How does NRS calculate readership?
The National Readership Survey estimates readership of newspapers and magazines by interviewing a representative sample of the population. The Survey is conducted on behalf of NRS by its appointed research agency, Ipsos-RSL. Every year, NRS interviews a sample of over 35,000 individuals aged 15+ in Great Britain. Interviewing takes place throughout the year, and is divided into 12 monthly assignments. Click here for more

What is the NRS definition of readership?
The NRS interview starts by asking which of these publications have you read or looked at for at least 2 minutes in the past 12 months? This provides the broadest definition of readership - Read Past Year, or RPY. However, RPY alone is of limited value, and NRS goes on to ask about readership in more detail. Click here for more

Does NRS measure overseas copies?
The NRS measures only readership within Great Britain, and primarily covers publications whose distribution is concentrated in this country. Therefore, copies distributed outside Great Britain are not included in the scope of NRS. Click here for more

How is social grade calculated?
Social grade is extensively used to define readers and define target audiences for advertisers. It has been in existence for over 50 years, but is still one of the most effective ways of distinguishing between one sector of the population and another. Social grade is based on the occupation of the Chief Income Earner (CIE) of the household. Click here for more

Does NRS measure newspaper sections?
NRS currently provides average issue readership estimates for some 43 newspaper supplements and sections, covering magazine and reviews, listings guides, personal finance sections and business/company news sections. Click here for more

How can readership go up if circulation is going down, and vice versa?
Circulation generates reading opportunities - 'readers per copy' - but this does not mean there is a fixed relationship between circulation and readership. It is possible for circulation to go in one direction and readership in the other, particularly if the focus is on short-term losses or gains. Over the long term, there is usually a correlation between the trends in a publication's circulation and its readership, but because NRS is an estimate based on a sample, the fluctuations in readership are usually more marked than those in circulation. Click here for more

How reliable is the NRS data?
The objective of NRS is to provide publishers and agencies with the best possible estimates of readership for newspapers and magazines. NRS is recognised as a world leader in readership research, with a thorough and rigorous attention to technique.
However NRS, like any survey based on a sample of the population, has its limitations. If NRS was to publish known levels of readership rather than estimates, it would have to interview everyone in the population, and that is clearly an impossibility. The solution is to interview a carefully selected sample of the population, designed to be as representative as possible of the population as a whole. This allows NRS to produce estimates of the number of readers of a publication, and the profile of that readership. Click here for more

How up to date is the NRS data?
The NRS is a continuous Survey. The full database is available to subscribers four times a year, containing rolling 6 months' and 12 months' data for periods ending March, June, September and December. Advance data is usually available to subscribers about 6 weeks following the end of the fieldwork period, with the full data available about a week later. This schedule will be revised for data periods ending June 2004 through to March 2005 due to NRS testing the Personalised Media List on half the sample. Click here for more

How can I get hold of NRS data?
Top line data for all publications is freely available on the open access section of www.nrs.co.uk. However, in order to benefit from the full wealth of NRS data, you have to be a subscriber. If you are an advertising agency, you need to subscribe via the IPA; if you are a magazine publisher, via the PPA; a newspaper via the NPA. Any other type of company or organisation should contact NRS directly. Click here for more

How does NRS calculate readership?
The National Readership Survey estimates readership of newspapers and magazines by interviewing a representative sample of the population. The Survey is conducted on behalf of NRS by its appointed research agency, Ipsos-RSL.
Every year, NRS interviews a sample of over 35,000 individuals aged 15+ in Great Britain. Interviewing takes place throughout the year, and is divided into 12 monthly assignments. A number of sampling points is selected throughout the country, and within each of those sampling points addresses are selected at random from the Postal Address File. Interviewers visit those addresses, and select individuals aged 15+ to interview, using a random procedure. Of all the individuals selected, currently 52% complete the interview.
Interviewers conduct the interview in respondents' homes using DS-CAPI - double screen personal interviewing. The interviewer has a laptop on which the interview is programmed: the laptop is radio-linked to a second screen which is shown to the respondent. The interview program automatically dispatches the relevant image to the respondent's screen.
The NRS interview asks which of some 280 newspapers and magazines respondents have read for at least two minutes in the past 12 months. To make the interview as concise and efficient as possible, respondents are initially shown a sequence of 47 screens each containing six publications, and asked to identify those screens which contain one or more of the publications they have read. The interviewer can then return to those screens and ascertain which individual publications the informant has read.
Following the readership questions, there are further classification and lifestyle questions that tell us more about the respondent, helping NRS users to profile the readership of newspapers and magazines.
All this data is weighted up to population estimates by region, sex, age and social grade. Therefore, if 10% of the NRS sample reads The Daily Planet, NRS estimates that it is read by 10% of the population. As the profile of the weighted NRS population matches the known profile of the Great Britain population, the readership profile of The Daily Planet reflects as closely as possible the correct ratio of Midlands to Scotland, men to women, age 15-24 to age 25-34 etc.

What is the NRS definition of readership?
The NRS interview starts by asking which of these publications have you read or looked at for at least 2 minutes in the past 12 months? This provides the broadest definition of readership - Read Past Year, or RPY. However, RPY alone is of limited value, and NRS goes on to ask about readership in more detail.
Firstly, there is reading recency: respondents are asked when they last read an issue of that publication other than today, and this is coded as yesterday, the last 7 days, the last 2 weeks, the last 4 weeks, the last 2 months, the last 3 months or longer ago.
Recency provides the basis for Average Issue Readership - AIR - the estimate of the number of people who read any given issue of a publication. This is defined as anyone who read that publication within the interval of time between one issue and next i.e. read yesterday for a daily newspaper, read in the past 7 days for a Sunday newspaper or weekly magazine, read in the past 4 weeks for a monthly magazine, the past 2 months for a bi-monthly magazine and the past 3 months for a quarterly magazine.
Secondly, there is reading frequency: respondents are asked how often they read that publication, and this is defined as almost always (at least 3 out of 4 issues), quite often (at least 1 out of 4) or only occasionally (less than 1 out of 4). Frequency provides the basis for Cumulative Readership, the estimated number of people who will be exposed to 1,2, 3, 4, 5, 6 etc. issues of that publication.
Lastly, there are the 'quality of reading' questions: source of copy and how disappointed. Source of copy asks how the most recent issue was obtained (e.g. newsstand, home delivery, postal subscription, passed on by someone else, work copy, read in public place) and whether it was obtained primarily for the informant, or for someone else. This is used to define Primary Readership and Secondary Readership. NRS then asks respondents how disappointed they would be if for some reason they were unable to obtain a copy of that newspaper or magazine: very disappointed, quite disappointed or not at all disappointed.

Does NRS measure overseas copies?
The NRS measures only readership within Great Britain, and primarily covers publications whose distribution is concentrated in this country. Therefore, copies distributed outside Great Britain are not included in the scope of NRS.

How is social grade calculated?
Social grade is extensively used to define readers and define target audiences for advertisers. It has been in existence for over 50 years, but is still one of the most effective ways of distinguishing between one sector of the population and another.
Social grade is based on the occupation of the Chief Income Earner (CIE) of the household. This information is collected by interviewers during the course of the interview. The allocated social grade is based on the job title, and factors such as the type size of the establishment, number of staff responsible for etc. In very broad terms the definitions are:

Social Grade	Social Status	CIE's Occupation
A	Upper Middle Class	Higher managerial, administrative or professional
B	Middle Class	Intermediate managerial, administrative or professional
C1	Lower Middle Class	Supervisory or clerical and junior managerial, administrative or professional
C2	Skilled Working Class	Skilled manual workers
D	Working Class	Semi and unskilled manual workers
E	Those at the lowest levels of subsistence	State pensioners or widows (no other earner), casual or lowest grade workers

Does NRS measure newspaper sections?
NRS currently provides average issue readership estimates for some 43 newspaper supplements and sections, covering magazine and reviews, listings guides, personal finance sections and business/company news sections. Respondents are shown screens for each of the newspapers they claim to read, containing a representation of that newspaper's sections. The current scope of the measurement is separate 'standalone' sections, and does not include those sections that consist of integral pages within the main part of the newspaper. Any attempt by NRS to extend the scope of the sections it measures will require an examination of alternative methods of collecting that information from respondents.

How can readership go up if circulation is going down, and vice versa?
Readership and circulation are two separate and different measures. Circulation is a know: the average number of copies of a publication old over a period of time, independently audited and verified by ABC, the Audit Bureau of Circulations. Readership can never be known, it has to be estimated, and it is the task of NRS to provide the best possible estimate, an estimate acceptable to publishers and agencies alike.
Circulation generates reading opportunities - 'readers per copy' - but this does not mean there is a fixed relationship between circulation and readership. It is possible for circulation to go in one direction and readership in the other, particularly if the focus is on short-term losses or gains. Over the long term, there is usually a correlation between the trends in a publication's circulation and its readership, but because NRS is an estimate based on a sample, the fluctuations in readership are usually more marked than those in circulation.
There are factors inherent in the measurement technique that may affect the relationship between readership and circulation. Although ABC data is more transparent and detailed than it has been in the past, the published figures may still conceal changes in methods and type of distribution. NRS readership estimates like any estimates generated by a sample survey are subject to a degree of uncertainty, so sudden changes in readership may not all be due to 'real' events, but are a product of sampling variation (see 'How reliable is the NRS data?' ). NRS has to constantly evolve to meet the changing needs of its markets, and any change in methodology is also likely to have some impact on the readership estimates.
There are also factors largely in the control of the publisher that may affect the relationship between readership and circulation. Editorial changes can narrow or broaden the appeal of a publication to secondary readers. Special promotions can have a similar effect, particularly in those cases which involve coupons being collected and redeemed. If a person who previously read someone else's copy now buys their own, circulation will go up but readership remains the same. Secondary readers -those who don't buy their own copy - are particularly susceptible to competitive activity. If a secondary reader of publication A is persuaded instead to read publication B, then publication A loses no circulation, but it does lose readership.
Lastly, there are demographic and social factors that can affect the relationship between readership and circulation. For example, there has been a decline in the average household size which restricts opportunities for pass on readership within the household. In the case of households with young adults living with parents, the incidence of pass on readership may have been reduced by the decline in reading newspapers and magazines amongst the young.
All these factors - and others - can change over time, and effect individual publications or genre of publication differentially.

How reliable is the NRS data?
The objective of NRS is to provide publishers and agencies with the best possible estimates of readership for newspapers and magazines. NRS is recognised as a world leader in readership research, with a thorough and rigorous attention to technique.
However NRS, like any survey based on a sample of the population, has its limitations. If NRS was to publish known levels of readership rather than estimates, it would have to interview everyone in the population, and that is clearly an impossibility. The solution is to interview a carefully selected sample of the population, designed to be as representative as possible of the population as a whole. This allows NRS to produce estimates of the number of readers of a publication, and the profile of that readership.
Inevitably, there is a degree of uncertainty attached to all the figures published by NRS, and this uncertainty is known as the sampling variation. It is possible, using statistical theory, for NRS to calculate the degree of sampling variation that is probably involved in any particular case. In turn, this allows NRS to apply standard statistical tests, either to say how confident it is about a single readership estimate, or to calculate the odds against the difference between two figures being nothing more than a chance occurrence.
NRS publishes the margin of sampling variation for all publications based on the 95% confidence level. For example, the Daily Planet has a published estimate of 5,800,000 average issue readers, with a margin of + or - 200,000 at the 95% confidence level. That means that there is only a 1-in-20 chance (5% = 100%-95%) that the readership of the Daily Planet is lower than 5,600,000 or higher than 6,000,000. The margin of sampling variation is relatively greater the smaller the sample base of the publication or the audience under scrutiny.
Similar tests can be conducted to determine whether the difference in the readership between one publication and another, or between one NRS period and another, is statistically significant i.e. is the difference likely to be a 'real' one, or the result of sampling variation.

How up to date is the NRS data?
The NRS is a continuous Survey. The full database is available to subscribers four times a year, containing rolling 6 months' and 12 months' data for periods ending March, June, September and December. Advance data is usually available to subscribers about 6 weeks following the end of the fieldwork period, with the full data available about a week later. This schedule will be revised for data periods ending June 2004 through to March 2005 due to NRS testing the Personalised Media List on half the sample.

How can I get hold of NRS data?
Top line data for all publications is freely available on the open access section of www.nrs.co.uk. However, in order to benefit from the full wealth of NRS data, you have to be a subscriber. If you are an advertising agency, you need to subscribe via the IPA; if you are a magazine publisher, via the PPA; a newspaper via the NPA. Any other type of company or organisation should contact NRS directly.