#Loading all packages with Pacman
pacman::p_load(dplyr, tidyr, knitr, ggplot2, kableExtra, BSDA, car, psych, binom, beepr, readr, pwr)

Hypothesis 1

Individuals with a history of sexual assault will report poorer relationship quality than individuals without a history of assault.

pwr.anova.test(k = 2, n = NULL, f = .10, sig.level = .05, power = .80)
## 
##      Balanced one-way analysis of variance power calculation 
## 
##               k = 2
##               n = 393.4057
##               f = 0.1
##       sig.level = 0.05
##           power = 0.8
## 
## NOTE: n is number in each group

or, if we want to be able to detect an effect size of .15, the sample size decreases.

pwr.anova.test(k = 2, n = NULL, f = .15, sig.level = .05, power = .80)
## 
##      Balanced one-way analysis of variance power calculation 
## 
##               k = 2
##               n = 175.3847
##               f = 0.15
##       sig.level = 0.05
##           power = 0.8
## 
## NOTE: n is number in each group

Hypthesis 2

Revictimized individuals (i.e., victimized in childhood and adulthood) will report worse relationship quality than CSA survivors, ASA survivors, and individuals without a history of sexual assault.

pwr.anova.test(k = 3, n = NULL, f = .10, sig.level = .05, power = .80)
## 
##      Balanced one-way analysis of variance power calculation 
## 
##               k = 3
##               n = 322.157
##               f = 0.1
##       sig.level = 0.05
##           power = 0.8
## 
## NOTE: n is number in each group

According to Smith et al. (2018), 21% of women are victims of attempted rape or rape (ASA).

According to Walker et al. (2019), the mean prevalence rate of revictimization is 48%.

According to CDC-Kaiser Ace Study, 25% of women are survivors of CSA.

We want to recruit 300 people in each group. Given prevalence rates, if we pre-screen 3,000 people, we can anticipate that 300 will be ASA only survivors (3,000 x .20 x .5), ~300 will be in revictimized group (3,000 x .20 x .5), and that 375 will be CSA only survivors ((3,000 x .25) x .5).

``` If we want to detect an effect of .15, then sample size decreases.

pwr.anova.test(k = 3, n = 300, f = NULL, sig.level = .05, power = .80)
## 
##      Balanced one-way analysis of variance power calculation 
## 
##               k = 3
##               n = 300
##               f = 0.1036095
##       sig.level = 0.05
##           power = 0.8
## 
## NOTE: n is number in each group