Types of publications, market definition, and questionnaire topics were the main issues discussed by the ComBase tri-partite committee on August 10.

The committee met in Toronto to discuss a variety of issues relating to ComBase, the principal readership research conducted on behalf of the community newspaper industry in Canada. The committee focused its main discussion on the pilot research project, which is scheduled to begin Sept. 10.

The purpose of ComBase is to provide consistent and accurate market-by-market information to assist in the buying and selling of community newspaper advertising space.

Preliminary results from the pilot study should be available in early 2002. ComBase aims to roll out the national study in 2002.

The discussion around publication types focused on what kinds of publications ComBase should measure. In particular, discussion centered on whether or not publications of interest should be measured.

ÔÇ£There was some disagreement on whether those types of publications should be measured ÔÇô the question centered on what we wanted ComBase to be,ÔÇØ said ComBase Manager Elena Dunn. ÔÇ£The agreement was that we should measure everything that is a newspaper, as long as it is published in English.

ÔÇ£We looked at the ethnic papers, but since we were only interviewing in English anyway, it seemed pointless to ask someone if they had read a Chinese-language paper, for instance,ÔÇØ she said.

Dunn said that the decision to include all English-language papers would have an effect on the length of each interview, and therefore on the response rate. She said people are more likely to respond to a shorter interview that a longer one.

The interview for ComBase is estimated to be between 13 and 15 minutes, which Dunn said is still within the range of being considered ÔÇÿshort.ÔÇÖ

The lengthier interview would also affect cost. ÔÇ£The more work you create for the research firm, the more expensive it is going to be,ÔÇØ said Dunn. ÔÇ£Longer interviews mean less respondents and more cost, but our supplier says were are still doing well on that basis.ÔÇØ

The entire pilot study, up until the implementation of the national study, is expected to cost almost $400,000.

Thomson-Lightstone has been chosen as the supplier to conduct the five-market pilot study. The pilot study markets will be Toronto (ON), Red Deer (AB), Yorkton (SK), Killarney (MB), and Campbellton (NB). The markets were chosen to represent a variety of population groups and geographical areas.

Dunn said that StatsCan defines clear geographies throughout Canada and studies such as NADbank define their markets based on those geographies. Market definitions therefore include Census Metropolitan Areas (CMA) ÔÇô markets with a population of 100,000 or more ÔÇô and Census Agglomerations (CA) ÔÇô markets with a population of between 10,000 and 100,000. Only about 140 markets in Canada fall into these two categories.

ÔÇ£In the smaller, rural centers, youÔÇÖre going to have a town of perhaps 1,000 people, but you are also going to have people driving into that town from as far away as 200 kilometres,ÔÇØ said Dunn. ÔÇ£Whatever community newspapers serve that town ÔÇô regardless of how big it is ÔÇô their geography is going to be different because they are covering more than just that town.

ÔÇ£And almost all of our members are in this type of situation, so the committee decided a couple of rules: one is that we canÔÇÖt go below the Enumeration Area (EA) ÔÇô we start with it and build up,ÔÇØ she said.

An EA, developed for electoral purposes, is defined as the route a single electoral officer can cover, whether it is 4,000 households in an urban core or 100 households in a rural setting. An EA is the smallest area available from StatsCan that has population projections.

ÔÇ£The second thing the committee decided was that where there was a NADbank study, we woul