Converting bird count data into a meaningful summary can be tricky. Here, we illustrate why it is important to get the number-crunching process right, using the Garden Bird Survey as an example.
The Garden Bird Survey (GBS) is a national survey of garden birds carried out each year during the first week of July. Volunteers are asked to record the type and number of birds they detect over a one hour period. The primary aim of the GBS is to answer the question of whether garden bird populations are increasing, decreasing or remaining stable.
However in order to reliably answer this question, the raw results need to be turned into meaningful summaries. In general, the raw data of bird counts per garden for a given year are averaged to give a value that represents the mean (i.e. average) number per garden for that year. A species that is increasing over time will have an average count that also increases over time, and vice versa.
In principal, calculating both the annual average and the trend is straightforward, however as we will explain, a number of challenges are present.
Regional differences in the abundance of a species coupled with the changing proportion of survey returns from each region can make it appear as though the species has changed over time, even though it is only the proportion of survey returns that is changing!
When determining whether an observed increase is real, we must separate the signal from the noise — an increasing trend does not necessarily mean that the population is increasing.