Data Release Rules

The following rules determine, for each publication on the survey, the minimum base period for which PAMCo will publish readership estimates (i.e. six, twelve or twenty four months).

Publications added to the survey for the first time will be subject to these rules, but the addition of new publications sometimes raises unexpected issues. First estimates for publications new to the survey are therefore individually reviewed by the PAMCo Technical Committee and, where appropriate, by the PAMCo Board, prior to any new estimates being released.

Estimates for newspaper supplements are released on the same base as estimates for their parent newspapers, as per the standard Data Release Rules.

[accrow title="Release of estimates"]

NRS estimates for any title are normally based on a minimum period of 6 months, if the title has accumulated a sufficient number of unweighted Average Issue Readership (AIR) respondents.

However, in the case of titles newly added to the survey, the intention is to publish a first estimate after as short a period of measurement as possible.

This is achieved under the following rules:

  • No estimates for any title new to the survey will be released on less than one month’s fieldwork. Where a title is newly launched, no estimates for that title will be released until at least four consecutive issues have been published.
  • Subject to 1 above, estimates for a title newly added to the survey will first be released when the title achieves a minimum of 140 unweighted AIR respondents. If such an estimate is released on less than a 6-month sample, subsequent estimates will be based on the cumulative monthly sample achieved by the title until a 6-month base has been accumulated.
  • Thereafter estimates for such a title will be released each quarter on a rolling 6-month base, and rolling 12-month estimates will also be released as soon as a 12-month base has been accumulated.
  • If this minimum of 140 unweighted AIR respondents is not reached after 6 months, but is reached after between 7 and 12 months of measurement, the first estimate released will relate to the period in which the sample threshold is reached (e.g. 9 months January-September). Once such an estimate is released, further estimates for the title will be based on the cumulative monthly sample until a 12-month base is achieved. Subsequent estimates will then be based on a rolling 12-month base.
  • If this minimum of 140 unweighted AIR respondents is not reached after 12 months, estimates can still be released if the title achieves 225 unweighted AIR respondents after 24 months. However because of the data processing procedures involved, no interim estimates can be released even though the 225 minimum may be achieved after 12 but before 24 months.
  • If this target of 225 unweighted AIR respondents is not achieved after 24 months, no estimates will be released.

[/accrow]

 

[accrow title="Changing to shorter periods"]

Once a title has had estimates published on a 12 or 24-month base, the period on which subsequent estimates are based may change, as follows:

  • From 24 to 12 months
    Estimates for a title on a 24-month base will be transferred to a 12-month base if the title achieves at least 140 unweighted AIR respondents in the most recent 12-month period ending in June.
  • From 12 to 6 months 
    Estimates for a title on a 12-month base will also be released on a 6-month base if the title achieves 140 unweighted AIR respondents in each of two consecutive 6-month periods, this 12-month period ending in June. Estimates on a 12-month base will continue to be released.

In all cases the media list will be reviewed each year when data to June becomes available, and the above changes will be made where appropriate.

[/accrow]

 

[accrow title="Changing to longer periods"]

Once a title has had estimates published on a 6 or 12-month base, the period on which subsequent estimates are based may change, as follows:

  • From 6 to 12 months 
    Estimates for a title on a 6-month base will cease to be released if the title’s unweighted AIR respondent count falls below 140 for 2 consecutive 6-month periods, this 12-month period ending in June. 12-month estimates will continue to be released.
  • From 12 to 24 months 
    Estimates for a title on a 12-month base will cease to be released, and be replaced by estimates on a 24-month base, if the title’s 12-month unweighted AIR respondent count falls below 140 for 2 consecutive 12-month periods, this 24-month period ending in June.

In all cases the media list will be reviewed each year when data to June becomes available, and the above changes will be made where appropriate.

[/accrow]

 

[accrow title="Cessation of estimates"]

If the estimated readership of any title falls below a certain level the estimate becomes unreliable and must be withdrawn from publication.

This level is defined as the lower limit of the 95% confidence interval of a theoretical all-adult readership estimate of 0.3% of the adult population.

For example, on a total sample of 36,000, with an estimated adult population of 49 million, an estimated adult readership of 109,000 (0.224%) would fall below this level.

If the estimated readership of any title falls below this level for 2 consecutive 24-month periods ending in June, the last estimate released will be for the period ending that December.

The intention of this rule is to avoid any title being removed from the survey if a shortfall against the required minimum could be the result only of sample variation, rather than because of any real diminution of readership below 0.3% of all adults.

[/accrow]