Dataford
Popular roles
Software EngineerData AnalystData ScientistData EngineerBusiness AnalystAI EngineerMachine Learning EngineerProduct Manager
Browse
Browse All RolesEvery role hub, from analyst to MLBrowse All CompaniesCompany-specific interview loopsAll Interview GuidesThe full guide library
Top questions by role
Software EngineerData AnalystData ScientistData EngineerBusiness AnalystAI EngineerMachine Learning EngineerProduct Manager
Top questions by skill
SQLPythonStatisticsMachine LearningA/B TestingSystem DesignGenerative AIProduct SenseMetricsBehavioral
Browse all questions →Try a mock interview
ExperiencesCoaching
New
Practice
Mock InterviewsTimed interview simulations with feedbackSuccess PathYour 6-week structured planModulesCurated lessons by topicWebinarsTalks from ex-Big Tech data leadsPlaygroundA free-form scratch editor
Learn
BlogInterview strategy and career adviceTech Job Market ReportHiring trends across data and AI rolesFor UniversitiesDataford for career centersAbout DatafordWho we are and how we build
Pricing
Get started

Explain P-Value and Power

MediumStatistics & Probability00:00
Practice interviewer
In session
5 left
00:00

Whenever you're ready, walk me through how you'd approach this.

You need to log in / sign up to chat or submit.

Problem

Business Context

Splice tested a new call-to-action on the Sounds browse page to increase the share of visitors who start a free trial. A Product Growth Analyst needs to explain the result to a non-technical PM using both intuition and the underlying numbers.

Problem Statement

You need to explain what a p-value means and what statistical power means in plain English, then compute both for this experiment and say whether the test was strong enough to trust.

Given Data

MetricControlTreatment
Visitors20,00020,000
Free-trial starts1,8001,980
Conversion rate9.0%9.9%

Additional assumptions for power analysis:

ParameterValue
Significance level0.05
Test typeTwo-sided
Minimum effect of interest0.8 percentage points
Baseline conversion rate for planning9.0%

Requirements

  1. State the null and alternative hypotheses.
  2. Compute the observed lift, pooled proportion, standard error, z-statistic, and p-value.
  3. Explain the p-value in plain English for a non-math audience.
  4. Estimate the test's statistical power to detect a true lift of 0.8 percentage points at the given sample size.
  5. Explain power in plain English and assess whether this experiment was adequately powered.
  6. Give a business recommendation for whether Splice should roll out the new CTA on the Sounds browse page.

Assumptions

  • Users were randomly assigned and counted once.
  • The normal approximation for proportions is appropriate.
  • Ignore segmentation and multiple-testing issues for the core calculation, but mention them in interpretation.