Report 8 - Access to Information Review Task Force

HOW JOURNALISTS USE THE FEDERAL ACCESS TO INFORMATION ACT

Published: January 2002

Paul Attallah and Heather Pyman

Table of Contents

1. Executive Summary
2. Introduction
3. Method
4. Content Analysis
5. Interviews With Journalists
6. Conclusions
7. Appendix A – Coding Grid
8. Appendix B – Questions to Journalists
9. Appendix C – Relevant statistical information
10. Appendix D – The News Stories Analyzed

Executive Summary

1. Statistics and interviews were used to construct an overview of how journalists use the Access to Information Act and how they view its implementation.

2. Journalistic use of the Access to Information Act is increasing and stories using the Act have become more numerous and more complex.

3 The rise in use is attributable to increased experience with the Act, journalistic professionalism, media competition, and key trigger events.

4. The issue areas most frequently covered by newspaper stories using the Act are politics, the military, and general interest stories.

5. The primary intention most frequently associated with newspaper stories using the Act is to show the workings of government.

6. The social outcome most frequently associated with newspaper stories using the Act is to provide general information to the public.

7. Journalists say that the Act contains too many exemptions which are too broadly applied, that requests are needlessly delayed, and that no sanctions apply for missing deadlines

8. Journalists disagree about the need for fees and about their levels.

9. Journalists say that the function of disclosing information should be separate from and outside the departmental chain of command.

10. Journalists say that use of the Access to Information Act is an indispensable tool of the trade and have developed "best practices" as a result.

Introduction

This document consists of the Report itself, the coding grid used to generate the statistics (Appendix A), the interview questions to journalists (Appendix B), the relevant statistical results on which the content analysis was based (Appendix C), and the newspaper stories which were analyzed (Appendix D).

Method

In order to answer the question "How do journalists use the federal Access to Information Act?", two major types of investigation were conducted. First, a descriptive content analysis was performed on a random sample of newspaper stories drawn from the years 1985, 1990, 1995, and 2000. The content analysis indicated, amongst other factors, overall trends in use of Access to Information (ATI), key events which triggered its use, as well as major themes and shifts in the nature of stories. Second, the content analysis was supplemented by interviews with selected print and broadcast journalists concerning their use of the ATI Act. The interviews provided information not only on the strategies and practices of journalists but also on their views about the Act, its stated goals and their implementation. Together, the content analysis and the interviews provide an overview of how journalists use ATI and of their views for its reform. The steps of the investigation are presented separately.

Content Analysis

The statistics generated by content analysis describe thematic trends in journalistic use of Access to Information. These trends need to be interpreted contextually and the statistics need to be understood both within their descriptive power and their limitations.

They were derived from a random sample of stories drawn from the years 1985, 1990, 1995 and 2000. The stories were obtained from the following electronic sources: Dow Jones Interactive, Virtual News Library, and Canadian NewsDisc. The electronic sources archive the following English-language newspapers: Calgary Herald, Cambridge Reporter, Charlottetown Guardian, The Edmonton Journal, Financial Post, Fredericton Daily Gleaner, Globe and Mail, Guelph Mercury, Halifax Chronicle, Hamilton Spectator, Kitchener-Waterloo Record, Moncton Times Transcript, Montreal Gazette, Ottawa Citizen, St John's Telegram, Sunday Herald, Toronto Star, Toronto Sun, Vancouver Province, Vancouver Sun, and Victoria Times- Colonist.

The electronic sources allow full-text searches. Hard copies also allow full text searches but using them would have required a time frame greatly in excess of our deadline. The electronic sources were therefore queried with the keywords "access to information", "access-to-information", and "access information". They yielded the general results shown in Table 1.

Table 1: Results of Querying Electronic Archives for "access to information"

It should be noted that Dow Jones Interactive archives exclusively the Globe and Mail. The numbers shown, therefore, represent all stories corresponding to the search criteria which appeared in the Globe and Mail during the sample years. Virtual News Library archives the Toronto Star and many smaller newspapers. The numbers shown, therefore, represent the stories corresponding to the search criteria which appeared in several newspapers of which the major one is the Toronto Star. Canadian NewsDisc archives mainly Southam Newspapers. As a result, its relatively high number of stories corresponding to the search criteria also reflects the fact that, within the Southam chain, the same story (with minor variations) can appear in several newspapers. Additionally, newspaper ownership has changed considerably between 1985 and 2000. It is likely that archival practices changed as well. For example, in the early 1990s, the Globe and Mail's archive was sold to the American firm, Dow Jones. Southam newspapers underwent a change of ownership in the period under study and added the National Post which is not reflected in the statistics. Under the best of circumstances, therefore, the statistics show thematic trends which need to be interpreted contextually.

From the archival sources, a random sample of some 500 news articles was drawn. Once irrelevant stories were eliminated (i.e., software which provides "access to information"), the sample consisted of 269 news articles of varying lengths and types. These were coded (see Appendix A) and the codes entered into a statistical database program which yielded the results (see Appendix C) on which the analysis below is based.

Before proceeding, it may be useful to compare Table 1 (above) which shows stories produced with Table 2 (below) which shows requests made. The years recorded by the two tables overlap only partially and Table 2 comes with its own caveats: (a) not every request results in a story and there are likely more requests than stories, (b) depending on the recording practices of the departments, and the way in which media requesters identify themselves, not all media requests may be counted or even recognized as requests emanating from the media.

Source: Treasury Board

Table 2: ATI Requests by Requester and Year

Nonetheless, Treasury Board Secretariat reports the same general rising trend in ATI requests as reported in Table 1. In particular, the leaps between 1990 and 1995 and between 1995 and 2000 (remembering that Table 1 shows inter-newspaper duplication) are reasonably consistent. Additionally, media requests appear to represent consistently approximately 10% of all ATI requests.

What, therefore, do the statistics allow us to say?

Clearly, use of the Access to Information Act has increased over the years. For example, the database search for 1985 revealed a maximum of 101 stories for all newspapers searched. By the year 2000, however, there were more than 2000 articles corresponding to the search criteria.

The increase in numbers is attributable to several factors operating in tandem. First, journalists and news organizations have become better acquainted with the Act and its use. Second, certain events acted as triggers to spur increased use of the Act. The main events in the years examined were the Somalia inquiry, the tainted blood scandal, and the matters surrounding HRDC and job grants. Other events - the launch of the National Post with its strong interest in Access stories, the diligence of opposition parties - falling outside our time frame also acted as triggers. The evidence of their impact can be found in the sheer volume of ATI requests and in the nature of the stories generated. Third, any 'good' story invites competition and imitation; as a result, whenever a story became big in one medium, there was a tendency for other media to follow up with the same story or different aspects of the story, thereby generating more use of the Act.

The nature of the stories also changed over time. In the earlier part of the sample, many stories are very punctual, in the sense that they report information yielded by ATI requests without attempting to locate the information within a larger context. Examples are those early stories which report the results of polls conducted by the government, of expense accounts, of bureaucratic irregularities, etc. These stories tend to report one fact yielded by an ATI request and end there. The request is not integrated into a larger story arc. While this type of story never disappears, as it is both legitimate and corresponds to a genuine public appetite, other types of stories built around ATI requests also begin to appear in the later part of the sample. This second type of story is longer and more investigative and uses ATI requests as only one resource amongst many in order to build a news report. In these later stories, the Act itself and the information it yields are not directly foregrounded; rather, they are simply another research tool. Examples include stories about former spies and collaborators residing in Canada; these stories provided very general historical background whose function was not to show that the Act worked or to catch out any public officials but rather to construct a more nuanced picture of Canadian history. Other such stories grew out of the trigger events mentioned above. In the tainted blood and Somalia stories, for example, one ATI request typically led to questions in the House and the press, which led to further 'revelations', which generated further ATI requests, and so on. The point of these stories was not so much to use the Act to uncover a single or specific piece of information as to integrate its results into a larger ongoing investigation.

The stories also clustered at certain times of year around certain issues (see Table 3). For example, February and March 2000, contain many stories about HRDC, minister Jane Stewart, job grants, and general administrative difficulties (i.e., cheques bearing the wrong phone number).

Table 3: Stories by Month and by Year

Likewise, November 1985 contains numerous stories about government polls and how they might inform federal strategy (on free trade, international affairs, defense matters, etc.)

The content analysis also shows the issues on which the media focused (see Table 4). The leading issue was "politics", with 31.2% of all ATI stories. Politics refers to stories whose focus was principally, though rarely exclusively, the actions of a particular political or political party. Hence, stories about secret polls were primarily political even though the poll in question might have concerned health care because the focus of the story was that the poll furthered the strategic planning of a party, politician or department. Likewise, stories about government grants to specific ridings, contracts, rule-bending, etc., are "political". That having been said, no single politician or political activity stands out sharply though stories concerning HRDC and job training grants tend to predominate. This, however, is an artefact of the years sampled. Had other years been sampled, another issue might have predominated. The point is that "political" stories constitute the largest single issue area covered by ATI requests and that the range of such requests is fairly broad.

The second major issue area at 16.0% is "Other". Over the years, this category has covered a very wide range of issues which are difficult to categorize. Some of the stories concerned, for example, KGB activities in Canada during the 1960s. Others concerned the fact that the Royal Canadian Mint's template for printing new currency had been stolen or revealed an internal study about women in the Navy, about airport precautions to deal with windsheer, about the Queen's visit to Canada, and so on. It is fair to say that many of these stories did not have immediate or even obvious political outcomes. Rather, they reflect the general texture of life at the time with its multiple facets and possibilities. Many of these stories simply convey interesting facts which a news organization would be unlikely to foreground but which it would equally be remiss to leave unreported. If nothing else, the sheer diversity of topics indicates that in their use of ATI, the media discover a very wide range of topics which they deem of interest to the public.

Table 4: Stories by Primary Issue

The third major issue area is the military with 14.1% of ATI stories. These stories revolved around the Somalia inquiry, mismanagement (senior officials using military equipment for personal purposes), and the destruction of documents.

Interestingly, the 4th most frequent issue area concerned the Act itself (see "About the Act" and "Act Referred to"). These stories concern the Act, its application, potential amendments, wishes for its extension, criticism of exemptions, and so on. The very existence of the Act is referred to as a good thing but the use of exemptions - even legitimate exemptions foreseen by the Act - is perceived by some as contrary to the spirit of openness and transparency. In addition, when individuals or organizations state that they are not covered by the Act (i.e., Crown Corporations), the perception is that they may have something to hide. This perception, in turn, raises calls for an extension of the Act. Willingness to be covered by the Act tends in itself to become a measure of public honesty.

The stories can have several apparent intentions: to expose patronage, self-dealing, inefficiency and waste, explain the workings of government, or other. Overwhelmingly, though, the majority of stories - 61.3% - were concerned simply with describing the workings of government. Such stories typically mention a report obtained under Access to Information which projects fish stocks, plans budgetary undertakings, discusses tax shelters, recommends investigations of irregularities, reports opinion data, etc.

Table 5: Stories by Primary Intention

The second most frequent intention, at 16.7%, was to expose government inefficiency or waste. The third most frequent intention, at 8.9%, was "other". This category included stories which "exposed" the accuracy of gas pumps and supermarket scanners, the environmental hazard of shotgun shells, airport weather systems, spy rings of the 1960s, etc. They did not foreground any political message or significance.

The social outcome of the stories is consistent with the above. The function of any Access to Information Act in any country is to make information held by the government available to the public, with certain exemptions. But "informing the public" or what we may call the social outcome of the Act means at least three separate things which we have tried to distinguish.

1. Informing citizens: the public can be informed with a view to creating an informed electorate or citizenry capable of making enlightened political decisions. This is the most ancient meaning of "public information" and is the bedrock of liberal democracies. The role of the media, in this view, is to watch over government in the name of the public and to ensure transparency. The social result is an informed citizenry capable of judging the activities of government (how it spends, uses power, organizes itself) and in making enlightened electoral choices.

2. Generating new information: the public can also be informed by making available new information which has no specific political bearing but which nonetheless tells us about our society. For example, stories about spies operating in Canada in the 1960s or the criminal intimidation of prison guards and jurors have no obvious or immediate political or electoral bearing. They do not shape a classically enlightened citizenry but they do provide a richer and more nuanced understanding of ourselves and could conceivably inform some future action.

3. Participating in policy debate: some information (i.e., how the bank rate is set, how food is tested, whether opinion polls should be released during election campaigns) does not always have immediate political relevance but nonetheless tells citizens about important issues. It is directed not at informing the public about political events and platforms but at setting the agenda for public debate. It invites the public to participate not directly in making laws but in reflecting upon the issues which should be considered in making laws, shaping political platforms or reaching political decisions. Unlike the information in point 1 which is concerned with enlightened political choices (i..e, through voting), the information in point 3 invites citizens to participate not in the apparatus of the state but rather in civil society or associative life, through debate, through consideration of broad questions and orientations, through generalized reflection which may have no immediate political outcome.

Table 6 shows the primary social outcomes of the stories analyzed. Overwhelmingly, the stories are concerned simply with "informing the public" as described under point 1 (above). Fully 59.5% of stories fall into that category. 24.9% of stories are concerned with generating new information as described under point 2 (above). Only 4.8% of stories are invitations to participate in the "policy process" by submitting frequently difficult or complex information to general public consideration. Finally, 8.9% of stories have as their primary social outcome the wish to hold a person or institution to account.

Table 6: Stories by Primary Social Outcome

Plainly, though, it is impossible to claim that stories which "generally inform" the public have no impact on accountability whatsoever, just as it is illogical to claim that stories focused on accountability have no purely informational function. These outcomes are closely intertwined and difficult to separate even for analytical purposes. Consequently, stories were also coded for their secondary social outcome (see Table 7). When secondary social outcomes are considered, the purely informational function of news stories drops to 33.1% (from 59.5%), new information remains stable, and participation in policy debate, as a secondary social outcome, rises significantly to 13.5% (from 4.8%). Finally, accountability is a secondary social outcome for 26.3% of stories (up from 8.9%).

Table 7: Stories by Secondary Social Outcome

Next Page

Return to Table of Contents