IB Data Table and Analysis Tool

DATA
DISPLAY
ANALYZE
EXPORT TABLE
EXPORT for GRAPHING

✓ Copied Processed Data!

Your data is on your clipboard! You can paste it into a spreadsheet, or head directly to our graphing tool.

Go to Graphing Tool 📊

Comparative Statistical Analysis

Five-Number Summary

Method: ?

Correlation Statistics

Calculate stats to see trend data.

Frequency Distributions

This groups your raw data into ranges (bins) so you can plot a Histogram to see the shape of your data.

Calculate stats to see distribution data.

IA Analysis Assistant

Use this guide to structure your Data Analysis section. To score well on the IB rubric, you must go beyond just listing numbers—you must interpret what those numbers mean for your specific research question.

👑 The Golden Rules of IB Science Writing

  • Never use the word "Prove": Science provides evidence; it does not prove. Use terms like "supports," "suggests," or "indicates."
  • No "Naked" Claims: If you claim a trend exists, you must immediately back it up by citing specific Means and Standard Deviations in the exact same sentence.
  • Context is Everything: Do not just write out the math. Explain what the math means in the real-world context of your independent and dependent variables.
  • Reference Your Figures: Always explicitly direct the examiner to your tables and graphs (e.g., "As demonstrated in Figure 2...").

1. Describing Your Trend (Means & Spread)

Before moving on to complex stats, ensure your opening analysis paragraph answers these questions:

  • Did you explicitly state which group had the highest and lowest average results?
  • Did you include the ± Standard Deviation (SD) or Standard Error (SEM) immediately after every mean you stated?
  • Did you discuss the spread of your data? (High SD = high variation/low precision; Low SD = tightly clustered data/high precision).
  • If you graphed error bars and they overlap, did you mention that the differences between those specific groups might not be significant?
  • Handling Anomalies: Did you identify any outliers in your raw data? If you excluded a point from your mean, did you explicitly justify why?

2. Discussing Statistical Significance

When reporting p-values (T-Tests, ANOVA), focus on the null hypothesis and the Scientific Why.

❌ Weak (Robotic):

"The p-value is 0.03. This is less than 0.05, so the null hypothesis is rejected."

✅ Strong (Context + Theory):

"An ANOVA yielded a p-value of 0.03 (p < 0.05), rejecting the null hypothesis. This confirms a significant increase in reaction rate as temperature rises, which aligns with Collision Theory as particles possessed greater kinetic energy."

3. Discussing Correlation (Trendlines & R²)

❌ Weak:

"The graph goes up, so they are correlated. The R-squared is 0.92."

✅ Strong:

"There is a strong positive correlation between light intensity and plant growth. The R² value of 0.92 indicates that 92% of the variance in growth can be directly predicted by changes in light intensity."

4. Linking to Your Evaluation

Your statistics should fuel the Evaluation section of your IA. Use your stats to identify exactly where your methodology lacked precision.

  • If you had a very large Standard Deviation in one specific trial, explain why (e.g., an uncontrolled variable or equipment error that happened that day).
  • If your error bars are massive, critique the precision of your measuring instruments.
  • Qualitative Data: Did you support your numbers with observations? (e.g., "The high standard deviation at 50°C is likely because the solution began to violently boil, making the meniscus difficult to read accurately.")

Validation Note

This calculator seamlessly isolates Raw Trial Data from either layout mode, matching Excel's exact statistical output.

IB Statistics Glossary

  • SD (Standard Deviation): Measures the amount of variation or dispersion in your data. A low SD indicates data is tightly clustered around the mean.
  • SEM (Standard Error of the Mean): Estimates how far your sample mean is likely to be from the true population mean. It accounts for sample size.
  • p-value: The probability that your results happened by random chance. A value p < 0.05 is generally considered statistically significant.
  • Levene's Test: Assesses the equality of variances across your groups. If p > 0.05, your data's variance is homogenous (equal spread), which guides the calculator to pick the most accurate test type.
  • T-Test: A statistical test used to compare the means of exactly two distinct groups.
    • Student's T-Test: The classic version, used when variances are equal.
    • Welch's T-Test: The robust alternative, used automatically when variances are unequal.
  • ANOVA (Analysis of Variance): A "Single-Factor" (or One-Way) test used instead of multiple T-Tests to compare three or more groups simultaneously across one independent variable.
    • Fisher's ANOVA: The classic version, used when variances are equal.
    • Welch's ANOVA: The robust alternative, used when variances are unequal.
  • Post-Hoc Test: A follow-up test conducted only if an ANOVA finds a significant difference, used to determine exactly which specific groups differ from each other.
  • Bonferroni Correction: A strict mathematical penalty applied during Post-Hoc testing to prevent false positives (Type I errors) when making multiple pairwise comparisons.

💡 Stats Tip: The Post-Hoc Paradox

Can an ANOVA be significant, but the Post-Hoc test finds no differences? Yes! ANOVA looks at the "big picture" of all groups combined, giving it the power to spot subtle overall trends. Post-Hoc tests (like Bonferroni) look at groups locally, two at a time, and apply a massive penalty to prevent false positives. If your ANOVA is only barely significant (e.g., p = 0.04), that heavy penalty can wipe out the significance of individual pairs!

Citable Sources for Your Report:

🧪 The Stats Wizard

What is the goal of your analysis?

Are you looking for differences between specific groups (e.g., control vs. treated), or a relationship/trend between two variables?