Accessing City-County Data Book via DBASE HI: Census CD-ROMs from the Ground Up Fred Gey Data Archivist UC Data Archive & Technical Assistance (UC DATA) (formerly State Data Program) University of California Berkeley, CA 94720 Presented to the 1990 Conference of the International Association of Social Science Information Service and Technology (IASSIST90), Poughkeepsie, New York, May 30 to June 2, 1990. The Bureau of the Census has published the 1988 City-County Data Book data files on a CD-ROM disk. Physically the data are organized as DBASE-HI database files and the Bureau has supplied a computer program to display profiles of particular areas. However, accessing the data for analysis purposes (such as finding the ethnicity of American counties) can only be done by directly using DBASE-UI, and doing so is a multi-step process. This paper describes how to use DBASE-UI direcdy on CCDB88 to select subsets of data for statistical analysis, and compares and constrasts the CCDB88 structure with the 1982 Census of Agriculture files on Test Disk 2. Some suggestions are made for how the Bureau might organize CD-ROM data and access software to better facilitate individual access and prepare for the 1990 Census. Fall/Winter 1990 Accessing City-County Data Book with DBASE III 1. Introduction In the 1990s the Bureau of the Census will utilize the CD-ROM as a major publishing medium for current and future census data. This process has been underway for several years and the Bureau has been through several iterations of test disks before settling on a de-facto standard for distribution file formats of the DBASE-HI database system by Ashton-Tate Corporation. While some measure of standardization is imposed by this decision by the Bureau, other standards and capabilities need to be added for the convenience of public data users of census data. It is the purpose of this paper to discuss the kinds of tasks which might be commonplace by data users and the effort required to accomplish them, as well as to draw some conclusions as to the structure the Bureau might impose to ease the burden of accessing this data. 2. What is CD-ROM? CD-ROM is a new data storage medium based on audio compact disk technology. Each cd- rom disk holds about 650 megabytes of data (about the equivalent of 4 computer magnetic tapes as we know them on the IBM mainframe). Since the production process is identical to that of compact disks, the cost of production is about the same (less than $5 per disk). CD-ROM can be conceived of as an extremely large and slow disk that you can't write on. It operates like a cross between a hard disk (because it's so big) and a floppy (because the disks are removable). On the IBM-AT machine in our archive, we have attached a CD-ROM drive from Denon Corp, and designated it to be drive F: Thus if we have the City-County Data Book disk inserted in the drive we can do a directory command to see what files are on the drive: D:\FRFD> dir f: *.dbf/w Directory listing Vo 1me in d r i ve F ii CCDB 1988 of database files Directory of F:\ on CCDB CD- CIF01 DBF CIF01DCT DBF CIF02 DBF CIF02DCT DBF CIF03 DBF ROM CIF03DCT DBF CIF04 DBF CIF04DCT DBF CIF05 DBF CIF05DCT DBF CIF06 DBF CIF06DCT DBF CIF07 DBF CIF07DCT DBF CIF08 DBF CIF08DCT DBF O0F0 1 DBF COF01DCT DBF COF02 DBF COF02DCT DBF COF03 DBF CDF03DCT DBF COF04 DBF COF04DCT DBF COF05 DBF COF05DCT DBF COF06 DBF COF06DCT DBF COF07 DBF COF07DCT DBF COF08 DBF COF08DCT DBF COF09 DBF COF09DCT DBF COF1 DBF COF10DCT DBF COF1 1 DBF COF1 1DCT DBF COFI2 DBF COFI2DCT DBF COF1 3 DBF COF13DCT DBF COF1 4 DBF COF1 4DCT DBF COF1 5 DBF COF15DCT DBF COF16 DBF CDF16DCT DBF COFI7 DBF COF17DCT DBF COFI8 DBF CDF18DCT DBF DCT COU DBF DCT CTY DBF DCT PLC DBF DCT STA DBF PLF01 DBF PLF01DCT DBF STFO 1 DBF STFO 1DCT DBF STF02 DBF STF02DCT DBF STF03 DBF STF03DCT DBF STF04 DBF STF04DCT DBF STF05 DBF STF05DCT DBF STF06 DBF STF06DCT DBF STF07 DBF STF07DCT DBF STF08 DBF STF08DCT DBF STF09 DBF STF09DCT DBF STF10 DBF STF10DCT DBF STF1 1 DBF STFllDCT DBF STF12 DBF STF12DCT DBF STF13 DBF STF13DCT DBF 84 Fil e(f) by ei free D:FRED> 36 IASSIST Quarterly 3. What data is available? We currently have data from the Census Bureau, and will soon obtain a great deal more. The following data files are on hand: Disk File Geography TEST1 AHS85 TEST2 CCDB 1980 Census, STF3 1985 American Housing Survey, National Core 1982 Census of Agriculture 1982 Census of Retail Trade 1988 City-County Data Book zipcode nat/state state/county zipcode state/county/ city/place TABLE 1 : CD-ROM Databases at UC DATA 4. Accessing CD-ROM data: City-County Data Book The Census Bureau has supplied some access software (in the form of canned profiles which can be applied to generate reports for particular areas), all of which is located in the CDROM subdirectory on drive C in the archive. One of these is the CCDB profile program, which can be accessed as follows: C:\ cd cdrom Census Bureau C:\CDRGM> dir CD-ROM access software „ , . . ~ . ^rjr\ -130686 Vo 1 ume in drive " Direct •« C:\CDRCM AG82 7-12-88 9:34a RETAIL 7-12-88 9:35a ZIPS 8-02-89 8:56a README 1 666 8-08-88 10:41a RHACME 2 747 8-08-88 10:46a README BAT 58 6-23-88 10:23a TAB36 DBF 35815 6-29-88 9:54a AGR EXE 179692 7-12-88 l:20p CCDB EXE 202788 8-24-89 11:26a CDREADER EXE 39067 8-08-88 10:30a RETAIL EXE 83026 7-09-88 l:20p DATA SCR 4008 4-28-88 9:17a MENU SCR 4103 6-08-88 4:19p CALIF DBF 276 2-04-90 2:39p 16 File(s) 4771840 bytes free Start CCDB profile C:\CDRCM> CCdb Fall/Winter 1990 37 Cursor moves to choose California then Bakersfield then Land Area ================ COUNTY AND CITY DATA BOOK 1988 ===== I STATES ^ I Cities of 25,000 or more alakazarSIcoctdedcfl LA?u i GA HI ID IL IN IA KS KY LA ME f Bakersfielft .. MDMAMINNMSMDMTNENVNH I Baldwin Park NJ NM NY NC ND OH OK OR PA RI I Bell SC SD TO TX OT VT VA W W WI I Be 1 1 flowe r W : CALIFORNIA " ( I Bell Gardens SUBJECTS ww,^_^„ Xand,Ar«»,,«tid PopuTai ion Vital Statistics and Health Social Welfare Programs (* not covered in this geographic area) Crime and Education Money Income and Poverty Status Personal Income (* not covered in this geographic area) Hous ing Civilian Labor Force and Employment Agriculture (* not covered in this geographic area) Manufactures Construction (* not covered in this geographic area) Wholesale and Retail Trade -Cursor < Enter-Select PgUp-Page Up PgDn-Page Down Esc-Reiet Choosing the city of Bakersfield and the Land Area and Population subject (as shown by shading above) will give the following data about Bakersfield: Data display for city of Bakersfield, CA COUNTY AND CITY DATA BOOK 1988 STATES __„ I Cities of 25,000 or moi AL AK AZ AR'CA*CO CT DE DC FL LArusa ^_„^^^ GA HI ID IL IN IA KS KY LA ME i> Bakersfield M0MAMIMVMSM3MTNENVNH r "Baldwin Park ' NJ NM NY NC ND OH OK OR PA RI I Bell SC SD TN TX UT VT VA WA WV WI I Be 1 1 flowe r •-• ^ufowIiaI n-" ^-- j - I Bell Ga SUBJECTS 'Land '"kTiY~aZZ"^"~~"'~~~~T~~ " LAND AREA. 1985 (SQUARE MILr/ax LAND AREA. 1980 (SQUARE MI Leo-' TOTAL PERSONS, 1986 .'.'.'.'.'.'.'." RANK OF CITY POPULATION, 1986 PERSONS PER SQUARE MILE, 1986 TOTAL PERSONS, 1980 (CORRECTED) NET CHANGE, 1980-1986 PERCENT CHANGE, 1980-1986 POPULATION CHARACTERISTICS, 1980: PERCENT WHITE PERCENT BLACK . . . . PERCENT AMERICAN INDIAN. ESKIM3, AND ALEUT. 78.3 73.6 150,400 109 1,921 105,611 4*.789 42.4 76.5 10.6 1.3 Cursor P-Print PgUp-Page Up PgDn-Page Down Esc-Reset Fl-FIag Legendl City county data book has data for four levels of geography: states, counties within states, cities (of 25,000 population or greater), and census designated places (including unincorporated towns of 2,500 population or greater for which census data has been tabulated). The example above retrieved the first screen of items available for the city level of geography. 38 IASSIST Quarterly 4.1. CCDB individual data As part of the UC DATA operations and services on CCDB, the following tasks might be desired: • Obtain a data file of all cities and towns in California containing population and per- capita income. • Determine which counties in the United States have Hispanic population greater than 15 percent of total population, and rank them by percent Hispanic. In order to achieve these tasks we must use DBASE-ITI directly. Assuming we have started DBASE correctly, and then set the drive to F, and the following screen should appear: DBASE-HI In ASSIST mode Set Up Create Updite Poiition Retrieve Organ i ze Mod i fy Tools 01 132 152 pm l==========_-,j I. Database file . 1 1 ============= f------ -• I I CIF01 .DBF I Format for Screen II CIF01DCT.DBF I Query II CIF02.DBF I II CIFOI.DBF I Catalog II CIF03.DBF I View II CIF03DCT.DBF I II CIFOI.DBF I Quit dBASE III PLUS II CIF04DCT.DBF I=============== || CI F05 . DBF I CIF05DCT.DBF I CIF06.DBF I CIF06DCT.DBF I CIF07.DBF I CIF07DCT.DBF I CIF08.DBF I CIF08DCT.DBF I CDF01.DBF Conmandl USE Fl ASSIST l
l Position - lOptl 1/84 ciion bar - Select - DY. Select a database file. 4.1. Choosing California Towns and Cities In order to obtain data for towns and cities in California, we need to know (from the Census Bureau Documentation [CENSUS 89]) that the data file we are looking for is the Place file which contains incorporated and unincorporated (census designated) places with 2,500 or greater population in 1980. We must choose this file as our DBASE data file. This is easiest done by pressing the ESC key until we get the DBASE command prompt (the dot in column 1) and typing the "use " command, and then using the DBASE copy command to actually create the new file. Fall/Winter 1990 fvmftpmtMr^'g Start place level display structured database and Structure for database: f:plf01.dbf show structure Number of data records: 9593 Date of last update : 06/06/89 Field Field Name Type Width Dec 1 STOO Character 5 2 MCDCODE Character 3 3 PLACECCO Character 4 4 LEVEL Character 1 5 AREANAME Character 36 6 \NY93080 Numeric 9 7 NNY93086 Numeric 9 8 MNY92079 Numeric 7 9 MVY92085 Numeric 7 10 MMY93186 Numeric 6 1 11 NNY92185 Numeric 6 1 *• Total •• 94 r - - ' - ":-. - - " i v copy to d:plf01ca.dbf for stco='06' * Cop; California 401 records copied subset Corrmand Line ::PLF01 :Rec: EOF/9593 : Enter a dBASE III PLUS corrmand. and then we can browse the new file to see its contents. to browse p use dtplfOIca; -browse ,.,.. i California 1 1 ================ I ================ ========= I ===========| extract file II CURSOR <•- --> 1 UP DOW DELETE 1 Insert Model Ins 1 1 1 Char 1 RecordI *X "Y Charl Del 1 Ex i t 1 End 1 II Field 1 Home End 1 Page 1 PgUp PgDn Fieldl "Y 1 Abortl Esc 1 II Panl •<- *•> 1 Helpl Fl RecordI "U 1 Set Optionsl "Home 1 1 1=========== 1 ================ 1 - I ISTPL-- LEVEL MSA- PMSA CENTRAL STATE PLAC1 • AREANAME - STATE 1060000 1 06 0000 CALIFORNIA CA 1060010 2 7362 5775 06 0010 Al ameda CA 1,060025 2 4472 4480 06 0025 Alhambr a CA 1060085 2 4 , 472 036° ' 06 007° Anahe im CA 1060105 2 4472 ITto °6 ° 085 1060175 2 4472 4480 Ii ° 105 1060180 3 0680 1 06 l\l 5 Ant ioch Arcadi a Azusa CA CA CA 1060185 2 4472 4480 06 0185 1060210 2 4472 4480 06 0210 1060215 2 4472 4480 06 0215 1 "-^rsfield Ba ldwm . _ u Bell Bel 1 flower CA CA CA «4. 1 IBRCWSE lICIF01 1 iRecl 40/1008 1 1 ' View and edi t fields. 4.3. Selecting Hispanic Counties mile obtaining California places only presented a mild challenge, the process of discovering US counties with substantial Hispanic population concentration requires significantly more detective work. The Census Bureau does not make it easy because they don't include a comprehensive codebook to document the CCDB CD-ROM file, and so we must search for the data element (field in DBASE terminology) which has Hispanic population. We can begin by examining what documentation the Bureau does provide, as shown in the following table- IASSIST Quarterly Table B. Counties - Population Characteristics and Households PopuUbon T^ace^jX3-Co<\ Hnmahqala IBM-Con. 1980 1SS5 two '—- Percent— Percent 198» l 585 now Percent— County Under 5 5 10 15 10 2« 25 10 34 yoara 35 10 44 45 54 55 to S4 S5 10 74 yaara 75 rearj and Over can Indian. EsJuno. and Alaui Paaftc tatanoer Ha- pane' Female fanwy One- person1 14 15 IS 17 IS IS 20 21 22 23 24 25 28 27 2S 29 30 31 " 'Hisptiruc paraora irwy b* of any i-icsl 7No spous* pr«»«nt. 'HotAonohMr flying tfasn*. TABLE 2 : CCDB County Data Elements the numbers above the column definitions refer to a field called ITEM in DBASE, where is replaced by the actual column number. Looking at the table, we find that percent Hispanic population is ITEM25. Unfortunately, however, we don't know which of the 18 DBF files that 1TEM25 is to be found. We can guess that it's probably in COF02.DBF, and display structure for this data subset, as shown below: p use f:cof82jibf -J 1 1 display structure ; structure of Structure for database: f:cof02.dbf second county Number of data records: 3191 dbf file Dale of last update : 06/16/S9 Field Field Name Type Width Dec 1 STCD Character 5 2 LEVEL Character 1 3 NBA Character 4 4 P\GA Character 4 5 AREANAME Character 36 6 FLAG14 Numeric 1 7 ITEM14 Numeric 6 8 FLAG15 Numeric 1 9 ITEM15 Numeric 6 10 FLAG16 Numeric 1 11 ITEM16 Numeric 6 12 FLAG17 Numeric 1 13 ITEM17 Numeric 6 14 FLAG18 Numeric 1 15 ITEM18 Numeric 6 16 FLAG19 Numeric 1 Press any key to continue... note number of Comnand Line ::CDF02 :8ec: t/3191 7 county records Enter a dRASE III PLUS conmand. But as we can see, this isn't the case. Fall/Winter 1990 Fortunately, the Census Bureau has not left us completely in the lurch. Several dictionary files have been constructed which describe the contents of the data items and where they are located in the many DBF files. ;r:DCT OOU Enter a dBASE III PLUS cotrmand. I 1 CURSOR II Char: II Field: II Pan: UP Record : Page: PgUp Help: Fl PgDn I DELETE I Insert Mode: Ins I Char : Del I Ex i t : "Endl I Field: "Y I Abort: Esc I I Record: "U I Set Options: "Home I I ITEM DESC FILE I ITEM22 PERCENT 75 YEARS AND OVER COF02 IFLAG22 GDF02 IITEM23 IITEM24 IITEM25 I IITEM25A TOTAL POPULATION - USED FOR COMPUTING PERCENT'S FOR RACE AND AGE POPULATION CHARACTERISTICS. 1980: PERCENT AMERICAN INDIAN. ESKIMO, AND ALEUT PERCENT ASIAN AND PACIFIC ISLANDER PERCENT HISPANIC TOTAL POPULATION (STF-1) - USED FOR COMPUTING PERCENTS FOR RACE AND HISPANIC HOUSEHOLDS : COF03 COF03 GOF03 ::DCT COU :Rec: 36/391 View and edit fields. Using this dictionary file tells us what data base file to use (COF03.DBF) and which item (ITEM25) to use to search on and create our restricted file. IASSIST Quarterly We can now begin the process of determining which counties had high concentrations of Hispanics according the the 1980 Census. Open the dbf containing percent hlspanic use f:cofl)3.dbfY ' .list stco^reaname,itenj25 « Copy to a new file If more than IS percent hlspanic s t co 00000 01000 01001 01003 01005 01007 01009 01011 01013 01015 01017 01019 01021 01023 01025 CotTTnand Line a r c a name United States ALABAMA. Autauga, AL Baldwin, AL Barbour, AL Bibb, AL Blount, AL Bullock. AL Bu 1 1 e r , AL Calhoun, AL Chambers , AL Cherokee, AL Chilton, AL Choctaw, AL Clarke. AL ::COF03 item25 6.45 0.86 1.13 1.01 1.08 1.12 0.56 1.51 1.33 1.11 0.91 0.53 0.55 0.97 Rec: EOF/3191 Enter a dBASE III PLUS corrmand. copy to hispanic.dbf fields stco,level,msa,pmsa,areanarae,item23,item24, item25, item2Sa for item25>15.0 197 records copied Once we have obtained this data file we must sort in descending order using the DBASE sort command, and then we can list the contents to find the highest Hispanic concentrated counties in the United States. Use DBASE sort to tiispsrtdhf on ttem25/d '? sort command 100% Sorted 197 Records sorted to sort to a new * use hispart\dbf "."* " '* "^v.„ rj^"•" file in list stcojrean2me,item25aatem2S,Heiii244teni23J descending order Records' stco areaname item25a i tem25 item24 i tem23 1 48427 Starr. TX 27266 96.93 0.05 0.10 2 48479 Webb. TX 99258 91.52 0.09 0.11 3 48247 Jim Hogg. TX 5168 90.54 0.02 0.00 4 48323 Maverick, TX 31398 90.34 0.11 2.35 5 48507 Zavala. TX 11666 89.03 0.03 0.08 6 35033 Mora. >M 4205 86.56 0.05 0.17 7 48047 Brooks. TX 8428 85.99 0.04 0.06 8 48131 Duval. TX 12517 85.76 0.0 Corrmand Line ::HISPSRT :Rec: 1/197 Enter a dBASE III PLUS corrmand. 5. 1982 Census of Agriculture A task which one might wish to undertake is to draw information from the 1982 Census of Agriculture for counties with high Hispanic concentrations and combine it with the County data book information just obtained. Some of this information can be found in CCDB itself, but more detailed information would lead us to Census Bureau's CD-ROM TEST DISK 2 which contains the entire 1982 Census of Agriculture. If we pop this disk into our CD-ROM Fall/Winter 1990 drive, we can see how the files are organized. Directory of * display files like ag*.dbf ' ag82 data files AG82 0J.DBF AG82_02.DBF AG82_03.DBF AG82 04. DBF on Test Disk 2 AG82 05. DBF AG82_06.DBF AG82~07.DBF AG82 08. DBF AG82 09. DBF AG82~10.DBF AG82~11.DBF AG82~I2.DBF AG82~13.DBF AG82~14.DBF AG82 15. DBF AG82 16. DBF AG82 17. DBF AG82J8.DBF AG82 19. DBF AG82_20.DBF AG82 21. DBF AG82~22.DBF AG82 23. DBF AG82~24.DBF AG82 25. DBF AG82 26. DBF AG82~27.DBF AG82 DOC. DBF AG82_GEO.DBF 128747612 bytes in 29 files. bytes remaining on drive. Comnand Line :: : Enter a dBASE III PLUS comnand. Open the first % use ag82_0Ldbf dbf file and % display structure 1 examine its Structure for database: F: ag82_01 .dbf structure Number of data records: 3177 Date of last update : 02/03/88 Field Field Name Type Width Dec 1 TAB_01_001 Numeric 12 2 TAB"01~002 Numeric 12 3 TAB~01 003 Numeric 12 4 TAB~01~004 Numeric 12 5 TAB~01~005 Numeric 12 Press any key to continue... 122 TAB_02_010 Numeric 12 123 TAB_02~011 Numeric 12 124 TAB~02~012 Numeric 12 125 TAB~02~013 Numeric 12 126 TAB~02_014 Numeric 12 127 TAB_02~015 Numeric 12 128 TAB~02~016 Numeric 12 •• Total ** " 1537 Comnand Line : :AG82_01 :Rec: 1/3177 Enter a dBASE III PLUS comnand. What we find is that the Agriculture data files, unlike the CCDB data files, don't have any geographic codes in their records! The sole place for the geographic codes is in a separate file AG82_GEO.DBF. What this means to the unwary analyst is that the DBASE command JOIN can't be used to merge files from these two databases; a substantially more complex program will have to be constructed if we wish to put together data from these two sources, even though they are ostensibly collected for the same counties. We can look at this geographic file and find out what it contains. IASSIST Quarterly Browse the geographic reference dbf file for 1982 Census of Agriculture iise ag82_geadbf1 ..browse &ii*kiS 11 CURSOR <---->! UP DOAN 1 DELETE 1 Insert Mode: Ins 1 || Char: 1 Record: 1 Char: Del 1 Exit: "End 1 II Field: Home End 1 Page: PgUp PgDn 1 Field: "Y 1 Abort: Esc 1 II Pan: "<- "-> 1 Help: Fl 1 Record: "U 1 Set Options: "Home 1 | | ======== 1====== 1===== 1 ======= 1 01 000 ALABAVft 01 001 AUTAUGA COUNTY 01 003 BALDWIN COUNTY 01 005 BARBOUR COUNTY 01 007 BIBB COUNTY 01 009 BLOUNT COUNTY 01 011 BULLOCK COUNTY 01 013 BUTLER COUNTY 01 015 CALHOUN COUNTY 01 017 CHAMBERS COUNTY 01 019 CHEROKEE COUNTY Note number of BROASE ::AGS2_GEO :Rec-. ' 1 /3177 : counties View and edi t fields . Notice that the geographic codes don't have the same name for the two databases. State and county codes are separate on the 1982 Agriculture geography file, while they are concatenated into a single STCO code in the CCDB database riles. This further complicates the task of merging the two files. Finally, the above screen shows 3177 counties in the data file, while looking at the CCDB screen at the bottom of page 6 shows 3191 counties in the CCDB file. While this difference can be explained by the independent cities in Virginia and by the Bureau's not releasing Agricultural data for counties with fewer than ten farms, these facts further complicate the database compatibility problem. Thus our preliminary investigation shows merging City-County Data Book with 1982 Census of Agriculture to be a complicated process beyond the scope of this paper. The key to making this process easier will be the development of standards for geographic coding. 6. Conclusions and Recommendations While the Census Bureau has gone a long way toward making census data available on inexpensive personal computers, certain additional features will make accessibility of CD- ROM census data to the average planner or statistician using these data. The Census Bureau, in constructing CD-ROM products, should provide • A comprehensive codebook for data dictionaries which not only names and describes each data item, but also gives its universe, so items are not inadvertantly combined. a) An effort should be made to combine items having the same universe into the same DBASE file. Fall/Winter 1990 • Uniformity of file structure across databases: a) one record per geographic area b) The same geographic units over all databases (e.g. either one file for all counties in the U.S. or one file per state) c) The same geographic naming structure within each file and across databases (e.g. STCO as in City-County Data Book or STATE, COUNTY as in 1982 Census of Agriculture). This is so that the DBASE 'JOIN' command can be used to connect data from different files or the 'LOCATE FOR' commands will use the same selection sequence for files being connected. This purpose of this paper has been to give a glimpse of the effort necessary to do more than trivial tasks in retrieving data from the City-County Data Book on CD-ROM. In doing so we have uncovered some of the issues in access to census data. The availability of 1990 Census data on CD-ROM will surely force social science information specialists to confront these issues in this new medium. The application of standards will resolve some problems, but others can be relieved with additional machine-readable documentation from the Census Bureau. IASSIST Quarterly The Merit Networking Seminars MAKING YOUR NSFNET CONNECTION COUNT Merit/NSFNET Information Services is commit- ted to providing current information on na- tional networking to all users of the NSFNET backbone. Toward this end we will sponsor a two-day seminar in Ann Arbor, Michigan, May 20 and 21. "Making Your NSFNET Connection Count" will be an informative seminar focusing on issues of interest to campus computing leaders, information systems and networking administrators, educational liaisons, librarians, and educators who want to learn more about national networking. Day 1, "Real People Doing Real Things," will feature a numberof presentations concerning network applications in education from the elementary grades through the college level. The Day 1 activities will begin with a keynote address by Paul Evans Peters, Director, Coalition for Networked Information and will close with a tour of the Merit Network Operations Center. Day 2, "How to Get Connected and Stay Connected" will provide local, state, mid- level, and national networking perspectives from the experts. Day 2 will also be comprised of presentations on internetworking, information/user services, and network operations. The seminar will be held at the Tenneco Auto- motive Training and Development Center in Ann Arbor. Microcomputers connected to regional and national networks will be avail- able on-site sothat attendees may access net- work resources discussed in the presentations. The registration fee is $395. An early-bird fee of $345 will be charged for registrations received before April 15, 1991. This fee includes the two- day seminar, a reception on Sunday evening, lunch on Monday and Tuesday, all seminar material, and an optional tour of the Network Operations Center. For further information send an electronic message to seminar@merit.edu or telephone 1-800-66-MERIT. 3480 cartridges. For those interested in 3480 cartridges: NTIS (PB8 233135) is selling for $12.95 National Archives Technical Information Information Paper No. 4 "3480 Class Tape Certridge Drives and Archival Tape Storage: Technology Assessment Report." Call 703/487-4600 or write to Document Sales NTIS, Springfield, VA 22161 . This paper covers the "mechanical & technological future of the systems" & "provides valuable information to data center managers data librarians, and archivists, in fact to all who are con- cerned about the long-term storage of machine-readable data." Fall/Winter 1990