Chapter 36 Exercises: Planning Your Future in Data Science
How to use these exercises: These are self-assessment and planning exercises. Unlike other chapters, there are no "right answers" — the goal is honest reflection and concrete planning. Treat this as a working session with your future self as the client.
Difficulty key: ⭐ Foundational | ⭐⭐ Intermediate | ⭐⭐⭐ Advanced | ⭐⭐⭐⭐ Extension
Part A: Self-Assessment ⭐
Exercise 36.1 — The honest skills audit
Rate yourself 1-5 on each skill area below (1 = "I've been exposed to this but can't do it independently" to 5 = "I can do this confidently and could teach it to someone else"). Be honest — this assessment is for you, not for a grade.
| Skill Area | Self-Rating | Evidence (specific project or exercise) |
|---|---|---|
| Python programming (variables, functions, control flow) | ||
| pandas (loading, filtering, grouping, merging) | ||
| Data cleaning (missing values, duplicates, type issues) | ||
| Data visualization (matplotlib, seaborn) | ||
| Descriptive statistics (mean, median, standard deviation) | ||
| Hypothesis testing (t-tests, chi-square, ANOVA) | ||
| Linear regression | ||
| Classification (logistic regression, decision trees) | ||
| Model evaluation (train/test, cross-validation, metrics) | ||
| Written communication of data findings | ||
| Git and GitHub | ||
| SQL |
After completing the table, identify: (a) your three strongest areas, (b) your three weakest areas, and (c) one area where your self-rating surprised you (higher or lower than expected).
Guidance
The "evidence" column is critical. A self-rating without evidence is just a feeling. If you rate yourself 4 on visualization but can't point to a specific chart you're proud of, reconsider the rating. If you rate yourself 2 on SQL but have never actually studied SQL, that's a 1 — you can't rate yourself on something you haven't tried. Honest self-assessment is a professional skill in itself.Exercise 36.2 — What did you learn that surprised you?
Reflect on your journey through this book. Write 3-5 sentences answering each question:
- What topic was harder than you expected?
- What topic was easier or more enjoyable than you expected?
- What single concept or skill has changed how you think about the world?
- What do you wish the book had covered that it didn't?
Guidance
Common answers to question 1: data cleaning (everyone underestimates how time-consuming it is), hypothesis testing (the logic of null hypotheses is genuinely counterintuitive), or writing up results (harder than the analysis itself). Common answers to question 3: "correlation is not causation" (once you see it, you can't unsee misleading causal claims), or "the question comes before the data" (a mindset shift that changes how you approach every problem). Your answers are personal — there are no wrong ones.Exercise 36.3 — The skills you didn't know you'd need
Think about your capstone project. List three skills you used during the capstone that you didn't expect to need when you started the book. For each one, explain how you acquired it (was it taught in the book? did you look it up? did you figure it out through trial and error?).
Guidance
Common unexpected skills: writing clear Markdown narrative (many students don't realize how much writing data science involves), debugging (reading error messages is a skill that develops through practice, not instruction), and making judgment calls under uncertainty (no chapter teaches you exactly how to handle *your* specific missing data problem — you have to reason through it). These "hidden skills" are often the most valuable ones for professional work.Exercise 36.4 — Revisiting Chapter 1
Go back and read your answers to the Chapter 1 exercises — particularly Exercise 1.1 (your definition of data science) and Exercise 1.5 (a question you want to answer with data). Then answer:
- How has your understanding of data science changed since Chapter 1?
- Is the question you articulated in Chapter 1 still interesting to you? Could you answer it now?
- What would you tell your Chapter-1 self about the journey ahead?
Guidance
This exercise is about recognizing growth. Your Chapter 1 definition of data science was probably vague — "using data to answer questions" or "statistics plus coding." Your definition now should be richer and more nuanced, informed by 35 chapters of actual data science practice. If the question you asked in Chapter 1 is one you could answer now — with real data, real code, and real analysis — that's powerful evidence of how far you've come.Part B: Career Exploration ⭐⭐
Exercise 36.5 — Job posting analysis
Find three real job postings for data roles that interest you (use LinkedIn, Indeed, Glassdoor, or a company's careers page). For each posting, record:
| Field | Posting 1 | Posting 2 | Posting 3 |
|---|---|---|---|
| Title | |||
| Company / Industry | |||
| Required skills (list the top 5) | |||
| "Nice to have" skills | |||
| Years of experience required | |||
| Skills you HAVE from this book | |||
| Skills you need to LEARN |
Then answer: What patterns do you notice across the three postings? What skill appears most frequently that you haven't learned yet?
Guidance
This exercise grounds your learning plan in actual market demand. You'll likely notice that SQL appears in nearly every posting, that communication skills are listed more often than you'd expect, and that "3+ years of experience" is often listed for roles that are actually open to entry-level candidates with strong portfolios. Don't be discouraged by requirements lists — they're wish lists, not minimum thresholds.Exercise 36.6 — Career path comparison
Using the four career paths described in Section 36.2, create a personal comparison:
| Factor | Data Analyst | Data Scientist | ML Engineer | Data Engineer |
|---|---|---|---|---|
| How excited am I? (1-5) | ||||
| How well do my current skills match? (1-5) | ||||
| How much additional learning is needed? (estimate months) | ||||
| What's the job market like in my area? | ||||
| Would I enjoy the typical day described? |
Based on this analysis, identify your top one or two target paths and explain your reasoning in 3-4 sentences.
Guidance
Excitement and skill match are both important but serve different functions. Excitement sustains motivation through the hard learning ahead. Skill match determines how quickly you can become job-ready. If your excitement and skill match point to different paths, lean toward excitement — you can always build skills, but you can't manufacture genuine interest.Exercise 36.7 — The informational interview plan
Identify three people you could reach out to for informational interviews — data professionals whose careers you admire or who work in roles you're targeting. For each person:
- Who are they? (Name and role, or a description if you haven't identified a specific person yet)
- How will you find/contact them? (LinkedIn, meetup, alumni network, mutual connection)
- What three specific questions would you ask them?
Write a draft outreach message (3-4 sentences, polite and specific) that you could send to one of these people.
Guidance
Good informational interview questions are specific and demonstrate that you've done your homework: "How did you transition from academic research to industry data science?" or "What skills do you use most frequently that you didn't expect to need?" Avoid questions that Google could answer ("What does a data scientist do?"). A good outreach message is short, specific about why you're reaching out to *them* specifically, and asks for a defined time commitment (15-20 minutes).Exercise 36.8 — Day-in-the-life simulation
Choose your target career path and write a realistic "day in the life" scenario — a description of what a typical Tuesday might look like if you had that role. Include: what you'd work on in the morning, what meetings or collaborations you'd have, what tools you'd use, and what deliverables you'd produce by end of day.
Then write 2-3 sentences about which parts of that day excite you and which parts concern you.
Guidance
Be realistic, not idealized. A data analyst's day involves a lot of SQL queries, dashboard updates, and responding to ad-hoc requests from stakeholders — not building exciting models. A data scientist's day often involves 70% cleaning data and only 10% actual modeling. An ML engineer's day might involve more debugging deployment pipelines than building new models. If the realistic version of the role still excites you, that's a great sign.Part C: Learning Plan ⭐⭐
Exercise 36.9 — The three-skill priority
Based on your career path choice and skills audit, identify the three skills you most need to develop. For each skill:
- Why is this skill important for your target role?
- What's your current level? (None / Beginner / Intermediate)
- What's your target level in 6 months?
- What specific resource will you use to learn it? (Book, course, tutorial — be specific)
- How will you demonstrate this skill? (Portfolio project, certification, etc.)
Guidance
The "how will you demonstrate it" question is the most important. Learning without demonstration is invisible — and invisible skills don't help you get hired. For every skill you learn, plan a portfolio project or tangible output that proves you learned it. "I completed a SQL course" is less convincing than "I built a project analyzing public transit data entirely in SQL."Exercise 36.10 — The six-month roadmap
Create a detailed learning plan for the next six months. Use this template:
Month 1: ___ - Primary learning goal: - Resource: - Time commitment: ___ hours/week - Deliverable by end of month:
Month 2: ___ - Primary learning goal: - Resource: - Time commitment: ___ hours/week - Deliverable by end of month:
Month 3: ___ - Primary learning goal: - Resource: - Time commitment: ___ hours/week - Deliverable by end of month:
Month 4: ___ - Primary learning goal: - Resource: - Time commitment: ___ hours/week - Deliverable by end of month:
Month 5: ___ - Primary learning goal: - Resource: - Time commitment: ___ hours/week - Deliverable by end of month:
Month 6: Review and Recalibrate - Self-assessment: What did I learn? What gaps remain? - Portfolio update: What new projects can I add? - Career progress: What applications, interviews, or connections have I made? - Next six months: What comes after this roadmap?
Guidance
Be realistic about time commitments. If you're working full-time, 5-10 hours per week of focused learning is achievable and sustainable. If you're a student, you might have more time. The key is consistency — 1 hour per day, five days a week, is more effective than 10 hours crammed into one weekend day. Each month should have a concrete deliverable — something you can point to and say "I made this." Vague goals ("learn more SQL") get abandoned; specific goals ("complete chapters 1-8 of Practical SQL and build a project analyzing public transit data") get done.Exercise 36.11 — The accountability structure
Learning alone is hard. Design an accountability structure that will help you follow through on your roadmap:
- Learning partner: Is there someone who could learn alongside you? A friend, classmate, or online study group?
- Public commitment: Would posting your roadmap publicly (blog, LinkedIn, Twitter) help you stay committed?
- Weekly check-in: What day and time will you review your progress each week?
- Reward system: What will you do to celebrate completing each month's goal?
Guidance
Research on habit formation consistently shows that accountability mechanisms dramatically improve follow-through. The specific mechanism matters less than having *something*. A weekly check-in with a study partner, a public learning log, or even a private journal where you track your progress — any of these is better than relying on willpower alone. The reward system is also important: learning is hard work, and recognizing your own progress keeps motivation alive.Part D: Reflection and Synthesis ⭐⭐⭐
Exercise 36.12 — The portfolio gap analysis
Review your current portfolio (from Chapter 34 and the capstone) against the three-project structure:
| Slot | Current Project | Skill Gaps Visible | Planned Improvement |
|---|---|---|---|
| Deep Dive | |||
| Domain Project | |||
| Technical Demo |
Then answer: If a hiring manager looked at your portfolio today, what would they conclude about your strengths? What gaps would they notice? What one project would most improve your portfolio?
Guidance
Be critical but fair. Your capstone is likely your Deep Dive. You may need a Domain Project that shows genuine interest in a specific field, or a Technical Demo that shows a skill like web scraping, dashboard building, or SQL analysis. The "one project that would most improve your portfolio" should address your most visible gap. If all your projects use the same technique, add one that demonstrates range. If none show SQL skills, build a SQL-based project.Exercise 36.13 — Writing your professional story
Write a 200-word professional summary (suitable for LinkedIn or a personal website) that tells the story of your data science journey. Include:
- Where you started (your background before data science)
- What you've learned and built (reference specific projects)
- What excites you about data science (be genuine)
- What you're looking for next (target role or learning goal)
Guidance
The best professional stories are specific and genuine. "I'm passionate about data science" is generic. "I started as a biology teacher who loved looking at test score patterns, taught myself Python, and built a capstone project investigating what drives global vaccination disparities. Now I'm looking for data analyst roles where I can bring my domain knowledge in education and public health to bear on real policy questions" is specific, memorable, and human.Exercise 36.14 — The learning resource evaluation
Find three learning resources (courses, books, or tutorials) for one of the skills in your learning roadmap. Evaluate each against these criteria:
| Criterion | Resource 1 | Resource 2 | Resource 3 |
|---|---|---|---|
| Name and type (book/course/tutorial) | |||
| Cost | |||
| Estimated time to complete | |||
| Is it project-based? | |||
| Does it match my current level? | |||
| Are reviews positive? (check multiple sources) | |||
| Is the content current? (check publication date) | |||
| Would I have a portfolio artifact afterward? |
Choose one and commit to starting it within the next week.
Guidance
The "project-based" and "portfolio artifact" criteria are underrated. A video course that you watch passively produces no evidence that you learned anything. A course that requires you to build a project produces a portfolio piece *and* deeper learning. When in doubt, choose the resource that requires you to produce something.Exercise 36.15 — The conference and community plan
Research three communities or events you could participate in:
- A local meetup — find one on Meetup.com, Eventbrite, or your local tech community calendar
- An online community — Reddit, Discord, Slack, or a forum relevant to your interests
- A conference or virtual event — upcoming data science conferences, hackathons, or workshops
For each, note: what it is, when/where it meets, what the cost is, and what you'd hope to gain from participating.
Guidance
Many people skip this exercise because networking feels uncomfortable. But the data on career success consistently shows that community involvement — even minimal involvement — accelerates learning and opens doors. You don't have to speak at a conference or present at a meetup. Showing up, listening, and asking one question counts. Posting a thoughtful response on Reddit counts. The bar is lower than you think.Part E: Looking Back, Looking Forward ⭐⭐⭐
Exercise 36.16 — Letter to your future self
Write a letter (200-300 words) to yourself, to be read in six months. Include: - What you've accomplished so far (celebrate it) - What you're planning to learn next (from your roadmap) - What you're worried about (be honest) - What you're excited about - A question you want to answer in six months ("Did I follow through? What surprised me?")
Save this letter somewhere you'll find it in six months — set a calendar reminder.
Guidance
This isn't busywork — it's a powerful accountability and reflection tool. Six months from now, reading what you wrote today will either affirm your progress or provide a gentle nudge to get back on track. Be honest about your worries (imposter syndrome, motivation, time management) because naming them reduces their power.Exercise 36.17 — Letter to a future student
Write a letter (200-300 words) to someone who is about to start this book. What would you tell them? What advice would you give? What would you want them to know about the journey ahead?
Guidance
This exercise serves two purposes: it helps you synthesize what you've learned (teaching is the best way to learn), and it could actually be useful — consider sharing it on a blog, a course forum, or with a friend who's considering learning data science. The best advice is usually the most honest: "Chapter 8 is harder than it looks," or "Don't skip the exercises — they're where the real learning happens," or "It gets easier around Chapter 15, I promise."Exercise 36.18 — The one-year vision
Imagine yourself one year from today. You've followed your learning roadmap, built new projects, and made progress toward your career goals. Describe your one-year-from-now self in specific, concrete terms:
- What role are you in (or actively pursuing)?
- What new skills have you added?
- What does your portfolio look like?
- What community are you part of?
- What's the most interesting data project you've completed?
Guidance
Specificity makes visions actionable. "I want to be a data scientist" is a wish. "I'm working as a junior data analyst at a healthcare company, using SQL and Tableau daily, continuing to build Python projects on the side, and attending the local PyData meetup monthly" is a plan. Even if the specifics change, having a clear vision gives direction to your daily choices.Exercise 36.19 — Gratitude inventory
List three to five people, resources, or experiences that helped you most during your data science learning journey. For each, write one sentence about what they contributed. If any of them are people, consider telling them — a brief thank-you message costs nothing and means everything.
Guidance
Learning rarely happens in isolation. Maybe a study partner helped you debug a frustrating error. Maybe a YouTube video finally made p-values click. Maybe a family member gave you time and space to study. Recognizing the people and resources that supported your journey is both personally grounding and practically useful — these are the people and resources you'll turn to in the next phase of learning.Exercise 36.20 — The final question
In Chapter 1, Exercise 1.5, you articulated a question that data science could help you answer. Now, 35 chapters later, articulate a new question — something you're genuinely curious about that you'd like to investigate with data.
This question should be more specific, more nuanced, and more technically informed than the one you asked in Chapter 1. It should demonstrate how your thinking has evolved.
Then write 3-5 sentences describing how you would investigate this question: what data you'd need, what methods you'd apply, and what you'd expect to find.