How far can you get with data and stats?

PSYC 11: Laboratory in Psychological Science

Jeremy R. Manning
Dartmouth College
Spring 2026

Reflecting on the pitches

On Wednesday you presented your pitches and filled out your reflections. Let's hear from a few of you:

  • What was the most surprising thing you observed during the pitches?
  • Did any pitch change your mind about something?
  • What made the most effective pitches work?

What is the point of this lab?

  • We're trying to gain insights into what makes an effective "pitch"
  • This is directly relevant to the Introduction section of scientific articles
  • Also relevant to presenting and describing your science more generally

What tools do we have?

  • Ratings data from the pitch evaluations (clarity, interest, efficiency, format)
  • Statistical tests and analytic tools
  • Visualization tools (figures!)
  • Our intuitions from having been in the room

Planning your analysis

With a partner, discuss:

  • What specific questions can you answer with the ratings data? (e.g., "Did group A score higher on clarity than group B?")
  • What questions can't you answer with the data alone? (e.g., "Why was group C's pitch more interesting?")
  • How would you test whether the differences between groups are "real" vs. due to chance?

Be ready to share one insight with the class.

The power of data

  • Reveal patterns we might not see on our own
  • Quantify how confident we should be in our conclusions
  • Compare groups, test hypotheses, make predictions
  • For this lab: compare mean ratings, rank groups, test whether differences are statistically significant
  • The "truth" about which presentation was actually best
  • Specific insights into why a pitch was effective (or not)
  • How to account for every source of bias— presentation order, audience mood, confounding variables, etc.
  • Whether the ratings reflect the quality of the idea or the quality of the pitch (remember: this is intentionally ambiguous!)

Data wrangling: getting the data into shape

Data wrangling means organizing or transforming your data into a format that is more convenient for you to work with.

  • Are there any challenges to analyzing the pitch ratings in their current form?
  • What format do you want the data in?
  • How might you "wrangle" the dataset into a more convenient format?
  • How might you use GenAI to help?

Data + Intuition = ❤️

  • Don't throw your intuitions out the window
  • You were there— you saw the pitches, felt the energy, noticed things the data can't capture
  • Use common sense to interpret results and understand limitations
  • At the same time, keep an open mind and be willing to revise your intuitions based on the data
  • Data and intuition work best together
  • The goal is to communicate your best understanding of the truth

Now: analyze the data with your group

  • Open the pitch ratings analysis notebook (also linked via QR code below)
  • Work with your group to wrangle, analyze, and visualize the data
  • Remember your predictions from Step 5 of the lab— how do the actual results compare?
  • Use GenAI to help you explore different analyses and visualizations

Pitch ratings analysis notebook (Google Colab)

Questions? Want to chat more?

📧 Email me
💬 Join our Slack
💁 Come to office hours
  • Your written report on the pitch ratings analysis is due on Monday!
  • Next week: Effective explaining + Picture (drawing) lab starts
  • Please read the lab instructions before Monday