A database codebook is the documentation of all of the data entry rules and coding schemes used in a given database. And code books usually contain examples of every kind of variable and level of measurement you need to teach your students during the first two weeks of Intro Stats. You can use any code book from any database relevant to your own scholarship as an example in class. Or perhaps you can find a code book particularly relevant to the students or majors you are teaching.
Here, I will describe how to use Global Terrorism Database’s code book for this purpose. The Global Terrorism Database is housed at the University of Maryland and has been tracking national and international terrorism since 1970 and has collected information on over 170, 000 attacks. So, the database in and of itself could be useful in class. But, I will focus on just the code book for now, as I think this example cuts across disciplines and interests as all of our students are aware of terrorism and this particular code book doesn’t contain much technical jargon.
Here are just a few of the examples contained within it: Dichotomous, mutually exclusive response options ( Yes = 1, No = 0) can be found on page 14, in response to the question ”The violent act must be aimed at attaining a political, economic, religious, or social goal.”.
Another dichotomous response is used to indicate whether or not an attack was intended as a suicide attack (p. 26).
You can demonstrate nominal coding with the response options used to identify the country where the attack occurred (Brazil = 30, Cambodia = 36, etc., on p. 17, ) or by looking at the coding scheme for different kinds of terrorist attacks (3 = Bombing/Explosion, p. 22).
Ratio scale of measurement is used to enter the number of perpetrators for a given terrorist act (p. 44). You can also discuss what is lost or gained by reporting (and analyzing) the cost of damages in either categorical and ordinal (number represents one of four ranges of dollars) or qualitative and ratio (enter the amount of loss in USD) format (on pages 49 and 50).
This codebook and the actual database also draw attention to attempts to better understand big, scary life problems via systematic data collection and analysis. We need first responders and law enforcement officers and the bravery of regular citizens thrust into terrible situations in order to deal with terrorist events. We also need sharp, statistical minds to look for the patterns in these attacks in an attempt to prevent future attacks.
For a more advanced statistics class, you can also point out the naming conventions used in this database, and how naming conventions are good practice and need to be sorted out prior to data collection.
Here, I will describe how to use Global Terrorism Database’s code book for this purpose. The Global Terrorism Database is housed at the University of Maryland and has been tracking national and international terrorism since 1970 and has collected information on over 170, 000 attacks. So, the database in and of itself could be useful in class. But, I will focus on just the code book for now, as I think this example cuts across disciplines and interests as all of our students are aware of terrorism and this particular code book doesn’t contain much technical jargon.
Here are just a few of the examples contained within it: Dichotomous, mutually exclusive response options ( Yes = 1, No = 0) can be found on page 14, in response to the question ”The violent act must be aimed at attaining a political, economic, religious, or social goal.”.
Another dichotomous response is used to indicate whether or not an attack was intended as a suicide attack (p. 26).
You can demonstrate nominal coding with the response options used to identify the country where the attack occurred (Brazil = 30, Cambodia = 36, etc., on p. 17, ) or by looking at the coding scheme for different kinds of terrorist attacks (3 = Bombing/Explosion, p. 22).
Ratio scale of measurement is used to enter the number of perpetrators for a given terrorist act (p. 44). You can also discuss what is lost or gained by reporting (and analyzing) the cost of damages in either categorical and ordinal (number represents one of four ranges of dollars) or qualitative and ratio (enter the amount of loss in USD) format (on pages 49 and 50).
This codebook and the actual database also draw attention to attempts to better understand big, scary life problems via systematic data collection and analysis. We need first responders and law enforcement officers and the bravery of regular citizens thrust into terrible situations in order to deal with terrorist events. We also need sharp, statistical minds to look for the patterns in these attacks in an attempt to prevent future attacks.
For a more advanced statistics class, you can also point out the naming conventions used in this database, and how naming conventions are good practice and need to be sorted out prior to data collection.
Comments
Post a Comment