Module 4: The Final Manuscript

One Click to Truth

Author

A. C. Del Re

The “Raw Data” Reality

Most Psychology data comes from Qualtrics or SurveyMonkey. It is usually “wide”, messy, and full of text labels where you want numbers.

1. Simulating “Messy” Data

Let’s imagine we just downloaded this CSV from Qualtrics. It has:

Metadata rows: Qualtrics gives 2-3 header rows.
String Likert scales: “Strongly Agree” instead of 5.
Reverse-scored items: An anxiety scale where “I feel calm” needs to be flipped.

Table 1

ID	Meta	Item 1	Item 2 (Rev)
R_x12z	IP Address	Strongly Agree	Strongly Disagree
R_a99b	Completed	Agree	Disagree
R_c77d	Completed	Disagree	Agree
R_k22l	Spam	Strongly Agree	Strongly Disagree
R_p55m	Completed	Neutral	Neutral

Table 2: Raw Data Preview (Failsafe)

Raw Data Preview (Failsafe)
ResponseID	Status	Q1_Anxious	Q2_Calm
R_x12z	IP Address	Strongly Agree	Strongly Disagree
R_a99b	Completed	Agree	Disagree
R_c77d	Completed	Disagree	Agree
R_k22l	Spam	Strongly Agree	Strongly Disagree
R_p55m	Completed	Neutral	Neutral

The Goal

We want a single file (manuscript.qmd) that contains everything: data loading, cleaning, analysis, and writing. When you render it, you get a submission-ready Word document.

Here is exactly how to do it.

1. The Raw QMD

Copy this code into a new Quarto file. This is your template.

---
title: "Untitled"
format: docx
---

```{r, echo=FALSE, warning=FALSE}
library(dplyr)
library(ggplot2)
library(flextable)

# 1. IMPORT DATA
# d <- read.csv("my_data.csv")
d <- data.frame(
  ID = 1:50,
  Sleep_Hours = rnorm(50, 7, 1.5),
  Anxiety = rnorm(50, 10, 3)
)

# 2. ADD ALL THE PROCESSING CODE HERE
clean_dat <- d %>%
  filter(Sleep_Hours > 0) %>%
  mutate(Group = ifelse(Sleep_Hours < 6, "Low", "High"))

# 3. CALCULATE STATS
mean_val <- round(mean(clean_dat$Anxiety), 1)
n_total <- nrow(clean_dat)
```

# Method

We analyzed data from `r n_total` students.

# Results

The mean anxiety score was `r mean_val`.

add other chunks for plots etc

2. The Logic Explained

A. The Setup Chunk

This is the most important part. It needs echo: false so it doesn’t show up in your paper.

Load Libraries: dplyr, ggplot2, flextable.
Data Wrangling: Do ALL your filtering and recoding here (clean_dat).
Pre-Calculation: Save specific numbers (like n_total or p_val) as variables here. This makes your text cleaner.

B. Inline Code

Instead of typing “The mean age was 20.2”, you write:

Code: “The mean age was r round(mean(age), 1).”

Result: “The mean age was 20.”

C. Tables & Plots

We put the code for tables in their own chunks. We use flextable because it formats perfectly for Word.

3. The Output (What it looks like)

When you click Render, this is what appears in your Word doc:

Introduction

Sleep deprivation is a common issue on college campuses.

Table 3

Raw Data Preview
ID	Sleep (Hrs)	Anxiety
1	3.46	14.89
2	7.56	12.03
3	9.36	8.47
4	7.25	9.70
5	7.83	10.22
6	9.42	5.35

Method

We analyzed data from 50 students.

Results

The mean anxiety score was 9.8. We found a significant negative correlation between sleep and anxiety, \(r\) = -0.03, \(p\) = 0.833.

Table 1: Descriptive Statistics

Table 4

Sleep Status	N	Anxiety (M)
Deprived	13	10.00
Sufficient	37	9.69

Figure 1

Course Complete

You have learned the Modern Psychologist’s workflow. 1. Clean with tidyverse. 2. Format with flextable. 3. Visualize with ggplot2. 4. Publish with Quarto.

< Back to Syllabus