Overview

This analysis explores survey data collected from restaurant servers for MKT 4000. The survey examines whether drawing a smiley face on the bill affects tip outcomes, along with demographic and workplace context variables.

Setup

# Run this once in the Console if you haven't already:
# install.packages(c("tidyverse", "scales"))

library(tidyverse)
library(scales)

Load Data

Qualtrics exports three header rows: short column names, full question text, and internal import IDs. We use row 1 as column names and drop rows 2 and 3 before analyzing.

raw <- read_csv(
  "BensusenMKGT4000_April 8, 2026_17.22.csv",
  show_col_types = FALSE
)

# Drop the two extra Qualtrics header rows
df_raw <- raw %>% slice(-(1:2))

Select and Clean Variables

df <- df_raw %>%
  select(
    consent       = Q2,
    gender        = Q3,
    experience    = Q4,
    satisfaction  = Q5,
    tip_share     = Q6,
    establishment = Q7,
    draws_smiley  = Q8,
    tip_impact    = Q9_1
  ) %>%
  filter(consent == "1") %>%
  mutate(across(everything(), as.numeric))

cat("Total consented responses:", nrow(df), "\n")
## Total consented responses: 30

Recode Variables

df_labeled <- df %>%
  mutate(
    gender = recode(gender,
      `1` = "Male",
      `2` = "Female",
      `4` = "Non-binary / Other"
    ),
    experience = recode(experience,
      `1` = "Less than 1 year",
      `2` = "1-3 years",
      `3` = "3-5 years",
      `4` = "5+ years"
    ),
    satisfaction = recode(satisfaction,
      `1` = "Very dissatisfied",
      `2` = "Dissatisfied",
      `3` = "Satisfied",
      `4` = "Very satisfied"
    ),
    tip_share = recode(tip_share,
      `1` = "Yes",
      `2` = "No"
    ),
    establishment = recode(establishment,
      `1` = "Chain restaurant",
      `2` = "Locally owned"
    ),
    draws_smiley = recode(draws_smiley,
      `1` = "Yes",
      `2` = "No"
    ),
    tip_impact = recode(tip_impact,
      `1` = "Tips increased",
      `2` = "No change",
      `3` = "Tips decreased"
    )
  ) %>%
  mutate(
    experience   = factor(experience,   levels = c("Less than 1 year","1-3 years","3-5 years","5+ years")),
    satisfaction = factor(satisfaction, levels = c("Very dissatisfied","Dissatisfied","Satisfied","Very satisfied")),
    tip_impact   = factor(tip_impact,   levels = c("Tips increased","No change","Tips decreased"))
  )

Frequency Tables

cat("--- Gender ---\n");        print(table(df_labeled$gender))
## --- Gender ---
## 
##             Female               Male Non-binary / Other 
##                  4                 21                  1
cat("\n--- Experience ---\n");  print(table(df_labeled$experience))
## 
## --- Experience ---
## 
## Less than 1 year        1-3 years        3-5 years         5+ years 
##                6                8                4                7
cat("\n--- Satisfaction ---\n");print(table(df_labeled$satisfaction))
## 
## --- Satisfaction ---
## 
## Very dissatisfied      Dissatisfied         Satisfied    Very satisfied 
##                15                 8                 2                 1
cat("\n--- Tip share? ---\n");  print(table(df_labeled$tip_share))
## 
## --- Tip share? ---
## 
##  No Yes 
##  21   5
cat("\n--- Establishment ---\n");print(table(df_labeled$establishment))
## 
## --- Establishment ---
## 
## Chain restaurant    Locally owned 
##                8               18
cat("\n--- Draws smiley? ---\n");print(table(df_labeled$draws_smiley))
## 
## --- Draws smiley? ---
## 
##  No Yes 
##  12  14
cat("\n--- Tip impact ---\n");  print(table(df_labeled$tip_impact))
## 
## --- Tip impact ---
## 
## Tips increased      No change Tips decreased 
##             12             10              3

Charts

Shared Theme

survey_theme <- theme_minimal(base_size = 13) +
  theme(
    plot.title    = element_text(face = "bold", size = 14),
    plot.subtitle = element_text(color = "grey40", size = 11),
    axis.title    = element_text(size = 11),
    legend.position = "none"
  )

Figure 1: Gender Identity

df_labeled %>%
  count(gender) %>%
  mutate(pct = n / sum(n)) %>%
  ggplot(aes(x = reorder(gender, n), y = pct, fill = gender)) +
  geom_col(width = 0.6) +
  geom_text(aes(label = paste0(n, " (", percent(pct, accuracy = 1), ")")),
            hjust = -0.1, size = 3.8) +
  scale_y_continuous(labels = percent_format(), limits = c(0, 0.85)) +
  scale_fill_manual(values = c("Male"               = "#4A90D9",
                               "Female"             = "#E88B6A",
                               "Non-binary / Other" = "#8BB56A")) +
  coord_flip() +
  labs(title = "Gender Identity of Respondents", x = NULL, y = "Percentage") +
  survey_theme

Figure 2: Server Experience

df_labeled %>%
  count(experience) %>%
  mutate(pct = n / sum(n)) %>%
  ggplot(aes(x = experience, y = pct, fill = experience)) +
  geom_col(width = 0.6) +
  geom_text(aes(label = paste0(n, " (", percent(pct, accuracy = 1), ")")),
            vjust = -0.4, size = 3.8) +
  scale_y_continuous(labels = percent_format(), limits = c(0, 0.60)) +
  scale_fill_manual(values = c("#D4E6F5","#7BB8E0","#3A88C5","#1A5C8A")) +
  labs(title = "Server Experience in the Industry", x = NULL, y = "Percentage") +
  survey_theme

Figure 3: Job Satisfaction

df_labeled %>%
  count(satisfaction) %>%
  mutate(pct = n / sum(n)) %>%
  ggplot(aes(x = satisfaction, y = pct, fill = satisfaction)) +
  geom_col(width = 0.6) +
  geom_text(aes(label = paste0(n, " (", percent(pct, accuracy = 1), ")")),
            vjust = -0.4, size = 3.8) +
  scale_y_continuous(labels = percent_format(), limits = c(0, 0.65)) +
  scale_fill_manual(values = c("#D9534F","#F0A05A","#5BA85A","#2E7D32")) +
  labs(title = "Job Satisfaction Among Servers", x = NULL, y = "Percentage") +
  survey_theme

Figure 4: Tip Impact by Smiley Face Drawing Behavior

This is the core research question — comparing servers who draw a smiley face on the bill versus those who do not, and the tip outcomes they report.

df_labeled %>%
  filter(!is.na(draws_smiley), !is.na(tip_impact)) %>%
  count(draws_smiley, tip_impact) %>%
  group_by(draws_smiley) %>%
  mutate(pct = n / sum(n)) %>%
  ggplot(aes(x = tip_impact, y = pct, fill = draws_smiley)) +
  geom_col(position = "dodge", width = 0.65) +
  geom_text(aes(label = percent(pct, accuracy = 1)),
            position = position_dodge(width = 0.65),
            vjust = -0.4, size = 3.5) +
  scale_y_continuous(labels = percent_format(), limits = c(0, 1.05)) +
  scale_fill_manual(values = c("Yes" = "#E88B6A", "No" = "#4A90D9"),
                    name = "Draws smiley face?") +
  labs(title = "Tip Impact by Smiley Face Drawing Behavior",
       subtitle = "Among servers who draw vs. don't draw a smiley on the bill",
       x = "Reported tip impact", y = "Percentage within group") +
  survey_theme +
  theme(legend.position = "right")

Figure 5: Smiley Face Drawing by Establishment Type

df_labeled %>%
  filter(!is.na(establishment), !is.na(draws_smiley)) %>%
  count(establishment, draws_smiley) %>%
  group_by(establishment) %>%
  mutate(pct = n / sum(n)) %>%
  ggplot(aes(x = establishment, y = pct, fill = draws_smiley)) +
  geom_col(width = 0.55) +
  geom_text(aes(label = percent(pct, accuracy = 1)),
            position = position_stack(vjust = 0.5), size = 3.8, color = "white") +
  scale_y_continuous(labels = percent_format()) +
  scale_fill_manual(values = c("Yes" = "#E88B6A", "No" = "#7BB8E0"),
                    name = "Draws smiley?") +
  labs(title = "Smiley Face Drawing by Establishment Type",
       x = NULL, y = "Percentage") +
  survey_theme +
  theme(legend.position = "right")

Figure 6: Tip Sharing by Establishment Type

df_labeled %>%
  filter(!is.na(establishment), !is.na(tip_share)) %>%
  count(establishment, tip_share) %>%
  group_by(establishment) %>%
  mutate(pct = n / sum(n)) %>%
  ggplot(aes(x = establishment, y = pct, fill = tip_share)) +
  geom_col(width = 0.55) +
  geom_text(aes(label = percent(pct, accuracy = 1)),
            position = position_stack(vjust = 0.5), size = 3.8, color = "white") +
  scale_y_continuous(labels = percent_format()) +
  scale_fill_manual(values = c("Yes" = "#5BA85A", "No" = "#D9534F"),
                    name = "Tip share?") +
  labs(title = "Tip Sharing by Establishment Type",
       x = NULL, y = "Percentage") +
  survey_theme +
  theme(legend.position = "right")

Cross-Tabulation: Job Satisfaction vs. Tip Impact

df_labeled %>%
  filter(!is.na(satisfaction), !is.na(tip_impact)) %>%
  count(satisfaction, tip_impact) %>%
  pivot_wider(names_from = tip_impact, values_from = n, values_fill = 0)
## # A tibble: 4 × 4
##   satisfaction      `Tips increased` `No change` `Tips decreased`
##   <fct>                        <int>       <int>            <int>
## 1 Very dissatisfied                5           7                2
## 2 Dissatisfied                     5           3                0
## 3 Satisfied                        2           0                0
## 4 Very satisfied                   0           0                1

Disclaimer

The sample size used for this survey is small (n = 30 consented responses). Please do not generalize these results to the broader population. For generalization purposes, a minimum of 30 responses per subgroup is typically recommended for most statistical analyses.

References

Wickham, H. (2016). ggplot2: Elegant Graphics for Data Analysis. Springer-Verlag. https://ggplot2.tidyverse.org

Wickham, H., & Grolemund, G. (2017). R for Data Science. O’Reilly. https://r4ds.had.co.nz/

R Markdown. https://rmarkdown.rstudio.com/