Assignment 1 — Data visualisation and descriptive analytics report¶

Act as an analytics consultant for a restaurant: produce two summary tables and three figures from 244 restaurant bills in Python, then write a four-section business report that turns them into staffing and revenue recommendations.

Adapted from the DATA 300 syllabus, Assignment 1 (Data visualization and descriptive analytics report). Brief rewritten for this classroom; the analysis runs offline in local JupyterLab.

Graded work: Assignment 1 (6%, CLO 1, 2 and 3)

• Set: Wednesday 11 November (Day 2), after the Tableau lecture.
• Due: Sunday 15 November 2026, 23:59 China Standard Time.
• Submit: one PDF, YourLastName_DATA300_Assignment1.pdf, to the DATA 300: Assignment 1 drop-box on GAP (Moodle).
• Work: individual. Online support is available over the weekend.

This notebook is both the brief (read it all first) and your starter notebook. Save your own copy before you change anything: File → Save Notebook As… → YourLastName_DATA300_Assignment1.ipynb, in the same folder as this notebook so that data/tips.csv is found.

The scenario¶

You have joined Bistro Analytics, a consulting firm that helps restaurants run better. A busy client restaurant has given you its transaction records: for each bill, the total bill, the tip, the party size, the shift (lunch or dinner), the day of the week, and whether the bill payer was a smoker and their sex. There are 244 bills; amounts are in US dollars.

The owner wants evidence for two decisions:

Your job is descriptive analytics: summarise what happened, show it clearly, and turn it into practical recommendations. You do not need to prove causes, and you should not claim them.

What you must do¶

You are not expected to write a lot of new code. Each cell below already runs and gives you a starting point; the text above it says exactly what the finished table or figure must show. The changes are small: the pandas you used on Day 1 (groupby, mean, sum) and the chart rules from Day 2.

Part 1: The analysis¶

Setup¶

Run this cell first. It loads the libraries and the data from the bundled file, so no internet connection is needed.

import matplotlib.pyplot as plt
import pandas as pd
import seaborn as sns

df = pd.read_csv("data/tips.csv")

# The restaurant opens Thursday to Sunday. Keep the days in week order in every table and chart.
DAYS = ["Thur", "Fri", "Sat", "Sun"]
df["day"] = pd.Categorical(df["day"], categories=DAYS, ordered=True)

print("Rows (bills), columns:", df.shape)
print("Missing values in any column:", int(df.isna().sum().sum()))
df.head()

Rows (bills), columns: (244, 7)
Missing values in any column: 0

Table 1: Average transaction metrics by day¶

The finished table must show, for each day (Thursday to Sunday, in that order): the average total bill, the average tip and the average party size, rounded to 2 decimals, with the column names Avg Total Bill ($), Avg Tip ($) and Avg Party Size.

Starting point: the cell computes only the average total bill. Add the other two columns and rename all three.

print("--- Table 1: Average Transaction Metrics by Day ---")
day_summary = df.groupby("day", observed=True)[["total_bill"]].mean().round(2)
day_summary.columns = ["Avg Total Bill ($)"]
day_summary

--- Table 1: Average Transaction Metrics by Day ---

Table 2: Total revenue by meal shift¶

The finished table must show the total (sum) of total_bill for Lunch and for Dinner, rounded to 2 decimals, in a column named Total Revenue ($).

Starting point: the cell counts the number of bills in each shift. Change it so it adds up the bill amounts instead.

print("--- Table 2: Aggregate Revenue Breakdown by Meal Shift ---")
time_summary = df.groupby("time", observed=True)[["total_bill"]].count()   # counts bills: change this
time_summary.columns = ["Number of bills"]
time_summary

--- Table 2: Aggregate Revenue Breakdown by Meal Shift ---

Figure 1: Total revenue by day of the week (bar chart)¶

The finished figure must show one bar per day (Thursday to Sunday) whose height is the total of total_bill for that day, with the title Figure 1: Total Gross Revenue Generated by Day of Week, the x-axis label Day of Week and the y-axis label Total Revenue ($).

Starting point: the chart shows the number of bills per day. Change what is plotted and the labels. Keep the bars starting at zero.

counts_by_day = df.groupby("day", observed=True)["total_bill"].count().reset_index()

plt.figure(figsize=(8, 5))
sns.barplot(data=counts_by_day, x="day", y="total_bill", color="#2a78d6")
plt.title("Figure 1: (starting point) Number of bills by day", fontsize=14, fontweight="bold")
plt.xlabel("Day of Week")
plt.ylabel("Number of bills")
plt.tight_layout()
plt.show()

Figure 2: Tips at lunch and dinner (box plot)¶

The finished figure must show a box plot of the tip amount for each shift (Lunch and Dinner), with the title Figure 2: Distribution of Server Tip Amounts: Lunch vs. Dinner, the x-axis label Meal Shift and the y-axis label Tip Amount ($).

Starting point: the box plot shows the total bill for each shift. Change the measure and the labels.

Reading a box plot

The line inside the box is the median; the box spans the middle half of the values (the interquartile range, IQR); the whiskers reach the furthest values within 1.5 × IQR of the box; points beyond them are drawn individually as possible outliers.

plt.figure(figsize=(8, 5))
sns.boxplot(data=df, x="time", y="total_bill", color="#9fc3ee")
plt.title("Figure 2: (starting point) Total bill by shift", fontsize=14, fontweight="bold")
plt.xlabel("Meal Shift")
plt.ylabel("Total bill ($)")
plt.tight_layout()
plt.show()

Figure 3: Bill size and tip (scatter plot with a trend line)¶

The finished figure must show each bill as a point, with total_bill on the x-axis and tip on the y-axis, plus a straight-line (least-squares) trend line. Use the title Figure 3: Linear Relationship: Total Bill Size vs. Tip Amount, the x-axis label Total Bill Amount ($) and the y-axis label Tip Amount ($).

Starting point: the cell draws the points but no trend line and placeholder labels. Replace the scatter with sns.regplot(...), which draws the points and the trend line together, and set the title and labels. Make the points semi-transparent so overlapping bills stay visible.

plt.figure(figsize=(9, 6))
plt.scatter(df["total_bill"], df["tip"], alpha=0.6, color="#2a78d6")
plt.title("Figure 3: (starting point) points only", fontsize=14, fontweight="bold")
plt.xlabel("x")
plt.ylabel("y")
plt.tight_layout()
plt.show()

Check before you write

• Restart the kernel and run every cell (Kernel → Restart Kernel and Run All Cells). Every table and figure must appear, with no errors.
• Each table and figure matches its specification above: right measure, right aggregation (average or total), right title and labels.
• The days run Thursday to Sunday.
• You may add a cell with an extra number you want to quote in the report (for example the slope of the trend line or a median from Figure 2). Keep it short and label it.

Part 2: The business report¶

Write the report in Word (or another editor that saves PDF). Use these four headings exactly, in this order. Write in full, connected paragraphs; only Section 4 uses bullet points. Refer to the evidence by name (Table 1, Figure 2…) and quote the numbers that support each point, with units.

1. Executive Summary¶

One paragraph. State the consulting context (Bistro Analytics and its client restaurant), the dataset (the tips data, N = 244 transactions) and the purpose: helping the owner schedule staff, grow revenue and keep good servers by acting on data.

2. Descriptive Data Overview¶

One or two paragraphs.

• From Table 1: which day has the highest average bill, and which has the highest average tip? Comment on party size where it helps explain the difference.
• From Table 2: compare total revenue at lunch and at dinner, and say how large the gap is.

3. Visual Insights & Graphical Trends¶

Two or three paragraphs.

• Figure 1: compare the weekdays (Thursday, Friday) with the weekend (Saturday, Sunday). Which days generate the most revenue?
• Figure 2: compare the medians and the spread (IQR) of tips at lunch and dinner. What does this mean for a server's earning potential on each shift? Comment on the high outliers above the upper whiskers.
• Figure 3: describe the direction and steepness of the trend line between bill and tip. What does this positive relationship suggest for upselling (for example appetisers, premium drinks, desserts)? Remember that an association is not proof that upselling causes higher tips.

4. Strategic Business Recommendations¶

Two detailed bullet points, each justified by the evidence above:

• Recommendation 1, staffing and labour: which days and shifts should get the strongest servers, so that peak demand is covered without slowing service?
• Recommendation 2, revenue and server training: a concrete upselling and service-training idea based on the bill–tip relationship, explaining how a higher average bill can benefit both the restaurant's revenue and the servers' tips.

Good practice from today's lectures

Lead with the finding, not the method ("Saturday and Sunday bring in most of the revenue…", not "Figure 1 is a bar chart that…"). Compare totals and averages carefully: a day can have a high average bill and still a low total if it has few bills. Write for the owner, a busy non-specialist.

Part 3: Export and submit¶

If you cannot combine PDFs on your computer, paste the notebook's tables and figures into the Word document after your report (as an appendix titled "Executed notebook output") and save everything as one PDF with the final file name.

How it is marked¶

The assignment is marked out of 100 with the course's assignment rubric, then scaled to its 6% weight.

Each criterion is judged on a scale from exemplary (full points) through excellent, proficient, developing and unsatisfactory; a missing submission scores 0.

Final checklist¶

• My notebook is saved as YourLastName_DATA300_Assignment1.ipynb and runs from top to bottom without errors.
• Table 1, Table 2, Figure 1, Figure 2 and Figure 3 match their specifications.
• My report has the four headings exactly, quotes numbers with units, and has two detailed recommendations.
• I have not claimed that anything causes higher tips.
• One PDF, YourLastName_DATA300_Assignment1.pdf, report first, uploaded to GAP by Sunday 15 November, 23:59 China Standard Time.