[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

RE: st: RE: labels on bar graphs

From   David Radwin <>
To   <>
Subject   RE: st: RE: labels on bar graphs
Date   Mon, 30 Jun 2008 13:08:16 -0700

I agree that Nick's prediction is true for the population of Stata users and would-be users whose next best alternative would be another bona fide statistical package like, perhaps, R or SAS.

But for the population of users and would-be users whose next best alternative would be a basic spreadsheet/graphing program like Excel, where you can designate percentage formats in graphs and tables with the click of a button, this convenience and others (like reliably copying and pasting tables to a Word document) would greatly outweigh the marginal benefit of another arcane estimation method.

The former group is apparently the perceived market for Stata and probably the large majority of current users. But the latter group is orders of magnitude larger.

Attracting even a tiny fraction of Excel users would no doubt increase Stata's user base significantly. I am not advocating that Stata should try to compete head-to-head with Excel (heaven forbid!). But a few conveniences that benefit a large number of users and particularly prospective users seems like a better investment than adding highly specialized statistical estimators destined to be used only by a limited and highly sophisticated set of users.

David Radwin

At 7:40 PM +0100 6/30/08, Nick Cox wrote:

 In a crunch I imagine that most users would
prefer StataCorp to make the impossible possible, rather than make the
awkward easy.
*   For searches and help try:

© Copyright 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index