In this chapter, you will bring together everything you have learned by completing a full A/B test workflow from start to finish. This comprehensive challenge will guide you through each stage, reinforcing your understanding and practical skills. Here is a checklist of the steps you should follow for a complete A/B test analysis:

- Define a clear hypothesis for your A/B test;
- Simulate or obtain sample data for both control and variant groups;
- Calculate and compare relevant metrics for each group;
- Perform a statistical test to determine significance;
- Compute confidence intervals for your key metric(s);
- Visualize the results for clear communication;
- Interpret your findings and summarize your conclusions in a concise report.

By following these steps, you will demonstrate your ability to conduct and communicate a robust A/B test analysis, preparing you for real-world experimentation scenarios.

import unittest
import user_code
import ast
import re   
import importlib
import csv
import numpy as np
import pandas as pd
import sys
from io import StringIO

class TestTask(unittest.TestCase):
    def test_simulate_ab_test_data_output(self):
        import user_code
        importlib.reload(user_code)
        # Use larger sample sizes to reduce variance
        data = user_code.simulate_ab_test_data(
            size_control=1000,
            size_variant=2000,
            conv_rate_control=0.1,
            conv_rate_variant=0.3,
            seed=123
        )
        _dynamic_test(
            self,
            isinstance(data, pd.DataFrame),
            "Function returns a DataFrame.",
            f"Expected pandas DataFrame, got {type(data)}."
        )
        control = data[data['group'] == 'control']
        variant = data[data['group'] == 'variant']
        _dynamic_test(
            self,
            len(control) == 1000 and len(variant) == 2000,
            "Correct sample sizes in each group.",
            f"Expected 1000 control and 2000 variant rows, got {len(control)} and {len(variant)}."
        )
        # Check conversion rates are within reasonable bounds (allow for statistical noise)
        conv_rate_c = control['conversion'].mean()
        conv_rate_v = variant['conversion'].mean()
        _dynamic_test(
            self,
            0.07 <= conv_rate_c <= 0.13,
            "Control group conversion rate is within reasonable range of input rate.",
            f"Expected ~0.1 (+/-0.03), got {conv_rate_c}."
        )
        _dynamic_test(
            self,
            0.26 <= conv_rate_v <= 0.34,
            "Variant group conversion rate is within reasonable range of input rate.",
            f"Expected ~0.3 (+/-0.04), got {conv_rate_v}."
        )

    def test_analyze_ab_test(self):
        import user_code
        importlib.reload(user_code)
        # Use deterministic (non-random) data for reproducible test
        data = pd.DataFrame({
            'group': ['control'] * 20 + ['variant'] * 20,
            'conversion': [0,1,0,0,0,1,0,0,0,0, 1,0,1,0,0,1,0,0,0,1,  # 6/20 = 0.3
                           1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1]   # 20/20 = 1.0
        })
        result = user_code.analyze_ab_test(data)
        _dynamic_test(
            self,
            isinstance(result, dict),
            "analyze_ab_test returns a dict.",
            f"Expected dict, got {type(result)}."
        )
        _dynamic_test(
            self,
            all(k in result for k in ['conv_rate_control','conv_rate_variant','difference','z_score','p_value','ci_low','ci_high']),
            "All required keys present in result.",
            f"Missing keys in result: {result.keys()}."
        )
        # Check conversion rates
        _dynamic_test(
            self,
            np.isclose(result['conv_rate_control'], 0.3),
            "Correct control group conversion rate.",
            f"Expected 0.3, got {result['conv_rate_control']}."
        )
        _dynamic_test(
            self,
            np.isclose(result['conv_rate_variant'], 1.0),
            "Correct variant group conversion rate.",
            f"Expected 1.0, got {result['conv_rate_variant']}."
        )
        # Difference
        _dynamic_test(
            self,
            np.isclose(result['difference'], 0.7),
            "Correct difference in conversion rates.",
            f"Expected 0.7, got {result['difference']}."
        )
        # p-value should be < 0.05 for this big difference
        _dynamic_test(
            self,
            result['p_value'] < 0.05,
            "p-value is significant for large difference.",
            f"Expected p-value < 0.05, got {result['p_value']}."
        )
        # CI should contain the difference
        _dynamic_test(
            self,
            result['ci_low'] < result['difference'] < result['ci_high'],
            "Difference inside confidence interval.",
            f"Difference is not inside CI: {result['ci_low']}, {result['difference']}, {result['ci_high']}."
        )

    def test_plot_ab_results(self):
        import user_code
        importlib.reload(user_code)
        import matplotlib
        matplotlib.use('Agg')
        data = pd.DataFrame({
            'group': ['control'] * 10 + ['variant'] * 10,
            'conversion': [0,1,0,0,0,1,0,0,0,0, 1,1,1,0,0,1,1,0,0,1]
        })
        # Should not raise
        try:
            user_code.plot_ab_results(data)
            _dynamic_test(
                self,
                True,
                "plot_ab_results runs without error.",
                "plot_ab_results raised an exception."
            )
        except Exception as e:
            _dynamic_test(
                self,
                False,
                "plot_ab_results runs without error.",
                f"plot_ab_results raised an exception: {e}"
            )

    def test_interpretation_and_reporting(self):
        import user_code
        importlib.reload(user_code)
        # Simulate data and run analysis (large sample for statistical stability)
        data = user_code.simulate_ab_test_data(
            size_control=2000,
            size_variant=2000,
            conv_rate_control=0.12,
            conv_rate_variant=0.15,
            seed=42
        )
        result = user_code.analyze_ab_test(data)
        # Capture print output
        buffer = StringIO()
        sys.stdout = buffer
        user_code.plot_ab_results(data)
        interpretation = (
            f"Control group conversion rate: {result['conv_rate_control']:.3f}\\n"
            f"Variant group conversion rate: {result['conv_rate_variant']:.3f}\\n"
            f"Difference: {result['difference']:.3f}\\n"
            f"95% Confidence Interval: [{result['ci_low']:.3f}, {result['ci_high']:.3f}]\\n"
            f"p-value: {result['p_value']:.4f}\\n"
        )
        print(interpretation)
        if result['p_value'] < 0.05:
            conclusion = "The result is statistically significant. The variant outperforms the control."
        else:
            conclusion = "The result is not statistically significant. There is no evidence the variant outperforms the control."
        print(conclusion)
        sys.stdout = sys.__stdout__
        output = buffer.getvalue()
        _dynamic_test(
            self,
            "Control group conversion rate" in output and "Variant group conversion rate" in output,
            "Interpretation summary includes conversion rates.",
            f"Interpretation summary missing conversion rates. Output: {output}"
        )
        if result['p_value'] < 0.05:
            _dynamic_test(
                self,
                "statistically significant" in output.lower(),
                "Correct conclusion for significant result.",
                f"Expected 'statistically significant' in output. Output: {output}"
            )
        else:
            _dynamic_test(
                self,
                "not statistically significant" in output.lower(),
                "Correct conclusion for nonsignificant result.",
                f"Expected 'not statistically significant' in output. Output: {output}"
            )

def _dynamic_test(test_case, condition, success_message, failure_message):
    if condition:
        test_case._testMethodName = success_message
        test_case.assertTrue(True, success_message)
    else:
        test_case._testMethodName = failure_message
        test_case.fail(failure_message)

def normalize_text(text):
    text = text.lower()
    text = re.sub(r"\\s{2,}", " ", text)
    text = re.sub(r"\\s*([,:?])\\s*", r"\\1 ", text)
    return text.strip()

def change_var(code: str, var_name: str, value: str) -> str:
    tree = ast.parse(code)
    lines = code.splitlines()
    changed = False
    # Collect all assignment nodes to modify
    assign_nodes = [
        (i, node)
        for i, node in enumerate(tree.body)
        if isinstance(node, ast.Assign)
        and any(isinstance(target, ast.Name) and target.id == var_name for target in node.targets)
    ]

    # If nothing to change, return unmodified code
    if not assign_nodes:
        return code

    # Perform replacements for all matching assignments (from last to first to not break line offsets)
    for i, node in reversed(assign_nodes):
        start_line = node.lineno - 1
        line = lines[start_line]
        indent = ' ' * (len(line) - len(line.lstrip()))
        lines[start_line] = f"{indent}{var_name} = {value}"
        next_line = len(lines)
        for next_node in tree.body[i+1:]:
            if hasattr(next_node, 'lineno'):
                next_line = next_node.lineno - 1
                break
        if next_line > start_line + 1:
            lines[start_line+1:next_line] = []
        changed = True

    return '\\n'.join(lines) if changed else code

if __name__ == "__main__":
    unittest.main()


test_main.py

Master the art and science of A/B testing using Python. This course covers the statistical foundations, experiment design, and practical implementation of A/B tests, with a strong emphasis on hands-on practice and real-world scenarios. Learn to analyze results, avoid common pitfalls, and communicate findings effectively.

Explore the essentials of A/B testing — from core terminology and the scientific method to practical workflows used in technology and marketing. This section provides a comprehensive foundation, including the historical context, the rationale behind experimentation, and the essential vocabulary needed for effective communication in data-driven environments.

Master the skills needed to design robust A/B tests: from hypothesis formulation and randomization to sample size calculation and metric selection. This section covers hypothesis creation, randomization, sample size determination, and metric selection in detail, with practical examples and hands-on exercises.

Dive into the statistical concepts that underpin A/B testing, including significance, p-values, error types, and confidence intervals. This section provides detailed explanations, practical code examples, and hands-on exercises to solidify your understanding.

Dive into the fundamentals of practical A/B test analysis — simulate data, interpret results, create compelling visualizations, and communicate findings to stakeholders. Learn to communicate findings effectively, create compelling visualizations, and avoid common pitfalls in interpretation.

Challenge: Full A/B Test Workflow

Solution