Home > Recreational > Canada > 1990 Survey of Recreational Fishing in Canada > Methodology

Recreational

Canada

1990 Survey of Recreational Fishing in Canada

Due to the administrative variability in licensing anglers across Canada, different approaches were used to conduct the survey across the 13 jurisdictions which participated in the survey. This appendix provides general information regarding the methodological procedures associated with the 1990 survey. Detailed information at the jurisdictional level will be provided on request. 

1. Sampling Procedures

In each jurisdiction, essentially two surveys were conducted - one covering residents and one covering nonresidents. Residents were defined as those anglers living in the jurisdiction under study. Nonresidents were defined as visitors from other jurisdictions in Canada or foreign visitors. Except for residents of Ontario, Quebec and Newfoundland, resident and nonresident samples were selected using stratified, systematic random sampling of licence bases with stratification based on licence category. In British Columbia (freshwater) and in the Northwest Territories, licence strata were further stratified by the management area in which the licences were sold. The sample sizes were determined from the reliability estimates for days fished for each stratum as calculated from the results of the 1985 Survey of Recreational Fishing in Canada.

Due to limitations in the licence bases available for residents of Ontario, Quebec and Newfoundland, sampling for the populations of anglers was conducted by using household samples of the general population. The households in each province were geographically stratified. The required sample of angling households was determined in consultation with representatives of each province using information on the distribution of angling households in the 1985 survey in combination with specific statistical criteria acceptable to each province. The overall sample size was then determined using the proportion of angling households to all households in each geographic area as estimated for 1990. The final sample size for residents in each province was restricted by funding availability, however, the proportionate distribution of households determined in the first phase was maintained.

Households were pre-screened by telephone to determine eligibility for inclusion in the survey. Profiles were obtained by determining for each selected household the age-group (under/over base age for "adult" anglers in the province) and sex of all individuals residing in the household. It was then further requested for those individuals identified as "adults", how many in the household fished for recreation in 1990. In households where anglers were identified, one was selected at random to respond to the questionnaire. For verification purposes, the selected respondent was asked for the number of days fished in 1990.

The actual pre-screening operations were handled by private research firms under joint contract with DFO and the respective provincial governments. The results of the pre-screening were provided to the responsible agency and, subsequently DFO for weighting procedures. Mailing labels for selected respondents to the survey were also provided in triplicate for the field work phase of the survey.

2. Field Procedures

The recommended field work approach was to send a first mailing in early January. This was to be followed one week later with a reminder card. After approximately three more weeks each agency was to assess the level of returns and determine an appropriate date for a second complete mailing to nonrespondents. The final cut-off date was based on the level of returns coming in from the second mailing.

Because the survey was conducted across 13 licensing jurisdictions, strict adherence to a specific schedule was not possible. Most jurisdictions had completed the field work phase by late April. Delays occurred in both Ontario and Quebec with pre-screening operations due to the selection process used for assessing bids from research firms. In each of these provinces, field operations began as soon as pre-screening results were received and mailings could be prepared. These delays resulted in the cut-off dates being extended in Ontario to late May and in Quebec to late August.

3. Manual Editing and Coding

Variations across questionnaires as well as specialized identification requirements dictated that all manual editing and coding be conducted by each participating agency. In some cases, all that was required after assessing completeness of the questionnaire was identification coding on each document. In others, far more involved coding was required ranging from special codes for sub-components of questions to geographic codes to allow for sub-provincial/territorial data analysis.

A major requirements of this phase was to verify that the information provided by anglers was both reasonable and sound for the jurisdiction involved. This included assessing species caught by area as well as the availability of the specific species, determining whether or not the catch levels indicated were within acceptable limits, assigning missing value codes where most primary information had been provided, cross-checking administrative lists to eliminate substitutions (someone responding other than the designated respondent), etc.

Once questionnaires had been completely assessed and coded, they were shipped to Ottawa for analysis. Problems in shipping occurred in only one province, New Brunswick. One of the boxes shipped did not reach Ottawa and, as a result, nonresident trout licence-holder returns were lost.

4. Data Processing

Each jurisdiction forwarded their completed documents to each mailing to the Surveys Unit in Ottawa for processing. Prior to datacapture, all documents were verified by Unit staff to determine if any should be removed due to incomplete coding, minimal information, etc. Datacapture procedures were written for each jurisdiction and were sent to Dominion Computer Support Services for programming. The company then keyed all documents with critical questions subject to key-verification.

All data from the survey was processed on an HP3000-917LX, using SPSS 3.0. Programming was done by Unit staff and consisted of three systematic editing procedures, development of substitution algorithms for missing data, preliminary data estimates for assessment by the participating agency staff, final data programs covering all data collected and special output programs for publication purposes.

5. Weighting Procedures

All output programs were weighted to reflect population estimates. This procedure for most jurisdictions where licence sales were known was a simple application of inverse weighting by stratum (population of licence-holders divided by resultant sample). For residents of Quebec, Ontario and Newfoundland a more complex procedure was required.

First, it was necessary to estimate the total population by age-group and sex in each stratum based on the pre-screening data. Household weights, determined from the estimated households in each stratum divided by the sample of households pre-screened, were used to generate estimates of population. Of primary interest were the estimated populations of "adult" anglers by sex. The definition of "adult" varied with Ontario using 18 years of age; Quebec, 15; and, Newfoundland, 18. Using the pre-screening data, it was possible to determine the proportion of anglers in each stratum.

The estimates of population in each stratum of interest were then compared to official statistics on population provided by each of the provinces. Population data was based on the 1986 census with post-censal estimates to June, 1990. The initial estimates of population and the respective estimates of anglers in each stratum were adjusted to reflect these statistics. Respondent weights for survey estimates were then derived using standard inverse weighting functions.

6. Post-Survey Adjustment Procedures

All participants were provided with the opportunity to request adjustments to survey results provided such adjustments were based on identifiable errors or changes required due to revised administrative information. While most jurisdictions had few adjustments, others required a considerable number of changes. Final revisions to the data were not completed until November 1993.

The area of primary concern to most jurisdictions was catch and retention of specific species by geographic area. Most errors were a result of incorrect identification of species and the changes were relatively straightforward. Another area where significant adjustments were required was refinement of population estimates. Most changes were made based on updated information on licence sales, however, in the three provinces where population samples had been used, a considerable number of refinements were requested to ensure that the estimated number of anglers was accurately reflected. The procedures for Quebec, in particular, were extremely detailed. Final estimates of the number of anglers in the province were derived from revised weighting procedures based on age-sex groups by administrative region, as estimated by the Quebec Bureau of Statistics.

7. Statistical Reliability

While the 1990 Survey of Recreational Fishing in Canada was designed to provide reliable information on angling activity in the country, it is not possible to provide tables of statistical reliability for all estimates. Two tables are provided at the end of this appendix in which the sample sizes for each province and the reliability of key variables are presented. Since differing methodologies were used for the survey, reliability has been estimated as if all information had been developed from samples selected using simple random sampling. Because all sampling was based on stratified, systematic random sampling, these estimates of reliability may be considered as conservative.

Information on reliability is presented in terms of the coefficient of variation of the mean, at one standard deviation. Because exact estimates of variance cannot be calculated, the coefficient of variation of the mean is used as a proxy for the assessing the variability of the data. As can be seen from the tables, the smaller the jurisdiction (and by extension, a smaller sample), the higher the variability of the data collected. Also true is that there may be higher or lower variability depending on the type of data being considered. For example, the number of days fished tends to have lower variability than the total amount spent on major purchases merely because the range of days fished is quite narrow whereas expenditures have extremely wide ranges depending on the purchases made. Similarly, direct expenditures tend to have lower variability since the expenditure patterns are directly related to the fishing activity which occurred.

The ranges of coefficient of variation below provide guidelines as to the use of data:

It should be noted that under Statistics Canada's guidelines, data with CV's in excess of 33.5% would not be published.

¹ coefficient of variation=(standard error of the mean / mean) * 100

¹ coefficient of variation=(standard error of the mean / mean) * 100