Scribd
Upload
90 views1 page

G.U.T.S. (Grand Unified Theory of Statistics)

The document outlines JB's Grand Unified Theory of Statistics (G.U.T.S.), which identifies three key considerations for statistical analysis: variable types, variable roles, and sample/population numbers. It then matches seven statistical scenarios involving different variable types and roles to the appropriate preliminary statistical model, method, procedure, or tool, such as one-sample z-test, two-sample t-test, chi-square test of independence, simple linear regression, and multiple regression. The goal is to help choose the right statistical analysis based on variable characteristics and sample structure.

Uploaded by

Manan Shah
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
90 views1 page

G.U.T.S. (Grand Unified Theory of Statistics)

The document outlines JB's Grand Unified Theory of Statistics (G.U.T.S.), which identifies three key considerations for statistical analysis: variable types, variable roles, and sample/population numbers. It then matches seven statistical scenarios involving different variable types and roles to the appropriate preliminary statistical model, method, procedure, or tool, such as one-sample z-test, two-sample t-test, chi-square test of independence, simple linear regression, and multiple regression. The goal is to help choose the right statistical analysis based on variable characteristics and sample structure.

Uploaded by

Manan Shah
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 1

JB’s Grand Unified Theory of Statistics [G.U.T.S.

]
Need to know three things:
1. types of variables: binary, categorical (>2 categories), quantitative
2. role of variables: predictor vs. outcome
3. number of samples/populations: one, two, many (perhaps pairing?)

Model or Method
Outcome Var. Predictor Var. or Procedure or Tool
Binary No Predictor: One-sample z-test
Prelim. – just an external
Quantitative standard One-sample t-test
(incl. paired t-test)
Categorical χ2 goodness of fit-test

1) Quantitative Binary Two-sample t-test


(e.g. Salary, $) (e.g. Gender M/F) of means

2) Binary Binary Two-sample z-test


(e.g. Text in class, (e.g. Gender M/F) of proportions Yes/No)
(or chi-square test)

3) Categorical Categorical Chi-square test of


(e.g. Smoking Status (e.g. SES independence
Current/Former/Never) High/Middle/Low)

4) Quantitative Quantitative Simple Linear


(e.g. C291 grade (e.g. Study Hours, Regression
0-100%) 0- ?? hrs) Chapter 14

5) Quantitative Many: Quant. & Multiple


(e.g. C291 grade Categ. & Binary Regression
0-100%) (e.g. Study Hrs, Quant. IQ Chapter 15
Class Attendance, Sleep,
Health, Location in Class)

6) Quantitative Categorical One-way


(e.g. Job Satisfaction, (e.g. Employment Analysis of
0-100 scale) Status: FT, PT, Casual) Variance
NOT IN C291 (See Ch. 16)

7) Binary Quantitative; or Simple/Multiple


(e.g. Purchase, Combination of Quant. & Logistic
Yes/No) Categorical & Binary Regression
NOT IN C291 (See Ch. 16)

You might also like