book

• Studenst for Urban Data Systems(SUDS) clup at Carnegie Mellon Univesrity's Heinz College
• Columbia Data Science Program
• Western PA Regional Data Center
• wikipedia tech portal

listStations

This station set offers exposure-focused activities and challenges which each relate to some facet of the emerging field of Data Science.

Station 1: People in data tables

1. Get to know one another by creating a data schema for data about the individuals in our class. Be sure to delimit the data types that each column can contain. Your choices for data types depend on the language or package you're using. For the purposes of this exercise, use whatever language you're most familiar with.

---

Station 2: Trends and predictions

Many businesses and organizations must make decisions based on predictions of conditions in the future. Anticipating cost changes over time of various inputs for a system influences planning efforts economy wide.

1. Download this spreadsheet of historic data of college prices. Review the source and vet it for reliabilty.
2. If a foundation is planning scholarship investments, how many more times expensive will private, four-year institutions be compared to public, two-year institutions 15 years from now?
3. State a level of confidence for your estimate. Qualify your estimate based on outside factors whose changes may impact your recommendations.
4. Print out or provide any backup info to justify your assertion.

Station 3: Domain knowledge

A chunk of data analysis competencies are rooted in command of field- or subject-specific knowledge. For example, Predicting changes in bus ridership due to economic shifts are made best by those who know transportation the best.

1. Visit this site exploring data in crime prediction algorithms. Summarize the findings from this study. Take moment to discuss with your group what data science has to contribute to this discussion.
2. Compare this data-related study to this one, by Mark Egge, who analyzed bus bunching patterns in Pittsburgh. What did Mark's study find?
3. How are the applications of data science different in these two studies? What standards of data reliability or integrity exist? How are they different?

Station 4: Gapminder Web!

In this station, you'll have the chance to explore a data visualization tool that has shaped the development of many online data visualization tools.

1. Visit the GapMinder web app. Devote about 5 minutes to just exploring the interface, tinkering with trends through time, etc.
2. The default chart shows income versus life expectancy by country. View the timeline video since 1800 with your team. Make a list of the factors that contribute to the major shifts and jumps seen by various country groups in this visualization.
3. Prioritize with your group the three factors which make this tool one of the most successful visualization endeavors of all of the internet.
4. Visualization babies per woman versus Income on the X axis. Discuss with your group the trends since 1800. What outliners exist? Why do they exist? How could the data be misinterpreted?
5. Explore the several other visualization modes other than "bubbles" such as "maps" and "trends". Which mode is the least intuitive for the lay user? Most intuitive?

Station 5: Final project review

1. All of the pilot course's final projects are uploaded to our shared directory. Find this shared directory by navigating back to technologyrediscovery.net >> DAT-102 >> SP18 Final Project Shared Directory
2. Some students chose more ambitious final projects than others. Dedicate a few minutes to reviewing at least 3 of the projects. Create a Plus-Delta table for the projects with your group: What did you think the students did well, overall? What areas of improvement could be made in their efforts?
3. Discuss with your group any burning ideas you have for final projects in DAT-102!

straightenNote card diagnostic

expand_moretoggle card:

playlist_add_checkMid-week ToDOs

sendProducts

cakeExtension exercises

bookResources

• Ambrsoft base converter and other goodies
• Google earth for the browser
• Graph theory on wolfram alpha
• Stanford repo of network data set
• Little Sis network data sets

listLesson sequence

1. Tree of data structures
2. Explore computer and human language relationship and data conversions with Matthew McConaughey
3. Encoding and decoding location data with computer conversion tools. Move from various bases with the Ambrsoft base converter
4. Break
5. Creating tree-type data structures and transferring that representation tabularly, and back to trees
6. Creating network type data sets and transferring that data to and from graphs
7. Representing and describing network data from LittleSis
8. Representing and describing network data from wikipedia

straightenNote card diagnostic

expand_moretoggle card:

playlist_add_checkMid-week ToDOs

Please create a model data set of both tree and graph structures relating to a possible topic/domain of your interest. Prepare to share those data representations next week with your peers.

bookResources

• Technology rediscovery spreadsheet module 1

listLesson sequence

1. Creating hierarchical data and informal versions of organizational charts
2. Developing a data set profile: Title, Synopsis, source, permissions, questions
3. Types of questions: Descriptive (basic), deeply analytic, and predictive/inferential
4. Crafting a data profile and brainstorming questions

playlist_add_checkMid-week ToDOs

1. Flesh out your data profile and secure one or more spreadsheet/tabular data set related to your field of interest. Bring a digital copy of this data with you to class next week.
2. Dust off your spreadsheet skills by creating a basic budget using the Technology rediscovery spreadsheet module 1 as a guide
3. Review the concept of mean, median, mode, variance, and standard deviation
4. Practice using the spreadsheet functions in your chosen tool for computing the previous metrics
5. Be prepared for a little "spreadsheet quiz" like thing next class.
6. Read about histograms (frequency distributions) in your chosen statistics textbook

cakeExtension exercises

For those already familiar with spreadsheet, please run the essential descriptive statistics on your variables of interest using the table you have already gathered.

bookResources

• Inquiry flowchart (editable) and PDF version

listLesson sequence

1. Investigate the application of statistical reasoning and modern aviation: View clip from Flight with Denzel Washington
2. With modern aviation serving as a gold standard of using statistics and rigorous investigation to promote transportation safety (According to the NTSB): Investigate population definitions
3. Dutch Elm Examples

listPrimary source listing

Number	In-text	Full citation	Access link & info
0	Chervany 1980	Chervany, N. L., Anderson, J. C., Benson, P. G., & Hill, A. V. (1980). A Management Science Approach to a Dutch Elm Disease Sanitation Program. Interfaces, 10(2), 108–114. Retrieved from https://ezproxy.ccac.edu/login?url=https://search.ebscohost.com/login.aspx?direct=true&db=bth&AN=6692483&site=eds-live	Access through CCAC libraries
1	Kozak 2015	Kozak, M., Krzanowski, W., Cichocka, I., & Hartley, J. (2015). The effects of data input errors on subsequent statistical inference. Journal of Applied Statistics, 42(9), 2030–2037. https://doi-org.ezproxy.ccac.edu/10.1080/02664763.2015.1016410	Access through CCAC libraries
2	Bille 2013	Billé, R., Kelly, R., Biastoch, A., Harrould-Kolieb, E., Herr, D., Joos, F., … Gattuso, J.-P. (2013). Taking Action Against Ocean Acidification: A Review of Management and Policy Options. Environmental Management, 52(4), 761–779. https://doi-org.ezproxy.ccac.edu/10.1007/s00267-013-0132-7	Access through CCAC libraries
3	Strayer 2013	David L. Strayer, Joel M. Cooper, Jonna Turrill, James Coleman, Nate MedeirosWard, and Francesco Biondi (University of Utah). Measuring Cognitive Distraction in the Automobile (June 2013). AAA Foundation for Traffic Safety	Free online PDF via AAA Foundation
4	Sugianto	Arif Sugianto, Reza Jaya Wardhana, Nanang Yulian, I Gede Kusuma Jaya Wardana, Muchtar Karokaro4,Hariyati Purwaningsih. Failure Analysis of a First Stage High Pressure Turbine Blade in an Aero Engine Turbine on PK-GSG Boeing B747-400. Undergraduate Paper. ITS	Free online PDF via its.ac.id. Also see NTSB full report on AA383 engine failure
5	Maecker 2016	Maecker, O., Barrot, C., & Becker, J. (2016). The effect of social media interactions on customer relationship management. Business Research, 9(1), 133–155. https://doi-org.ezproxy.ccac.edu/10.1007/s40685-016-0027-6	Access through CCAC libraries

playlist_add_checkMid-week ToDO: Assignment 1

1. With the help of a librarian or data expert, select an academic article which provides a statistical analysis of an area of interest with respect to your projects.
2. Print the article.
3. Digest the abstract/summary of the article. Read the first few sentences of each section. Read the conclusion. Review the flow chart. Read it more carefully again. As you read, annotate the paper for the following:
   • Managerial concerns
   • Managerial questions
   • Statistical questions
   • Statistical analysis techniques
   • Statistical claims
   • Managerial answers
   • Managerial implementation
4. Recreate your own version of the inquiry cycle as discussed in class either by hand, on a computer, or both. Create a reader-friendly digestion of each step in the cycle drawn from your article or inferred using available resources.
5. Locate an artifact related to one of the stages in your study and prepare to discuss it and your article with your peers next week at the beginning of class next week. Prepare by brainstorming 2-3 discussion questions related to our course content. Examples of artifacts include:
• An accompanying agency report related to the statistical data somehow
• A news story reporting on either the statistics themselves or an event about which the data relate.
• A summary of a personal experience or condition of your own
• Notes from a discussion of a subject matter expert you know or found in the area
• Notes from background research you did on the subject
6. Package your work for sharing by creating the following files:
• A word processing document citing your academic article and artifact with links to Internet versions if possible. Please make note of access restrictions (i.e. you must log in to MyCCAC to get full PDF, etc.)
• A digital copy, photographed, or scanned version of your inquiry process diagram based on your article

sendFinding and citing academic journal articles at CCAC

Step 1: Naviate to the CCAC libraries home page

Step 2: Use OneSearch

Step 3: Login

Step 4: Click on the article title to pull up the full listing

Step 5: Access the full PDF, grab the permalink

sendReference Diagrams

Download the PDF of these two images for references:

Classification tree of sources of statistical data

Sources of error in statistical inquiry

bookResources

• Ida Mae Darsow Interest Inventory Image
• Stanley Migram's 1963 paper on obedience

straightenNote card diagnostic

expand_moretoggle card:

listLesson sequence

1. Note card diagnostic: Ida Mae Darsow's professional skills inventory
2. Round-robin sharing of articles and artifacts from last week
3. Data tree classification introduction: Tree structure and leaf categories
4. Classifying sample articles from last week
5. Work time: Classify your article, double check inquiry cycle extractions with peers, and build feature article banner
6. Break
7. Introduction to error in statistical studies: Stanley Milgram's experiments (Youtube link)
8. Exploration of two key factors to defend generalization of results: 1) "large N" and 2) random selection from population of interest
9. Rating potential error sources in existing articles
10. Identifying and rating error sources in chosen article
11. Graphically depicting relative weightings of error sources by article

playlist_add_checkMid-week TODOs

This week's TODOs invite your to consider the generalizability of two or more journal articles assembled for last class in addition to your selected journal article related to your topic.

Step 1-Setup: Create your data gathering tool using the following empty table as a guide

Article citation	Population of interest	Large N?	Randomization?	Assessment of generalizability
Article X from last week
Article Y from last week
Your own article

Step 2 - Learn about Central Limit Theorem: Dig out the statistics text your bought for this class. Look up the Central Limit Theorem (used in calculus) and read about the criteria required for making statistical inference based on sample data.

Step 3 - Classify: Dig into your chosen article and two articles from the previous week to assess for random draws from the population of interest and sufficiently large n.

sendProducts

A thoughtfully completed data table from the midweek TODO assessment of 3 journal articles

cakeExtension exercises