"There are a lot of small data problems that occur in big data. They don't disappear because you've got lots of stuff. They get worse." is a quote by British biostatistician David J. Spiegelhalter (1953 - ). The quote may be found in a March 28, 2014 article in the Financial Times written by Tim Hartford entitled "Big data: are we making a big mistake?"
A joke to be used in teaching about the use of randomization in experiments or about the Pearson correlation coefficient. The idea for the joke came from Lawrence Mark Lesser of The University of Texas at El Paso in 2012.
This is a chapter on data wrangling excerpted from a book on data science. The book is “Modern Data Science with R,” and the authors are Benjamin J. Baumer, Daniel T. Kaplan, and Nicholas J. Horton. It contains the R code needed to do basic things with data such as sorting, arranging, and summarizing data.
This is a chapter on ethics excerpted from a book on data science. The book is “Modern Data Science with R,” and the authors are Benjamin J. Baumer, Daniel T. Kaplan, and Nicholas J. Horton. The chapter presents several ethical dilemmas, then a framework to use when evaluating ethical issues. Then it discusses the dilemmas again, now resolving them.
This site is a lesson on using SQL. It starts with a simple SELECT query. The user must type in the correct command to select certain columns from a database. Once the user has completed the first lesson, then he or she may continue to more complicated lessons.
This site is a government-run repository of information on current and completed clinical trials. Users can search for clinical trials by disease type and also by whether the trial is currently recruiting. Then a detailed description of the trial is given. This can be used in a classroom setting to discuss design issues and ethical issues with clinical trials.
This website is a summary of a randomized controlled trial of a metropolitan police department's body-worn camera program. It is useful in class to talk about the design of the experiment and also to talk about how they state their results. Their results are given as confidence intervals for differences.
A song that can be used in discussing lurking variables - unobserved variables that may drive the relationships seen in the data. The lyrics were written by Mary McLellan from Aledo High School in Aledo, Texas as one of several dozen songs created for her AP statistics course. The song may be sung to the tune of the 2013 song “Happy,” written by Pharrell Williams for the animated movie Despicable Me 2. Also, an accompanying video may be found at
https://www.youtube.com/watch?v=iWf-8_UjUyg
A song that may be used in discussing how confounding variables may provide alternate explanations for the data making causal interpretations difficult. The lyrics were written by Mary McLellan from Aledo High School in Aledo, Texas as one of several dozen songs created for her AP statistics course. The song may be sung to the tune of the mid-20th century folk song 99 Bottles of Beer. Also, an accompanying video may be found at
https://www.youtube.com/watch?v=-daUPdUV8C4
A song that may be used in discussing the difference between cluster sampling and stratified sampling and the value of them when you have groups known to be homogeneous for the variable under study. The lyrics were written by Mary McLellan from Aledo High School in Aledo, Texas as one of several dozen songs created for her AP statistics course. The song may be sung to the tune of Mike Chapman and Nicky Chin’s 1979 song "Kitty", popularized by Toni Basil in her 1981 recording of the song as "Mickey". Also, an accompanying video may be found at https://www.youtube.com/watch?v=rX9erSAH9WY