The program can handle two different kinds of ego network data, which I label row-wise and column-wise. Two sample files are included with the
program.
Row-Wise Format
In the row-wise format, the data are recorded as two grids or matrices. First matrix gives the attributes of each ego in the study. So it is a matrix
in which rows are egos and columns are attributes (e.g., age sex etc.).
This is followed by a second matrix that indicates who each ego has ties with, what kind of ties these are, and attributes of these alters (from
ego's point of view). So the rows are ego-alter pairs, and the columns are variables that describe ego's perception of the alter (e.g., age,
occupation etc.) and/or ego's relation to the alter (e.g., friend, lover, coworker, etc).
At present, E-NET accepts row-wise data only in the form of a VNA file, which is a text file that looks like this:
*ego data
ID age sex
01 32 male
02 67 female
03 ….
*alter data
From To Friends lovers Age
01 1-1 1 0 15
01 1-2 1 1 30
02 2-1 0 1 50
…
*Alter-alter data
From to knows
1-1 1-2 1
…
Column-Wise Format
In the column-wise format, all the data are in a single matrix in which rows are egos and columns are variables. The variables include both ego
attributes, and relational information. For example, the survey might have allowed for up to 5 alters for each ego. So the data might look
something like this:
EGOID EGOAGE EGOSEX Alter1YrsKnown Alter1Age Alter2YrsKnown Alter2Age Alter3YrsKnwon …
001 52 male 4 45 1 89 20
At present, the program assumes that column-wise data are held in an Excel file.