Difference in Means

Part of the Statistics for ML section of Coddy's Machine Learning journey. Lesson 42 of 64.

Is a new model, page or keyboard better? Compare the groups by their difference in means: one group's mean minus the other's. Ten typists each tried one layout:

typing = pd.DataFrame({
    "layout": ["Qwerty", "Dvorak", "Qwerty", "Dvorak", "Qwerty", "Dvorak", "Qwerty", "Dvorak", "Qwerty", "Dvorak"],
    "wpm": [68, 76, 74, 65, 61, 82, 79, 70, 66, 74],
})
means = typing.groupby("layout")["wpm"].mean()
print(np.round(means["Dvorak"] - means["Qwerty"], 2))

This prints 3.8: Dvorak typists averaged 3.8 more words per minute. Qwerty minus Dvorak gives -3.8, so say which group comes first.

Five Qwerty typists and five Dvorak typists as dots spread from 61 to 82 words per minute, with each group's mean as a bar, 69.6 and 73.4, and a red bracket marking the gap of 3.8 between them

Is 3.8 big? Single typists spread over 20 words per minute, so the luck of who got which keyboard could make this gap alone.

Challenge

Medium

Students practiced with one of two math apps. math_apps.csv has app (algo or numo), score and sets; the program reads one number column's name.

Compute the numo mean minus the algo mean of that column.
Let its sign pick the app ahead, and print the line below: the gap positive, the leader's mean first, all to 2 decimals.

For other students and the input score:

algo is ahead on score by 4.0 (151.0 vs 147.0)

Try it yourself

import numpy as np
import pandas as pd

students = pd.read_csv("math_apps.csv")
column = input()

# Write your code below

Test yourself

This lesson includes a short quiz. Start the lesson to answer it and track your progress.

All lessons in Statistics for ML

2Distributions

Density Histograms Shapes of Data The Normal Curve The 68-95-99.7 Rule Z-Scores Recap: Battery Life

5Outliers

Box Plots The IQR Rule The Z-Score Rule Handling Outliers Recap: Sensor Glitches

8Comparing Groups

Box Plots by GroupDifference in MeansCould It Be Chance?P-Values The t-Test Recap: A/B Test

11Exploring a Dataset

A First Look Every Feature at Once Features Against the Label Two Best Features Recap: Diabetes Data

3Probability by Simulation

Probability as Frequency The Law of Large Numbers Sums of Dice Expected Value Conditional Probability Recap: Dice Stall Odds

6Relationships

Reading Scatter Plots Correlation Correlation Matrix Correlation Heatmaps Correlation Is Not Cause Recap: Ice Cream and Sun

9Feature Scaling

Why Scale Features Min-Max Scaling Standardization StandardScaler Recap: Climbing Partners

12Final Challenges

Train Delays Air Quality Wine Chemistry Clinic Survey

Practice on your own: Online Python compiler