The All Media Survey 2002
published: Monday | July 21, 2003
Stephen Vasciannie
THE ALL Media Survey for the Year 2002 was compiled by Market Research Services Ltd. on the basis of fieldwork conducted in the six-month period from April 28, 2002 to October 23, 2002. The report indicates that it was compiled for the Media Association of Jamaica, and it was submitted in April 2003. On its face, the survey is a serious piece of statistical research, designed to provide the type of information that should be of value to media managers, advertisers and the wider public.
The survey indicates that it is based on random sampling of the population, and various technical means were applied to ensure randomness and accuracy. Overall, a total of 2,997 persons, aged four years and above, were interviewed, persons were drawn from all parishes of Jamaica, the person in a particular house to be interviewed was determined by identifying the last person in that house to have celebrated a birthday, and interviews were conducted and verified continuously throughout the six-month period of the study. Forty interviewers and eight supervisors were involved in the study, and Market Research Services argues that the margin of error has been reduced to three per cent.
OLD DATA?
The Survey has been criticised as relying on old data, this being a reference to the fact that the six-month period began in April 2002. Whether this criticism is valid will depend on the question one wishes to ask. If the pollster is seeking to identify media patterns for the six-month period identified in the Survey, then no particular problem arises from starting the study in April 2002. If, however, the pollster wishes to use data for the six-month period in 2002 as a way of predicting viewing patterns in 2003, then, clearly, the 2002 results can provide only broad guidance as to the situation in 2003.
But this is inherent in the nature of process: the poll is a guide, based on the past, as to current and future activity, not a definitive statement about the present or the future. It could hardly be otherwise.
RESULTS
Now to some results for the period April 28, 2002 to October 23, 2002. As to radio listenership (based on the total number of different people who listen on a particular day), IRIE FM and RJR 94 were in close competition at the upper end of the scale. The results are set out in Table 1.
TABLE 1
Share of radio
listenership(2002)
IRIE FM 26.11%
RJR 94 25.01%
LOVE FM 13.01%
FAME FM 11.01%
RADIO II 9.02%
POWER 106 6.77%
HOT 102 5.65%
KLAS FM 3.42%
When compared with figures obtained by the same polling agency for the Year 2000, the share of radio listenership for 2002, in Table 1, represented a decline of 3.38 per cent for IRIE FM, an improvement of 1.99 per cent for RJR 94, a slight decline of 0.26 per cent for LOVE FM, an improvement of 0.32 per cent for FAME FM, and a decline of 2.53 per cent for RADIO II (from among the top set of stations in Table 1). As to newspapers, the results for total readership of select newspapers are set out in Table 2.
TABLE 2
Total Readership, Selected Newspapers (2002)
Sunday Gleaner 580,000
Daily Star 512,000
The Gleaner 491,000
Daily Observer 335,000
Weekend Star 328,000
Sunday Observer 295,000
Sunday Herald 131,000
X-News 105,000
In the case of The Gleaner, and the Daily Observer, the figures given in Table 2 are an average of the number of readers from Monday to Saturday. For instance, the Wednesday Gleaner had a reported readership of 548,000, while for the Friday Gleaner the corresponding figure was 480,000. For the television sector, the overall "tube share", for Monday to Sunday in the specified period was as set out in Table 3.
TABLE 3
Tube Share, Monday to Sunday (2002)
Cable 36.6%
TVJ 35.6%
CVM TV 27.1%
LOVE TV 0.8%
The 36.6 per cent share indicating the popularity of Cable TV was surprising to me, and prompted the thought that because it may be a "status thing" to have cable, the statistical results concerning cable could benefit from closer scrutiny. When reference is made to viewership among local television stations (without cable) from Monday to Friday, TVJ had 56.5 per cent of the market share, CVM TV 42.3 per cent and LOVE TV 1.2 per cent; the corresponding figures for Saturday and Sunday reflect a rough similarity to the Monday to Friday breakdown.
DISAGGREGATION
The All Media Survey also provides detailed disaggregation of listenership and viewing patterns for the different stations. In the case of radio, the time segments are broken down into 15-minute periods, while for television the breakdown is on a half-hourly basis.
On the basis of this disaggregation, and with reference to advertising charges for various radio and television segments, the survey presents interesting results on the cost of reaching 1,000 listeners by advertising on different stations. This cost varies significantly from station to station, but that may not be decisive in all cases for advertisers: among other things, advertisers also need to know the social class composition of the listeners and viewers of particular programme if they wish to maximise the advertising dollar.
PERPLEXING
Finally, although the methodology of the Survey appears to be sound, some results do not ring true. Take, for instance, the question of listenership to early morning talk shows. According to the detailed breakdown, the listenership pattern for the KLAS FM programme First Edition went like this on Monday mornings. At 6 a.m. the programme started out with 6,000 listeners; by 6:15 listenership rose to 17,000, and held steady until 7 a.m. when it rose again to 23,000; at 8 a.m. it fell to 17,000, by 8:15 it was down to 11,000, and at 8:30 only 6,000 listeners were tuned in, up to 9 a.m.
Now compare this with Tuesdays. First Edition started out with fewer than 6,000 at 6 a.m.; from 6:45 to 7:30, 6,000 listeners were on board, between 7:30 and 8 a.m., the figure rose to 11,000, and then at 8 a.m., there was a dramatic jump to 22,000. The 22,000 listeners then stayed until 9 a.m. Among the discrepancies here is the fact that 6,000 people were on board on Mondays at 9 a.m., versus 22,000 one day later - when, in fact, the programme format did not really vary in the last half an hour on Mondays or Tuesdays. What could explain this, one wonders.
Similarly, as to the Breakfast Club, the listenership patterns may best be described as perplexing, if the survey is to be believed. Between 7 a.m. and 8 a.m. on Wednesdays, only 6,000 people listened to Anthony Abrahams and Beverly Anderson-Manley; but on Mondays, between 7 and 8 a.m., 28,000 people listened, and on Tuesdays between 17,000 and 22,000 listened for the same time period. Come now to Thursdays, 7 to 8 a.m.: between 28,000 and 34,000 people tuned in to Abrahams and Anderson-Manley, and by Friday, the figure reached a splashing 56,000 for the same 7 to 8 a.m. period. Pray tell, Market Research Services: could it be the club members?
Stephen Vasciannie is Professor of International Law at the University of the West Indies and Consultant in the Attorney-General's chambers.
