
## MM.23 0.79 0.79 0.81 0.36 3.9 0.015 0.022 0.31
## MM.24 0.76 0.76 0.77 0.31 3.2 0.017 0.016 0.29
## MM.34r 0.76 0.76 0.77 0.31 3.1 0.018 0.019 0.29
##
## Item statistics
## n raw.r std.r r.cor r.drop mean sd
## MM.3r 430 0.70 0.70 0.66 0.58 4.6 2.0
## MM.6r 427 0.71 0.71 0.69 0.59 5.5 1.9
## MM.12r 425 0.70 0.70 0.66 0.57 5.3 2.0
## MM.14r 419 0.61 0.60 0.50 0.45 5.8 2.1
## MM.16r 426 0.52 0.51 0.38 0.35 4.9 2.0
## MM.23 425 0.50 0.51 0.39 0.35 5.9 1.8
## MM.24 422 0.68 0.68 0.65 0.55 5.3 1.9
## MM.34r 426 0.71 0.71 0.67 0.59 5.8 2.1
##
## Non missing response frequency for each item
## 1 2 3 4 5 6 7 8 9 miss
## MM.3r 0.05 0.09 0.14 0.31 0.11 0.10 0.10 0.09 0.02 0.02
## MM.6r 0.02 0.04 0.10 0.19 0.14 0.16 0.16 0.15 0.04 0.02
## MM.12r 0.02 0.06 0.10 0.25 0.11 0.15 0.13 0.14 0.04 0.03
## MM.14r 0.02 0.05 0.08 0.18 0.13 0.10 0.14 0.20 0.10 0.04
## MM.16r 0.03 0.08 0.15 0.23 0.14 0.13 0.10 0.10 0.03 0.03
## MM.23 0.01 0.03 0.06 0.16 0.10 0.19 0.25 0.17 0.04 0.03
## MM.24 0.03 0.05 0.08 0.21 0.17 0.20 0.12 0.13 0.03 0.03
## MM.34r 0.02 0.04 0.08 0.23 0.09 0.13 0.15 0.19 0.08 0.03Creating smaller data frame
Below I’m creating a data frame that only has the variables I’m interested in so that it’s easier to view and manage data.
Removing men from analyses
#Creating new data frame called 'women' that removed men from analyses
women<- subset(df, Sex==2)
length(women$Participant)## [1] 805#Now there is 805 casesHypothesis 1
H1: Women with a history of sexual victimization will be less likely to be in a romantic relationship than women without a history of sexual victimization.
StatusbySViolence<- table(women$RStatust1, women$SViolence)
head(StatusbySViolence)##
## -1 1
## 1 302 135
## 2 283 85# 1 = in relationship, 2 = not in relationship
# -1 = no violence; 1 = s violence
(chisqTest<- chisq.test(StatusbySViolence))##
## Pearson's Chi-squared test with Yates' continuity correction
##
## data: StatusbySViolence
## X-squared = 5.7252, df = 1, p-value = 0.01672chisqTest$expected##
## -1 1
## 1 317.5714 119.4286
## 2 267.4286 100.5714#Expected values are over 5, does meet assumptions
#Calculating Odds ratio
addmargins(StatusbySViolence)##
## -1 1 Sum
## 1 302 135 437
## 2 283 85 368
## Sum 585 220 805# OR = Odds given SV victim / Odds given not SV victim
135/85 # Victim odds of being in relationship## [1] 1.588235302/283 # Non Victim odds of being in a relationship## [1] 1.067138(135/85)/(302/283)## [1] 1.488313Contrary to our hypothesis, we would reject the null hypothesis and conclude that women with a history of sexual abuse or assault are more likely to be in a romantic relationship.
The odds of being in a romantic relationship if a woman was a victim of sexual violence was about 1.49 times higher than being in a romantic relationship if a woman was not a victim of sexual violence.
Frequencies and Descriptive Statistics
#Creating subset for sample that's in a relationship
Rwomen<- subset(women, RStatust1==1)
# Now 437 casesSee below for frequencies and descriptive statistics for women who reported being in a romantic relationship (sample used for primary analyses).
Frequencies
#Need to find frequencies for categorical variables
#### Race
Rwomen$Race<- as.numeric(Rwomen$Race)
Rwomen<- Rwomen %>% #Abby and pat recoded 'other' responses in Other.race
mutate(Race.recoded = case_when(
Rwomen$Participant==1134 ~ 5,
Rwomen$Participant==6030~ 7,
Rwomen$Participant==1141 ~ 7,
Rwomen$Participant==5111 ~ 7,
Rwomen$Participant==6049 ~ 6,
Rwomen$Participant==6128 ~ 7,
Rwomen$Participant==6026 ~ 7,
.default=Rwomen$Race
))
Rwomen$Race.recoded<- as.numeric(Rwomen$Race.recoded)
Rwomen$Race.recoded <- factor(Rwomen$Race.recoded,
levels=c(1, 2, 3, 4, 5, 6, 7),
labels=c("African American or Black", "Middle Eastern", "Asian or Asian American", "Native American", "Hispanic or Latinx", "Multiracial", "White or European American"))
table(Rwomen$Race.recoded)##
## African American or Black Middle Eastern
## 14 3
## Asian or Asian American Native American
## 55 2
## Hispanic or Latinx Multiracial
## 10 13
## White or European American
## 340prop.table(table(Rwomen$Race.recoded))##
## African American or Black Middle Eastern
## 0.032036613 0.006864989
## Asian or Asian American Native American
## 0.125858124 0.004576659
## Hispanic or Latinx Multiracial
## 0.022883295 0.029748284
## White or European American
## 0.778032037## Age
Rwomen$Age<- as.numeric(Rwomen$Age)
Rwomen$Age <- factor(Rwomen$Age,
levels=c(1, 2, 3, 4),
labels=c("18-21", "22-30", "31-40", "Over 40"))
table(Rwomen$Age)##
## 18-21 22-30 31-40 Over 40
## 307 103 17 10prop.table(table(Rwomen$Age))##
## 18-21 22-30 31-40 Over 40
## 0.7025172 0.2356979 0.0389016 0.0228833## School
Rwomen$School<- as.numeric(Rwomen$School)
Rwomen$School <- factor(Rwomen$School,
levels=c(1, 2),
labels=c("UMN", "Normandale"))
table(Rwomen$School)##
## UMN Normandale
## 329 108prop.table(table(Rwomen$School))##
## UMN Normandale
## 0.7528604 0.2471396## Partner Gender
Rwomen$partnerGender<- as.numeric(Rwomen$partnerGender)
Rwomen$partnerGender<- factor(Rwomen$partnerGender,
levels=c(1, 2, 0),
labels=c("Man", "Woman", "Unknown"))
table(Rwomen$partnerGender)##
## Man Woman Unknown
## 416 19 2prop.table(table(Rwomen$partnerGender))##
## Man Woman Unknown
## 0.951945080 0.043478261 0.004576659## Relationship type
Rwomen$RType<- as.numeric(Rwomen$RType)
Rwomen$RType<- factor(Rwomen$RType,
levels=c(1, 2, 3, 0),
labels=c("Dating", "Engaged", "Married", "Unknown"))
table(Rwomen$RType) ##
## Dating Engaged Married Unknown
## 364 32 40 1prop.table(table(Rwomen$RType))##
## Dating Engaged Married Unknown
## 0.83295195 0.07322654 0.09153318 0.00228833## Adult Sexual Assault
# 0=no; 1=yes
kableExtra::kable(table(Rwomen$ASA1), booktabs = TRUE, col.names = c("Adult Sexual Assault", "Frequency")) %>%
kableExtra::kable_styling(bootstrap_options = "striped", full_width = TRUE) | Adult Sexual Assault | Frequency |
|---|---|
| 0 | 384 |
| 1 | 53 |
## CSA same age
# 0 = no; 1=yes
kableExtra::kable(table(Rwomen$CSA.sameage), booktabs = TRUE, col.names = c("CSA Same Age", "Frequency")) %>%
kableExtra::kable_styling(bootstrap_options = "striped", full_width = TRUE) | CSA Same Age | Frequency |
|---|---|
| 0 | 403 |
| 1 | 34 |
## CSA Older
# 0=no; 1=yes
kableExtra::kable(table(Rwomen$CSA.Older), booktabs = TRUE, col.names = c("CSA Older", "Frequency")) %>%
kableExtra::kable_styling(bootstrap_options = "striped", full_width = TRUE) | CSA Older | Frequency |
|---|---|
| 0 | 400 |
| 1 | 37 |
## Adolescent CSA
# 0=no; 1=yes
kableExtra::kable(table(Rwomen$CSA.adolescent), booktabs = TRUE, col.names = c("Adolescent CSA", "Frequency")) %>%
kableExtra::kable_styling(bootstrap_options = "striped", full_width = TRUE)| Adolescent CSA | Frequency |
|---|---|
| 0 | 373 |
| 1 | 64 |
## Sexual Violence
# -1 = no; 1 = yes
kableExtra::kable(table(Rwomen$SViolence), booktabs = TRUE, col.names = c("Sexual Violence", "Frequency")) %>%
kableExtra::kable_styling(bootstrap_options = "striped", full_width = TRUE) | Sexual Violence | Frequency |
|---|---|
| -1 | 302 |
| 1 | 135 |
prop.table(table(Rwomen$SViolence))##
## -1 1
## 0.6910755 0.3089245## CSA
# -1 = no; 1 = yes
kableExtra::kable(table(Rwomen$CSAtotal), booktabs = TRUE, col.names = c("Any CSA", "Frequency")) %>%
kableExtra::kable_styling(bootstrap_options = "striped", full_width = TRUE) | Any CSA | Frequency |
|---|---|
| FALSE | 331 |
| TRUE | 106 |
Descriptive Statistics
# Creating relationship length variable that includes year and month information
Rwomen$RYears[is.na(Rwomen$RYears)] <- 0
Rwomen$RMonths[is.na(Rwomen$RMonths1)] <- 0
RYears<- as.numeric(Rwomen$RYears)
RMonths<- as.numeric(Rwomen$RMonths)
Rwomen$relLength<- RYears + (RMonths/12)
summary(Rwomen$relLength)## Min. 1st Qu. Median Mean 3rd Qu. Max. NA's
## 0.08333 0.50000 1.41667 2.23325 2.75000 26.66667 44# So average relationship length is 2.23 years.
# Descriptives
Rwomen %>%
dplyr::select(PRQC1, PTrust, Security, IOS.current, Responsiveness, AAQ.AV, AAQ.ANX, Neuroticism)%>%
psych::describe(na.rm=TRUE) %>%
as.data.frame() %>%
dplyr::select("n", "mean", "sd", "median", "min", "max", "range", "skew", "kurtosis") %>%
kableExtra::kable(caption= "Descriptive Statistics for Study Variables", digits = 2) %>%
kable_styling(bootstrap_options = "striped", full_width = TRUE)| n | mean | sd | median | min | max | range | skew | kurtosis | |
|---|---|---|---|---|---|---|---|---|---|
| PRQC1 | 426 | 6.04 | 0.88 | 6.23 | 2.89 | 7.00 | 4.11 | -1.23 | 1.23 |
| PTrust | 426 | 5.53 | 0.92 | 5.65 | 2.06 | 7.00 | 4.94 | -0.75 | 0.41 |
| Security | 429 | 6.12 | 0.87 | 6.39 | 2.78 | 7.00 | 4.22 | -1.30 | 1.18 |
| IOS.current | 434 | 5.00 | 1.68 | 5.00 | 1.00 | 7.00 | 6.00 | -0.47 | -0.76 |
| Responsiveness | 427 | 7.31 | 1.37 | 7.61 | 2.44 | 9.00 | 6.56 | -1.03 | 0.80 |
| AAQ.AV | 431 | 3.28 | 1.14 | 3.25 | 1.00 | 6.62 | 5.62 | 0.31 | -0.37 |
| AAQ.ANX | 431 | 3.35 | 1.10 | 3.33 | 1.00 | 6.78 | 5.78 | 0.21 | -0.32 |
| Neuroticism | 425 | 4.61 | 1.27 | 4.62 | 1.25 | 7.88 | 6.62 | 0.07 | -0.11 |
sum(Rwomen$AAQ.ANX>3.5)/length(Rwomen$AAQ.ANX) #percentage of people above midpoints## [1] NAHistograms and QQ Plots
par(mfrow=c(1, 2))
# Avoidance
hist(Rwomen$AAQ.AV, sub=paste("Skewness:",
round(e1071::skewness(Rwomen$AAQ.AV, na.rm=TRUE), 2)))
qqnorm(Rwomen$AAQ.AV, pch = 1, frame = FALSE, main="QQ Plot of Attachment Avoidance")
qqline(Rwomen$AAQ.AV, col = "hotpink", lwd = 2)
# Anxiety
hist(Rwomen$AAQ.ANX, sub=paste("Skewness:",
round(e1071::skewness(Rwomen$AAQ.ANX, na.rm=TRUE), 2)))
qqnorm(Rwomen$AAQ.ANX, pch = 1, frame = FALSE, main="QQ Plot of Attachment Anxiety")
qqline(Rwomen$AAQ.ANX, col = "hotpink", lwd = 2)
# IOS
hist(Rwomen$IOS.current, sub=paste("Skewness:",
round(e1071::skewness(Rwomen$IOS.current, na.rm=TRUE), 2)))
qqnorm(Rwomen$IOS.current, pch = 1, frame = FALSE, main="QQ Plot of IOS")
qqline(Rwomen$IOS.current, col = "hotpink", lwd = 2)
# PRQC Total Scores
hist(Rwomen$PRQC1, sub=paste("Skewness:",
round(e1071::skewness(Rwomen$PRQC1, na.rm=TRUE), 2)))
qqnorm(Rwomen$PRQC1, pch = 1, frame = FALSE, main="QQ Plot of PRQC Total Scores")
qqline(Rwomen$PRQC1, col = "hotpink", lwd = 2)
# Felt Security
hist(Rwomen$Security, sub=paste("Skewness:",
round(e1071::skewness(Rwomen$Security, na.rm=TRUE), 2)))
qqnorm(Rwomen$Security, pch = 1, frame = FALSE, main="QQ Plot of Felt Security")
qqline(Rwomen$Security, col = "hotpink", lwd = 2)
# Partner Trust
hist(Rwomen$PTrust, sub=paste("Skewness:",
round(e1071::skewness(Rwomen$PTrust, na.rm=TRUE), 2)))
qqnorm(Rwomen$PTrust, pch = 1, frame = FALSE, main="QQ Plot of Partner Trust")
qqline(Rwomen$PTrust, col = "hotpink", lwd = 2)
# Responsiveness
hist(Rwomen$Responsiveness, sub=paste("Skewness:",
round(e1071::skewness(Rwomen$Responsiveness, na.rm=TRUE), 2)))
qqnorm(Rwomen$Responsiveness, pch = 1, frame = FALSE, main="QQ Plot of Responsiveness")
qqline(Rwomen$Responsiveness, col = "hotpink", lwd = 2)
# Neuroticism
hist(Rwomen$Neuroticism, sub=paste("Skewness:",
round(e1071::skewness(Rwomen$Neuroticism, na.rm=TRUE), 2)))
qqnorm(Rwomen$Neuroticism, pch = 1, frame = FALSE, main="QQ Plot of Neuroticism")
qqline(Rwomen$Neuroticism, col = "hotpink", lwd = 2)
Correlations between Study Variables
Rwomen$AnySA [Rwomen$AnySA=="TRUE"]<-1
corrdata<- Rwomen %>%
select(SViolence, AnySA, PRQC1, PTrust, Security, IOS.current, Responsiveness, AAQ.AV, AAQ.ANX, Neuroticism)
cor<- cor(corrdata, use= "pairwise.complete.obs", method=c("spearman"))
round(cor,2)## SViolence AnySA PRQC1 PTrust Security IOS.current Responsiveness
## SViolence 1.00 1.00 0.04 -0.03 -0.03 -0.01 -0.05
## AnySA 1.00 1.00 0.04 -0.03 -0.03 -0.01 -0.05
## PRQC1 0.04 0.04 1.00 0.67 0.72 0.51 0.70
## PTrust -0.03 -0.03 0.67 1.00 0.79 0.34 0.76
## Security -0.03 -0.03 0.72 0.79 1.00 0.41 0.76
## IOS.current -0.01 -0.01 0.51 0.34 0.41 1.00 0.42
## Responsiveness -0.05 -0.05 0.70 0.76 0.76 0.42 1.00
## AAQ.AV 0.17 0.17 -0.32 -0.32 -0.28 -0.16 -0.32
## AAQ.ANX 0.15 0.15 -0.33 -0.51 -0.51 -0.21 -0.42
## Neuroticism 0.08 0.08 -0.18 -0.29 -0.30 -0.07 -0.24
## AAQ.AV AAQ.ANX Neuroticism
## SViolence 0.17 0.15 0.08
## AnySA 0.17 0.15 0.08
## PRQC1 -0.32 -0.33 -0.18
## PTrust -0.32 -0.51 -0.29
## Security -0.28 -0.51 -0.30
## IOS.current -0.16 -0.21 -0.07
## Responsiveness -0.32 -0.42 -0.24
## AAQ.AV 1.00 0.39 0.27
## AAQ.ANX 0.39 1.00 0.45
## Neuroticism 0.27 0.45 1.00rcorr(as.matrix(corrdata), type = c("spearman"))## SViolence AnySA PRQC1 PTrust Security IOS.current Responsiveness
## SViolence 1.00 1.00 0.04 -0.03 -0.03 -0.01 -0.05
## AnySA 1.00 1.00 0.04 -0.03 -0.03 -0.01 -0.05
## PRQC1 0.04 0.04 1.00 0.67 0.72 0.51 0.70
## PTrust -0.03 -0.03 0.67 1.00 0.79 0.34 0.76
## Security -0.03 -0.03 0.72 0.79 1.00 0.41 0.76
## IOS.current -0.01 -0.01 0.51 0.34 0.41 1.00 0.42
## Responsiveness -0.05 -0.05 0.70 0.76 0.76 0.42 1.00
## AAQ.AV 0.17 0.17 -0.32 -0.32 -0.28 -0.16 -0.32
## AAQ.ANX 0.15 0.15 -0.33 -0.51 -0.51 -0.21 -0.42
## Neuroticism 0.08 0.08 -0.18 -0.29 -0.30 -0.07 -0.24
## AAQ.AV AAQ.ANX Neuroticism
## SViolence 0.17 0.15 0.08
## AnySA 0.17 0.15 0.08
## PRQC1 -0.32 -0.33 -0.18
## PTrust -0.32 -0.51 -0.29
## Security -0.28 -0.51 -0.30
## IOS.current -0.16 -0.21 -0.07
## Responsiveness -0.32 -0.42 -0.24
## AAQ.AV 1.00 0.39 0.27
## AAQ.ANX 0.39 1.00 0.45
## Neuroticism 0.27 0.45 1.00
##
## n
## SViolence AnySA PRQC1 PTrust Security IOS.current Responsiveness
## SViolence 437 437 426 426 429 434 427
## AnySA 437 437 426 426 429 434 427
## PRQC1 426 426 426 424 426 426 425
## PTrust 426 426 424 426 426 426 425
## Security 429 429 426 426 429 429 427
## IOS.current 434 434 426 426 429 434 427
## Responsiveness 427 427 425 425 427 427 427
## AAQ.AV 431 431 424 425 426 430 425
## AAQ.ANX 431 431 425 426 427 430 426
## Neuroticism 425 425 421 422 423 424 422
## AAQ.AV AAQ.ANX Neuroticism
## SViolence 431 431 425
## AnySA 431 431 425
## PRQC1 424 425 421
## PTrust 425 426 422
## Security 426 427 423
## IOS.current 430 430 424
## Responsiveness 425 426 422
## AAQ.AV 431 430 424
## AAQ.ANX 430 431 425
## Neuroticism 424 425 425
##
## P
## SViolence AnySA PRQC1 PTrust Security IOS.current
## SViolence 0.0000 0.3934 0.5189 0.4748 0.8132
## AnySA 0.0000 0.3934 0.5189 0.4748 0.8132
## PRQC1 0.3934 0.3934 0.0000 0.0000 0.0000
## PTrust 0.5189 0.5189 0.0000 0.0000 0.0000
## Security 0.4748 0.4748 0.0000 0.0000 0.0000
## IOS.current 0.8132 0.8132 0.0000 0.0000 0.0000
## Responsiveness 0.2889 0.2889 0.0000 0.0000 0.0000 0.0000
## AAQ.AV 0.0003 0.0003 0.0000 0.0000 0.0000 0.0007
## AAQ.ANX 0.0025 0.0025 0.0000 0.0000 0.0000 0.0000
## Neuroticism 0.0823 0.0823 0.0003 0.0000 0.0000 0.1370
## Responsiveness AAQ.AV AAQ.ANX Neuroticism
## SViolence 0.2889 0.0003 0.0025 0.0823
## AnySA 0.2889 0.0003 0.0025 0.0823
## PRQC1 0.0000 0.0000 0.0000 0.0003
## PTrust 0.0000 0.0000 0.0000 0.0000
## Security 0.0000 0.0000 0.0000 0.0000
## IOS.current 0.0000 0.0007 0.0000 0.1370
## Responsiveness 0.0000 0.0000 0.0000
## AAQ.AV 0.0000 0.0000 0.0000
## AAQ.ANX 0.0000 0.0000 0.0000
## Neuroticism 0.0000 0.0000 0.0000mean(Rwomen$AnySA, na.rm=T)## [1] 0.3089245sd(Rwomen$AnySA, na.rm=T)## [1] 0.4625795Mean Centering Predictors
Now I am mean centering the predictor variables subset by gender and relationship status.
Rwomen$cANX<- Rwomen$AAQ.ANX - mean(Rwomen$AAQ.ANX, na.rm=TRUE)
Rwomen$cAV<- Rwomen$AAQ.AV - mean(Rwomen$AAQ.AV, na.rm=TRUE)
Rwomen$cNeuroticism<- Rwomen$Neuroticism - mean(Rwomen$Neuroticism, na.rm=TRUE)
Rwomen$cResponsiveness<- Rwomen$Responsiveness - mean(Rwomen$Responsiveness, na.rm=TRUE)
Rwomen$cBSI<- Rwomen$BSI - mean(Rwomen$BSI, na.rm=TRUE)Winsorizing Variables
## PRQC
PRQCmin <- mean(Rwomen$PRQC1, na.rm=T) - (3*(sd(Rwomen$PRQC1, na.rm=T)))
PRQCmax <- mean(Rwomen$PRQC1, na.rm=T) + (3*(sd(Rwomen$PRQC1, na.rm=T)))
Rwomen$PRQC1[which(Rwomen$PRQC1 < PRQCmin | Rwomen$PRQC1 > PRQCmax)] # 6 outliers## [1] 3.222222 2.890000 3.166667 3.277778 2.890000 2.888889Rwomen$PRQC.win<- winsor(Rwomen$PRQC1, trim=.15, na.rm=T) #winsorizing to 3sds
# Partner Trust
PTrustmin <- mean(Rwomen$PTrust, na.rm=T) - (3*(sd(Rwomen$PTrust, na.rm=T)))
PTrustmax <- mean(Rwomen$PTrust, na.rm=T) + (3*(sd(Rwomen$PTrust, na.rm=T)))
Rwomen$PTrust[which(Rwomen$PTrust < PTrustmin | Rwomen$PTrust > PTrustmax)]## [1] 2.060000 2.588235 2.058824Rwomen$PTrust.win<- winsor(Rwomen$PTrust, trim=.15, na.rm=T) #3 outliers
#Security
Securitymin <- mean(Rwomen$Security, na.rm=T) - (3*(sd(Rwomen$Security, na.rm=T)))
Securitymax <- mean(Rwomen$Security, na.rm=T) + (3*(sd(Rwomen$Security, na.rm=T)))
Rwomen$Security[which(Rwomen$Security < Securitymin | Rwomen$Security > Securitymax)]## [1] 3.333333 3.222222 3.444444 2.777778 2.888889Rwomen$Security.win<- winsor(Rwomen$Security, trim=.15, na.rm=T) # 5 outliers
# Responsiveness
Responsivenessmin <- mean(Rwomen$cResponsiveness, na.rm=T) - (3*(sd(Rwomen$cResponsiveness, na.rm=T)))
Responsivenessmax <- mean(Rwomen$cResponsiveness, na.rm=T) + (3*(sd(Rwomen$cResponsiveness, na.rm=T)))
Rwomen$cResponsiveness[which(Rwomen$cResponsiveness < Responsivenessmin | Rwomen$cResponsiveness > Responsivenessmax)] # 3 outliers## [1] -4.754546 -4.865657 -4.587879Rwomen$Responsiveness.win<- winsor(Rwomen$cResponsiveness, trim=.15, na.rm=T)
# Avoidance
AVmin <- mean(Rwomen$cAV, na.rm=T) - (3*(sd(Rwomen$cAV, na.rm=T)))
AVmax <- mean(Rwomen$cAV, na.rm=T) + (3*(sd(Rwomen$cAV, na.rm=T)))
Rwomen$cAV[which(Rwomen$cAV < AVmin | Rwomen$cAV > AVmax)]## numeric(0)Rwomen$AV.win<- winsor(Rwomen$cAV, trim=.15, na.rm=T) #0 outliers
#Anxiety
ANXmin <- mean(Rwomen$cANX, na.rm=T) - (3*(sd(Rwomen$cANX, na.rm=T)))
ANXmax <- mean(Rwomen$cANX, na.rm=T) + (3*(sd(Rwomen$cANX, na.rm=T)))
Rwomen$cANX[which(Rwomen$cANX < ANXmin | Rwomen$cANX > ANXmax)]## [1] 3.430611Rwomen$ANX.win<- winsor(Rwomen$cANX, trim=.15, na.rm=T) # 1 outlier
# Neuroticism
Nmin <- mean(Rwomen$cNeuroticism, na.rm=T) - (3*(sd(Rwomen$cNeuroticism, na.rm=T)))
Nmax <- mean(Rwomen$cNeuroticism, na.rm=T) + (3*(sd(Rwomen$cNeuroticism, na.rm=T)))
Rwomen$cNeuroticism[which(Rwomen$cNeuroticism < Nmin | Rwomen$cNeuroticism > Nmax)]## numeric(0)Rwomen$Neuroticism.win<- winsor(Rwomen$cNeuroticism, trim=.15, na.rm=T) # 0 outliersResponsiveness Regression
group_by(Rwomen, SViolence) %>%
summarise(
count = n(),
mean = mean(Responsiveness, na.rm = TRUE),
sd = sd(Responsiveness, na.rm = TRUE),
median = median(Responsiveness, na.rm = TRUE),
IQR = IQR(Responsiveness, na.rm = TRUE)
)## # A tibble: 2 × 6
## SViolence count mean sd median IQR
## <dbl> <int> <dbl> <dbl> <dbl> <dbl>
## 1 -1 302 7.39 1.26 7.67 1.56
## 2 1 135 7.14 1.57 7.5 2.15Rwomen$SViolence<- as.factor(Rwomen$SViolence)
#Regression
Responsiveness.step3<- lm(Responsiveness.win ~ SViolence*cNeuroticism + SViolence*cAV + SViolence*ANX.win, data=Rwomen)
summary(Responsiveness.step3)##
## Call:
## lm(formula = Responsiveness.win ~ SViolence * cNeuroticism +
## SViolence * cAV + SViolence * ANX.win, data = Rwomen)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.38024 -0.75941 0.06469 0.61780 1.94230
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.07180 0.05346 1.343 0.17994
## SViolence1 0.04338 0.09610 0.451 0.65192
## cNeuroticism -0.07429 0.04910 -1.513 0.13105
## cAV -0.13562 0.05184 -2.616 0.00922 **
## ANX.win -0.43512 0.07254 -5.998 4.36e-09 ***
## SViolence1:cNeuroticism 0.08024 0.08209 0.977 0.32895
## SViolence1:cAV -0.07240 0.09003 -0.804 0.42176
## SViolence1:ANX.win 0.06779 0.13495 0.502 0.61571
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.8932 on 413 degrees of freedom
## (16 observations deleted due to missingness)
## Multiple R-squared: 0.2181, Adjusted R-squared: 0.2049
## F-statistic: 16.46 on 7 and 413 DF, p-value: < 2.2e-16lm.beta(Responsiveness.step3)##
## Call:
## lm(formula = Responsiveness.win ~ SViolence * cNeuroticism +
## SViolence * cAV + SViolence * ANX.win, data = Rwomen)
##
## Standardized Coefficients::
## (Intercept) SViolence1 cNeuroticism
## NA 0.02015651 -0.09366078
## cAV ANX.win SViolence1:cNeuroticism
## -0.15302338 -0.36115727 0.06547126
## SViolence1:cAV SViolence1:ANX.win
## -0.04774068 0.03205279confint(Responsiveness.step3, level=0.95)## 2.5 % 97.5 %
## (Intercept) -0.03327637 0.17688035
## SViolence1 -0.14552266 0.23228604
## cNeuroticism -0.17081325 0.02223050
## cAV -0.23752811 -0.03371133
## ANX.win -0.57771407 -0.29252094
## SViolence1:cNeuroticism -0.08113318 0.24160440
## SViolence1:cAV -0.24936727 0.10457232
## SViolence1:ANX.win -0.19748140 0.33305335#Non-parametric test
wilcox.test(Responsiveness ~ SViolence, data=Rwomen, alternative=c("two.sided"), conf.int=T, conf.level=.95)##
## Wilcoxon rank sum test with continuity correction
##
## data: Responsiveness by SViolence
## W = 20887, p-value = 0.2886
## alternative hypothesis: true location shift is not equal to 0
## 95 percent confidence interval:
## -0.1110771 0.4444385
## sample estimates:
## difference in location
## 0.1111571wilcoxonRG(x = Rwomen$Responsiveness, g = Rwomen$SViolence, verbose=TRUE)##
## Levels: -1 1
## n for -1 = 302
## n for 1 = 135
## Mean of ranks for -1 = NA
## Mean of ranks for 1 = NA
## Difference in mean of ranks = NA
## Total n = 427
## 2 * difference / total n = NA## rg
## NAwilcox_effsize(PRQC1 ~ SViolence, data=Rwomen)## # A tibble: 1 × 7
## .y. group1 group2 effsize n1 n2 magnitude
## * <chr> <chr> <chr> <dbl> <int> <int> <ord>
## 1 PRQC1 -1 1 0.0414 292 134 smallwilcox_test(Responsiveness ~ SViolence, data=Rwomen, conf.level=.95)##
## Asymptotic Wilcoxon-Mann-Whitney Test
##
## data: Responsiveness by SViolence (-1, 1)
## Z = 1.0617, p-value = 0.2884
## alternative hypothesis: true mu is not equal to 0Regression I: PRQC Total
# Descriptives by group
group_by(Rwomen, SViolence) %>%
summarise(
count = n(),
mean = mean(PRQC1, na.rm = TRUE),
sd = sd(PRQC1, na.rm = TRUE),
median = median(PRQC1, na.rm = TRUE),
IQR = IQR(PRQC1, na.rm = TRUE)
)## # A tibble: 2 × 6
## SViolence count mean sd median IQR
## <fct> <int> <dbl> <dbl> <dbl> <dbl>
## 1 -1 302 6.02 0.889 6.22 1.07
## 2 1 135 6.09 0.865 6.28 1.19# Box Cox transformation
bP.win<- boxcox(lm(Rwomen$PRQC.win ~ 1)) #PRQC.win
lambdaP.win <- bP.win$x[which.max(bP.win$y)]
Rwomen$bcPRQC.win<- ((Rwomen$PRQC.win^lambdaP.win-1)/lambdaP.win)
PRQC.bc.lin<- lm(bcPRQC.win ~ SViolence*Responsiveness.win + SViolence*Neuroticism.win + SViolence*AV.win + SViolence*ANX.win, data=Rwomen) #model with transformed outcome variable does not improve residual plots
# GLM with gamma distribution. (Link of log vs inverse changes interpretation)
Rwomen$InversePRQC<- 8-Rwomen$PRQC1 ### Reverse scoring PRQC so it fits gamma distribution
Rwomen$InversePRQC.win<- winsor(Rwomen$InversePRQC, trim=.15, na.rm=T) #winsorizing to 3 sds from mean
PRQC.gamma.win<- glm(formula = InversePRQC.win ~ SViolence*Responsiveness.win + SViolence*Neuroticism.win + SViolence*AV.win + SViolence*ANX.win,
family = Gamma(link = "log"),
na.action=na.exclude,
data = Rwomen)
PRQC.gamma<- glm(formula = InversePRQC ~ SViolence*cNeuroticism + SViolence*cAV + SViolence*cANX,
family = Gamma(link = "log"),
na.action=na.exclude,
data = Rwomen)
confint(PRQC.gamma, level=0.95)## Waiting for profiling to be done...## 2.5 % 97.5 %
## (Intercept) 0.65086910 0.74610771
## SViolence1 -0.21282204 -0.04038919
## cNeuroticism -0.02616955 0.06210264
## cAV 0.04044081 0.13625212
## cANX 0.05812642 0.16455709
## SViolence1:cNeuroticism -0.11269856 0.04438639
## SViolence1:cAV -0.03313549 0.12997287
## SViolence1:cANX -0.13329345 0.05652916lm.beta(PRQC.gamma)##
## Call:
## glm(formula = InversePRQC ~ SViolence * cNeuroticism + SViolence *
## cAV + SViolence * cANX, family = Gamma(link = "log"), data = Rwomen,
## na.action = na.exclude)
##
## Standardized Coefficients::
## (Intercept) SViolence1 cNeuroticism
## NA -0.06689187 0.02566734
## cAV cANX SViolence1:cNeuroticism
## 0.11247187 0.13894807 -0.03160024
## SViolence1:cAV SViolence1:cANX
## 0.03612613 -0.02963318# Winsorized Linear model
PRQC.win.lin<- lm(PRQC.win ~ SViolence*Responsiveness.win + SViolence*Neuroticism.win + SViolence*AV.win + SViolence*ANX.win, data=Rwomen)
PRQC.lin<- lm(PRQC1 ~ SViolence*cResponsiveness + SViolence*cNeuroticism + SViolence*cAV + SViolence*cANX, data=Rwomen)
PRQC.step1<- lm(PRQC.win ~ SViolence, data=Rwomen)
PRQC.step2<- lm(PRQC.win ~ SViolence + cNeuroticism + cAV + ANX.win, data=Rwomen)
PRQC.step3<- lm(PRQC.win ~ SViolence*cNeuroticism + SViolence*cAV + SViolence*ANX.win, data=Rwomen)
#Results
summary(PRQC.step1)##
## Call:
## lm(formula = PRQC.win ~ SViolence, data = Rwomen)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.00434 -0.52267 0.08886 0.60677 0.74899
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 6.13990 0.03618 169.708 <2e-16 ***
## SViolence1 0.03111 0.06451 0.482 0.63
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.6182 on 424 degrees of freedom
## (11 observations deleted due to missingness)
## Multiple R-squared: 0.0005483, Adjusted R-squared: -0.001809
## F-statistic: 0.2326 on 1 and 424 DF, p-value: 0.6298lm.beta(PRQC.step1) #standardized coefficients##
## Call:
## lm(formula = PRQC.win ~ SViolence, data = Rwomen)
##
## Standardized Coefficients::
## (Intercept) SViolence1
## NA 0.02341585summary(PRQC.step2)##
## Call:
## lm(formula = PRQC.win ~ SViolence + cNeuroticism + cAV + ANX.win,
## data = Rwomen)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.3521 -0.4624 0.0180 0.4222 1.1936
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 6.088467 0.033577 181.330 < 2e-16 ***
## SViolence1 0.150148 0.060165 2.496 0.013 *
## cNeuroticism -0.008298 0.024474 -0.339 0.735
## cAV -0.138557 0.026773 -5.175 3.55e-07 ***
## ANX.win -0.193365 0.038593 -5.010 8.06e-07 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.563 on 415 degrees of freedom
## (17 observations deleted due to missingness)
## Multiple R-squared: 0.181, Adjusted R-squared: 0.1731
## F-statistic: 22.93 on 4 and 415 DF, p-value: < 2.2e-16lm.beta(PRQC.step2)##
## Call:
## lm(formula = PRQC.win ~ SViolence + cNeuroticism + cAV + ANX.win,
## data = Rwomen)
##
## Standardized Coefficients::
## (Intercept) SViolence1 cNeuroticism cAV ANX.win
## NA 0.1129453 -0.0169915 -0.2525076 -0.2594603summary(PRQC.step3)##
## Call:
## lm(formula = PRQC.win ~ SViolence * cNeuroticism + SViolence *
## cAV + SViolence * ANX.win, data = Rwomen)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.37186 -0.46477 0.01461 0.42302 1.21437
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 6.08718 0.03381 180.065 < 2e-16 ***
## SViolence1 0.15469 0.06070 2.548 0.011189 *
## cNeuroticism -0.01971 0.03095 -0.637 0.524573
## cAV -0.12065 0.03288 -3.669 0.000275 ***
## ANX.win -0.21414 0.04603 -4.652 4.44e-06 ***
## SViolence1:cNeuroticism 0.02661 0.05180 0.514 0.607661
## SViolence1:cAV -0.05378 0.05692 -0.945 0.345322
## SViolence1:ANX.win 0.06116 0.08532 0.717 0.473898
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.5639 on 412 degrees of freedom
## (17 observations deleted due to missingness)
## Multiple R-squared: 0.1842, Adjusted R-squared: 0.1704
## F-statistic: 13.29 on 7 and 412 DF, p-value: 1.742e-15lm.beta(PRQC.step3)##
## Call:
## lm(formula = PRQC.win ~ SViolence * cNeuroticism + SViolence *
## cAV + SViolence * ANX.win, data = Rwomen)
##
## Standardized Coefficients::
## (Intercept) SViolence1 cNeuroticism
## NA 0.11635903 -0.04035652
## cAV ANX.win SViolence1:cNeuroticism
## -0.21986747 -0.28733341 0.03517679
## SViolence1:cAV SViolence1:ANX.win
## -0.05744708 0.04684573confint(PRQC.step3, level=0.95)## 2.5 % 97.5 %
## (Intercept) 6.02072543 6.15363081
## SViolence1 0.03535982 0.27401162
## cNeuroticism -0.08054311 0.04112529
## cAV -0.18528081 -0.05601289
## ANX.win -0.30462350 -0.12365151
## SViolence1:cNeuroticism -0.07520370 0.12842984
## SViolence1:cAV -0.16568013 0.05811723
## SViolence1:ANX.win -0.10656180 0.22888162wilcox.test(PRQC1 ~ SViolence, data=Rwomen, alternative=c("two.sided"), conf.int=T, conf.level=.95)##
## Wilcoxon rank sum test with continuity correction
##
## data: PRQC1 by SViolence
## W = 18556, p-value = 0.393
## alternative hypothesis: true location shift is not equal to 0
## 95 percent confidence interval:
## -0.16669375 0.05556911
## sample estimates:
## difference in location
## -0.05554931wilcoxonRG(x = Rwomen$PRQC1, g = Rwomen$SViolence, ci=T)## rg lower.ci upper.ci
## 1 NA -0.121 0.117wilcox_effsize(PRQC1 ~ SViolence, data=Rwomen)## # A tibble: 1 × 7
## .y. group1 group2 effsize n1 n2 magnitude
## * <chr> <chr> <chr> <dbl> <int> <int> <ord>
## 1 PRQC1 -1 1 0.0414 292 134 smallwilcox_test(PRQC1 ~ SViolence, data=Rwomen, conf.level=.95)##
## Asymptotic Wilcoxon-Mann-Whitney Test
##
## data: PRQC1 by SViolence (-1, 1)
## Z = -0.85454, p-value = 0.3928
## alternative hypothesis: true mu is not equal to 0Regression II: Partner Trust
# Descriptives by group
group_by(Rwomen, SViolence) %>%
summarise(
count = n(),
mean = mean(PTrust, na.rm = TRUE),
sd = sd(PTrust, na.rm = TRUE),
median = median(PTrust, na.rm = TRUE),
IQR = IQR(PTrust, na.rm = TRUE)
)## # A tibble: 2 × 6
## SViolence count mean sd median IQR
## <fct> <int> <dbl> <dbl> <dbl> <dbl>
## 1 -1 302 5.57 0.854 5.71 1.06
## 2 1 135 5.44 1.05 5.56 1.51# GLM with gamma distribution. Link of log vs inverse changes interpretation
Rwomen$InversePtrust<- 8-Rwomen$PTrust ### Reverse scoring Ptrust so it fits gamma distribution
Rwomen$InverseTrust.win<- winsor(Rwomen$InversePtrust, trim=.15, na.rm=T)
Trust.gamma.win <- glm(formula = InverseTrust.win ~ SViolence*Responsiveness.win + SViolence*Neuroticism.win + SViolence*AV.win + SViolence*ANX.win,
family = Gamma(link = "log"),
data = Rwomen)
Trust.gamma<- glm(formula = InverseTrust.win ~ SViolence*cResponsiveness + SViolence*cNeuroticism + SViolence*cAV + SViolence*cANX,
family = Gamma(link = "log"),
data = Rwomen)
# Linear models
Trust.win.lin<- lm(PTrust.win ~ SViolence*Responsiveness.win + SViolence*Neuroticism.win + SViolence*AV.win + SViolence*ANX.win, data=Rwomen)
Trust.lin<- lm(PTrust ~ SViolence*cResponsiveness + SViolence*cNeuroticism + SViolence*cAV + SViolence*cANX, data=Rwomen)
Trust.step1<- lm(PTrust.win ~ SViolence, data=Rwomen)
Trust.step2<- lm(PTrust.win ~ cNeuroticism + cAV + ANX.win, data=Rwomen)
Trust.step3<- lm(PTrust.win ~ SViolence*cNeuroticism + SViolence*cAV + SViolence*ANX.win, data=Rwomen)
# Results
summary(Trust.step1)##
## Call:
## lm(formula = PTrust.win ~ SViolence, data = Rwomen)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.07358 -0.60120 0.06961 0.63230 0.93358
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 5.60299 0.04021 139.34 <2e-16 ***
## SViolence1 -0.06598 0.07170 -0.92 0.358
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.6871 on 424 degrees of freedom
## (11 observations deleted due to missingness)
## Multiple R-squared: 0.001993, Adjusted R-squared: -0.0003604
## F-statistic: 0.8469 on 1 and 424 DF, p-value: 0.358lm.beta(Trust.step1)##
## Call:
## lm(formula = PTrust.win ~ SViolence, data = Rwomen)
##
## Standardized Coefficients::
## (Intercept) SViolence1
## NA -0.04464768summary(Trust.step2)##
## Call:
## lm(formula = PTrust.win ~ cNeuroticism + cAV + ANX.win, data = Rwomen)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.41034 -0.48726 -0.00925 0.40480 1.50488
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 5.56426 0.02853 195.056 < 2e-16 ***
## cNeuroticism -0.03886 0.02538 -1.531 0.12652
## cAV -0.08025 0.02742 -2.927 0.00361 **
## ANX.win -0.36013 0.03992 -9.021 < 2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.5837 on 417 degrees of freedom
## (16 observations deleted due to missingness)
## Multiple R-squared: 0.2865, Adjusted R-squared: 0.2813
## F-statistic: 55.81 on 3 and 417 DF, p-value: < 2.2e-16lm.beta(Trust.step2)##
## Call:
## lm(formula = PTrust.win ~ cNeuroticism + cAV + ANX.win, data = Rwomen)
##
## Standardized Coefficients::
## (Intercept) cNeuroticism cAV ANX.win
## NA -0.07140014 -0.13173191 -0.43441565summary(Trust.step3)##
## Call:
## lm(formula = PTrust.win ~ SViolence * cNeuroticism + SViolence *
## cAV + SViolence * ANX.win, data = Rwomen)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.38444 -0.47729 0.00009 0.40664 1.52790
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 5.541857 0.035059 158.071 < 2e-16 ***
## SViolence1 0.076331 0.063000 1.212 0.226
## cNeuroticism -0.035479 0.032153 -1.103 0.270
## cAV -0.080006 0.033976 -2.355 0.019 *
## ANX.win -0.361956 0.047753 -7.580 2.3e-13 ***
## SViolence1:cNeuroticism -0.007740 0.053788 -0.144 0.886
## SViolence1:cAV -0.013705 0.059005 -0.232 0.816
## SViolence1:ANX.win -0.004173 0.088560 -0.047 0.962
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.5855 on 413 degrees of freedom
## (16 observations deleted due to missingness)
## Multiple R-squared: 0.2891, Adjusted R-squared: 0.2771
## F-statistic: 24 on 7 and 413 DF, p-value: < 2.2e-16lm.beta(Trust.step3)##
## Call:
## lm(formula = PTrust.win ~ SViolence * cNeuroticism + SViolence *
## cAV + SViolence * ANX.win, data = Rwomen)
##
## Standardized Coefficients::
## (Intercept) SViolence1 cNeuroticism
## NA 0.051595017 -0.065194946
## cAV ANX.win SViolence1:cNeuroticism
## -0.131329341 -0.436622863 -0.009188421
## SViolence1:cAV SViolence1:ANX.win
## -0.013147506 -0.002870745confint(Trust.step3, level=0.95)## 2.5 % 97.5 %
## (Intercept) 5.47293982 5.61077395
## SViolence1 -0.04750906 0.20017116
## cNeuroticism -0.09868344 0.02772526
## cAV -0.14679221 -0.01321901
## ANX.win -0.45582561 -0.26808692
## SViolence1:cNeuroticism -0.11347195 0.09799128
## SViolence1:cAV -0.12969328 0.10228319
## SViolence1:ANX.win -0.17825840 0.16991197# Just looking at SV and partner trust
wilcox.test(PTrust ~ SViolence, data=Rwomen, alternative=c("two.sided"), conf.int=T, conf.level=.95)##
## Wilcoxon rank sum test with continuity correction
##
## data: PTrust by SViolence
## W = 20326, p-value = 0.5186
## alternative hypothesis: true location shift is not equal to 0
## 95 percent confidence interval:
## -0.1176517 0.2500349
## sample estimates:
## difference in location
## 0.05887059wilcoxonRG(x = Rwomen$PTrust, g = Rwomen$SViolence, ci=T)## rg lower.ci upper.ci
## 1 NA -0.124 0.124wilcox_effsize(PTrust ~ SViolence, data=Rwomen)## # A tibble: 1 × 7
## .y. group1 group2 effsize n1 n2 magnitude
## * <chr> <chr> <chr> <dbl> <int> <int> <ord>
## 1 PTrust -1 1 0.0313 292 134 smallwilcox_test(PTrust ~ SViolence, data=Rwomen, conf.level=.95)##
## Asymptotic Wilcoxon-Mann-Whitney Test
##
## data: PTrust by SViolence (-1, 1)
## Z = 0.64595, p-value = 0.5183
## alternative hypothesis: true mu is not equal to 0# No sig. differences at bivariate level
mean(Rwomen$SViolence)## Warning in mean.default(Rwomen$SViolence): argument is not numeric or logical:
## returning NA## [1] NARegression III: Felt Security
# Descriptives by group
group_by(Rwomen, SViolence) %>%
summarise(
count = n(),
mean = mean(Security, na.rm = TRUE),
sd = sd(Security, na.rm = TRUE),
median = median(Security, na.rm = TRUE),
IQR = IQR(Security, na.rm = TRUE)
)## # A tibble: 2 × 6
## SViolence count mean sd median IQR
## <fct> <int> <dbl> <dbl> <dbl> <dbl>
## 1 -1 302 6.16 0.816 6.39 1.02
## 2 1 135 6.04 0.987 6.37 1.22# GLM with gamma distribution. Link of log vs inverse changes interpretation
Rwomen$InverseSecurity<- 22-Rwomen$Security ### Reverse scoring felt security so it fits gamma distribution
Rwomen$InverseSecurity.win<- winsor(Rwomen$InverseSecurity, trim=.15, na.rm=T)
Security.gamma.win <- glm(formula = InverseSecurity.win ~ SViolence*Responsiveness.win + SViolence*Neuroticism.win + SViolence*AV.win + SViolence*ANX.win,
family = Gamma(link = "log"),
data = Rwomen)
Security.gamma<- glm(formula = InverseSecurity ~ SViolence*cResponsiveness + SViolence*cNeuroticism + SViolence*cAV + SViolence*cANX,
family = Gamma(link = "log"),
data = Rwomen)
# Linear models
Security.win.lin<- lm(Security.win ~ SViolence*Responsiveness.win + SViolence*Neuroticism.win + SViolence*AV.win + SViolence*ANX.win, data=Rwomen)
Security.lin<- lm(Security ~ SViolence*cResponsiveness + SViolence*cNeuroticism + SViolence*cAV + SViolence*cANX, data=Rwomen)
Security.step1<- lm(Security.win ~ SViolence, data=Rwomen)
Security.step2<- lm(Security.win ~ SViolence + cNeuroticism + cAV + ANX.win, data=Rwomen)
Security.step3<- lm(Security.win ~ SViolence*cNeuroticism + SViolence*cAV + SViolence*ANX.win, data=Rwomen)
# Results
summary(Security.step1)##
## Call:
## lm(formula = Security.win ~ SViolence, data = Rwomen)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.0296 -0.5296 0.1502 0.5750 0.7417
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 6.25187 0.03586 174.365 <2e-16 ***
## SViolence1 -0.04908 0.06415 -0.765 0.445
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.6158 on 427 degrees of freedom
## (8 observations deleted due to missingness)
## Multiple R-squared: 0.001369, Adjusted R-squared: -0.00097
## F-statistic: 0.5853 on 1 and 427 DF, p-value: 0.4447lm.beta(Security.step1)##
## Call:
## lm(formula = Security.win ~ SViolence, data = Rwomen)
##
## Standardized Coefficients::
## (Intercept) SViolence1
## NA -0.03699641summary(Security.step2)##
## Call:
## lm(formula = Security.win ~ SViolence + cNeuroticism + cAV +
## ANX.win, data = Rwomen)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.16931 -0.41852 0.05819 0.36758 1.20878
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 6.19613 0.03082 201.015 <2e-16 ***
## SViolence1 0.06963 0.05537 1.258 0.2093
## cNeuroticism -0.05465 0.02247 -2.432 0.0154 *
## cAV -0.04132 0.02458 -1.681 0.0935 .
## ANX.win -0.33878 0.03541 -9.568 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.5185 on 417 degrees of freedom
## (15 observations deleted due to missingness)
## Multiple R-squared: 0.2969, Adjusted R-squared: 0.2902
## F-statistic: 44.03 on 4 and 417 DF, p-value: < 2.2e-16lm.beta(Security.step2)##
## Call:
## lm(formula = Security.win ~ SViolence + cNeuroticism + cAV +
## ANX.win, data = Rwomen)
##
## Standardized Coefficients::
## (Intercept) SViolence1 cNeuroticism cAV ANX.win
## NA 0.05262069 -0.11240385 -0.07578554 -0.45719458summary(Security.step3)##
## Call:
## lm(formula = Security.win ~ SViolence * cNeuroticism + SViolence *
## cAV + SViolence * ANX.win, data = Rwomen)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.16578 -0.43829 0.04954 0.35622 1.17367
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 6.196205 0.031045 199.587 < 2e-16 ***
## SViolence1 0.073423 0.055869 1.314 0.1895
## cNeuroticism -0.070393 0.028390 -2.479 0.0136 *
## cAV -0.027470 0.030149 -0.911 0.3628
## ANX.win -0.342709 0.042186 -8.124 5.26e-15 ***
## SViolence1:cNeuroticism 0.043950 0.047646 0.922 0.3568
## SViolence1:cAV -0.043843 0.052361 -0.837 0.4029
## SViolence1:ANX.win 0.002907 0.078486 0.037 0.9705
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.5195 on 414 degrees of freedom
## (15 observations deleted due to missingness)
## Multiple R-squared: 0.2993, Adjusted R-squared: 0.2875
## F-statistic: 25.26 on 7 and 414 DF, p-value: < 2.2e-16lm.beta(Security.step3)##
## Call:
## lm(formula = Security.win ~ SViolence * cNeuroticism + SViolence *
## cAV + SViolence * ANX.win, data = Rwomen)
##
## Standardized Coefficients::
## (Intercept) SViolence1 cNeuroticism
## NA 0.055488001 -0.144777621
## cAV ANX.win SViolence1:cNeuroticism
## -0.050387575 -0.462490982 0.058299454
## SViolence1:cAV SViolence1:ANX.win
## -0.046999518 0.002234492confint(Security.step3, level=0.95)## 2.5 % 97.5 %
## (Intercept) 6.13517900 6.25723022
## SViolence1 -0.03639967 0.18324485
## cNeuroticism -0.12619981 -0.01458628
## cAV -0.08673449 0.03179476
## ANX.win -0.42563456 -0.25978408
## SViolence1:cNeuroticism -0.04970867 0.13760804
## SViolence1:cAV -0.14676911 0.05908403
## SViolence1:ANX.win -0.15137418 0.15718794#Bivariate test
wilcox.test(Security ~ SViolence, data=Rwomen, alternative=c("two.sided"), conf.int=T, conf.level=.95)##
## Wilcoxon rank sum test with continuity correction
##
## data: Security by SViolence
## W = 20616, p-value = 0.4744
## alternative hypothesis: true location shift is not equal to 0
## 95 percent confidence interval:
## -0.05557498 0.16670765
## sample estimates:
## difference in location
## 0.05555086wilcoxonRG(x = Rwomen$Security, g = Rwomen$SViolence)## rg
## NAwilcox_effsize(Security ~ SViolence, data=Rwomen)## # A tibble: 1 × 7
## .y. group1 group2 effsize n1 n2 magnitude
## * <chr> <chr> <chr> <dbl> <int> <int> <ord>
## 1 Security -1 1 0.0346 295 134 smallwilcox_test(Security ~ SViolence, data=Rwomen, conf.level=.95)##
## Asymptotic Wilcoxon-Mann-Whitney Test
##
## data: Security by SViolence (-1, 1)
## Z = 0.7158, p-value = 0.4741
## alternative hypothesis: true mu is not equal to 0Regression IV: IOS
# Descriptives by group
group_by(Rwomen, SViolence) %>%
summarise(
count = n(),
mean = mean(IOS.current, na.rm = TRUE),
sd = sd(IOS.current, na.rm = TRUE),
median = median(IOS.current, na.rm = TRUE),
IQR = IQR(IOS.current, na.rm = TRUE)
)## # A tibble: 2 × 6
## SViolence count mean sd median IQR
## <fct> <int> <dbl> <dbl> <dbl> <dbl>
## 1 -1 302 5.02 1.66 5 2
## 2 1 135 4.97 1.71 5 3IOS.lin.win<- lm(IOS.current ~ SViolence*Responsiveness.win + SViolence*Neuroticism.win + SViolence*AV.win + SViolence*ANX.win, data=Rwomen)
IOS.lin<- lm(IOS.current ~ SViolence*cResponsiveness + SViolence*cNeuroticism + SViolence*cAV + SViolence*cANX, data=Rwomen)
IOS.step1<- lm(IOS.current ~ SViolence, data=Rwomen)
IOS.step2<- lm(IOS.current ~ SViolence + cNeuroticism + cAV + ANX.win, data=Rwomen)
IOS.step3<- lm(IOS.current ~ SViolence*cNeuroticism + SViolence*cAV + SViolence*ANX.win, data=Rwomen)
summary(IOS.step1)##
## Call:
## lm(formula = IOS.current ~ SViolence, data = Rwomen)
##
## Residuals:
## Min 1Q Median 3Q Max
## -4.0201 -1.0201 -0.0201 1.9799 2.0296
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 5.02007 0.09702 51.744 <2e-16 ***
## SViolence1 -0.04970 0.17395 -0.286 0.775
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.678 on 432 degrees of freedom
## (3 observations deleted due to missingness)
## Multiple R-squared: 0.0001889, Adjusted R-squared: -0.002125
## F-statistic: 0.08162 on 1 and 432 DF, p-value: 0.7752lm.beta(IOS.step1)##
## Call:
## lm(formula = IOS.current ~ SViolence, data = Rwomen)
##
## Standardized Coefficients::
## (Intercept) SViolence1
## NA -0.01374417summary(IOS.step2)##
## Call:
## lm(formula = IOS.current ~ SViolence + cNeuroticism + cAV + ANX.win,
## data = Rwomen)
##
## Residuals:
## Min 1Q Median 3Q Max
## -4.2104 -1.0773 0.1638 1.3663 2.8586
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 4.92638 0.09631 51.151 < 2e-16 ***
## SViolence1 0.12028 0.17319 0.694 0.487772
## cNeuroticism 0.05572 0.06977 0.799 0.425002
## cAV -0.17572 0.07650 -2.297 0.022104 *
## ANX.win -0.40361 0.11079 -3.643 0.000303 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.622 on 418 degrees of freedom
## (14 observations deleted due to missingness)
## Multiple R-squared: 0.06195, Adjusted R-squared: 0.05297
## F-statistic: 6.901 on 4 and 418 DF, p-value: 2.186e-05lm.beta(IOS.step2)##
## Call:
## lm(formula = IOS.current ~ SViolence + cNeuroticism + cAV + ANX.win,
## data = Rwomen)
##
## Standardized Coefficients::
## (Intercept) SViolence1 cNeuroticism cAV ANX.win
## NA 0.03353455 0.04247143 -0.11924990 -0.20084908summary(IOS.step3)##
## Call:
## lm(formula = IOS.current ~ SViolence * cNeuroticism + SViolence *
## cAV + SViolence * ANX.win, data = Rwomen)
##
## Residuals:
## Min 1Q Median 3Q Max
## -4.1853 -1.0658 0.1369 1.4192 2.8456
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 4.92966 0.09708 50.779 < 2e-16 ***
## SViolence1 0.12592 0.17487 0.720 0.47190
## cNeuroticism 0.02025 0.08787 0.230 0.81786
## cAV -0.15925 0.09371 -1.699 0.08999 .
## ANX.win -0.37467 0.13210 -2.836 0.00479 **
## SViolence1:cNeuroticism 0.11240 0.14859 0.756 0.44980
## SViolence1:cAV -0.05904 0.16356 -0.361 0.71831
## SViolence1:ANX.win -0.11961 0.24577 -0.487 0.62675
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.627 on 415 degrees of freedom
## (14 observations deleted due to missingness)
## Multiple R-squared: 0.06354, Adjusted R-squared: 0.04775
## F-statistic: 4.023 on 7 and 415 DF, p-value: 0.0002812lm.beta(IOS.step3)##
## Call:
## lm(formula = IOS.current ~ SViolence * cNeuroticism + SViolence *
## cAV + SViolence * ANX.win, data = Rwomen)
##
## Standardized Coefficients::
## (Intercept) SViolence1 cNeuroticism
## NA 0.03510828 0.01543672
## cAV ANX.win SViolence1:cNeuroticism
## -0.10807126 -0.18644626 0.05497920
## SViolence1:cAV SViolence1:ANX.win
## -0.02333781 -0.03390305confint(IOS.step3, level=0.95)## 2.5 % 97.5 %
## (Intercept) 4.7388311 5.12049104
## SViolence1 -0.2178306 0.46966960
## cNeuroticism -0.1524816 0.19298224
## cAV -0.3434534 0.02495162
## ANX.win -0.6343408 -0.11499795
## SViolence1:cNeuroticism -0.1796794 0.40448464
## SViolence1:cAV -0.3805520 0.26247317
## SViolence1:ANX.win -0.6027273 0.36350408# Bivariate test
wilcox.test(IOS.current ~ SViolence, data=Rwomen, alternative=c("two.sided"), conf.int=T, conf.level=.95)##
## Wilcoxon rank sum test with continuity correction
##
## data: IOS.current by SViolence
## W = 20464, p-value = 0.8132
## alternative hypothesis: true location shift is not equal to 0
## 95 percent confidence interval:
## -2.957400e-05 4.856633e-05
## sample estimates:
## difference in location
## 4.575258e-05wilcoxonRG(x = Rwomen$IOS.current, g = Rwomen$SViolence)## rg
## NAwilcox_effsize(IOS.current ~ SViolence, data=Rwomen)## # A tibble: 1 × 7
## .y. group1 group2 effsize n1 n2 magnitude
## * <chr> <chr> <chr> <dbl> <int> <int> <ord>
## 1 IOS.current -1 1 0.0114 299 135 smallwilcox_test(IOS.current ~ SViolence, data=Rwomen, conf.level=.95)##
## Asymptotic Wilcoxon-Mann-Whitney Test
##
## data: IOS.current by SViolence (-1, 1)
## Z = 0.23669, p-value = 0.8129
## alternative hypothesis: true mu is not equal to 0Notes
Based on conversations with stats professor and LATIS consultants, decided to not do multivariate multiple regression as this is the same as running multiple multiple regressions and need to do more complex analyses to truly take the correlations betwee DVs into account. I conducted a multiple regression using PRQC total as DV and SV, attachment, and neuroticism (and interactions between SV and these variables) as predictors, but the residuals were not normally distributed. I attempted a Box Cox transformation to address the left skew of the data, but this did not improve the residual plots. We added responsiveness as predictor to better understandwhy SV may be associated with better relationship quality. To address residual plot issues, we reverse-coded the dependent variables so that they were right skewed and therefore similar to gamma distributions. We then ran several general linear model regressions using a gamma distribution using the subscale totals as dependent variables.
Exploratory: To understand why negative SV impacts reported in qualitative data were not matching up to quantitative measures, we conducted several exploratory analyses.
```