LogoSigma
Statistics Exam Prep โ€” Personalized to You

Which chapter of this story are you in?

Sigma builds a study plan around your exam, your gaps, and the days you have left โ€” not a generic syllabus.

4,200+
students prepped
+18 pts
avg score lift
94%
pass rate
Scroll
The Ordinary World

The dread is real. So are the specific topics causing it.

Most stats anxiety isn't about math โ€” it's about never having someone explain the why behind each test.

Student sitting at desk late at night with open textbooks, looking overwhelmed by complex statistics problems

"I opened my biostat syllabus and genuinely thought it was in another language."

โ€” Priya M., pre-med sophomore

Biostatistics

The exact topics tripping them up:

  • 1Chi-square independence tests
  • 2Kaplan-Meier survival curves
  • 3Logistic regression odds ratios
  • 4Confidence intervals for proportions

Sigma covers all of these โ€” in your exam's exact format.

Professional adult student staring at laptop screen showing complex regression output with a confused expression

"I haven't used a formula since 2014. Now my MBA program is asking me to interpret regression output."

โ€” Derek T., MBA candidate

Business Analytics

The exact topics tripping them up:

  • 1Multiple regression interpretation
  • 2Multicollinearity diagnostics
  • 3ANOVA for business decisions
  • 4Forecasting with time series

Sigma covers all of these โ€” in your exam's exact format.

Graduate student surrounded by printed research papers and SPSS output at a late-night study session

"My committee asked me to explain my ANOVA assumptions. I smiled and said nothing for four seconds."

โ€” Camille R., psychology PhD candidate

Research Methods

The exact topics tripping them up:

  • 1Sphericity assumption testing
  • 2Post-hoc comparisons (Tukey, Bonferroni)
  • 3Mixed-design ANOVA
  • 4Effect size reporting (ฮทยฒ)

Sigma covers all of these โ€” in your exam's exact format.

The Mentor Appears

Every tutor has a stats-struggle origin story.

They didn't start out fluent. That's exactly why they're so good at explaining it.

Dr. Nalini Krishnaswamy, biostatistics tutor, smiling warmly in a professional academic setting
Dr. Nalini Krishnaswamy
Biostatistics & Clinical Research
Survival AnalysisLogistic RegressionClinical Trial Design
"

I failed my first biostatistics exam in medical school. Not by a little โ€” spectacularly. I remember sitting in the hallway afterward thinking: nobody ever explained *why* we use a log-rank test instead of a t-test for survival data. That gap between "here is the formula" and "here is when and why" is exactly what I teach into.

โœ“PhD Biostatistics, Johns Hopkins
โœ“8 years teaching pre-med students
โœ“340+ students prepped
Marcus Webb, business analytics tutor, sitting at a desk with laptop and business analytics materials
Marcus Webb, MS
Business Analytics & MBA Prep
Regression AnalysisBusiness ForecastingDecision Theory
"

I spent three years in consulting before going back for my master's. When I sat in my first stats class, I kept thinking: why won't anyone just tell me what this actually means for a real decision? I translate every concept into a scenario you've already lived through โ€” because that's the only way it sticks.

โœ“MS Statistics, University of Chicago
โœ“Former McKinsey analyst
โœ“5 years MBA tutoring
Dr. Yuki Tanaka, psychology research methods tutor, in an academic office surrounded by research books
Dr. Yuki Tanaka
Psychology Research & Thesis Defense
ANOVA & MANOVAFactor AnalysisSPSS & R
"

My dissertation defense was three weeks away when I realized I couldn't fully explain my own ANOVA assumptions. I called my advisor at 11pm. She talked me through it for two hours. That conversation changed everything โ€” I now give every one of my students that same 11pm conversation before they need it.

โœ“PhD Psychology, Northwestern
โœ“Dissertation committee member
โœ“200+ thesis defenses coached
Trials & Tests

Try one. Feel the click.

Tap any problem below. Read the question. Then reveal the worked solution โ€” and notice how it feels different when someone explains the why.

Escalating complexity โ†’
ProbabilityBasic Probability
+

A diagnostic test for a disease has 95% sensitivity and 90% specificity. In a population where 1% have the disease, what is the positive predictive value (PPV)?

๐Ÿ’ก Use Bayes' theorem. PPV = (sensitivity ร— prevalence) / ...
Worked Solution
PPV = (0.95 ร— 0.01) / [(0.95 ร— 0.01) + (0.10 ร— 0.99)] = 0.0095 / (0.0095 + 0.099) = 0.0095 / 0.1085 โ‰ˆ 8.8% Only 8.8% of positive tests are true positives โ€” even with a highly accurate test, rare diseases produce mostly false positives. This is the base rate fallacy in clinical practice.
Concepts:Bayes' Theorem ยท Positive Predictive Value ยท Base Rate Fallacy
Hypothesis Testingp-values & Significance
+

A study reports p = 0.04 for a new drug's effect. Your colleague says "there's a 4% chance the drug doesn't work." What's wrong with this interpretation?

๐Ÿ’ก Think about what p-value actually measures. It's a prob...
Worked Solution
The p-value (0.04) is the probability of observing data at least this extreme *if the null hypothesis were true* โ€” not the probability that the null hypothesis is true. The correct interpretation: "Assuming the drug has no effect, there is a 4% chance of seeing a result this large (or larger) by random chance alone." It says nothing about the probability the drug works.
Concepts:p-value interpretation ยท Null Hypothesis ยท Type I Error
RegressionMultiple Regression
+

In a multiple regression predicting salary from years_experience (ฮฒ=2,400) and has_MBA (ฮฒ=8,500), what does the ฮฒ for has_MBA actually mean?

๐Ÿ’ก Multiple regression coefficients are interpreted while ...
Worked Solution
ฮฒ = 8,500 for has_MBA means: holding years of experience constant, having an MBA is associated with a $8,500 higher predicted salary on average. This is the key insight of multiple regression โ€” it isolates each predictor's contribution by statistically controlling for the others. Without the multiple regression, you couldn't separate the MBA effect from the experience effect.
Concepts:Coefficient interpretation ยท Ceteris paribus ยท Partial effects
ANOVAOne-Way ANOVA
+

You run a one-way ANOVA comparing anxiety scores across 3 therapy conditions and get F(2, 57) = 4.23, p = .019. What does this tell you โ€” and what doesn't it tell you?

๐Ÿ’ก ANOVA's F-test is an omnibus test. Think about what "om...
Worked Solution
What it tells you: At least one group mean differs significantly from the others (p < .05). The F-ratio compares between-group variance to within-group variance. What it does NOT tell you: Which specific groups differ. F = 4.23, p = .019 only says "somewhere in these three groups there's a real difference." You need post-hoc tests (Tukey's HSD, Bonferroni) to find where. Effect size (ฮทยฒ) is also missing โ€” statistical significance โ‰  practical importance.
Concepts:Omnibus F-test ยท Post-hoc testing ยท Effect size ฮทยฒ

That clarity? That's what every session feels like.

We build your plan around exactly these moments โ€” the ones where the concept finally lands.

The Return

Score distributions from our last cohort.

Not testimonials. Actual before-and-after score distributions from 412 students, Spring 2025.

Before Sigma (n=412)
After Sigma (n=412)
+18.4
avg point gain
40โ€“50
51โ€“60
61โ€“70
71โ€“80
81โ€“90
91โ€“100
Score range (exam percentile)

Key shift: Before Sigma, 54% of students scored below 70. After Sigma, 71% scored above 80. The distribution doesn't just shift โ€” it concentrates in the upper range.

"Went from a 61 to an 84 on my biostatistics final. The p-value explanation alone was worth every session."

Priya Mehta
Pre-med sophomore, UC San Diego
+23 pts

"I hadn't thought about regression in 11 years. Marcus made it make sense in 3 sessions. Passed my MBA analytics exam with a B+."

Derek Torres
MBA candidate, Kellogg
+19 pts

"My committee asked about my ANOVA assumptions and I actually answered. That has never happened before."

Camille Rousseau
Psychology PhD, Northwestern
Defended โœ“
Find Your Study Plan

Three questions. A plan built for you.

Tell us your exam, your current confidence level, and where to send your plan.

Which exam are you preparing for?

How do you feel about stats right now?

No judgment. This helps us calibrate where to start.

Where should we send your plan?

You'll get a personalized study schedule + your first session recommendation within 10 minutes.

Exam:
Confidence:

No spam. Unsubscribe any time.

Not ready to commit? Start with a free diagnosis.

The "Diagnose My Weak Spots" mini-quiz takes 4 minutes and tells you exactly which topics to focus on first.