The researcher had difficulty in collecting data. There were several reasons for this. First, little statistical data on the media is generated in the country. The under-resourced research unit of the Zambia National Broadcasting Corporation (ZNBC) produces most of the available statistics on broadcasting trends. What little data the Central Statistical Office (CSO) has generated, mainly covers how often people consume media products, largely through its Demographic and Educational/Health Surveys (CSO, 1998; 2002; CSO, Central Board of Health & ORC Macro, 2003). Few private sector organisations have undertaken serious, long-term statistical research. The Panos Institute Southern Africa and the Zambia Institute of Mass Communication have done some issue-focused studies, which have produced percentages on, for example, how women are covered in the media (PISA, 2004; Gender Links & Zamcom, 2004).
On the international front, Steadman Research Services undertook a study in 2003 to monitor various media trends, such as statistics for readership, listening and viewing. In 2002, ORC Macro conducted an Intermedia National Survey in Zambia, which it repeated in 2004 (ORC Macro, 2002; 2004). Although these surveys generated useful statistics on the media in Zambia, their main aim seems to have been to capture statistical data about consumption patterns of international media, including the Voice of America (VoA), the British Broadcasting Corporation (BBC) World Service, Radio Netherlands, Channel Africa, and so on. Thus, the research data generated has tended to be so commodified that its availability in the public domain is severely limited. Moses Odhiambo, the manager of Steadman Research Services in Zambia, confirmed this in an interview with this researcher.
Another difficulty with the data is that it tends to be represented in a ‘standardised’ way that is inconsistent with the statistical categories developed by the AMDI research group. For instance, the number of telephone lines is usually expressed as a proportion per 1,000 of the population. Data therefore had to be manipulated to fit the research categorisation used in this study.
Finally, the data is not available in one place, making it a tedious task to collect and collate such data from myriad sources. Had the researcher had more time, it is possible that more data could have been located. Also, what data is available may not be reliable because of the lack of sustained effort in Zambia to collect such information.