---
title: "DASS Data"
author: "Dustin Haraden"
date: "2025-12-01"
output: html_document
editor_options: 
  markdown: 
    wrap: 72
  chunk_output_type: console
---

```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = TRUE)
library(tidyverse) #pipe, mutate, filter, select (dplyr) ggplot 
library(here) #Creates relative file paths
library(psych) #describe()
library(rio) #generic import function
library(datawizard) #renaming and computing sums/means
library(sjPlot) #Nice tables

set.seed(42) #keep randomness consistent 

dass <- import(here("Data", "data.csv")) %>% 
  slice_sample(n = 400) %>% 
  janitor::clean_names()


```

DASS Scoring: <https://www2.psy.unsw.edu.au/dass/down.htm>

## Validity Checking

Using the codebook we are able to identify participants who may not be
responding appropriately

The following items were presented as a check-list and subjects were
instructed "In the grid below, check all the words whose definitions you
are sure you know":

VCL1 boat VCL2 incoherent VCL3 pallid VCL4 robot VCL5 audible VCL6
cuivocal VCL7 paucity VCL8 epistemology VCL9 florted VCL10 decide VCL11
pastiche VCL12 verdid VCL13 abysmal VCL14 lucid VCL15 betray VCL16 funny

A value of 1 is checked, 0 means unchecked. The words at VCL6, VCL9, and
VCL12 are not real words and can be used as a validity check.

```{r}
# filter based on if it is checked on the three items
check1 <- dass %>% 
  filter(vcl6 == 0, 
         vcl9 == 0, 
         vcl12 == 0)

check2 <- dass %>% 
  filter(vcl6 == 1 | 
         vcl9 == 1 | 
         vcl12 == 1)

check3 <- dass %>% 
  mutate(
    check = vcl6 + vcl9 + vcl12
  ) %>% 
  filter(check == 0)

check4 <- dass %>% 
  select(vcl6, vcl9, vcl12) %>% 
  filter(vcl6 == 0, 
         vcl9 == 0, 
         vcl12 == 0)

table(check3$check)
```

We will use `check1` as our dataset moving forward.

DASS Scoring:

Stress: 1, 6, 8, 11, 12, 14, 18, 22, 27, 29, 32, 33, 35, 39

Depression: 3, 5, 10, 13, 16, 17, 21, 24, 26, 31, 34, 37, 38, 42

```{r}
final <- check3 %>% 
  mutate(
    dass_stress = row_sums(.,
      select = c(q1a, q6a, q8a, q11a, q12a, q14a,
                 q18a, q22a, q27a, q29a, q32a, q33a,
                 q35a, q39a), 
      min_valid = 0.8
    )
  )
```

## Renaming variables

Or use `data_rename()` from datawizard

```{r}
demo_names <- check3 %>% 
  select(1:5)

#new names
list <- c("item1", "item2", "item3", "item4", "item5")

names(demo_names) <- list
names(demo_names)
```

## Correlation Table

```{r}
descrip <- final %>% 
  select(q1a:q42a, age, familysize) %>% 
  select(contains("a"))

descrip %>% 
  select(age, familysize, q1a, q6a, q8a, q11a, q12a, q14a,
                 q18a, q22a, q27a, q29a, q32a, q33a,
                 q35a, q39a) %>% 
  tab_corr(triangle = "lower", 
           title = "Table 1 - Correlations", 
           var.labels = c("Age", "Family Size", "DSS1", "DSS6", "DSS8",
                          'DSS11', 'DSS12', 'DSS14',
                 'DSS18', 'DSS22', 'DSS27', 'DSS29', 'DSS32', 'DSS33',
                 'DSS35', 'DSS39'))
```

## Regression

Does age and family size impact self-reported stress?

```{r}

stress.reg <- lm(dass_stress ~ age + familysize, data = final)
summary(stress.reg)

```

To examine the impact of age and family size on self-reported stress, we
ran a multiple linear regression. Age significantly predicted
self-reported stress (*b* = -0.17, *p =* .010) while holding family size
constant, suggesting that the older participants have less self-reported
stress. Family size was not statistically significant (*p* = .88).

## ANOVA

Does handedness impact self-reported stress? Are there group differences
in self-reported stress by handedness?

```{r}

stress.aov <- aov(dass_stress ~ hand, data = final)
summary(stress.aov)



```