Computers in newsrooms of Michigan’s newspapers

Computers in newsrooms of Michigan’s newspapers

Lucinda Davenport

A state-wide survey of daily newspapers finds 80 percent are using electronic information sources, with all using electronic morgues and most using two or more electronic sources.

It’s hard to imagine that only 10 years ago few newspapers seriously considered buying computers for anything other than word processing (if they seriously considered computers at all). Furthermore, few reporters would have believed that one day soon they would be routinely pulling information from those computers that would be used in stories. But it’s happened, and quickly.

The idea of using computers for writing text, formulating statistics and presenting graphics was slow to sink in, but it has now attained a firm hold in news organizations. And, although the idea of journalists using computers to connect to other computers to find information sounded like science fiction, more journalists are today regularly going online to surf the net.

As Charles Hill, Michigan AP bureau chief noted, “Computer-assisted projects have won Pulitzer Prizes for six straight years. Reporters and editors are routinely using these skills not only for in-depth projects, but also for spot news, beat reporting and enterprise stories.”‘

But, what exactly does computer-assisted mean in journalism? For some people, the term is a general one that simply means using a computer to produce a news product. To others, like Hill, it is the skill of using a computer in all ways – from CD-ROMs to electronic morgues – to gather information to write a news story. And, still to others, the phrase means a very specific skill, such as the method of getting information off a commercial online database service such as Lexis/Nexis or the procedure of analyzing data using spreadsheet software.

The types of computer-assisted journalism skills and their rates of adoption are intriguing issues for news professionals who are trying to balance economics and competition. These points are also important to journalism educators, who are trying to keep abreast of the industry and creatively squeeze yet more information into the same amount of class time.

Thus, the overall objective of the present study is three-fold. First, it is a longitudinal study in that it follows up on an older state study to find out newspapers’ adoption rate of computerized information sources. Second, it picks apart the phrase computer-assisted journalism to discover just exactly what methods or combination of methods journalists are using as they gather computerized information. Third, the study seeks to illuminate the reasons journalists have for using these different electronic sources.

Few, if any, studies have been able to trace the adoption of these information sources over time in a single state. And, none have dissected computer-assisted journalism into seven specific areas of gathering computerized information: commercial online databases and videotex services, electronic bulletin board services (BBS’s), the Internet, compact disks-read only memory (CD-ROMs), electronic morgue of only that newspaper’s past issues, in-house topical databases that journalists develop, and the analyses of electronic public records.

In addition, this study looks at when newspapers of different circulation size adopted each of these computerized sources, and why they did so in terms of organizational benefits. At the level of the individual journalist, the study also explores how frequently different newspaper personnel use the various computerized sources, and for what newsgathering purpose. Background

Previous studies have slowly but significantly added to a composite picture of newspapers’ use of computers in gathering information for news stories. A review of previous research indicates that despite some variations, it is safe to say that commercial online databases have been the central focus. The first studies, sporadically spanning mostly the early- to late-1980s, were surveys to find out which newspapers were using, and the benefits of using, commercial online databases (noting, for instance, Knight-Ridder’s previous newspaper database VU/TEXT) and videotex services (such as CompuServe’s late competition, The Source). Initial studies that showed editors experiencing electronic information overload anxiety soon gave way to studies that looked at reasons why some newspapers were not as able as others to subscribe to online database services. Evidence of the growing body of research showed, however, that newspapers were moving from a tentative trial stage to a wider adoption phase.2 In the early 1990s, studies generally looked at the accessibility and use

of online databases by reporters and news librarians for news stories. Their conclusions prompted some journalism professors and professionals to reexamine the need to train reporters to use online databases when it seemed as though news librarians were experts already.3 As it turns out, there is collective (reporter and librarian) and distinctive (reporter) electronic news work.4

By 1991, Jean Ward and Kathleen Hansen introduced the term computer-assisted reporting in the results of their study to mean the additional use of personal computers for creating databases and analyzing data generally.5 They didn’t dwell on a definition, but Margaret DeFleur and Lucinda Davenport did.6Their definition included a variety of computer-information gathering skills that journalists need in order to: 1) search for information in commercial online databases from publicly available online databases within electronic bulletin board services (BBS’s) or from a newspaper’s own electronic morgue; 2) analyze public agencies’ electronic records; and 3) build their own customized topical databases. Thus, the phrase computer-assisted reporting is a reference to all computer information-gathering skills, not just to commercial online databases.

With this new phrase came expanded research initiatives in the earlyto mid-1990s. Studies broadened their computer-assisted reporting objectives to include analyses of relational databases (using database management programs) and electronic records (using nine-track tapes and spreadsheet programs), and the computer hardware and software used to do this type of reporting.

Unfortunately, computer-assisted reporting has now become a confusing term, used in different ways in professional discussions and in academic research. Although study objectives have been useful, comparisons are difficult. Some researchers use the term to mean a specific type of computer skill, while others mean it as a general reference for several computer information skills. For example, Cecilia Friend used the phrase generally to mean “both accessing online background material and processing quantitative databases.”8 Conversely, Brian Brooks and Yai-en Yang said it specifically “refers to the extraction of data, usually from government agencies, into relational databases.”9

On one hand, Bruce Garrison’s research looked at the adoption of various computer skills, and grouped them all under the general title, ComputerAssisted Reporting. On the other hand, his online news research was a separate variable from CAR tools. And, popular CAR subjects required at least relational database skills. Therefore, it was difficult to know if the CAR desk represented a specific skill (analyzing relational databases) or was a general reference to all computer information skills.lo And, more confusion: A Texas study asked a multiple-choice question on the “benefits of computer-assisted reporting,” but omitted any mention of the analysis of electronic records, relational databases or quantitative databases.ll

The distinction of the term is important because many newspapers are employing one or more skills used to gather computerized information. Some newspapers and journalism schools may be excluded from the arena of computer-assisted reporting because they use computers one way for gathering information, while others use theirs a different way. These are equally important skills used on different computer information systems requiring different hardware and software and needing different start-up and maintenance costs. Yet, all are a part of computer-assisted reporting.

Ideally, journalists should be skilled in obtaining all types of computerized information.

The present study

No study has asked newspapers about these different sources and the skills needed to use them, so one of the purposes of the present study is to identify the different types of computerized information being used by newspaper reporters. Results should show which ones are being used and for what reasons, but also which ones are used most often and least frequently. In addition, identifying standard combinations of computerized sources for different purposes is important.

No previous studies except Garrison have included the Internet or CD-ROM technology in their quest to find out where journalists are obtaining computerized information.12Interestingly, many newspapers that cannot afford commercial online databases have journalists who are becoming proficient at searching the Internet or CD-ROMs instead. Thus, including these under-estimated sources is another objective of the present study.

Also, as DeFleur and Davenport indicated, the quantitative pattern of computer-assisted reporting adopted in newsrooms can only be inferred because of the limited number of studies reporting the use of online databases around the country. “No one has systematically gathered and reported the numbers of newspapers who adopted year by year.”13 Thus, an additional objective of the present study was to follow up on a 1986 (presented in 1987) Michigan study to gauge Michigan daily newspapers’ rate of adoption.

The initial study that this research updates – by Stan Soffin and others – found that only two (the Detroit Free Press and the Detroit News) of 51 newspapers in the state used commercial online databases. More than 60 percent considered skills in searching online databases as not important. Many respondents didn’t know what an online database was. In addition, some form of in-house electronic morgue was operated by four newspapers – one large (the Detroit Free Press), one medium (the Oakland Press, circulation 71,000) and two small (the Marshall Chronicle, circulation 2,000; and the Albion Recorder, circulation 3,500, both owned by the Bedient family).14

Thus, the present study is able to illuminate change in this situation. In addition, this study focuses in depth on a single state whereas most previous studies except a Texas one”s are of a national scope. Finally, most previous studies had not surveyed smaller daily newspapers to access the degree to which they have kept up or been left behind in high tech reporting. Overall, this study will help classroom and newsroom professionals to prioritize their understanding, training and adoption of computerized information technologies.

Research questions

1: How many dailies in the state of Michigan use one or more of the seven electronic information sources?

2: What is the average number of such sources used by dailies, and what are the most frequent combinations?

3: How is circulation size related to use of such sources?

4: When and for what organizational reasons did newspapers adopt such sources, and how frequently are they used?

5: For what journalistic reasons are such sources used?

6: How likely are reporters to use such sources?


Following the Soffin et. al. Michigan newspaper census, each daily newspaper in Michigan was called and an appropriate editor identified who could provide information about electronic information sources. The editor was asked first if the daily used one or more of the seven electronic information sources: commercial online databases, bulletin board systems, the Internet, CDROMs, electronic morgues, newspaper-developed databases and electronic public records.

If the newspaper used one or more of the sources, the person the editor identified as the most common user of the source was sent a follow-up mail survey asking for more detail.?The two-page questionnaire asked when a particular information source was obtained and the reasons the newspaper obtained it, such as keeping up with competition or efficiency. The questionnaire also asked about eight journalistic reasons for using the source, including finding background statistics and getting quotes. Mail respondents were asked to identify the frequency with which the sources were used and whether reporters were among those who used the sources. Results

Forty-six of the state’s 51 dailies (90 percent) responded to the phone survey to identify daily newspapers using one or more of the seven electronic information sources.

Thirty-seven of these 46 dailies (80 percent) responding to the phone survey indicated they used one or more of the information sources, and were subsequently sent a follow-up mail survey seeking more information. Twenty of these 37 electronic information source users (54 percent) responded with additional information through the follow-up mail survey.l7

1: How many dailies in the state of Michigan use one or more of the seven electronic information sources?

Electronic information sources are now widely used by daily newspapers in Michigan compared to findings from the 1986 census. Phone survey responses indicated that the most commonly used sources were newspaperdeveloped databases and electronic public records. (see Table 1) The 19 users of online database and videotex services represented a nine-fold increase from findings obtained in the 1986 study. The least commonly used sources were an electronic morgue and the Internet. Bulletin boards and CD-ROMs figured right in the middle of commonly-used sources.

2: What is the average number of such sources used by dailies, and what are the most frequent combinations?

The phone survey data showed that the average number of different electronic information sources used by these 37 newspapers was about 3.4. Only four newspapers (11 percent of the total) used all seven sources, with three of those four newspapers having circulations exceeding 50,000. Some 11 newspapers (30 percent of the total) used only one of the sources, with half of those newspapers having circulations of less than 10,000.

The rest of the newspapers used various combinations. For example, three newspapers used every source except for an electronic morgue, representing the next most frequent combination found in the data. The only other combinations found more than once were at two newspapers using an electronic morgue and a newspaper-developed database.

3: How is circulation size related to use of such sources?

A four-category circulation breakdown of all phone respondents indicated that many of the sources – with a few notable exceptions – are now more widely available to small as well as big news organizations. (see Table 2) In the case of all seven sources, more than half of the adopters are newspapers with circulations below 50,000. However, 47 percent of the newspapers adopting both BBS and Internet systems are from newspapers of more than 50,000. Interestingly, electronic morgues are most likely to be adopted by the smaller newspapers, with 38 percent of that group coming from newspapers with under 10,000 circulation and another 23 percent coming from newspapers of between 10,000 and 25,000 circulation.

4: When and for what organizational reasons did newspapers adopt such sources, and how frequently are they used?

The follow-up (mailed) survey responses indicate that most newspapers adopted various information sources relatively recently. With the exception of electronic morgues, the vast bulk of the adoption of the various electronic sources occurred after 1990. (see Table 3) In the case of BBS systems, Internet and CD-ROMs, between two-thirds and 100 percent of this adoption occurred after 1993. Half the users of electronic public records reported adopting such sources after 1993. Interestingly, the adoption years for online databases indicate a steady but slow growth across the time period. The growth in use of newspaper-developed databases and electronic public records show an almost cyclical pattern.

The primary reason newspaper given for adopting these sources was to maximize personnel efficiency, followed by affordability and competition concerns. (see Table 4) Efficiency was overwhelmingly the most common reason cited for adopting electronic information sources generally, but also for each individual source. After efficiency, affordability was the reason for adopting the Internet, an electronic morgue and electronic public records. Also after efficiency, affordability and competition concerns were cited equally as reasons for adopting online databases, CD-ROMs and newspaper databases. After efficiency, competition concerns were rated as a second concern alone only for BBS’s. The other category was large for some sources because respondents added many unique reasons for adopting these sources.

Given that efficiency was the most often cited reason for adopting a source, it might be expected that such sources would be used frequently in the newsroom. This expectation was confirmed for some of the sources but not for others. (see Table 5) Sources used on a daily or weekly basis by most newspapers are the electronic morgue (100 percent of newspapers that have an electronic morgue), BBS’s (66 percent of newspapers that have a BBS) and online databases (54 percent of newspapers that have online databases).

Other sources were used less frequently: The internet, CD-ROMS and newspaper-developed databases are usually used several times a month to once a month (60 percent of newspapers, 55 percent of newspapers and 54 percent of newspapers that have these computerized sources, respectively). Finally, electronic public records are usually used only several times a year by 70 percent of the newspapers that have them.

5: For what journalistic reasons are such sources used?

Online databases, the Internet and electronic morgues were most frequently cited by the newspapers for obtaining background statistics (83 percent, 90 percent and 80 percent, respectively). (see Table 6) As might be expected, newspaper-developed databases and electronic public records were used mostly for obtaining raw data (64 percent and 73 percent). The most frequent reasons to use BBS’s included finding information that lent context to stories and finding sources (44 percent each). And, CD-ROMs were used to find general information (64 percent). In addition, newspapers used their electronic morgues to find general information, specific information on people and contextual information (60 percent each).

6: How likely are reporters to use such sources?

Reporters were the most likely to use all of these sources directly, with the exception of online databases and CD-ROMs. (see Table 7) Others, such as editors or librarians were cited as primary users of online databases (75 percent) and CD-ROMs (64 percent), obtaining the information for reporters.

Discussion and implications

This study has captured an accelerating revolution in the way news organizations gather information. At its inception, two centuries ago, American journalism relied on the direct observation of events by reporters. The telegraph and telephone later extended the scope of information gathering, but at the expense of removing reporters from the scene. But a primary information gathering tool of reporters – the interview – was facilitated by these technologies. Reporters could talk to many people by phone in the time it would take to sit through a trial or cover a meeting.

Now, the advent of electronic information sources has vastly expanded the scope of the reporter’s information search while almost entirely removing the reporter’s direct involvement with the sources of that information. The relative importance of the interview must decline in proportion to the use of these sources. While direct observation and the interview are certainly likely to continue in journalism, the use of electronic sources will require reporters with a different mix of skills to gather information.

And these changes seem to be happening quickly. The picture of computer-assisted reporting skills being adopted in Michigan newsrooms, if that is typical of the nation, is one of rapid growth, especially within the past two years. In 1986, only two newspapers in the state used online databases and four had electronic morgues. By mid-1994, 37 newspapers had adopted one to seven different types of electronic information sources. Results show that most newspapers began using five of these sources in 1993 (keeping in mind that this study was done halfway into 1994). If this study had asked how many newspapers were planning to adopt various computerized information sources, no doubt the numbers would have been much higher. In addition, Michigan’s total adoption rate generally correlates with the national rate.18 More than half of the Michigan adopters are newspapers with a circulation of less than 50,000, suggesting that this trend is hardly confined to just the biggest and richest news organizations. In fact, one or more of these electronic sources is used by nearly three-quarters of Michigan dailies.

The primary reason editors said they adopted these sources was to maximize personnel efficiency, such as reporters’ time. Finding or verifying information with a computer is a lot faster than driving to several places, making a lot of calls or sitting with a calculator re-adding large columns of data. With electronic sources, the reporter can write more stories within the same amount of deadline time.

But newsrooms also appear to be experimenting with these sources, attempting to find some optimal mix that suits their need. Although it may seem that different sources are used for uniquely different reasons, this is not always the situation. Editors often gave several different reasons for using the same source, and also the same reason for using different sources.

Finally, reporters in this study had almost universal access to these sources. The only sources that are not accessed primarily by reporters are online databases and CD-ROMs, probably because of expense and quantity, respectively. Expensive commercial online database usage is usually controlled because it entails subscription fees, search charges and printing surcharges. An information expert, such as a trained librarian, is usually able to find the needed information in less time using fewer searches, thus expending less money. In the case of CD-ROMs, publishing houses usually send one per newspaper. Its whereabouts and longevity usually can be controlled more easily if one person, such as a librarian, is in charge. However, as more newspapers network their reporters’ terminals to remotely access a CD-ROM carousel, then newsroom personnel will obtain CD-ROM information themselves.

These findings also have implications for undergraduate and graduate training of students in journalism and mass communication programs. Students will need to be trained in ways formerly considered more relevant for social scientists and librarians than for journalists. And schools and departments of journalism will have to upgrade their own technologies to meet the expectations newsrooms make of their graduates.


1. Charles Hill, Letter to Michigan Associate Press members on computer-assisted reporting workshop at Michigan State University, April 11-13, 1995. 2. John Ullmann, Large Newspaper Use of Commercial Data Bases. IDEAS; Research You can Use from the Missouri School of Journalism, 1983, pp. 11-20; Tim Miller, Information, Please, and Fast: Reporting’s Revolution-Databases. Washington Journalism Review, September 1983, pp. 51-53; Lanny McDonald, Commercial Database Survey. Bulletin of the Newspaper Division of the Special Libraries Association, Summer 1984, p.14; Frederic F. Endres, Daily Newspaper Utilization of Computer Data Bases. Newspaper Research Journal, Fall 1985, pp. 29-35; Tim Miller, The Data-Base Revolution. Columbia Journalism Review, September/October 1988, pp. 35-38; John Kerr and Walter E. Niebauer, Use of Full Text, Database Retrieval Systems by Editorial Page Writers. Newspaper Research Journal, Spring 1987, pp. 21-32; Stan Soffin et al., Online Databases and Newspapers: An Assessment of Utilization and Attitudes. Paper presented to the Newspaper Division of the Association for Education in Journal

ism and Mass Communication, August 1987; Kathleen A. Hansen et al., Role of the Newspaper Library in the Production of News. Journalism Quarterly, Winter 1987, pp. 714-720; Kathleen A. Harness et al., Effects of the Electronic Library on News Reporting Protocols. Journalism Quarterly, Winter 1985, pp. 845-852; Thomas L. Jacobson and John Ullman, Commercial Databases and Reporting: Opinions of Newspaper Journalists and Librarians. Newspaper Research Journal, Winter 1989, pp.15-25. 3. D.P. Wolfe, Newspaper Use of Computer Databases and Guidelines for Access: A Case Study: The St. Petersburg Times. Unpublished master’s thesis, University of South Florida, Tampa, 1989; Cynthia De Riemer, Commercial Database Use in the Newsroom. Unpublished paper, March 1991 (later published version is A Survey of Vul Text Use in the Newsroom. Journalism Quarterly, Winter 1992, pp.960-970; Jean Ward and Kathleen A. Hansen, Journalist and Librarian Roles, Information Technologies and Newsmaking. Journalism Quarterly, Fall 1991, pp. 491-498. 4. Jean Ward and Kathleen Hansen, op. cit., p. 496. 5. Ibid., p. 496.

6. Margaret H. DeFleur and Lucinda D. Davenport, Computer-Assisted Journalism in Newsrooms vs Classrooms: A Study in Innovation Lag. Journalism Educator, Summer 1993, pp. 26-36.

7. Cecilia Friend, Daily Newspaper Use of Computers to Analyze Data. Newspaper Research Journal, Winter 1994, pp.63-71; Brian S. Brooks and Yai-en Yang, Patterns of Computer Use in Newspaper Newsrooms: A National Study of U.S. Dailies. Paper presented to the Newspaper Division, Association for Education in Journalism and Mass Communication, Kansas City, August 1993; Bruce Garrison, Computer-Assisted Reporting at U.S. Daily Newspapers, 1994 Study. Presented to Society of Professional Journalists, Nashville, Tennessee, October 1994. Part of this study was turned into an article, Online Services as Reporting Tools: Daily Newspaper Use of Commercial Databases in 1994. Newspaper Research Journal, Fall 1995, pp. 74-86; Dave Mayes, Survey Finds Texas Newspapers Gearing Up to Move Onto Information Superhighway. Sent by Kathleen Davis as Internet message, article from Getting on the Information Superhighway in the Newsroom: A Survey of Texas Newspapers. Sponsored by the Texas Daily Newspaper Association, Texas Press Association, Texas A&M University’s Department of Agricultural Communications, and Texas A&M Public Policy Resources Institute, Nov. 1994.

8. Cecilia Friend, op. cit, p.65. 9. Brian Brooks and Tai-en Yang, op. cit., p.12. 10. Bruce Garrison, op. cit., lecture handout. 11. Dave Mayes, op. cit. p.20. 12. Bruce Garrison, 1994.

13. Margaret H. DeFleur and Lucinda D. Davenport, op. cit., p.27. 14. Stan Soffin et al., op. cit p.4-5. 15. Dave Mayes, 1994.

16. A follow-up mail survey was used for several reasons. First, the researchers wanted to give editors an opportunity to find information that they might not have access to while on the phone. Second, if newspapers used several information sources, interview time may have exceeded what is generally considered acceptable. Third, the researchers were confident that respondents would be interested enough in the study to be willing to respond with more detail at a later time.

17. This follow-up response included 63 percent of online database users, 53 percent of BBS users, 67 percent of Internet users, 58 percent of CD-ROM users, 36 percent of electronic morgue users, 44 percent of newspaper-developed database users and 52 percent of electronic public record users. 18. Bruce Garrison, 1994.

Lucinda Davenport, Frederick Fico & David Weinstock

Davenport is associate professor and Fico professor in the School of Journalism at Michigan State University in East Lansing. Weinstock is a doctoral student at Michigan State.

Copyright E.W. Scripps School of Journalism Summer/Fall 1996

Provided by ProQuest Information and Learning Company. All rights Reserved