Library and Archives Canada
Symbol of the Government of Canada

Institutional links


Archived Content

This archived Web page remains online for reference, research or recordkeeping purposes. This page will not be altered or updated. Web pages that are archived on the Internet are not subject to the Government of Canada Web Standards. As per the Communications Policy of the Government of Canada, you can request alternate formats of this page on the Contact Us page.

Audits and Evaluations

Condition of Archival Records in Federal Institutions (CARFI)

2.1 Participation Rate – General Counts for Questionnaire Distributed and Completed

Although LAC originally planned to survey 210 Records Disposition Authorities covering archival records in 61 institutions, some RDAs had to be dropped and others needed to be replaced because the Authorities were no longer in force, they had been superseded by a newer Authority, responsibility for the records had been delegated to another GC institution not included in the sample, the archival records had already been transferred to LAC, or no archival records were identified in the Authority because the Authority gave permission to destroy specified non-archival records. (See Annex D for a List of Problematic RDAs.)

The CARFI project held information/launch sessions in order to answer any questions and to encourage participation prior to distributing the on-line version of the survey. On 26 February 2007, 173 surveys were distributed electronically to 54 Government of Canada institutions. Seven of these 54 institutions did not participate. In the end, 104 surveys were received from 47 institutions - a response rate of 60%. Of these 58% contained useable answers. (See Annex E for CARFI Surveys Completed.)

The institutions were given four weeks to submit their completed questionnaires on-line; the deadline was subsequently extended to April 13. Up to 7 May 2007, LAC accepted completed questionnaires in Microsoft Word or in hard copy in the hope of encouraging institutions to complete what they had started.

The CARFI project had aimed for a response rate of 80% or better from the GC institutions who were included in its sample. Reasons for non-participation were given by the five institutions (Endnote 6) that informed LAC in writing of their decision not to participate in the survey, as well the remaining GC institutions that were sent the on-line version were asked to complete a post-survey evaluation form that requested comments on the survey process in general, the ease of completing the questionnaire and the information asked for in the questions, as well as for any additional comments. The reasons for not participating cited by the five institutions included staff shortages, year end commitments and time constraints. However, the majority of the post-survey evaluation respondents answered that they had sufficient time to complete the survey, the survey was of value to their institutions, and they would participate in future surveys of this kind. A further discussion of the participation rate is found in the Findings and Observations, section 2.4 of this report. (See Annex H for a copy of the post-survey evaluation form and a summary of the responses.)

2.2 Responses to Survey Questions

The questionnaire was divided into three parts:

  • Part I General Requirements for Recordkeeping
  • Part II Specific Criteria for Records Storage, and
  • Part III Requirements for Specific Media.

A comments box appeared at the end of each part to give participants a chance to explain their responses and/or comment on the questions. Annex C is a table which summarizes the counts of completed CARFI surveys, broken down by part and section as well as location.

In terms of what proportion of the survey was completed by GC institutions, the following information is of note. LAC received 98 answers to the first part on recordkeeping. For Part II, the survey asked institutions to provide answers for each of their three (3) top locations where archival records are stored. We received 96 responses for the first location, 36 for the second and 15 for the third. Part III was divided into 7 sections, representing each media type as indicated below:

  • Paper
  • Electronic
  • Microform
  • Photographs
  • maps, plans and technical drawings
  • Audio-video
  • Art, and
  • Motion picture film

We received insufficient responses for art and motion picture film. There were few valid RDAs to choose from in the stratified sample and thus we were counting on a small number to provide us with sufficient information to draw conclusions. Unfortunately, we cannot provide any comments on the condition of archival records for those media types. However, we have adequate data to comment on paper, electronic, photographic, audio-video, maps, plans and technical drawings, and microform archival records.

The majority of the data (about 80%) can be derived from the first location where archival records are stored since the figures indicate that most records are stored with the responsible institution on site in a headquarters building. This suggests that institutions store large amounts of archival records in expensive office space. Considering all three locations where archival records are stored, only about 10 % of archival records are stored in Regional Service Centres across Canada. Less are stored in private facilities such as those operated by Iron Mountain.

Overall response rates for the survey were about 60% which was lower than expected but which still allowed for analysis of issues with regard to recordkeeping and storage and the identification of various risk factors. Response rates by characteristics of the authorities were sufficient in most categories. Response rates were lower for older (pre-1993) authorities and those where the status indicated that they needed to be replaced. It may have been that institutions had difficulty in locating these records and could not complete the questionnaire. With regard to the various media types the results should be used with caution since the number of responses is often quite small.

See Annex F for Response Rates by Characteristics of Authorities.
See Annex G for CARFI Survey Questions and Results.

2.3 Interpreting Survey Results

Based on about 100 responses from a population of approximately 1000 valid RDAs the standard error of the sample is about 4.5%. This means that we can be 95% confident that the true values lie within an interval that is plus or minus 9% of the estimated values. In practical terms if 16% answered yes to a particular question then we are 95% confident that the true value lies between 7% and 25%. This does not of course apply to the media specific questions where far fewer answers were provided. In this area our confidence in the results would be much lower and should be treated as notional only (for example, more said yes than no) and not as statistically significant.

How has risk been defined for purposes of the CARFI survey?

  • Risk is potential for realization of unwanted, adverse consequences to (collections) archival records. (Endnote 7)
  • Two aspects of risk identified in the CARFI Survey are:
    1. Recordkeeping-Intellectual Control
    2. Storage Conditions and Environments-Physical Management

Can the CARFI survey results be used to identify risk factors affecting archival records in federal institutions?

For the purposes of our analysis, the CARFI Working Group identified from a larger series of questions approximately 20, in relation to both recordkeeping and storage, where negative answers would indicate risk to archival records. These questions were then grouped under six factors that would contribute to increased risk. The table below shows the six factors and their risk rating.

Risk Rating
1) The archival records cannot be transferred to LAC and they stay indefinitely in the GC institution LOW
(2 questions)
2) Problems with maintaining control over the archival records in the GC institution MEDIUM
(7 questions)
Risk Rating
3) GC institution is not prepared to handle potential incidents HIGH
(2 questions)
4) Storage environments are not designed for the preservation of archival records HIGH
(5 questions)
5) Incidents where potential damage and loss have incurred MEDIUM
(2 questions)
6) More than one media type is stored in the same storage container MEDIUM
(1 question)

Note the highest risks are in emergency preparedness and recovery (factor 4), as well as storage environments not designed for preservation (factor 3). Respondents were asked to answer questions in section II and III of the survey for up to 3 locations. Results indicated that the bulk of records are stored in headquarters buildings, primarily in the National Capital Region.

It is important in practice to distinguish between identifying potential risks to archival records and cases where damage had already occurred or where damage was narrowly averted. There were some survey questions that were designed to ascertain if archival records had been damaged as a result of various types of incidents. Once these were identified it was found that there was no correlation between these incidents and unfavourable conditions or risk factors that could lead to potential damage of archival records. For example, no correlation was found between institutions that reported water damage to records and the storage conditions that one would expect to heighten the risk of water damage (such as records stored in rooms with overhead pipes, etc.) The frequency of incidents was reported at around 10%, and the incidents in which archival records were damaged were even lower.

Importantly, analysis of the survey responses indicates that there is no single factor that contributes to a risk but a combination of factors; nor is there any overall pattern that would lead to an easy identification of records at risk. However, the CARFI survey responses resulted in a rich data source from which we can identify areas of concern. In the event of an incident, the survey results indicate that the greatest potential risk for damage to archival records will occur because GC institutions are not prepared to handle potential incidents (54% for factor 3). The next highest source of risk to archival records is from storage environments that are not designed for the preservation of archival records (35% for factor 4). See Annex H for a complete set of questions identified as indicating risk factors and the risk rating scale for these risks.

Can the CARFI survey results identify those archival records which are now at risk in GC institutions?

  • From the survey results we cannot identify those specific records in specific GC institutions that are at risk, but we can identify types of records that are potentially at risk.
  • For reasons of confidentiality of survey responses, the identity of individual institutions is not linked to responses.
  • Since the survey was based on a sample of RDAs in government institutions the methodology does not allow for a comprehensive list of GC institutions or RDAs in which records are at risk.

What types of archival records are potentially at risk in GC institutions?

  • Electronic Records
    • There is a lack of electronic corporate filing in GC institutions.
    • There is a continued reliance on the paper record as the official corporate record.
    • There is inconsistent migration of electronic data in systems containing archival records.
  • Paper Records
    • There are preservation issues, arising from the low use of permanent paper in paper records. These records will require treatment to ensure long-term preservation once transferred to LAC
  • Photographs (nitrate film and colour film)
    • There are preservation issues, arising from the quantity and improper storage of nitrate film and colour film. These records will require treatment to ensure long-term preservation once transferred to LAC.

2.4 Findings and Observations

Findings and observations are based on analysis of the survey responses and fall under the following categories: General Observations; Recordkeeping; Record Storage; and Media Specific.

General Observations

2.4.1 While a significant portion of the archival records are in institutions with high standards and practices, some are clearly at risk and others face enough risks to suggest that there is significant room for improvement.

As a result of an analysis of the various risk factors identified from the survey questions it was found that in about 25% of the cases where archival records at low risk for both recordkeeping and storage factors (see table below):

Levels of risk to archival records
(number and % of RDAs)
Recordkeeping   Storage
Low Med High Total
Low 24 (25) 22 (23) 12 (13) 58 (61)
Med 10 (11) 12 (13) 5 (5) 27 (29)
High 1 (1) 7 (8) 1 (1) 9 (10)
Total 35 (37)   41 (44)   18 (19)   94 (100)  

It can also be seen that a slightly higher proportion (28%) faced high risks with regard to either one factor or the other, with the remainder (47%) facing predominantly medium levels of risk.

2.4.2 Overall response rates to the survey were moderate (60%) which was sufficient for analysis of recordkeeping and storage issues, however, results for the various media types should be used with caution.

Despite efforts to get a higher response rate to the survey through the participation of an advisory group, an official launch that was well attended, an on-line survey, answers provided for those who stored records at Regional Service Centres, and an extension to the deadline to include surveys that had been completed in other formats, the overall response rate was only 60%. While this permitted analysis of the larger questions on recordkeeping and storage it was more problematic once the responses were divided into the various media types and this data should be used with caution.

2.4.3 Gathering information on older authorities was challenging suggesting that many are out of date and need to be replaced or revoked.

While about 40% of the original sample included older (pre-1993) Records Disposition Authorities, it turned out to be difficult to achieve this proportion of responses. Further investigation of the continuing validity of these agreements demonstrated that many were no longer valid because the records had already been transferred, the agreements no longer identified archival records, or the custody of the records no longer resided with the institution named in the agreement. As a result, many could not be replaced in the sample. This situation was compounded further by lower response rates for this group of RDAs than for the more recent ones perhaps because the institutions themselves were having difficulty in locating the records associated with the agreements. As a result, only 21% of the survey responses involve older agreements.

2.4.4 Processes do not appear to be in place to protect archival records once they cease to be of business value.

While there was no response to any particular question that led to this finding, there were indications in a number of areas that the treatment of archival records benefits from the fact that they are normally of business value and this treatment declines once they are no longer so. The fact that so many archival records are stored in headquarters locations suggests that this is the case. These records do over time make their way to areas of these buildings where they are exposed to various risks (basements, unclean areas, water pipes, etc). With various media types it was found that procedures were in place to keep the records accessible, however, there were seldom any procedures to look after their long-term preservation. Some examples below include the lack of processes to refresh and migrate electronic records, the lack of playback equipment for audio-video records and the number of paper records that continue to be photocopied.


2.4.5 The overall situation with regard to recordkeeping is mixed, with some aspects such as applying RDAs and setting retention periods well done and others such as capture and management of records not so.

Of those who responded to the survey, many (84%) indicated that they had no problems in applying the Record Disposition Authority in question and that retention periods had been set (83%) for the archival records. In other areas the situation was not so good. Only 59% of the archival records were captured in a single records management system standard to the institution while a slightly higher proportion (68%) was captured in a single classification system standard to the institution.

2.4.6 Those responsible for handling and storage of archival records in institutions are often working without up-to-date guidelines or are not well trained.

In only 65% of the RDAs do the institutions have up-to-date documented guidelines for storage and handling of the archival records. In the remaining cases the guidelines either do not exist or do not reflect current practice. Those who access the records are not always trained. In only 58% of the cases are 80% or more of the records management personnel trained, and in only 22% of the cases are other personnel trained.

Record Storage

2.4.7 Archival records are most often stored in buildings that were designed for people and not records.

Based on answers provided to the survey questions it was concluded that most archival records (about 80%) are stored in buildings which are not designed to house records exclusively but are primarily working offices. Often these were the headquarters buildings of the various institutions. Aside from issues of the cost of space it should be pointed out that the storage of records in the midst of people and equipment means that the needs of the records often come last. This is illustrated by such responses that show that physical security of the buildings is strong while emergency response plans, training in and testing of such plans, and recovery priorities for archival records are weak. As well, HVAC systems tend to be run during working hours which satisfies the needs of staff but can cause temperature fluctuations for records. Records are often stored in locations that workers don't want to occupy such as basements and areas exposed to plumbing (34% indicated that over 80% of their records were exposed to potential water damage). Finally there are signs that records storage areas are approaching full capacity (in fact 16% indicated that the capacity has already been exceeded).

2.4.8 Specific incidents (such as pests and contaminants as well as water damage) while not at an alarming level occur frequently enough to cause concern.

While the number of reported incidents was in the acceptable range (about 10% for insects, rodents, mould, and water damage, less for light and fire), they do occur with enough frequency to cause concern. To date, few of these incidents have resulted in the loss of any significant number of archival records but if they continue to occur it is only a matter of time, given the number of records exposed to potential water damage in particular. Related to this finding and the previous one is that storage areas are cleaned less than monthly in 24% of the cases and food and drinks are allowed in the storage areas in 60% of the cases.

2.4.9 While most institutions do monitor temperatures in the storage areas many do not measure humidity levels.

The questions regarding the monitoring of temperatures and humidity levels were asked at several points in the survey with regard to storage conditions and the treatment of various media types. In general, temperatures are monitored and fall within acceptable ranges, however, it became clear from the pattern of answers provided that humidity levels are simply not monitored on an on-going basis and it cannot be concluded whether they fall within acceptable ranges or not.



2.4.10 While most paper archival records are in folders and filled properly they are not boxed, and originals are often used to make copies.

Responses to the survey revealed that several good practices are in place such as storing paper records in folders and filling those folders properly, however, it was also found that the folders are seldom stored in boxes on shelves (only 28% said that over 80% of the records were stored this way) and that making photocopies is the most frequently used method of copying (86%) the records for use.

2.4.11 A surprisingly high portion of archival paper records is not on permanent paper.

In response to the question about the portion of archival records printed on low-acid or permanent paper some 45% indicated that less than 20% of the archival paper records were printed on such paper. Although this is balanced somewhat by others (31%) who indicated that over 80% were on permanent paper, on the whole it would appear that less than half of all paper archival records are printed on permanent paper, which could result in significant preservation issues when those records are transferred to LAC in the future.


2.4.12 While many electronic archival records are stored on servers and backed up off-site, many institutions also reported that there were no procedures in place to migrate/refresh these records.

Based on responses to the survey it would appear that the vast majority of archival electronic records are stored on servers, backed up on servers off-site and in the same format as the originals, all of which are positive results. One troubling result, however, is that 75% indicated that they had no procedures in place to refresh and migrate these electronic records to ensure that they continue to be readable. While 90% indicated that they believed the records could still be read there is still cause for concern in today's rapidly changing environment.

2.4.13 While the widespread presence of ERDMS was encouraging most archival records still remain outside these systems.

While most respondents (65%) indicated that they have an Electronic Records/Document Management System (ERDMS) in place they also indicated that staff do not file their archival records there. As a result it is estimated that less than half of the archival records are currently in these systems.


2.4.14 Microform records are usually stored in appropriate containers but equipment is not well maintained.

Respondents indicated that microform records were stored in appropriate containers (78% indicated that over 80% of records were in appropriate containers). The institutions also indicated that viewing equipment was available (89%), however, those with viewing equipment did not clean and maintain it regularly (only 37%).


2.4.15 A surprisingly high portion of photographs are on nitrate film

One of the most surprising findings was the one-third of those institutions with archival photos indicated that the photographs were on nitrate film, and the majority of these indicated that nitrate film represented more than 10% of the holdings of archival photographs.

2.4.16 The high proportion of colour film held in institutions requires special storage conditions that are not currently being met.

Based on responses from the survey it is estimated that approximately 66% of archival photos are in colour. Long-term preservation of colour film, however, requires storage conditions which are much colder than normal room temperature (18-24 degrees), a condition which is not currently being met by any of the institutions in the sample.

Maps, Plans and Technical Drawings

2.4.17 Most of the archival records in this category are supported and/or covered and are rarely handled for access.

Respondents have indicated that a maps, plans and technical drawings are stored in a variety of ways, the most frequent being rolled (53%). They are however supported (60%) and covered (70%) and rarely handled for access (60%).


2.4.18 While these records are often stored in containers and under appropriate conditions, the existence and on-going maintenance of playback equipment is an area of concern.

The majority (64%) of respondents with archival audio-video records indicated that a significant portion of these records (over 80%) are stored in protective containers, however, 29% also indicated that they do not have procedures to ensure that the records remain accessible which is further supported by the response that 36% do not have playback equipment, and of the 64% that do, 44% indicated that they do not clean and maintain this equipment.

Table of Contents