Computer-Assisted Reporting Resources in Newsrooms, 1994-96

Bruce Garrison

School of Communication, University of Miami

P.O. Box 248127, Coral Gables, FL 33124-2030

305-284-2846 (v) and 305-284-3648 (f)

bruce@miami.edu 

A paper presented to the Investigative Reporters and Editors and Council of Affiliates interest group, Association for Education in Journalism and Mass Communication, for judging for presentation at Chicago, July-August, 1997.

Computer-Assisted Reporting Resources in Newsrooms in 1994-96

ABSTRACT

This paper focuses on trends in use of personal computer hardware and software for computer-assisted reporting (CAR) at U.S. daily newspapers over a three-year period ending in early 1996. Findings revealed that the leading hardware included Pentium processors, DOS/Windows, and use of more than one CD-ROM drive. Software category leaders were Microsoft Word, Microsoft Excel, Borland Paradox, MapInfo, SPSS, askSam, Procomm Plus, and Visual Basic. The most desired additions were hardware upgrades or new hardware and analytical mapping software.

Computer-Assisted Reporting Resources in Newsrooms in 1994-96

Personal computers have altered the practice of investigative journalism and are beginning to have a significant impact on routine reporting as well. This statement seems particularly applicable to the in-depth projects and other public records newsgathering tasks of journalists, not just the routine production and distribution of information. The types of daily breaking news, investigations, and special projects being published and broadcast by news organizations have changed in recent years. The speed of gathering what was once hard-to-access information has also improved. The breadth and depth of public information has increased. Even the type of information used for these stories is sometimes different. In fact, there are numerous technological effects that can be attributed to the use of computers in newsgathering (Garrison, 1996a; Garrison, 1995a; Houston, 1996; Paul, 1996; Ross & Middleberg, 1997).

Elliot Jaspin, a Cox News Service Washington bureau editor and founder of the National Institute for Computer-Assisted Reporting (NICAR), observed that he somewhat serendipitously discovered the benefits of computers while working on a story for the Providence Journal. It changed the way he looked at reporting, he said. Jaspin, recognized as a founder of computer-assisted reporting (CAR), argued that the computer has modified how reporters look at public information, such as records, in this decade. With computers, public records are more interesting, he feels. Jaspin also said that there are "much better stories" brought to publication through the use of computers (Jaspin, 1994).

Other journalists agree with Jaspin. Dave Davis, a Cleveland Plain Dealer reporter, stated, "I think one of the things it [the computer] has done is taught me to be aware, to look out for opportunities when they come knocking" (Cline, 1994, pp. 13-15). Los Angeles Daily News reporter David Bloom has stated that building his own databases for routine beat stories has made significant differences in how he approaches his work. "Spreadsheets can be incredibly useful in lots of ways for ongoing beat coverage, involving fairly small investments of time, as well as for understanding a few numbers on … deadline" (Bloom, 1994, p. 11). He recommended a gradual approach that leads to data that cannot be obtained in electronic form elsewhere. This, he contended, has changed how he handles his routine reporting on public affairs beats, such as law enforcement, city hall, and politics.

There are two approaches to computer-assisted reporting. CAR includes (a) both online-based newsgathering that uses specialized commercial services and Internet-based services, such as the World Wide Web, and (b) database-oriented analysis using existing and originally created databases from both the public and private sectors (Garrison, 1995a; Garrison, 1995b; Houston, 1996). CAR has been increasingly adopted by reporters as subjects have dictated it (Garrison, 1996a).

Rose Ciotta, CAR director at the Buffalo News and a member of the Investigative Reporters and Editors board of directors, recently wrote that journalists "are tapping into data and producing high-impact stories on topics ranging from criminals among nursing home workers and school teachers to unsafe elevators and the influence of political contributions" (Ciotta, 1996, p. 36). Ciotta noted that a number of CAR experts have acknowledged the revolution in investigative and other news reporting techniques occurring in this decade.

The effects of the revolution seem clear: Reporters save time, but gather more information. In doing so, reporters also save their companies money while they are more productive (Garrison, 1996e; Ruberry, 1996). It is one of the few areas in newsrooms in this decade where jobs are being created (Ciotta, 1996). Interest in CAR has grown rapidly, as has recognition of its effectiveness through numerous high-level awards since 1990 (Reavy, 1996a). One of the reasons for these computer-driven benchmark changes, simply stated, is that American society has changed. Business has been equally, perhaps even more, influenced by personal computing. Education, government, and almost all other segments of modern America have been affected by computing. Governments, for example, have kept records on computers for almost a half century, but the extent of the electronic record keeping has geometrically increased in the past decade (Houston, 1996; Garrison, 1995a).

Thus, it seems natural that journalists would begin using computers more often to study those records and to report about newsworthy information contained in them. The modern tradition of investigative journalism using public information has simply extended its reach with these new tools. Computer-based journalism is also viewed as a way to revitalize the newspaper industry and "restore public confidence in journalism," one media analyst recently observed (Foley, 1996, p. 8). Taking the issue further, another CAR expert said the revolutionary use of CAR is a "matter of survival" (Ruberry, 1996, p. F-2).

Brant Houston (1996), managing director of the National Institute for Computer-Assisted Reporting since 1994, has argued that computers change the way journalists think about their stories – computers require users to be more methodical and somewhat less intuitive in the beginning. But, he noted, the results are frequently better reporting skills and better stories. "Most important, computer-assisted reporting is at the heart of good public service journalism" (p. 6).

Because of the growing importance of CAR in newsrooms, new attention is being given to on-the-job training and education (Garrison, 1996a). Most CAR training focuses on use of basic tools, such as operating systems, multimedia, spreadsheets, the Internet, database management, mapping, statistics, nine-track data tape drives, and programming (Reavy, 1996b). Houston (1996) has noted that there are three basic CAR tools – spreadsheets, database managers, and online services. But he also said that statistical software and mapping software are important categories that are rapidly growing in popularity and value. "Computer-assisted reporting uses other tools, some of which are complex and some of which are merely handy," Houston explained (p. 7). Feola (1995) reduced CAR basics to two areas, arguing that spreadsheets and online research services are the most important starting points for CAR.

Given the growing use and significance of computer tools in investigative reporting and other newsgathering, it is necessary to learn more about the extent of their use in newsrooms. Which categories of computer tools are commonly used? Which tools are most often used? Specifically, this study sought to determine the trends in use of hardware and software often critical to information gathering and analysis in investigative journalism.

RESEARCH QUESTIONS AND FOCUS

The present study focused on use of basic CAR hardware and software tools used during a three-year period: December 1993- February 1994, December 1994-March 1995, and January to March 1996. For practical purposes, the project conducted in 1993-94 will be referred to as "1994." The project conducted in 1994-95 will be referred to as "1995." Generally, the study sought to answer these research questions:

1. What were the most-often installed hardware configurations used for CAR in 1996? During the three-year period from 1994 to 1996?
2. What were the most-often installed software configurations for CAR in 1996? During the three-year period from 1994 to 1996?
3. What were the most-wanted new hardware tools for CAR in 1995 and 1996?
4. What were the most-wanted new software tools for CAR in 1995 and 1996?

THE STUDY METHOD

An on-going national research project studying the development and use of CAR has been underway since 1993. This paper analyzes findings from three national surveys:

• During the last week of December 1993, questionnaires were mailed and during January 1994 data were collected from a population of 514 Sunday and daily newspapers across the United States. A circulation minimum of 20,000 on Sundays was the minimum for inclusion. One follow-up mailing was sent in February 1994 and returned in February and March 1994.
• During the last week of December 1994, data collection began with a mailing to 510 Sunday and daily newspapers. Again, a circulation minimum of 20,000 on Sundays was required for inclusion. Two follow-up mailings were sent. The first follow-up was mailed in early February 1995 and the second was sent in mid March 1995.
• In early January 1996, the most recent mail survey was conducted with a mailing to 510 daily newspapers. Two follow-up mailings were used to increase participation.

Each follow-up mailing resulted in a five to ten percent increase in response rates in each year and specific figures are available from the author. Since this study involved surveying an entire population, significance tests are not reported. Circulation figures were obtained from the latest editions of the Editor & Publisher International Year Book (Anderson, 1994; Anderson, 1995; Anderson, 1996).

Editors were asked either to complete the questionnaire themselves or to forward it to the person in charge of online news research and/or CAR. In some cases, two or three persons completed portions of the questionnaires. Questionnaires were developed from discussions and interviews during the Investigative Reporters and Editors and NICAR conferences at Raleigh, N.C., in 1993, at San Jose, Calif., in 1994, and Cleveland in 1995. The instruments consisted of four sets of questions, including institutional and personal information, CAR tools, online news research, and field reporting use of computers. Data were processed using the Statistical Package for the Social Sciences for Windows, Version 6.1.3 (Norušis, 1995).

FINDINGS

In 1994, a total of 208 responses were received, a rate of 41 percent. In 1995, 287 responses were received, a rate of 56 percent. In 1996, 233 responses provided a rate of 45 percent. Table 1 shows general demographics of respondents in each year. Respondents in each of the three groups were typically managing editors, CAR supervisors or directors, news researchers, or reference supervisors.

What were the most-commonly installed hardware configurations used for computer-assisted reporting?

Newsrooms were upgrading hardware during the period studied. While the upgrades may have had other purposes – such as improved production – there was a potential impact on CAR and investigative reporting. Some news organizations have replaced dedicated word processing systems with PC-based networks, but others upgraded older PCs already in use. The current major overhaul of the newsroom computing systems at the Philadelphia Inquirer that began with hardware and software planning in late 1995 and continued with installations this spring is just one example (Borowski, 1997). As shown in Table 2, about one-third of respondents used Pentium-type processors in 1996, a growth of 33 percent since the processor was introduced in 1994. As upgrades occurred, the proportions of newsrooms using other processors dropped. The number of newsrooms using Macintosh computers remained flat at about nine percent for each of the three years.

Operating systems have gradually been converted from DOS to a Windows/DOS configuration used by 51 percent of newsrooms in 1996 and more than double the proportion two years earlier. Windows 95 use seems to be growing slowly.

The number of newsrooms using CD-ROM drives has also increased from 1994 to 1996. In 1994, only 40 percent used one or more CD-ROM drives. That figure grew to 58 percent in 1995 and to 62 percent in 1996. It is also interesting to note that the proportion using one CD-ROM drive has dropped while the proportion using two or more has increased dramatically.

Other hardware changes include modem upgrades to 28.8 Kbps (34 percent in 1996 and 16 percent in 1995), while another large group used 14.4 Kbps (35 percent in 1996 and 35 percent in 1995). Digital tape was the most widely used storage medium at 25 percent in 1996, up slightly from 24 percent in 1995. The mean hard drive capacity of computers used for CAR was 1.66 GB, but those in use ranged from 60 MB to 1,600 GB. A year earlier, mean hard drive capacity was 1.06 GB, ranging from 20 MB to 1,000 GB.

What were most-commonly installed software configurations for computer-assisted reporting?

There are a considerable number of software categories used for CAR – everything from word processing to advanced statistical packages. For this study, eight major software categories have been tracked for the past three years. There is general stability in the category leaders over the three years, but some shifts were noted. In general, Microsoft Office products are gaining wider acceptance as basic CAR tools, Table 3 data show.

Microsoft Word has become the preferred word processor at 34 percent in 1996, replacing XyWrite at a number of newspapers. Similarly, Microsoft Excel has not only maintained its leadership role, but it increased its dominance among spreadsheet programs at 43 percent. Borland Paradox, a long-time favorite among relational database managers and sold to Corel in 1996, continued its popularity at 24 percent. Microsoft FoxPro was also widely used at 21 percent. Microsoft Access, a relatively new relational database program developed for Windows in 1993, grew to a base of 14 percent.

Among more specialized and high-end CAR software, nonuse remained the most significant finding. High-end products include analytical mapping, statistical packages, text database managers / personal information managers, development and programming software, and communications software. However, the amount of newspapers not using more advanced CAR software was gradually declining.

MapInfo (14 percent) was the preferred analytical mapping tool in 1996, but 74 percent of respondents did not use mapping tools. This was a decline from 83 percent in 1994. SPSS grew slightly in its dominance as the preferred statistical package at 9 percent in 1996. However, nonuse of statistics programs remained steady at 87 percent, slightly higher than the 86 percent recorded in 1994 and 1995. askSam continued to be the leading text database management tool at 8 percent and use of text database managers and PIMs has declined from 93 percent in 1994 to 81 percent in 1996.

While most newspapers did not use development tools for preparing database front ends or other newsroom applications, the growth in nonuse has flattened in the past two years at 85 percent. Visual Basic is the leading tool among users, growing slightly from 2 percent in 1994 to 8 percent in 1996.

The dominance of Procomm Plus as the preferred non-proprietary access tool to online resources has grown steadily over the past three years. Starting at 31 percent in 1994, it has increased to 35 percent in 1995 and 39 percent in 1996.

What were most-wanted new hardware and software tools for using computer-assisted reporting in the past two years?

The most important new CAR resource that respondents listed was a hardware upgrade or new hardware (see Table 4). This improvement was noted by 22 percent of newspapers. Among software and services, newspapers sought mapping software (7 percent in 1996) and improved Internet access or new Internet access (6 percent).

CONCLUSIONS

Use of hardware and software tools for CAR has stabilized somewhat in the past three years. These hardware and software products have become part of the basic "tool kit" of most journalists who use CAR. The stability has developed because of two processes taking place at the same time. First, the computer software industry is consolidating in some ways. Earlier in the decade, there were more choices and more vendors for software products and services. Some of these have failed and some have been acquired by other companies and evolved into their current versions. Products have survived because of their success in accomplishing the jobs they were designed to do. Like their counterparts in the business world who use database-oriented software, journalists have gradually determined that certain products are more suitable than others are and have gravitated to them.

Hardware manufacturers have also evolved, but the basic CAR hardware tools have not changed remarkably from 1994 to 1996. There were several notable changes in this period, nevertheless, such as the introduction of the Pentium processor, widespread introduction and use of multi-media tools, such as CD-ROM drives, the increase in hard drive capacities, and the increase in the base amount of random access memory. Aside from these improvements, the basic hardware configuration in 1996 was not that different from that in 1994. Thus, there is general stability in what the hardware can do, but improvement in how it does that job each year.

Second, journalists were going through their own decision-making processes involving hardware and software products. Corporate decisions were filtering down to newsrooms that lead to some standardization. For example, several newspaper groups have adopted the Microsoft Office suite "Professional" edition for newsroom PC networks, resulting in use of Word, Excel, and Access for CAR by both those experienced and those just beginning to use CAR.

There was notable growth in entry-level users over the past three years. More mid-size and small dailies were using CAR in 1996 than in 1994 or 1995. The increases in software and hardware use each year point to this change.

However, while there was growth in "entry-level" use of such programs as spreadsheets and relational databases, there was little growth in use of more advanced CAR tools, such as statistical packages, analytical mapping programs, and programming-development tools. This is clearly the next generation of growth, a second-level approach to the development of CAR and, perhaps, investigative journalism in general. Furthermore, journalists have yet to discover utility database programs, such as personal information managers and text database managers. The nonuse of these tools shows advanced CAR tools and certain utility software categories remains one of the study’s significant findings. Because of the lack of use and lack of expertise in how to use these advanced CAR tools, there will be need for training and continuing education among journalists on the job as well as those entering the job market for the first time. Some professional organizations and universities are trying to meet the need, but the effort is only in its beginning stages.

There seems to be indicators pointing toward continued growth in use and improvement in the sophistication of CAR. The "wish lists" data showing the desire to upgrade or replace older hardware and the expressed need for advanced software tools, such as analytical mapping, suggest movement in this direction. There was also indication for continuing training and education with these new tools.

The study falls short in several respects and deeper probing into the subject matter may be needed. It would be helpful to analyze data by looking at characteristics of the newspapers, such as size, region, available CAR resources, computer literacy levels of journalists involved, and availability of data, for instance. There were other weaknesses. This study did not include wire services involved in CAR, nor did it look at specialty publications, such as news magazines, that were readily using CAR in their investigative reporting. Regardless, it is evident in these data that the availability of CAR hardware and software tools have grown and continue to be adopted in newsrooms. It would also be valuable to look at broadcast news media since it is apparent that an increasing number of local news organizations and network news teams were producing investigative projects that used numerous CAR tools.

TABLE 1: Respondent Demographics, 1994-96

TABLE 2: Hardware, 1994-96

TABLE 3: Software Use, 1994-96

*Percent change from 1995 to 1996 only.

TABLE 4: First Choice for New CAR Tools, 1995-96

REFERENCES

Anderson, Ian E., ed. (1994), Editor & Publisher international year book 1994, New York: Editor & Publisher.

Anderson, Ian E., ed. (1995), Editor & Publisher international year book 1995, New York: Editor & Publisher.

Anderson, Ian E., ed. (1996), Editor & Publisher international year book 1996, New York: Editor & Publisher.

Bloom, David. (1994, March-April). Analyzing data on the beat. IRE Journal, 17(2), pp. 10-11.

Borowski, Neill (1997, March 7). Personal communication with author.

Ciotta, Rose (1996, March). "Baby you should drive this CAR," American Journalism Review, 18(2), pp. 34-39.

Cline, Francesca (1994, January-February). "Perspectives on how computers change journalism," IRE Journal, 17(1), pp. 13-15.

Feola, Christopher J. (1995, March). "Making the cut: Enhance your job opportunities by developing your computer skills," Quill, 83(2), pp. 24-26.

Foley, Michael (1996, May 6). "The computer gets the story," The Irish Times, city edition, CompuTimes section, p. 8.

Garrison, Bruce (1995a). Computer-assisted reporting. Hillsdale, N.J.: Lawrence Erlbaum Associates.

Garrison, Bruce (1995b, Fall). "Online services as news reporting tools: Daily newspaper use of commercial databases in 1994," Newspaper Research Journal, 16(4), pp. 74-86.

Garrison, Bruce (1996a). Successful strategies for computer-assisted reporting. Mahwah, N.J.: Lawrence Erlbaum Associates.

Garrison, Bruce (1996b, August). "Newsroom tools for computer-assisted reporting in 1995," unpublished paper presented to the Communication Technology and Policy Division, Association for Education in Journalism and Mass Communication, Anaheim, Calif.

Garrison, Bruce (1996c, August). "Online services and the Internet: Computer-assisted reporting in newsrooms in 1995," unpublished paper presented to the Newspaper Division, Association for Education in Journalism and Mass Communication, Anaheim, Calif.

Garrison, Bruce (1996d, Winter/Spring). "Tools daily newspapers use in computer-assisted reporting," Newspaper Research Journal, 17(1-2), pp. 113-126.

Garrison, Bruce (1996e, April 27). "Interactive communications: Computer-assisted reporting tools," Editor & Publisher, 129(17), pp. 86-87, 102-104.

Norušis, Marija J. (1995). SPSS for Windows base system user’s guide, release 6.1. Chicago: SPSS, Inc.

Jaspin, Elliot G. (1994, January-February). Perspectives on how computers change journalism. IRE Journal, 17(1), pp. 13-15.

Paul, Nora (1996). Computer assisted research: A guide to tapping online information. 3^rd edition. St. Petersburg, Fla., The Poynter Institute (HTML diskette edition).

Reavy, Matthew M. (1996a, March). "How the media learned computer-assisted reporting," unpublished paper presented to the Newspaper Division, Association for Education in Journalism and Mass Communication, Southeast Colloquium, Roanoke, Va.

Reavy, Matthew M. (1996b, March). "Teaching CAR: Lessons from the National Institute for Computer-Assisted Reporting," unpublished paper presented to the Newspaper Division, Association for Education in Journalism and Mass Communication, Southeast Colloquium, Roanoke, Va.

Ross, Steven & Middleberg, Don (1997, March 23). The Media in Cyberspace III. http://www.mediasource.com/study/cont.htm.

Ruberry, William (1996, March 31). "Newsroom changing to keep pace with information revolution," Richmond Times Dispatch, city edition, Commentary section, p. F-2.