104 PART 3 Getting Down and Dirty with Data

Assigning participant study

identification (ID) numbers

Every participant in your study should have a unique participant study identifier

(typically called a study ID). The study ID is present in the participant’s data and

is used for identifying the participant on study materials (for example, laboratory

specimens sent for analysis). You may need to combine two variables to create a

unique identifier. In a single-site study that is carried out at only one geographical

location, the study ID can be a whole number that is two- to four-digits long. It

doesn’t have to start at 1; it can start at 100 if you want all the ID numbers to be

three-digits long without leading zeros. In multi-site studies that are carried out at

several locations (such as different clinics or labs), the number often follows some

logic. For example, it could have two parts, such as a site number and a local study

ID number separated by a hyphen (for example, 03-104), which is where you need

two variables to get a unique ID.

Organizing name and address data

in the study ID crosswalk

A research database should not include private identifying information for the

participant, such as the participant’s full name and home address. Yet, these data

need to be accessible to study staff to facilitate the research. Private data like this

is typically stored in a spreadsheet called a study ID crosswalk. This spreadsheet

keeps a link (or crosswalk) between the participant’s study ID and their private

data not to be stored in the research database. When you store names in the study

ID crosswalk, choose one of the following formats so that you can easily sort par-

ticipants into alphabetical order, or use the spreadsheet to facilitate study

mailings:»

» ^{A single variable: Last, First Middle (like Smith, John A)»}

» ^{Two columns: One for Last, another for First and Middle}

You may also want to include separate fields to hold prefixes (Mr., Mrs., Dr., and

so on) and suffixes (Jr., III, PhD, and so forth).

Addresses should be stored in separate fields for street, city, state (or province),

ZIP code (or comparable postal code).