John Qualls is Senior Economic Advisor, National Center for Financial and Economic Information, Ministry of Finance and National Economy, Riyadh, Kingdom of Saudi Arabia

He can be reached at jhqualls@yahoo.com

Back to the PC Corner/ Window on the Web page

PC Corner

by John Qualls
February 2001

My apologies for the long delay between columns. Frankly, I miss being hounded by Ed Mennis for my contribution each quarter, and it's far too easy to let things slide. One of my New Millennium resolutions earlier this year was to do a better job in putting out my column on a timely basis.

As always, I appreciate your feedback and ideas for new columns. My e-mail address is somewhere on the right side of this page, the exact placement depending upon the whims of the HTML gods. Given that the gods are not always pleased, here it is again:jhqualls@yahoo.com.

Results of NABE PC Hardware/Software Survey

This column reports on the results of the NABE PC hardware/software survey that appeared on the NABE web page back last spring. Unfortunately, the response was not overwhelming - only 51 members filled out the on-line form. This may have been due to the new on-line methodology used - in retrospect, we should have given you the chance to print out the form and mail or fax it to us. An e-mail options might also have helped. In addition, there were some difficulties with the wording of the questionnaire and the options offered to the person filling out the form. My apologies for not catching these before we put it out on the web page.

Because of the small sample, you should use caution in drawing any hard and fast conclusions from the data presented below. In addition, the survey was done last spring and summer, so it is somewhat out of date. However, it is an interesting snapshot of the NABE membership's computer usage. The questionnaire on our web page gathered statistics for both a work computer and a home computer. Most responders had both, but a surprising number of responders only reported on their work computer (37 out of 51). I seriously doubt if this was due to the lack of a computer at home; rather, it probably reflected a busy schedule and/or a lack of information about the home machine when they filled out the questionnaire.

A quick summary of the "average" work computer for our survey sample showed the following hardware configuration:

• a Dell computer with an Intel Pentium II chip running at 350 mh
• 64 megs of RAM and a 6.5 gig hard drive
• a HP laser printer, probably hooked up to a local area network (unfortunately, we didn't specifically ask if the printer was hooked up directly)
• for those without a LAN Internet connection, a 56K modem, and
• a monitor (size unspecified - my fault, I forgot to ask the question) running at 640x480 resolution (hard to believe, but more about this later).
  

The software running on our "average" work computer was as follows:

• a Windows 95 operating system (a surprising and questionable result, as will be seen),
• MS Office as the office suite software,
• Microsoft's Internet Explorer 5.x as the web browser, and
• EViews as the econometric package. However, 47 percent of respondents did not specify any econometric package - the highest percent of any "PC Corner" survey. This is a very interesting fact, one that reflects on the changing composition of our membership and their job functions.

The "average" home computer was identical to the work computer, with the following exceptions:

• the type of Pentium chip was typically unspecified, and the average speed was 333 mh
• the hard drive capacity was greater, at 9.7 gigs (probably reflecting the lack of a LAN file server at home)
• the printer was more likely to be an ink-jet, with HP as the preferred make, and
• Windows 98 was the preferred operating system (a much more believable result), and
• the vast majority of home computers (2/3) have no econometric package at all, again a reflection of the changing interests of our NABE members.

Following are the tables with the detailed results and some commentary

Dell was far and away the most popular brand mentioned in our survey. One surprise was the large number of custom computers at home. Could it be that our membership is more interested in assembling their own machines from off-the-shelf components than they are in econometrics?

The above table is somewhat misleading, in that the Pentium (unspecified) category undoubtedly included quite a few Pentium II and Pentium III models, a conclusion based on the chip speeds that were included. It is also interesting to note the greater popularity of AMD and Intel Celeron chips in home machines - undoubtedly reflecting the lower price of machines using these chips.

I must admit that I was somewhat surprised at the lack of the faster chips in these survey results. Quite a few of the survey respondents were using computers with relatively slow chip speeds of 300 mh or less, both at work and at home. In contrast, in our previous PC surveys, our members were using "state of the art" machines. To me, this indicates a shift away from the "number-crunching" nature of what we do - again, a clue to the changing nature of our jobs.

No surprises here. Most of the respondents were able to get by with 64 meg of memory, although almost as many work computers had 128 meg of memory.

At first, I was somewhat surprised to see the popularity of larger disks on the home computers. However, our responses also showed that 78 percent of the work computers were connected to a local area network. Given the large amount of storage on the LAN file servers, this probably lessened the need for large hard drives on the work machines.

It is not surprising that the overwhelming majority of work computers are connected to a laser printer. Unfortunately, I did not think to ask if the connection was via a LAN or a dedicated cable. Somewhat more surprising was the relative popularity of laser printers used with home computers, given the large cost differential.

Some things never change with our NABE members, and one of these is the continuing popularity of Hewlett-Packard printers, both laser and ink jet. Over 80 percent of the printers at work and 60 percent of the home printers were HP.

I was not surprised at the popularity of Windows NT in the work environment. However, I was astounded at the continued predominance of Windows 95 at work. I have reason to believe that we may have had a problem with a misplaced "dot" in the Windows 95 circle as the default choice on the questionnaire. Thus, respondents may have left it by accident.

This conclusion is buttressed by data collected from our NABE web server. Below is the frequency breakdown by type of operating system for those visiting our pages during the January 7 - February 22, 2001 period. Of course, this includes a large number of non-members accessing the public pages, but I would know of no reason why their operating system percentages should differ dramatically.

Note that Windows 95 usage is only around half that of Windows 98. Of course, this is data that is six months more current, but I still suspect that our work computer operating system responses were not accurate.

Microsoft Office dominates the workplace, and our survey reflected this dominance among NABE members. I was somewhat negligent in not specifying Microsoft Works as one of the "suites", as it is quite popular on many of the low-end computers. Typically, Works is combined with MS Word 97 or 2000, and several of our respondents may have figured that this is close enough to MS Office to qualify. In addition, there is some evidence that MS Office was the "default" answer if the respondents didn't bother to check another circle. However, there can be no doubt that MS Office dominates, at least in the office setting.

EViews continued to be the most popular econometric package among our respondents, but it appears to be hard-pressed by SAS in the workplace. However, the biggest story in these statistics was the large numbers of respondents without any econometric package. Almost half of the users at work and 2/3 of the home users had no econometric program on their machines.

Microsoft's Internet Explorer dominated as our respondents' favorite web browser, as it was on 65 percent of both the work and home machines. This figure is somewhat lower than the breakdown of the browser types used to access the NABE web site, as can be seen below.

The higher percent of Internet Explorer browsers using the site can be explained by the more current data in the NABE web site report. Netscape has been steadily losing market share, and this trend is evidently continuing. Another possibility might be that our respondents liked to "root for the underdog" and reported Netscape as their browser of choice when in actuality they used IE more of the time.

The above table on internet connection speed only includes the work responses from those who indicated that they did not use a LAN to access the Internet. I was surprised by two things in the table. One is the large number of people still using 28.8K (or less) modems. I suspect that 28.8K was the default answer in the questionnaire, which probably biased this statistics in an upward direction. The second surprise was that so few respondents were using cable modems and none were using DSL lines from home to access the Internet. However, I do want to know how several of the respondents are able to afford a T1 line from their home! Bottom line - I am very suspicious of the statistics in this table.

The final table is on the screen resolution being used by the respondent. I have a real problem with the fact that half of the respondents report using the lowest setting possible on their monitors - 640x480. Again, this was apparently the default in the questionnaire, and I fear that many people simply left it at that, since they might not know how to check and see what their monitor setting actually is.

Given this, I thought that I would end this "PC Corner" column with a suggestion that you try experimenting with your monitor resolution to see if a different setting might be better for you. The old 640x480 setting was intended for the original 13" IBM monitors, which were really horrible and are mostly out of service now. An 800x600 setting is the minimum acceptable setting for many Windows applications, in terms of being able to see all of the screen area without a lot of scrolling around. It gives the best results on 15" monitors. I have a 17" monitor and prefer the 1024x768 setting. For a 19" or 21" monitor, try the 1280x1024 setting (which I neglected to include on the survey).

To change your screen resolution, click on "Settings" in the start menu. Click on "Control Panel" and then on "Display". Click on the Settings tab in the Display Properties dialog box. In the lower right corner of the box, there is a Screen Area setting that can be changed by clicking on and moving the sliding lever. The actual settings that are available will vary from computer to computer - my computer has four settings - 640x480, 800x600, 1024x768, and 1280x1024. Try a different setting and see how it looks. Windows will give you a sample of how it will look and ask you if you want to keep the new setting. If you click on "No", it will restore the old setting.

I would encourage you to try different settings. It's easy to restore the old setting if you don't like the new one. Of course, there is a trade-off - the numbers, characters, and desktop icons are smaller with the higher resolutions, but you can see so much more information on the screen. Even with my presbyoptic eyes, I really do like the higher resolutions, although I have to use the 800x600 setting on the 15" monitor at work because of the small screen.

If you have any questions or comments about the survey, don't hesitate to write me at jhqualls@yahoo.com.

Webmasters Note-
The reason for the odd date range is that we switched servers on 1/7, and that server's stats program tracked the OS. It wasn't tracked on our old server.