Finding and Using Primary Documents

When starting a research project, one of the biggest mistakes students often make is haphazardly collecting and reading research materials as they look for information about their topic. Gathering information requires a planned method as you need to make distinctions between various kinds of materials such as primary, secondary and tertiary documents. In this chapter, we will examine what primary documents are, and how to collect and file these documents in an organized and systematic way. We will also examine how to initially survey sources for information on primary documents, and the best kinds of sources to use in the early stages of your research.

Screen Shot 2016-04-02 at 10.26.36 AM.png


  • The term document itself refers to any piece of writing such as letters, consultant reports and novels as well as non-written and/or oral texts like films, photos and chat room discussions, speeches or casual conversations (this also includes your notes on these items).

A. Primary Documents

  • Primary Documents: these are documents prepared and distributed by actors and would-be actors, either in the form of decisions, or as attempts to shape deliberations and decisions about public issues
  • These documents will often contain information about the events, actors, publications, deliberations and decisions that will be valuable in your research
  • Examples: laws, legal decisions, rules of procedure, consultant reports, government discussion papers, advocate websites, news media reports, archives, memoranda, meeting notes and conference agendas
  • In addition to providing you with information, primary documents are often designed to persuade or influence others, as they are often developed to tell people what they should do (i.e. a law) or what they ought to do (i.e. a government or consultant’s report)
  • Since these documents aim to persuade, they will often emphasize certain kinds of information and omit other things altogether in the desire to seek a specific outcome
  • As a researcher, your job will eventually be to determine whether the document is useful, or whether it needs to be set aside because you cannot determine the extent to which the persuasive purposes of the material overwhelm the information content.

B. Secondary Documents (for a greater in-depth discussion, see Chapter 4)

  • A. Secondary documents are materials which highly abstract and try to describe, depict or explain various situations in a comprehensive way
  • These source will explore relationships and issues not easily documented and do so by offering insight, arguments and different conclusions
  • Examples: academic books, scholarly articles, in-depth news reports in magazines

C. Tertiary Documents (for a greater in-depth discussion, see Chapter 5)

  • Tertiary materials describe events and situations but don’t provide any kind of in-depth explanation about an issue
  • Examples: tables, graphs, statistical material (not including any accompanying text), field notes


Can a document be both primary and secondary?


It is entirely possible for one single document to be both a primary and secondary document. Simply use each chapter guide as a reference for how to deal with your material. Think about it: a document may be primary if it is produced by an actor or would-be actor, and secondary if it tries to explain or explicate a given situation. For now, don’t worry about figuring out whether to put your document in your primary, secondary or tertiary piles as through the course of your research, you may find yourself reading the same document (or pieces of it) in different ways. For example, an academic writes and article for a journal that falls somewhere between a disciplinary publication and a government publication. The article is therefore a primary document if the author hopes to have influence based on what he or she has written, but it may also be a secondary document if the article provides an explanation of events and is published in an academic context.

Getting started on your research

Don’t spend too much time reading right away

  • When beginning your research, you will not be able to tell which documents are essential or absolutely relevant versus those that may simply be interesting but not entirely important to your research
  • That is why you should simply scan your documents and NOT carefully read any materials until you are well into your research and have a better idea of the story you want to tell
  • One reason why you should not carefully read materials right away goes back to the fact that primary documents contain information but are also designed to persuade
  • This means you need to make judgements about reading and extracting information in determining the reliability of your source and whether or not the style and purpose of the document has influenced the information in it
  • For example, ask yourself the following questions: would another publication, with a different editorial slant, provide the same information, albeit in support of different decisions? What information has been collected and left out? What information has been taken for granted as being true or not? What concepts are used and how are they defined?
  • Initially, you will collect ALL primary documents including those that may be unreliable, since biased documents will often reflect the position, interests, biases and intended audiences of the different actors involved in your story
  • Link: HOW TO SPEED READ infographic.


How can a document be both a reliable source of information yet not tell the whole story?


Think of a magazine like The Economist. While The Economist does have a very specific editorial slant designed to persuade readers, the information in their articles is usually very reliable and is fact-checked to ensure no mistakes are made.  Yet although each story might include credible information, think about things like what information is missing, how the article is organized and how concepts are used to make their point. For instance, if the Economist relies on using GDP as a standard measurement of economic growth, what other factors are they not considering such as length of a work-week or the unpaid work of caregivers?  These are factors that are not accounted for in GDP, and yet might be used to tell a very different story.

Collecting and Filing Documents

  • Instead of spending all your time carefully reading documents right away, it is far more important to create a system to collect and annotate this material so that you can easily find it later on, especially considering something may be relevant in the future
  • You need to organize your materials by creating a workable FILING SYSTEM (paper and/or digital)
  • A workable system does not include meticulously detailing and cross-referencing files (too time-consuming) or simply piling files on the floor (doesn’t tell you what information is missing)
  • You will create hundreds (maybe more) of files for your research (even if your time frame is only a few years) and it always a good idea to err on the side of having more folders than less
  • When you start your research, you will require THREE (3) kinds of files including a chronological list, actor files, and institution files
1. Chronological Lists
  • This list should include a chronology (timeline) of primary and secondary documents which is connected to the time frame and jurisdiction you are studying and is organized by DATE ONLY (you may include events in other time frames if you believe they could be useful to you later on)
  • A simple annotation (just a sentence or a few words) is all that is required for each entry and you should have hundreds of entries over time as you proceed with your research
  • Your list should include the following kinds of information: academic publications, deliberations, decisions, events, newspaper stories, meetings, laws passed, press releases, government documents, conference proceedings
  • Your chronology enters are the backbone of your research and you will likely have hundreds of entries
  • There are many reasons why it is important to create a chronology:
  • Reason #1: a chronology will provide you with a quick overview of your story line including such details as what happened, when and where, and what outcomes resulted from different events
  • Reason #2: by creating a chronology, you start to see not only what happened, but based on gaps in your material, what information still needs to be collected
  • Reason #3: if there are gaps, this allows you to further investigate whether or not another issue pre-empted yours in the eyes of decision-makers and/or the public
  • Reason #4: each entry in your chronology will help you see the evolution of own research as you continuously add information about your issues
  • Link: How to make a chronology
2. Actor Files
  • Any actor will get an individual file if they are relevant to your story
  • Over time, you will have a very large number of different actor files and later research will determine which actors are important and while files can be discarded
  • Actor files may include a government unit or department, trade associations, corporations, or even a particular newspaper that has covered your issue extensively
  • Authors of scholarly materials may also be labelled as actors if you think they have influenced a public issue in any way
  • You should organized your files alphabetically by each actor but DO NOT attempt to organize them any further as you are still not ready to decide what role they will play in your story
3. Institution Files
  • Each institution relevant to your research will also get an individual file
  • These files will contain mandate or purpose of the institution, rules of procedure, records, transcripts, committee proceedings, organizational minutes, conference proceedings (including academic book publications) and anything else that shows how deliberations took place and/or decisions were made
  • Like actor files, please keep these in alphabetical order by name of institution


What is the purpose of having separate file folders for actors and institutions? Will putting something in the wrong category negatively impact my research?


Having separate actor and institution files allows you start making connections between actors and institutions that might not always be noticeable right away. For example, a conference document may have the names of various actors (i.e. the people, groups and organizations, government departments and corporations) involved in your story that you now create separate actor files for. Conversely, with actor files, you will want to examine if an actor sponsors a meeting, conference or other area of decision-making, and therefore requires an institution file as well.

That said, please DON’T WASTE YOUR TIME trying to figure out if something should be categorized as an actor or institution. In the long run, what matters most is that all the relevant dates are listed in your chronology. You will eventually be reviewing and reading all your documents, and that means that sooner or later, your important documents will turn up no matter where they have been filed.

Writing down what you know

  • In addition to creating file folders, you should also write a summary of what you already know about your issue
  • Informally writing something down (whether it is a couple of paragraphs or ten pages) will allow you to identify what is interesting, missing or questionable about your public issue
  • Don’t be discouraged if you discover you don’t know as much as your thought as part of this process is about uncovering gaps in your own knowledge
  • Writing things down will also allow you to identify elements of your issue that might not be open to discussion or amenable to research

Have a look at the 15 ways to beat procrastination here

News Media Search


Where can I locate News Media Sources? Aren’t they “yesterday’s news” (so to speak) in today’s digital world?


Don’t be fooled into thinking that newspapers aren’t relevant for research. Much of the information you want is likely to be from previous years when newspapers were the public record. Even today, a lot of information that was once found exclusively in print media is now found on online news sites that theoretically, are “newspapers” regardless of their mode of distribution.

Surveying Documents

  • Surveying documents in advance of your research will let you know what material is likely to be available and accessible to you, what information is unavailable, what sources of information might be useful to you down the road and what will be the baseline of your research
  • You are essentially looking for sources of information for later research
  • This information is critical to know so that you don’t waste your time searching for material you will never be able locate or get access to
  • Perhaps most importantly, a document survey will likely play a huge role in how you shape or frame your research question


I am confused. How can a document survey shape my research question? I already know exactly what issue I want to explore!


It is never a good idea to be certain of your research question before you actually start looking for materials. For example, say you want study water policy in the Philippines. While reading academic literature is good start, you are also going to have to search for original documents, government and World Bank reports, minutes of community council meetings, and research currently being done in the Philippines. Well, while you might be able to easily access the World Bank records and the international organization reports in English, think about what you can’t access either because it is written in another language or can only be located within the country itself. This might include those council meeting minutes or Philippian government reports! If all of this material is now unavailable to you, you might have to re-think your research question. For example, go back to the material you can explore and this may help you find a new angle for your research. Thus instead of examining water policy in the Philippines, you may now be focusing on the role that outside institutions play in Filipino water policy instead.

Surveying Sources of Good Documents

1. Academic Scholars and Professors
  • A potentially valuable source of info are those individuals who either teach about or write academic literature about your subject.
  • Professors and academic researchers are likely to know which actors and institutions are involved in your story and what books are articles will be useful for your research.
  • They much also have their own collection of documents or be proficient in a foreign language.
  • Course reading lists may also provide you with names, dates and documents you need to look at.

Link to Academic research sites – academic resources

2. Bibliographies and References
  • Book and article bibliographies, report references and meeting participant biographies may also provide you with a great source for names, dates and documents
  • You should also survey graduate students’ theses and dissertations (especially the bibliographies) for primary and secondary source materials
3. Encyclopedias
  • Encyclopedias often contain peer-reviewed articles written by experts that will provide you with an overview of your topic and normally cite sources for further material
  • They are likely to contain many names, dates and events which are relevant to your research
  • In addition to traditional encyclopedias such as Britannica and Encarta, there are many topical ones (with online availability) such as: Ullmann’s Encyclopedia of Industrial Chemistry, American Medical Association Complete Medical Encyclopedia, Human Environments: A Cross-Cultural Encyclopedia, Medline Plus and the Encyclopedia of Gender and Information Technology


Is Wikipedia a valid encyclopedia?


Be very cautious and selective when using consulting Wikipedia. Wikipedia often gives you too much information or information that has not been verified. Your goal now is to find materials you will later use in your research. This means focusing on things like names and dates and not the story or explanation of the public issue you have chosen.

Link: Wikipedia explained

4. Internet subject searches
  • The Internet is simply a mechanism for search and delivery and while there is lots of good information available, it is often hard to find or incomplete
  • That is why an internet subject search should not be your first line of research as they often take a lot of time and often distract you from your goal of document survey now and research later
  • Don’t forget that much of the information needed for research on public issues is not available on the web, and there are many useless websites which will only cause you to waste time
  • Even when a government, corporate or advocate website has a search engine, don’t forget that there will be many documents that never find their way onto the website, especially considering info on the web is often changed continuously
  • Internet aggregators of data (i.e. Web of Science, Investopedia, Google News) can be sources of potentially useful information but they are only as good as the sources of information they use

Link: Helpful explanation of internet research from Mount Allison University

5. Blogosphere and User-Generated Materials
  • Since you are simply surveying information, tread carefully on these sites as you are likely to be drawn into a world of both reliable and unreliable information
  • Carefully read blogs and other social media only after your initial survey is done
  • Good sites include (a listing of popular blog sites), (legal blogs), (financial info), or (environmental blogs)
6. Libraries
  • Even though digitized collections are more readily available than ever before, on-site library searches are often necessary when you are searching for dated materials since many documents are still only available in hard copy
  • Often, a library search will include both a digital search as well as spending time “shelf-walking” as you look for documents
  • Reference librarians that will be a very valuable source when searching for primary documents
  • Libraries will often have access to archives and special and rare book collections as well but you often need an appointment or special permission to access your materials (see below)
  • There are many different kinds of libraries such as public libraries, government documents libraries, business libraries and law libraries


What is shelf-walking? Which libraries are good for shelf-walking?


Shelf-walking simply means you walk through the shelves of a library looking for documents that may have not made it on any list or search engine. For example, a government documents library will often organize material by jurisdiction (local, provincial, national, etc…) and at each level, the materials from a department or agency will be filed together. Say you want to find materials published by Health Canada. You will shelf-walk looking for reports and unpublished material that might not be on any official publications list. In addition, you might also find older materials, and even publications from now defunct departments and agencies. Please keep in mind that you cannot use a government documents library well unless you know the structure and organization of government during the relevant time. This can be done by consulting the reference librarian. It is also very useful to shelf-walk business libraries, as they often have volumes of information on various business organizations. Since only you know which reference volumes might be useful in your research, shelf-walking is critical for allowing you to peruse various titles on the shelves.

7.  Special Collections
8.  Archives
  • Archives are specialized libraries that deal with a wide assortment of documents kept on file that are not likely to be found elsewhere
  • All publications (reports, laws, treaties, manuscripts, old literary and religious texts, historical materials, etc…) are eventually archived and are available in person or on-line
  • In some cases, if you cannot go to the archives and the material is not available on-line, they may provide a service that will send you digital copies of the relevant material for a fee
  • Examples include:
    Archives Canada
    Ontario Archives, United States National Archives
    UK National Archives
    UNESCO’s archives portal
9. Research Centres and Think Tanks
  • Research centres are often connected to universities and can be found by visiting university websites
  • These centres are highly specialized fields (i.e. Caribbean Studies) that may have their own documents room or libraries which contain published and unpublished materials that are not listed in the library system
  • Examples of US think tanks include the Brookings Institute, RAND Corporation, and Woodrow Wilson International Centre for Scholars

Following Up


Why can’t I carefully read the materials I find during my document survey? Won’t it be a waste of time for me to go back and collect the actual information itself?


Often, when conducting your document survey, you are going to be really tempted to not only locate where your materials can be found, but to then find, read and file the actual documents at the same time. It will seem like you are wasting time if you have to go back and collect the actual information but this not the case at all. On the contrary, unless you have the somewhere to put your documents, you will lose track of each one. Without an overview, it is going to be hard to focus on what needs to be found, and without a media search, you have no foundation to build your research on. Finally, without a document survey, you won’t really know where the best sources of information can found. Once you have identified the best sources of material, you can follow-up by knowing where to look for materials to add to your files and chronology.

Case Study

The Birth of the Canada Council for the Arts

The Canada Council for the Arts is a federal agency mandated to encourage and support research in the arts.  The Council is not a federal department, but a semi-autonomous agency that exists at “arm’s length” from governmental control, despite receiving annual federal appropriations since the mid-1960s.

Researching how and why the Canada Council for the Arts was created presents a number of unique challenges.  Firstly, the agency changed its name (ever so slightly) in the 1990s, adding “for the Arts” to its original English-language name.  Record searches would, of course, have to be oriented towards the original name the agency was created with.  Additionally, in the 1970s, the mandate of the agency was changed, with a significant portion of its responsibilities being hived off to a new federal granting agency responsible for supporting social sciences and humanities research.  Considerations of the logic and pressures involved in creating the agency would have to consider the additional portions of the agency’s mandate that no longer apply.  Complicating research further, scholars have not tended to focus on the birth of the Council, meaning that researchers cannot depend on existing published research to lead them to the relevant materials.  Finally, unlike research dealing with ‘contemporary’ events, where a significant volume of government records are often digitized and publicly-available, most of the records generated during the creation of the Canada Council have not been digitized, and in many cases require special applications to be submitted in order to gain access to them.

My plans to investigate how the Canada Council was created, and why, first of all required basic familiarity with how decisions are made by the federal government.  A basic inquiry, such as consultation with the annual report of the Canada Council (for the Arts), indicated that the agency was created by federal legislation in 1957.  This discovery opened a meaningful point of entry into the historical records.  Federal legislation was accessed through consultation with the (Revised) Statutes of Canada.  Additionally, because legislation must be debated and voted on by the House of Commons and Senate and because both bodies produce written records of their debates, commonly referred to as Hansard (a name referring to the firm that originally produced verbatim records of the debates of the British House of Parliament), the Statutes for 1957, and the Hansard records for the same year were searched.  However neither hasnot been digitizesr so were consulted through a library that had a decent collection of Canadian government documents.

Consultation with the Canada Council Act (1957), while indicating the original conception for the agency, did not say much about the ‘how and why’ of the Council’s creation.  The Hansard record of the debates involved in passing the Council legislation was very informative, however, in outlining some contending visions for the role the agency was expected to play within the country, and the rationalization members of the government chose to give in support of creating the agency.  Consultation with Hansard also showed that the government announced the idea of creating the Council in the throne speech that opened the 1957 session of Parliament, and show that the idea was, in part, a response to the work of a royal commission, which recommended the creation of such an agency in 1951.

Royal commissions or their equivalents in other countries tend to produce significant volumes of written material, and depending on the breadth of their mandate and activities, provide an excellent indication of who was interested in the issue the commission was mandated to consider, and what the interests of these groups and individuals were.  This is certainly true of the Royal Commission on National Development of the Arts, Letters, and Sciences, which sat from 1949 to 1951.  Most academic libraries have a copy of the commission’s report.  In fact, this commission’s work has been of such interest, and was so voluminous, that its report, as well as a good sampling of briefs submitted to the commission, have been digitized by Library and Archives Canada located in Ottawa.  It would have been a mistake to forget about the materials that are not digitized, however.

Library and Archives Canada is the official repository for government documents, it is the most logical place to begin looking for primary documents.  Like most large archives, it has a digitized search engine that helped ascertain what the archive were available.  This search engine did not include a comprehensive list of every possible document within the archive, though.  Using it made it possible to gain a general sense of what was available.  For instance, I found the Council’s archival materials, and the records related to the Royal Commission, as well as materials on the Canada Council dating from the early 1950s up to the present that resided in the files of a number of areas of the government.  These included the Finance and Justice Departments, as well as the Cabinet, Prime Minister’s Office), and Privy Council Office .  The document collections related to each of these areas is usually referred to as ‘fonds’ (e.g., the Privy Council Office fonds, the Canada Council fonds, etc.).  Library and Archives Canada, as well as some other archives, produce individual documents related to each fonds, known as ‘finding aids’.  These provided a clearer record of what was included within the fonds, allowing more focus the research plan.  The finding aid indicated whether the records were open to the public, and if not, the basic procedure to be used to apply for access.

Some records, such as the minutes of Cabinet meetings, have been digitized up to the 1970s.  This saved a significant  amount of time, but did not, unfortunately, include the background documents (memos to Cabinet) that Cabinet members used to make their decisions.  These documents are available in hard copy at the Library and Archives.  Documents from the Prime Minister’s Office and Privy Council Office required application to the Access to Information Program officers at the Library and Archives.  Each application required submission of a small fee, and as clear a description of the requested files as possible (e.g., the name and number of a particular file or set of files; a query for something as general as ‘records related to the creation of the Canada Council’ will very likely generate a curt response from the ATIP office asking you to narrow your search).  As the Archives receives a multitude of Access to Information Program requests every day, it can sometimes take up to six months to receive a response, and one should not expect a response for at least eight to ten weeks.  The Canada Council historical records are generally open to public scrutiny, but the finding aid for the Council fonds had not been digitized.  A paper copy of the finding aid was available.

When researching the creation of the Canada Council, a visit to the Archives  in Ottawa proved essential.  Visits were made much more valuable by advance communication with Archives staff, and my early submission of a request for documents that I wanted to consult. Almost all of the records are not held in the research area of the Library and Archive, and required transportation from a building in the suburbs of Ottawa.  However, when an order was placed at least a week in advance, the desired items were made available on at the Archives.

Memos provided to Cabinet by relevant ministers and their assistants named important organizations pressuring the government, and indicated certain reservations and hopes members of the government had for the agency.  Similarly, correspondence between the Justice Department, Privy Council Office, and Prime Minister’s Office was very revealing.  Written in a far more informal tone than ‘official’ documents that were intended for public consumption, these letters offered unique insights into the thinking of various members of the government and civil service regarding the Canada Council.  Some of these letters also included draft versions of legislation, often with informative, hand-written marginalia, annotations, and revisions.  Other documents held in these collections provided insights into the type of pressure being applied by non-governmental institutions and individuals, and in some cases, records of the letters, briefs, and phone calls made by groups and individuals attempting to influence the government’s thinking.

Despite how valuable a search of official government records is when researching public policy development, it would have been a mistake to think that this avenue would provide everything needed for the research.  Research into government records provided  a list of prominent organizations and individuals, within the government, civil service, and outside of the government entirely, who seemed to have a stake in this particular debate.  Consultation with the records related to these institutions and individuals produced very informative materials that had not made it into the most obvious files related to the policy decision of interest.  For instance, in the case of the creation of the Canada Council, consultation with the records of various arts and education advocacy groups, as well as prominent government ministers, produced draft versions of policy briefs, personal letters communicating the results of meetings and personal conversations, as well as strategy planning on the part of organizations.  These documents were extremely useful in reconstructing a nuanced foundation and fuller context for policy deliberation.  Some non-governmental records were located at the Library and Archives, while others required investigation of the holdings of other archives.  The bibliographies and footnotes of related secondary studies were invaluable for identifying the archival whereabouts of non-governmental and personal deposits of papers.

                                                                                                        Greg Klages

Once you have read this chapter, you should be able to:

  1. Highlight the differences between primary, secondary and tertiary documents
  2. Classify a document as a primary source based on the criteria described in this chapter
  3. Make judgement calls when assessing the reliability of your primary sources
  4. Create and develop a filing system using the three categories of chronological lists, actor and institutional files
  5. Write a summary of what you know about your research topic to date
  6. Perform a news media search and use this information to add to your research files
  7. Recognize the difference between surveying sources versus carefully reading documents for information
  8. Identify different sources of information and determine the benefits and drawbacks of each method of research


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s