Wrap-up

1. Wrap-up

So what did we cover in this course and what R packages did we use?

2. R packages

Let's start with the R packages. We used survey for survey analysis, dplyr for data wrangling, and ggplot2 for data visualization. Okay. And, what concepts did we learn?

3. Course summary

Recall that we started by tackling the fundamentals of a sampling design, such as clusters and stratification. We learned that the survey weight for a sampled element is the number of population units it represents. Handling these weights properly turned out to be a key theme of the course! In Chapter 1, we practiced telling R about the structure of a survey with svydesign(). In Chapter 2, we explored one and two categorical variables. We summarized with svytable(), graphed with geom_col(), and conducted hypothesis tests with svychisq().

4. Course summary

We threw a quantitative variable into the mix in Chapter 3 and looked at useful summary stats, such as the mean, the total, and quantiles, computing these summary stats for different sub-populations using svyby(). We visualized the shape of a quantitative variable using geom_histogram() and geom_density(). We closed out Chapter 3 by testing for a difference in means with svyttest(). Chapter 4 was about modeling the trend between two quantitative variables. We produced survey-weighted scatter plots where we mapped the survey weights to the size, color, and opacity of the points. We added trend lines and then we built the linear regression model with svyglm().

5. Extensions

But of course, there is still so much more exciting survey analysis we could do. We could estimate more complex quantities, such as ratios, or we could build more complex models, such as logistic regression models.

6. Congratulations!

Sadly, we will have to save those topics for a future course! For now, it's time to go out and analyze your own survey data!

Create Your Free Account

or

By continuing, you accept our Terms of Use, our Privacy Policy and that your data is stored in the USA.

This exercise is part of the course

Analyzing Survey Data in R

IntermediateSkill Level

4.8+

Start Course for Free

Our exploration of survey data will begin with survey weights. In this chapter, we will learn what survey weights are and why they are so important in survey data analysis. Another unique feature of survey data are how they were collected via clustering and stratification. We'll practice specifying and exploring these sampling features for several survey datasets.

Exercise 1: What are survey weights?Exercise 2: Survey weights Exercise 3: Visualizing the weights Exercise 4: Specifying elements of the design in R Exercise 5: Designs in R Exercise 6: Stratified designs in R Exercise 7: Cluster designs in R Exercise 8: Comparing survey weights of different designs Exercise 9: Visualizing the impact of survey weights Exercise 10: NHANES weights Exercise 11: Tying it all together!

Now that we have a handle of survey weights, we will practice incorporating those weights into our analysis of categorical data in this chapter. We'll conduct descriptive inference by calculating summary statistics, building summary tables, and constructing bar graphs. For analytic inference, we will learn to run chi-squared tests.

Exercise 1: Visualizing a categorical variable Exercise 2: Summarizing a categorical variable Exercise 3: Interpreting frequency tables Exercise 4: Graphing a categorical variable Exercise 5: Exploring two categorical variables Exercise 6: Creating contingency tables Exercise 7: Building segments bar graphs Exercise 8: Summarizing with svytotal()Exercise 9: Interpreting svymean()Exercise 10: Inference for categorical variables Exercise 11: Running a chi squared test Exercise 12: Tying it all together!

Of course not all survey data are categorical and so in this chapter, we will explore analyzing quantitative survey data. We will learn to compute survey-weighted statistics, such as the mean and quantiles. For data visualization, we'll construct bar-graphs, histograms and density plots. We will close out the chapter by conducting analytic inference with survey-weighted t-tests.

Exercise 1: Summarizing quantitative data Exercise 2: Survey statistics Exercise 3: Estimating quantiles Exercise 4: Visualizing quantitative data Exercise 5: Bar plots of survey-weighted means Exercise 6: Output of svyby()Exercise 7: Bar plots with error Exercise 8: Survey-weighted histograms Exercise 9: Survey-weighted density plots Exercise 10: Inference for quantitative data Exercise 11: Survey-weighted t-test Exercise 12: Tying it all together!

To model survey data also requires careful consideration of how the data were collected. We will start our modeling chapter by learning how to incorporate survey weights into scatter plots through aesthetics such as size, color, and transparency. We'll model the survey data with linear regression and will explore how to incorporate categorical predictors and polynomial terms into our models.

Exercise 1: Visualization with scatter plots Exercise 2: Bubble plots Exercise 3: Survey-weighted scatter plots Exercise 4: Use of color in scatter plots Exercise 5: Visualizing trends Exercise 6: Line of best fit Exercise 7: Trend lines Exercise 8: Modeling survey data Exercise 9: Regression model Exercise 10: Regression inference Exercise 11: More complex modeling Exercise 12: Multiple linear regression Exercise 13: Tying it all together Exercise 14: Wrap-up

Current Exercise