Media monitors – whether electoral administrators, international observers, civic activists or academics –should address a number of practical and methodological questions before embarking on their project:

• What are the overall goals of the monitoring? Is the monitoring intended as a process of constructive intervention in election implementation – for example, to require the media to adhere to professional standards – or is it intended primarily to document whether media coverage was fair and balanced? Does the monitoring provide a broad look at the media’s contribution to free and fair elections, or more narrowly at certain aspects?

• Who is the target audience for results? What format will the results take? How often will results be reported? What resources are available or needed, and what is the best way to allocate them for the monitoring tasks?

• What media are to be monitored? Will it be just public media, or all media? Will it be just broadcasting outlets or print media? Will monitor social media be monitored? Will it be a selection of media or all major national outlets? Will we consider sub-national media?

• Which parts of the media output are to be monitored? Will it be specified news bulletins, all output during particular times of the day, or all output?

• What content will be monitored (and with what purpose): news, advertising, free direct access slots, special programming, voter education, or all of these?

• Will the monitoring seek to gather only data about how much time was allocated to the different parties or candidates or will it also look at other aspects of coverage, such as the use of language, the selection of news stories, and so on?

The answers to each of these questions have an important impact on the monitoring methodology that is adopted.

Most media monitors employ methodologies that utilise a technique known as ‘content analysis’. This kind of analysis is essentially quantitative in nature. In other words, it is concerned with elements of media output that can be measured and counted. Content analysis has sometimes been criticized for reducing media coverage to what is measurable, omitting important aspects such as tone and language, or identifiers such as audio or visual. There are many things that content analysis cannot do – most simply and obviously, it cannot reveal whether news coverage was accurate or inaccurate.

Typically, quantitative monitoring of media election coverage will focus time allocation according to the various parties and candidates. This may then be qualified by an assessment of whether the coverage is favourable or unfavourable. Although these measures may also be quantified, they are essentially qualitative judgments.

Some monitoring methodologies introduce other types of quantitative measures in an attempt to avoid relying on monitors’ assessments of whether coverage is positive or negative. They may, for example, count the sources that journalists use, assigning them to different political or social categories. This may be a more objective measure of balance. They may classify media items by topic. This can be useful since, in an election campaign, political parties often campaign not only with different positions but also on different issues. The media’s selection of topics may therefore be a sensitive indicator of their political sympathies.

Another aim of quantitative monitoring may simply be to measure the amount, and perhaps timing, of political advertising or free direct access programming. This may be to ensure that what is actually published or broadcast conforms to the laws or regulations governing direct access.