https://goo.gl/forms/0Osph3CZG7chm2yF3

You must be logged in to you Colorado google account to access the survey. This survey is NOT anonymous, your Colorado.edu email address will be automatically recorded.

]]>

You should read the assigned chapters (listed below) and complete the practice questions (listed below). Answers to practice questions are in back of book.

- Chapter 1 – NA
- Chapter 2 – Question 7 page 22
- Chapter 3 – Question 1 page 46
- Chapter 4 – Question 8 page 71, Question 4 page 73
- Chapter 5 – Question 1 page 88, Question 1 page 89
- Chapter 6 – Just read! No questions
- Chapter 8 – Questions 2 + 3 Page 134
- Chapter 9 – Question 2 page 152
- Chapter 10 -Question 2 page 167
- Chapter 11 -Question 1-2 page 197; Question 3 page 189; Question 3 page 193
- Chapter 12 – read!
- Chapter 16 – Question 1 + 2 page 277
- Chapter 17 – Question 3 page 303
- Chapter 18 – Question 2 page 319
- Chapter 19 – Question 11 page 350
- Chapter 20 – Question 2 page 366
- Chapter 21 – Question 5 page 380, Question 6 page 387
- Chapter 23 – Question 9 page 414, Question 5 page 420, Question 2 page 425
- Chapter 26 – Question 5 page 478, Question 4 page 481, Question 7 + 10 page 487
- Chapter 27 – Question 4 + 5 page 515

- Spatial Autocorrelation (from lecture)
- Weights matrix
- Scale and correlations, MAUP.

- Regression
- When it works and when it doesn’t
- Slope and intercept
- Residuals what they are, what their patterns should be.
- RMS of the Residuals
- Geographic patterns in residuals (from lecture)
- Deriving the coefficient of determination (r-squared, from lecture)
- Making predictions with a regression model.

- Simpsons Paradox.
- Controlled Experiments
- Observational Studies
- Confounding
- Average (concept + formula)
- Standard Deviation (concept + formula)
- Normal curve, understand its properties and the kinds of processes that generate it.
- Chance Error
- Bias (be able to contrast with chance error)
- The difference between population parameters and statistics
- Standard Error
- Contrast with SD
- Compute SE for mean, sum, and percent (formula)

- Tests of significance (hypothesis tests)
- One sample T-Test (formula + concept)
- Two sample T-Test (formula + concept)
- What’s the difference between z and t test?
- Correlation (concept + formula)
- Confidence Intervals and how they relate to the normal curve.

https://goo.gl/forms/4mJpCwKfmmR5YQKu2

]]>

Sales of new single-family houses in September 2017 were at a seasonally adjusted annual rate of 667,000, according to estimates released jointly today by the U.S. Census Bureau and the Department of Housing and Urban Development. This is 18.9 percent (±19.0 percent)* above the revised August rate of 561,000 and is 17.0 percent (±22.4 percent)* above the September 2016 estimate of 570,000.

]]>

You should read the assigned chapters (listed below) and complete the practice questions (listed below). Answers to practice questions are in back of book.

- Chapter 1 – NA
- Chapter 2 – Question 7 page 22
- Chapter 3 – Question 1 page 46
- Chapter 4 – Question 8 page 71, Question 4 page 73
- Chapter 5 – Question 1 page 88, Question 1 page 89
- Chapter 6 – Just read! No questions
- Chapter 8 – Questions 2 + 3 Page 134
- Chapter 9 – Question 2 page 152
- Chapter 16 – Question 1 + 2 page 277
- Chapter 17 – Question 3 page 303
- Chapter 18 – Question 2 page 319
- Chapter 19 – Question 11 page 350
- Chapter 20 – Question 2 page 366
- Chapter 21 – Question 5 page 380, Question 6 page 387
- Chapter 23 – Question 9 page 414, Question 5 page 420, Question 2 page 425
- Chapter 26 – Question 5 page 478, Question 4 page 481, Question 7 + 10 page 487
- Chapter 27 – Question 4 + 5 page 515

- Controlled Experiments
- Observational Studies
- Confounding
- Average (concept + formula)
- Standard Deviation (concept + formula)
- Normal curve, understand its properties and the kinds of processes that generate it.
- Chance Error
- Bias (be able to contrast with chance error)
- The difference between population parameters and statistics
- Standard Error
- Contrast with SD
- Compute SE for mean, sum, and percent (formula)

- Tests of significance (hypothesis tests)
- One sample T-Test (formula + concept)
- Two sample T-Test (formula + concept)
- What’s the difference between z and t test?
- Correlation (concept + formula)
- Confidence Intervals and how they relate to the normal curve.

#Trust levels in 2010 and 2015

p2010 <- .6

p2015 <- .56

#Sample size for the poll

n <- 1000

#standard erros for the poll in 2010 and 2015

SE2010 <- sqrt((p2010*(1-p2010))/n)

SE2015 <- sqrt((p2015*(1-p2015))/n)

#SE of the difference between polls

SEdiff <- sqrt(SE2010^2 + SE2015^2)

#Z-test

z_test <- (p2015 – p2010) / SEdiff

#P-value

pnorm(z_test)

#Box Model

pop <- 1000000 #number of tickets in the box

#writing 1’s and 0’s on the tickets

box <- c(rep(1, pop*p2010), rep(0, pop*(1-p2010)))

#running the model

diffs <- NA

for (i in 1:10000){

samp1 <- sample(x = box, size = n)

samp2 <- sample(x = box, size = n)

diffs[i] <- mean(samp1) – mean(samp2)

}

#Distribution of the differences

hist(diffs)

abline(v=-0.06, col = “red”)

length(diffs) #total number of differences we calculated

#number of times diff was more than what we observed

length(diffs[diffs >= (p2010 – p2015)])

#p-value value via box model

length(diffs[diffs >= (p2010 – p2015)]) / length(diffs)

**Sign up here. **

This change may have been due to cultural shifts in the reception of clergy, or it may be due to chance variation.

In each year the poll uses an independent simple random sample of a thousand Americans.

- Would you do a one sample or a two sample hypothesis test? Why?
- Formulate a null and alternative hypothesis.
- If you wanted to make a box model to answer this question would you use one or two boxes? Why? How many tickets would you put into each box? How many draws from each box? What would you write on the tickets?
- Is the difference between the perception of clergy in 2010 and 2015 due to random chance.

The Center for Research Data and Digital Scholarship will judge the submitted visualizations based on their ability to communicate insightful information, their overall design, technical merit and originality. The rules are simple: you need to be a CU Boulder student, your visualization must have been created this year, and you can only submit two entries. You will be required to submit a mini-essay detailing the insights you personally gained about the data during the visualization process (minimum 100-words), as well as a brief (50-word maximum) biography of each author. Image files submitted for entry must adhere to the following name convention: If the visualization is dynamic, please provide a screenshot and hyperlink. Learn more about the required formating and submit your entry by going to the Data Visualization Contest Entry Form.

Selected visualizations will be displayed in Norlin Library. Top prizes of $200, $100 and $50 will be awarded to the top three entries.

]]>

Here are the instructions, data, and variable descriptions for lab 1.

You are welcome to use the lab computers. I recommend bringing your own computer to KESDA if you can, though, both to make saving things easier and because it’s most likely the machine you will be using in the “real world.”

]]>