Example Final Memos

I’ve posted two memos from a previous class that were very strong and received As. One is longer than the other, although the shorter one was accompanied by more data analysis. But both do an excellent job of describing how the author found a theme or central question in the data and what kind of story that exploration might generate.

First Example
Second Example

Links for August 18

PDFs
Peace Corps College Rankings PDF, Search results

xPDF
FOIA Log
Bad PDF!

Mapping:

• GeoData.gov
• Geodata at Data.gov
• RouteMap
• Election maps

Geocoding

• Geocoder.us
• Using Google Spreadsheets
• Geocoding with Fusion Tables

Aug. 4 Homework Answers

1. SELECT COUNT(id) FROM senate_votes WHERE gop_yes=0 and dem_no=0 and gop_not_voting=0 and dem_not_voting=0
2. SELECT last_name, first_name, COUNT(senate_positions.position) FROM senate_votes JOIN senate_positions ON senate_votes.id=senate_positions.senate_vote_id JOIN members ON senate_positions.member_id = members.id WHERE party ='D' AND position = 'No' GROUP BY last_name, first_name ORDER BY COUNT(senate_positions.position) DESC
3. SELECT last_name, first_name, COUNT(members.id) FROM senate_votes JOIN senate_positions ON senate_votes.id=senate_positions.senate_vote_id JOIN members ON senate_positions.member_id = members.id WHERE (members.party)="R" AND senate_positions.position=senate_votes.gop_majority_position GROUP BY last_name, first_name ORDER BY COUNT(senate_positions.position) DESC
4. SELECT last_name, first_name, COUNT(members.id) FROM senate_votes JOIN senate_positions ON senate_votes.id=senate_positions.senate_vote_id JOIN members ON senate_positions.member_id = members.id WHERE position ='Not Voting' GROUP BY 1, 2 ORDER BY 3 DESC
5. SELECT COUNT(id) FROM senate_votes WHERE total_no=0
6. SELECT last_name, first_name, COUNT(senate_positions.position) FROM senate_votes JOIN senate_positions ON senate_votes.id=senate_positions.senate_vote_id JOIN members ON senate_positions.member_id = members.id WHERE senate_votes.total_no=1 and senate_positions.position='No' GROUP BY last_name, first_name ORDER BY 3 desc

Assignments for August 11

Review BankTracker before class.

Story Memos: Send me your logs and any accompanying data. You should have a solid understanding of what your memo will say by this point.

Critique: Milwaukee Journal-Sentinel story on Recovery Act spending on bridge repair. Write at least 500 words on this piece (and its accompanying graphics), being sure to mention how you would have approached the story and what you might have done differently or what questions arose in your mind that weren’t answered.

More Data:

Using data posted by the Nuclear Regulatory Commission here (be sure to check the Data Dictionary which explains what the fields are), download and import into SQLite the “Power Status (raw data text file) for the last 365 Days” file. Give it headers based on the data dictionary. Then briefly describe what question(s) you could ask of this data as a journalist? What kind of stories could you use it for?

Assignments for August 4

Browse the Post’s Congressional Vote Database before class.

Project:

You should have narrowed down a question to answer or explore for your topic area, and be gathering and interviewing your data. Start keeping a log of your actions – it can be a Word file, or blog posts, whatever – and send me the log.

SQLite:

Using the imported vote data, write SQL queries to answer the following questions. They may require 1, 2 or 3 tables to answer, so please look at the fields/data carefully to see what you have. If you’re not sure what a field name means, let me know.

1. How many votes have all Republicans voting “No” and all Democrats voting “Yes”?
2. Which Democrat has voted “No” most often?
3. Which Republican has agreed with the republican majority position the most?
4. Which senator has missed the most votes (the position is “Not Voting”)?
5. How many votes have received 0 “No” votes?
6. Which senators have been the only person to vote “No” on a vote, and how many times have they done it?

Send both SQLite files, plus your SQL queries, to me by email.

Reading: How the Bristol (VA) Herald Courier won a Pulitzer using databases.

Links for July 28

Two Table Joins

A Visual Explanation of SQL Joins

Assignments for July 28

SQLite: Continuing with the senate voting database, write SQL to answer the following questions. You will need to study the fields in your tables. Send me both your queries and your sqlite data file.

1. How many votes were decided by less than 10 votes?
2. Show each vote result and how often it occurred.
3. How many votes have all Republicans voting “No” and all Democrats voting “Yes”? (hint: all “No” votes means how many “Yes” votes?)
4. How many votes on cloture motions were agreed to?
5. What is the average number of Republican No votes on a vote On Passage of the Bill?

Find Your Topic/Data: Please take some time this week to identify a topic for your final project (it may already be in data form, or not) that you’d like to work with over the next few weeks. Preferably it should be government-related, but if you can make a good case for something else, I’ll entertain it (give me a heads-up before class if this is the case). If you’re searching for possible ideas, try the Census Bureau or FedStats or Data.gov.

Reading: Part 3 of the SQLite Tutorial. Also, please read my friend Matt Waite, creator of Politifact.com, on data ghettos. And this USA Today story on subsidies for college athletics.

Links for July 21

Senate Voting Data:

Members
Senate Votes
Vote Positions

State-to-State Migration
Google Fusion Tables

Assignments for July 21

Complete Part 2 of the SQLite tutorial. It’s long; start early.

Create a new database called fec (the .sqlite gets added on when you create it). Download this CSV file of contributors to Wisconsin Rep. Tammy Baldwin during the second quarter of 2011. Import it into the fec database, creating a table called baldwin. Look at the CSV file in Excel before you define the fields in SQLite – be sure to define the zip field as VARCHAR, not INTEGER. Just define the date field as VARCHAR for now.

Once you’ve done that, write queries to do the following:

Show the contributors from Hawaii
Show contributors from Philadelphia who gave at least $1,000
Show contributors who listed their occupation as CEO, with the largest amount first

Reading: This Tulsa World story on lottery ticket sales.
Send me both the fec.sqlite file and the SQL from your queries (I can check the results).

Links for July 14

SQLite
Firefox SQLite Manager
SQLite Tutorial, Part I

FEC Data catalog
Senate Votes sqlite