Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: RE: Creating a frequency distribution from matrix

From   Nick Cox <>
To   "''" <>
Subject   st: RE: Creating a frequency distribution from matrix
Date   Wed, 9 Nov 2011 17:47:02 +0000

It is a good idea to use Stata terminology in a Stata forum. As I understand it, you are not using a Stata matrix, but a set of variables (which you call columns) and a set of observations (which you call rows). 

Use -tabm- from -tab_chi- on SSC to look at the combined distribution of two or more comparable variables. 


Coffey,Daniel J

I am working on a dataset of internet user comments to online newspaper articles. In one database,  I have arranged the dataset such that each row represents a specific newspaper article and the columns represent coding of individual user comments. Each comment is coded on a 1-5 scale, so the cells all represent observations on the same coding scale. There are 200 cases (newspaper articles) but a varying number of comments per article (some stories less than 5 comments, while others have 500 or more comments). 

The question I have is simple but something I will need for future analysis of the dataset. That is, is there a way to quickly create a frequency distribution for the whole dataset? The columns aren't really variables, but just a way to track comments efficiently. So, I want STATA to treat all of the cells as one single variable without having to manually copy each row into a single column. 

*   For searches and help try:

© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index