The primary objective of the Joint National Readership Survey (JNRS) is to provide reliable estimates of readership of the main newspapers and magazines circulating in the Republic of Ireland.
The Joint National Readership Survey is designed to generate 2 types of readership measurement:

1. Average Issue Readership and
2. Reading Frequency.

1. Copyright
• All data derived from the Joint National Readership Survey (JNRS), are subject to copyright.
• Limited extracts from the contents of JNRS may be published without prior permission provided that the source of the extract is acknowledged as “JNRS/Lansdowne (date of report period)”
• Data obtained by special analysis may be published, provided the source is quoted as above. In general, published data should be based on a minimum sub-sample of 140 respondents.
• Research companies can only have access to special analysis of the data after a prior formal request has been granted by the Joint National Readership Survey Management Committee.
2. Reference to data used (if not AIR)
• Except in cases where the ‘average issue readership’ is quoted, reference should be made to the basis of the readership quoted with specific attention given to the fact that the figures quoted are not AIR.
• In all instances duplication of readers should be excluded from published data (otherwise the reference should clearly state that the data quoted is ‘not based on readers but ‘opportunities to see (OTS)’’
  For example:
  Source: JNRS/Lansdowne 2007/2008.
  NB: Readership figures quoted are not AIR. They are based on cumulative coverage of 6 insertions in Publication x and 1 insertion in Publication y. Duplication has been removed.
3. Accessing Data from less than 12 months fieldwork
• Data based on a period shorter than the recommended 12 month period may not be published or used in any advertising or other promotional material without the permission of the JNRS Management Committee.
• Permission to access to data for periods shorter than 12 months must be granted by the JNRS Management Committee and is solely for internal use.

JNRS Management Committee is committed to ensuring that information from the Survey is not misused. As guidelines only, mis-use may arise:

1. where results are based on inadequate sample sizes;
2. if the sample size and the universe to which the figures relate are not made clear;
3. where direct or implicit comparisons, in figures themselves, or accompanying text, are drawn between basically dis-similar sets of data;
4. where a difference between two periods is interpreted as a change in readership without checking whether the difference is large enough to be significant, bearing in mind the sample size.
5. Where comparisons are drawn between data which have been subject to methodological changes.

In any case of doubt regarding the acceptability of a special analysis, consultation with the JNRS Management Committee prior to publication is strongly advised.

There are potentially 3 separate situations that could arise with regards to a break in the ‘normal publishing cycle’. These are outlined below together with our recommendation on how best to address such situations (specific details of the recommendation is outlined over):

• Situation A: Publications unavailable for more than 4 consecutive publishing cycles. We recommend predictive modelling during the interviewing periods when the title is not available.
• Situation B: In cases where the publication misses 3-4 consecutive cycles we suggest that the relevant extended readership recency should be used for calculating the AIR. This would apply only to the interviews conducted during the relevant interviewing period.
• Situation C: In cases where a publication is unavailable for less than 3 of its normal publishing cycles we recommend no changes are made at the post analyses stage.

A more detailed account of each of the techniques that could be used is outlined below for your consideration and discussion.

The method would use predictive modelling techniques to ‘best estimate’ the readership levels during the period that the publication is not published. The resultant AIR would be a ‘best representation’ of the true AIR for the publication in that it would not suffer the decrease in AIR as a result of it not being available in the marketplace for a period during fieldwork data collection. A separate charge will be negotiated between Lansdowne and the User in the event that this method is required.

The question used to define AIR asks when each respondent ‘last read’ a copy of each publication. For Sunday and weekly publications it is ‘past week’ and for monthly titles, ‘past month’. During the interviewing period when the publication is not available we could simply code up the extended recency option of ‘past month’ for weekly titles and ‘past 3 months’ for monthly titles (new code would need to be added to the answer grid).

However, unless the period of ‘non-publication’ is relatively short, we feel that the publication is likely to suffer as result of ‘poor recall’ (and especially so if the title has a relatively short ‘shelf-life’).

In this situation the title would simply be published as if it were available for the whole of the research period. So besides respondents who may have read an ‘old copy’ of the publication no ‘new’ readership would be recorded during the period that the title is not published.

Out of interest, the NRS encounter this issue regularly as many publications do not publish for the 52 weeks of the year, however they treat them as if they do, and do not conduct any post analysis of the AIR results.

Bearing in mind that the number of publications measured averages around 362 and that there may be up to 70 changes to the main media list during any one reporting period it is not too surprising that NRS have implemented stringent rules in relation to their survey.

The onus would be on each publisher to notify Lansdowne of any changes to their respective titles on an ongoing basis. The above rules would apply only to weekly, fortnightly or monthly publications. It would not apply to daily publications.