The part of an email message or webpage that is visible without scrolling. Material in this area is considered more valuable because the reader sees it first. Also refers to a printing term for the top half of a newspaper above the fold. (IAB)
A summary measure made by adding two or more separate measures.
Ascription is a model that is used to achieve a complete data set. The technique ascribes missing data by assigning responses from respondents who did not complete the product questionnaire (but completed readership) by “marrying” it to a respondent with similar demographic characteristics who did.
A single piece of information known about a user and stored in a behavioural profile which may be used to match ad content to users. Attributes consist of demographic information (age, gender, geographical location), segment or cluster information (auto enthusiast), and retargeting information (visited Site X 2 days ago).
The total number of people or households usually expressed within a defined geographic area, who may receive an advertising message delivered by a medium or a combination of media.
The distribution of a publication as certified by a recognized independent industry circulation auditing body.
The demographic makeup of people represented in an audience with respect to income group, age, sex, geography, etc.
The measured overlap of potential exposure between different issues of the same publication or among issues of different publications.
Audience intelligence (AI)
The use of first- and third- party data to determine an advertiser’s audience.
Potential customers modelled after an advertiser’s 1st party data (usually data from their customers who visit and make purchases from their websites). Attributes of the advertiser’s customers are matched against a larger audience, creating a pool of highly targetable and ‘pre-qualified’ users. Some companies refer to this also as ‘pre-targeting’.
The most common average is the arithmetic mean. This is computed by adding a group of values together and dividing by the total number of values in the group.
Average age is based on all respondents who provided their age/ year of birth. Age was estimated for those who did not provide their year of birth by asking them to select an appropriate age range. The estimate was attained by assigning them to the mid-point of that range. The average age reflects the 18+ population as only adults are interviewed in the study.
Average number of times a consumer has been exposed to an advertisement in any medium that is part of an advertising campaign.
This is calculated by multiplying the total number of responses in each income range by the mid-point of the range and dividing by the total population. It does not include those who have no income. This is available for both household and personal income.
Average Issue Audience/Average issue readership (AIR)
An estimate of the number of people who read or looked into a surveyed issue of any given publication within its publication interval e.g. within the last seven days for a weekly publication.
A defined universe upon which analysis will be done, for example:
a) All people 12+, Total Canada
b) Women 18+, English Ontario
Data related to specific users as well as their historical patterns of interaction with media and advertising content.
A study conducted to obtain an originating “point in time” measurement. A benchmark is usually conducted prior to a change in stimuli. The results are then used as a standard for comparison.
Measurement effects which are the result of survey design or implementation. Also called “systemic error”.
Currently North America’s largest generation consisting of people born after World War II (1945-1965).
A measure or indication of the readiness with which a brand springs to mind (MRA).
A person who has abandoned the survey of their own free will. This is more common in self-administered surveys like online surveys than it is in interviewer assisted surveys.
Standard size/traditional newspaper.
Computer Assisted Personal Interviewing (CAPI) is an interviewing technique in which the respondent or interviewer uses a computer to answer the questions.
Is a Computer Assisted Telephone Interview. Responses are keyed directly into a computer and a specifically assigned program checks for invalid responses and will not accept responses outside prescribed limits, thus avoiding editing and keying in of data.
A Computer Assisted Web Interview (CAWI) is another name for online (or web) surveys.
A performance-based advertising model where payment is dependent upon an action that a user performs as a result of the ad. The action could be making a purchase, signing up for a newsletter, or asking for a follow-up call. An Advertiser pays a set fee to the Publisher based on the number of visitors who take action. Many affiliate programs use the CPA model. (IAB)
Also called Pay Per Click or PPC. Cost of advertising based on the number of clicks received. A performance-based advertising model where the Advertiser pays a set fee for every click on an ad. The majority of text ads sold by Search engines are billed under the CPC model. (IAB)
CPM (Cost per Thousands)
Advertisers’ cost per thousand readers exposed to a campaign or an advertisement. If the cost of the advertising campaign is $10,000 and the readership is 1,000,000, the CPM is $10.
1,000 X $10,000 ÷ 1,000,000
A repeat telephone call to a potential respondent for their participation in a survey, or a repeat call to someone who participated in an interview to correct an error.
Related to sampling, a cell typically represents the number of respondents who make up a sub-group (e.g. region) of the overall sample. The cell size constitutes the base which determines the error margins for the results of that sub-group; the larger the cell, the more reliable the results.
Cell Phone Only Households
Persons living in households with only cell phones (no landline access).
An official, usually periodic, enumeration of a population that systematically collects and records a variety of data from a representative sample of respondents.
Census agglomeration (CA)
Area consisting of one or more neighbouring municipalities situated around a major urban core. A census agglomeration must have an urban core population of at least 10,000.
Census metropolitan area (CMA)
Area consisting of one or more neighbouring municipalities situated around a major urban core. A census metropolitan area must have a total population of at least 100,000 of which 50,000 or more live in the urban core.
A census survey seeks to obtain information from all respondents in a given universe. In survey research, a representative sample of respondents is normally interviewed because this is far less costly, while providing results with an acceptable level of accuracy.
The average number of copies per issue of a publication that are (1) sold through subscription; or newsstands or news boxes; (2) distributed free to predetermined recipients, carried within other publications, or made available through retail stores or other outlets.
Classified advertising (Print)
Print advertising that is limited to certain classes of goods and services, and usually limited in size and content.
Click-through rate (CTR)
The rate (expressed in a percentage) at which users click on an ad. This is calculated by dividing the total number of clicks by the total number of ad impressions. For example, if an ad is displayed 100 times and is clicked on 2 times, that ad has a click-through rate of 2% (2/100). (IAB)
A classification of demographically or attitudinally similar groups into homogeneous groups. Each group represents distinctive lifestyles patterns and offers a basis for segmenting the market.
This has two distinct meanings: one in the market research area and the second in the media analysis area. Both relate to the practice of recording questionnaire information on the computer in a coded form (e.g. 1= Yes, 2= Brand A).
In market research, coding refers to the stage of data entry at which questionnaire answers are assigned a code.
In media analysis, coding refers to the process of translating the requirements of a media crosstab into computer language using the codes from the above.
Community newspapers are non-daily newspapers published at regular intervals under a common title and whose editorial content serves the information needs of a small geographic community.
Questionnaires that are completed through a pre-determined question sequence and are included in the final data set for the study.
Statistical measure of the number of times out of 100 that test results can be expected to be within a specified range. For example, a confidence level of 95% means that the result of an action will probably meet expectations 95% of the time.
Confidence limits define an upper and lower boundary within which the value of a particular population estimate is expected to lie. For example, the statement ” the mean is estimated to be x + y% at the 95% confidence limit ” can be interpreted to mean that if an experiment was repeated 20 times, the estimated mean of the population would lie between x + y% and x – y% in at least 19 of the 20 times.
Study of how people behave when obtaining, using and disposing of products and services.
A group of recruited survey respondents who have agreed to take part in surveys.
Data related to the content and context of the specific web page where advertisement is run.
Advertising on a website that is targeted to the specific individual who is visiting the website. Natural context places an ad on a relevant site such as a bank ad on a finance page. Contextual advertising scans the text of a website for keywords and targets ads based on those keywords. These ads can be text or images. Often referred to as “Targeting”.
The part of a publication’s circulation that is sent free or mainly free to individuals with specific qualifications, or within a geographic area.
These correct the sample composition to household statistics derived from Census Canada data. Conversion weights are sometimes also referred to as Adjustment weights.
A small amount of data sent to a computer’s browser from a web server and stored on the computer’s hard drive.
The efficiency of a media vehicle as measured by the relationship of audience, either potential or actual, to cost and expressed as a cost-per-thousand.
Cost rank (Crank)
Ranking of publications by their coverage, cost per thousand, audience composition, etc., against a defined target group.
Percentage of number of individuals in a specific target group/geographic area reached by an advertisement or a publication or combinations of publications.
Cross-Tab (X-Tab)/ Cross-Tabulation
A table which shows the number and/or percentage of respondents and their projection in the population who gave various answers to a question in a survey. The table can show the answers simultaneously for various sub-groups of respondents and characteristics.
The number of different people who read or are exposed to any medium, vehicle or group (schedule) within a defined population. Also known as Net Unduplicated Audience.
The net reach achieved by a publication over a given period of time. A mathematical model is used to calculate what proportion of the target group has the opportunity to see at least one advertisement over this period. (NRS)
A newspaper published daily, usually being unrestricted in its contents to a subject field, and containing actual information.
Gathering sought-after information, most often through the use of questionnaires, interviews and the implementation of sampling procedures to select respondents.
Data management platform (DMP)
A centralized system for gathering first-party data, integrating with third-party data, and applying this data to one’s advertising strategy. Advanced DMPs offer users the ability to create custom segments, forecast segment volumes, sync segments with other sources, overlay advanced analytics, and are often integrated with or part of DSP platforms.
Once interviewing has been completed, the information from the questionnaires must be put into a form suitable for rapid analysis. Data processing entails coding of open-ended questions (if any), data transfer, and programming and computer analysis
With regard to Vividata, a bank of information containing responses to questions asked by researchers.
Day-after recall (DAR)
A research method that tests consumers’ memories the day after they have seen an ad to assess the ad’s effectiveness.
Population or consumer statistics regarding socioeconomic factors such as age, income, sex, occupation, education, family size, etc.
These correct intentional disproportionate area or individual sampling or intentionally biased selection procedures. The design weight corrects for the over-representation of households in specific major urban communities (area disproportionality) such as Toronto and Montreal. Additionally, it corrects for the over-representation of households in DAs that have relatively high income levels. Disproportionality in both cases is pre-determined, and the number of DAs selected is derived from specific household counts in each sub-stratum prior to selection
The design or pre-weight also corrects for under or over-achievement of completions in each sub-stratum. While the theoretical sample design is based on completing between 6 and 7 interviews in each primary sampling unit, some variance inevitably occurs. Unless the expected completion rate in each of the high and low-income sub-strata is achieved, adjustments are made to bring the sample into line with the theoretical targets.
The specific individual in the household who is selected to be interviewed.
Where the basic identity and content are similar to the printed edition but the articles and advertising may differ. Free, unrestricted public access to a website does not qualify as a paid digital edition.
Where the advertising and editorial content exactly match the printed publication.
Promotions that permit or request consumers to directly respond to the advertiser, by mail, telephone, email, or some other means of communication.
The deliberate use of different sampling rates for various strata such as high income neighbourhoods. This can sometimes be used to reduce sampling error or to assure minimal numbers of respondents who might otherwise rarely appear in the sample.
Dissemination area (DA)
The dissemination area is a small, relatively stable geographic unit composed of one or more blocks. It is the smallest standard geographic area for which all census data are disseminated. DAs cover all the territory of Canada
Duplicated readership is derived through a cross tabulation of readership measures between two or more publications in a market to determine the percentage of the market who read two or more publications.
A geographical area comprised of a market and adjacent counties or census divisions as defined by Statistics Canada.
EMC (Extended Market Coverage)
Flyers distributed to non-subscribers of daily newspapers. Delivery is houses and/or apartments and one day a week, usually Friday or Saturday.
Measures of reader engagement indicate the readers’ relationship with a publication, and/or how that publication is read; for example Time Spent Reading. Also known as Quality of Reading. (NRS)
English Canada is defined as comprising all English speaking respondents in Canada including those who live in the province of Quebec, New Brunswick, Ottawa/Gatineau or Northern Ontario who claim to speak English “most often” at home.
Counting all units of the population as a basis for drawing a sample of those units; typically applied to a sample of Canada Census areas included in a door-to-door survey.
See Unduplicated readership.
Consumers who have seen a media vehicle, whether or not they paid attention to it.
The first three digits of the postal code (e.g. L6H) comprising of 4,000 to 20,000+ households. Urban codes A1 to A9, Rural codes A0.
Conforming to common sense or, depending on point of view, conforming to presumptions and prejudices.
A general term that refers to any data gathering process.
Focus group interview
A research method that brings together a small group of consumers to discuss the product or advertising under the guidance of a trained interviewer. This is a qualitative research method.
French Canada is defined as comprising respondents in the province of Quebec, New Brunswick, Ottawa/Gatineau or Northern Ontario who claim to speak French “most often” at home.
Number of times an average person or home is exposed to a media vehicle (or group of vehicles) within a given time period.
The number or percentage of individuals exposed to a publication or an advertising campaign 1 time, 2 times, 3 times, etc.
A set of numbers giving the relative value of different positions in the frequency distribution, i.e. of different numbers of insertions seen. More often known as a Response Function.
Audience that encompasses a medium’s total reach.
A sophisticated technique for integrating two independent databases while maintaining the integrity of both. Matches participants from one data source with another, based upon shared characteristics. It essentially creates a third database. (NRS)
The generation sandwiched between the “Baby Boomers” and the “Millennials” (Generation Y), sometimes referred to as the “Baby Bust” generation. This period between 1965 and 1984 is also known as a period of high divorce rates, working Moms and latchkey kids.
Geo-targeting allows Advertisers to specify where ads will or will not be shown, based on the visitor’s or searcher’s location, enabling more localized and personalized results. (IAB)
The combined audience of a combination of media or a campaign in a single medium. For example, if Medium A and Medium B have audiences of 700,000 and 600,000 respectively, their gross audience is 1,300,000. To convert from gross audience to net audience, one must subtract the duplicated audience.
The total number of exposures or opportunities to see an advertisement delivered by a media schedule. Net Reach x Average Frequency = Gross Impressions
Gross rating points
An aggregate of the total “ratings” of a schedule against the target group. % Reach x Average Frequency = GRP’s
Gross Audience expressed in percentage terms.
A summary of the total number of readership claims for a given group of publications. As some participants will read more than one title, gross readership may exceed 100%. (NRS)
A media rate that comes with a guarantee that the publication will achieve a certain circulation.
A form of response bias where a respondent carries an overall generalized positive or negative impression from one specific characteristic to the next, e.g. if a respondent considers a product to be excellent, then he or she is likely to rate the product highly on taste, appearance, and texture etc. (MRA)
Head of Household
Designated as the person who makes major financial decisions and/or who looks after any dependents in the household.
Refers to the language spoken most often or on a regular basis at home by the individual at the time of the census.
An occupied housing unit. An individual or group of individuals occupying a house, apartment, group of rooms, or single room.
The total number of households to which the sample is projected to represent. The household behaviour, rather than individual behaviour, would be used as the base for specific questions pertaining to television, grocery shopping, household products and services have and plan to buy, long distance expenditure, computer brand ownership, home improvements, automotive and household structure
Household Income (HHI)
Total annual income of all household members (including money received from all sources such as pensions, interest, dividends, as well as earnings.)
Household weight scheme
This scheme is designed to permit examination of household data by correctly reflecting the population of households in Canada. (It may not be used in conjunction with any personal characteristics.)
Household Usage of Products
Some parts of the product questionnaire contain questions relating to the use of products or services by the household (members) rather than a particular individual.
A measurement of responses from a web server to a page request from the user browser, which is filtered from robotic activity and error codes, and is recorded at a point as close as possible to opportunity to see the page by the user. If an ad is displayed 1,000 times, that is considered to be 1,000 impressions. (IAB)
Readership claimed to have occurred in the respondent’s place of residence.
The total number of completed responses tabulated in a survey.
The frequency of something occurring in the population. It usually refers to persons and is stated as a percentage (i.e. the percentage of people in the Canada, who ate at or visited a fast food restaurant in the past month). In marketing and opinion research, incidence is a measure of the level of effort needed to reach qualified or eligible respondents.
The percentage above or below the population average. A comparative measure used to point out the strengths or weaknesses in relation to a norm of 100%.
An advertisement, collection of advertisements or other promotional matter published by an advertiser or group of advertisers, to be inserted in a publication. It may be bound into the publication, or be inserted without binding.
Refers to an ad in a print publication.
Asking questions and recording answers via computer.
Publications which are issued other than on a regular daily, weekly, monthly, bimonthly, or quarterly basis.
A single copy of a newspaper, magazine or publication distributed at the same time and with the same data (date of issue).
A drawing that indicates the relative positions of the elements of an ad (e.g. headline, photo, logo, body copy, etc.).
Separating consumers into groups based on their hobbies, interests and other aspects of their lifestyles.
Margin of error is the + and – figure around the survey estimate that indicates the level of precision associated with that estimate. Error margins represent the statistically estimated range of accuracy at a given level of confidence within which results may be trusted. All other things being equal, for random samples, the larger the sample size, the smaller the margin for error.
The geographic area or areas in which a research project takes place. A market may be a census division (CMA, CA), a census subdivision (PMA, CSD), a province or territory or a sub-market.
A summary of the characteristics of a market, including information of typical purchasers and competitors, and often general information on the economy and retailing patterns of an area.
The systematic and objective identification, collection, analysis, and dissemination of information for the purpose of improving decision making related to the identification and solution of problems and opportunities in marketing.
The percentage of a product category’s sales, in terms of dollars or units, obtained by a brand, line, or company.
The number of people in the chosen target market.
Part of a page devoted to the official heading/name of the publication, usually positioned at the top of the page.
The sum of all items divided by the number of items. Also referred to as average.
A plan designed to select the proper demographics for an advertising campaign through proper media selection.
A two-dimensional graph profiling the media or product usage of demographic groups.
A number used to modify the measured readership of a publication. Can be applied to individual probabilities or to aggregate statistics. Used to reflect changed publication circumstances (e.g. circulation changes) or qualitative judgements of a publication’s performance (e.g. based on creative or other editorial factors).
The middle number in a sequence of numbers. Also referred to as midpoint.
The midpoint on a frequency distribution.
Medium (plural, Media)
A vehicle or group of vehicles used to convey information, news, entertainment, and advertising messages to an audience. These include daily newspapers, television, magazines, radio, etc.
A description of the way in which data is collected for part or all of a research project.
Used to determine the cost effectiveness of advertising in a newspaper; reached by multiplying the cost per agate line by one million, then dividing by the circulation. Also referred to as Milline.
Millennials (aka Millennial Generation or Generation Y)
Millennials are the demographic cohort following Generation X. There are no precise dates when the generation starts and ends. Researchers and commentators use birth years ranging from the early 1980s to the early 2000s. (Wikipedia)
Advertising which is aimed at a National Market as opposed to Local Advertising.
The proportion of the target group who have seen at least one of a given group of publications. (NRS)
Net unduplicated audience
The number of different people exposed to at least one of two or more insertions in one or more vehicles. Those people exposed to more than one insertion are only counted once.
A type of non-sampling error caused by some sub-groups of the sample responding less than the rest of the sample.
A symmetrical bell-shaped statistical distribution where most of the examples in a set of data are close to the mean or “average” while relatively few examples tend to be one extreme or the other.
An open-ended question allows for recording verbatim responses (e.g. “What do you think about the death penalty?”). These may be coded subsequently. Regular interview questions typically have a prelisted, straightforward response option (e.g. ‘yes/no’) which allows for pre-coding.
OPPORTUNITY TO SEE (O.T.S.)
Readership surveys usually measure exposure to a publication, not to particular pages within those publications. Thus ‘reading’ a publication gives an ‘opportunity to see’ the ads carried.
A technique for building the “optimum” schedule for a given target market. Needs a response function (or frequency weights) to calculate effective reach for alternative schedules. Rarely used today due to the difficulty of defining the ‘right’ frequency weights.
Over-sampling is used in order to increase the representation in the sample of individuals belonging to a specific group.
A group of recruited survey respondents who have agreed to take part in surveys.
A person who reads a publication that they or a member of their household did not purchase, subscribe to or receive in a primary manner.
The proportion of the target group who read a particular publication. Also known as Coverage or Reach. (NRS)
The percentage of a medium’s total audience that is part of a specific group. Example: if there are 100,000 women who read Magazine A, and 50,000 are 18 to 34 years old, then 50 percent of the total audience is composed of women 18 to 34.
In print media: the total audience of a publication as a percent of the total population. Or, the circulation of a publication as a percent of total homes. In broadcast media: the number of homes that are able to receive a signal of specific strength, but which do not necessarily tune to the station(s).
The number of people in the target market or the survey universe.
The total number of people to which the sample has been projected.
Primary market area (PMA)
A newspaper or related publication’s major area of editorial and advertising coverage. This represents a geographically defined area based on Statistics Canada boundaries.
Primary sampling unit
The units at the first stage of a multi-stage sample. In Vividata, these are the Census Dissemination Areas (DAs) defined by Statistics Canada.
Principal grocery shopper
Those who state that they are personally responsible for grocery shopping within the household are classified as Principal Grocery Shoppers. This allows for the possibility of male or female principal grocery shoppers and joint male and female responsibility.
Principal wage earner
The main wage or income earner in the household.
A measure of how likely it is that something will happen or that a statement is true. Probabilities are given a value between 0 (0% chance / will not happen) and 1 (100% chance / will happen). (NRS)
A method of identifying consumers by the amount of product usage, usually categorized demographically or psychographically.
The way readership breaks down across a single variable such as age, sex or social grade, e.g. what proportion of readers are aged 15-44 versus 45+. Profile data should add to 100%. (NRS)
The total number of people to which a sample size is projected to represent the population of a group of people.
Customized questions that are asked in a research survey and are paid for by the individual newspaper, advertiser or advertising agency. The answers to these questions are provided only to the individual newspaper, advertiser or advertising agency that paid for the questions.
A method of deriving from the known characteristics of measured publications, the audience or readership characteristics for a non-measured publication. Some databanks allow these “prototyped” estimates to be added to those of measured publications in a survey and through a “model” or random duplication to generate reach/frequency estimates for media combinations.
A more sophisticated form of demographics that includes information about the psychological and sociological characteristics of media consumers, such as attitudes, values, emotional responses and ideological beliefs.
Psychographic segmentation is based on traits, attitudes, interests, or lifestyles of potential consumer groups.
A notarized statement made by the publisher regarding total circulation, geographic distribution, methods of securing subscriptions, etc. These are issued between audited statements.
The time interval between issues of a publication. (NRS)
A method of research that emphasizes the quality of meaning in consumer perceptions and attitudes; for example, in-depth one to one interviews and focus groups.
Qualitative readership measures
In Vividata, questions are asked on the degree of involvement with the publication issue under study, including number of occasions looked into, time spent reading, and the degree of interest in the publication in general.
A method of research that emphasizes measurement of incidence of consumer trends within a population.
The printed or computer form or instrument used to ask specific questions in a survey.
A type of sampling where quotas are set for certain cells or demographics.
A division of a specified population into five equal groups, based on an ordered array of a specified quantity such as number of magazines read, income, number of ½ hours of television viewed. The order is from high (Quintile 1) to low (Quintile 5).
Sample in which every potential respondent in a given universe has an equal chance of being selected.
Random sample (stratified)
Population sample that requires the population to be divided into smaller groups called ‘strata”.
Information cards, provided by print media, which contain information concerning advertising costs, mechanical requirements, issue dates, closing dates, cancellation dates, and circulation data, etc.
In television, one percentage of all TV households who are viewing a particular station at a given time. In radio, one percentage of all listeners who are listening to a particular station at a given time.
Both instances vary depending on time of day.
Reach or net reach
The total number of adults reached by the schedule one or more times. Gross Impressions / Average frequency = Reach or Net Reach. Reach percent is defined as the percentage of the target group reached by the total number of insertions in the schedule. GRP’s / Average Frequency = Reach %.
The total audience reached by a media vehicle using successive issues. Also used to describe the total reach of a campaign or media schedule.
Defined as the number or group of qualified people who read a particular newspaper or magazine. For example, Vividata readership estimates constitute the average-issue-audiences or average-issue-readership (AIR) for the publications in the study. This is based on those claiming to have “read or looked into” a specific issue.
An evaluation of degree of interest. Also used to describe an evaluation of the relative level of general interest in different types of products.
Readers per copy (RPC)
The average number of persons who read each copy of an issue of a publication. Calculated by dividing readership of a publication by its confirmed circulation.
A measure of how much time it takes a publication to accumulate its full audience. (NRS)
The (demographic) profile of a publication’s readership.
In Vividata, the number of issues of a publication usually looked into by a respondent is used to determine probabilities for estimating cumulative issue coverage.
The technique which establishes readership (including Average Issue Readership) by asking participants when they last read or looked at any issue of a particular publication. (NRS)
A sample split into equal parts such that each part (or replicate) is a matched random sample of its companion replicate. Analysis of the data for any replicate should fall within the expected confidence limits.
A sample that contains units in the same proportion as the population of interest.
In each market the newspapers are classified as either resident or non-resident. Resident newspapers are the local daily newspapers in a particular market. Non-resident newspapers are newspapers from other markets that are available in the local market.
The person who is interviewed. No matter what type of survey is being conducted, the person being interviewed is always called the respondent.
When respondents are disinclined to continue participating in a research survey due to fatigue, which can lead to invalid responses.
The number of people in the survey universe represented by a respondent in the completed sample.
The number of completed interviews divided by the number of eligible units in a sample. Rates may vary depending on how partial interviews are considered and unknown eligibility is handled. The response rate for an online survey is usually the number of clicks on the URL divided by the number of invitations sent. For a mail or postal survey, it is the number of completed surveys returned divided by the number of deliverable pieces.
River sampling is an online sampling method that drives potential respondents to an online portal where they are screened for studies in real-time. Qualified respondents are then randomly assigned to a survey.
A method of ensuring survey issues are presented to respondents in close to random order to avoid potential bias due to questioning sequence.
Unlike a census, a sample constitutes a fraction of the universe under study (i.e. a group of respondents). It should entail a representative cross-section of the universe.
A sample in which every unit of the population has a calculable probability of being selected in the sample.
The actual number of people who respond to a question or group of questions. All other things being equal, the larger the sample or cell size, the lower the margin of error for the results (i.e. greater accuracy).
The probability distribution of sample estimates that is to be expected from a given probability sample design.
The error attributed to studying a fraction of the population rather than undertaking a census under the same general conditions.
Results from the survey of a sample of a population can be expected to differ from comparable results obtained by surveying the entire population. The expected differences are sometimes referred to as the sampling tolerance. Sampling tolerances are determined by sample size and by the homogeneity of the characteristics in the survey population. Other factors being equal, larger samples produce smaller tolerances; also, uniformly dispersed characteristics produce smaller sampling tolerances than “bunched” or clustered characteristics.
The sampling tolerance is usually expressed as a + around the survey estimate at a given confidence level.
Different values put on elements of the sample to restore them to the correct proportions within the population they represent.
A factor which adjusts a publication’s audience to account for changes in circulation or distribution. These factors may be used in computer analyses of Vividata. It should be noted that the use of “automatic scaling factors”, wherein a Vividata licensed software supplier provides an automatic circulation weighting on the database, is not permitted by Vividata.
In Vividata, respondents who answer “yes” or “maybe” to the screening question for a particular publication.
The portion of the interview that asks specific questions to determine the eligibility of the respondent.
Target “attributes” e.g. “employment criteria” are sometimes used to designate an eligible respondent within a representative sample. Only respondents meeting the screening criteria will be asked additional questions.
Search engine marketing (SEM)
A form of Internet marketing that seeks to promote websites by increasing their visibility in search engine result pages through the use of paid placement, contextual advertising and paid inclusion.
The process of dividing markets into groups of consumers who are similar to each other but different to the consumers in other groups.
Selected professional/business managers (SPBMs)
A Business Target Audience definition used by Vividata, based on responses to a series of questions regarding employment status, occupation and involvement in business.
Share of Voice
The percentage of all promotion in a specific product category that is accounted for by a particular brand. (MRA)
Refers to whether some research results genuinely reflect a population of interest in some way or whether the results could occur by chance. Statistical significance is determined by comparing the research results with the values defined by the confidence interval.
“Sub-Markets” are local markets or sub-divisions of a CMA. Sub-markets are weighted independently of the CMAs in which they are located. Sample size varies for each sub-market but is large enough to be analyzed on its own.
A sample of a sample, which may or may not be selected using the same approach as the original sample.
The methods and procedures used to measure people’s behaviour, attitudes or opinions in a scientific manner such that the results may be replicable. The broad area of survey research encompasses any measurement procedures that involve asking questions of respondents.
100% household coverage within a given area.
A size of newspaper that is roughly half the size of a broadsheet newspaper.
The process of placing a pixel on an advertiser’s website or search landing pages to “tag” users as having visited those pages so that they can be eligible for subsequent targeting/messaging.
Target market or audience
Group(s) of individuals to whom a communication is directed. Usually defined in terms of specific demographic (age, gender, income, geography, etc.), purchase, or ownership characteristics.
Time spent reading (Newspapers)
The average time spent is based on the number of minutes spent reading yesterday’s, Saturday’s or Sunday’s newspaper. Refused/Don’t know/Not Stated are excluded from the calculation. The total time spent is based on respondents who read yesterday’s issue or a weekend issue. It is calculated by multiplying each respondent’s time spent reading yesterday’s issue by the number of weekday issues read and adding the reading time for each weekend issue read.
Total net paid circulation
Total of all classes of a publication’s circulation for which the ultimate purchasers have paid in accordance with the standards set by AAM or CCAB. Includes single copy sales, mail subscriptions and specials.
The total number of qualified readers of an average issue of a publication.
A type of research study that follows the same group of subjects over an extended period of time. These types of studies track attitudes, awareness or buying habits.
Changes or patterns in attitudes and behaviours of a population. Trends can be measured by repeating a question to a representative sample of respondents over several successive surveys.
Unduplicated readership is derived through a cross tabulation between two or more newspapers or magazines in a market to determine the numbers of readers who read only a particular publication.
The population chosen for a research study. Universe estimates are the estimated number of actual households or adults from which the sample will be taken and to which data from the sample will be projected.
The actual number of respondents to a survey. Also expressed as sample size.
User (of a Product or Service)
The number of people who reported the use of a particular product or service within a specified time span.
A wave is a fielding period for a project in which there will be multiple time points in the fielding. Often results are compared from wave to wave.
A questionnaire that is displayed on a website.
In Vividata, this corrects the sample composition to statistics derived from Census Canada data.
Weighting is conducted at the data processing stage of a survey in order to restore target sub-groups to their proportion of the population under study. The weighting may be used to re-balance a sample subject to stratification, or to correct for various types of disproportionate over-sampling.
Working phone rate
The number of working or assigned residential telephone numbers as a proportion of the sample.
A readership technique in which respondents are asked whether they have read or looked into a publication on the previous day. Most commonly used in estimating average issue audiences of daily newspapers.