Reverse Engineer Your Topic
The first step to find data and statistics is to clearly define your broad topic and why you need data or statistics.
- Why -- are you looking for these data/statistics?
- Do you need data to support or contradict a hypothesis/argument for a paper, assignment, or project?
- Do you need data to practice statistical analysis and/or methods?
One you have your topic in mind, you should try to identify your Unit of Analysis by asking:
- Who/What -- is the population you want to study?
- People, organizations, things, etc.
- When -- is the time period you want to study?
- Point in time, current information, historic information, longitudinal information
- Where -- is the geography or place you want to study?
- Political Boundary
- Nation, state, country, city, etc.
- Statistical/Census Geography
- Census tracts, block groups, zip codes, etc.
Determining your Unit of Analysis and answering the questions above will help you determine which search strategies will be most effective.
Sample Data Unit of Analysis
Below is an example of how to reverse engineer a sample topic for finding data to create a Unit of Analysis
Sample Topic: The impact of various ecological factors on butterfly communities
- Why: I'm looking for data that would allow me to compare a previously collected data set to my own data and study how the weather, ecosystem, and habitat affect butterfly communities in the Midwest
Unit of Analysis
- Who/What: Butterflies
- When: Recent/current data is preferable since butterfly communities may be impacted by various outside sources such as weather, climate, etc.
- Where: United States and hopefully from the Western region of the United States, but I would be open to data from other regional areas
To summarize the Unit of Analysis for this sample topic: I am looking for a recently collected data set about butterflies located in the United States (more specifically the Western region of the United States) to compare to my own data set which examined butterflies in the Midwestern region of the United States.
Sample Statistics Unit of Analysis
Below is an example of how to reverse engineer a sample topic for finding statistics to create a Unit of Analysis
Sample Topic: The percentage of U.S. population that have graduated high school by state
- Why: I'm trying to show that poverty levels are lower in states with a higher percentage of high school graduates to support my argument that more educational programs are needed in states with higher poverty.
Unit of Analysis
- Who/What: People that have graduated from high school
- When: Most current information is preferable
- Where: United States broken down by state
To summarize the Unit of Analysis for this sample topic: I am looking for recently collected statistics about the number of high school graduates located in each individual state of the United States to show that poverty levels are lower in states with a higher percentage of high school graduates to support my argument that more educational programs are needed in states with higher poverty.