I ask this question in all sincerity, because the subject came up in conversation recently. It promoted me to dig out a survey I did in 2014 for a major university. Using my media connections, I was given access to the number of application clicks some of the main job-boards saw leaving their sites. By matching this with the number actually arriving on the specific job on the university's own website, as shown by Google Analytics, it was clear there is a bit of a discrepancy for many jobs. Some sample stats below, taken from three separate job-boards (two generalist and one specialist), should give an indication of the problem. You'll have to take my word that these are representative of the entire sample - which was based on a study of over 25% of all jobs posted in the previous 12 months, a statistically robust sample from which to draw conclusions.
Media Apply Clicks Actual Apply Clicks
Job 1 20 15
Job 2 2 0
Job 3 69 33
Job 4 18 12
Job 5 10 9
Job 6 12 3
Job 7 10 9
Job 8 32 18
Job 9 32 6
Job 10 78 42
In addition, the actual conversion rate of applications to successful candidates per job-board made for very interesting reading, with the number of jobs filled being generally a low, single figure percentage of the number of applications from each job-board or other source. However, that's for another day.
The reason for publishing this now is because there is a huge amount of interest in the number of jobs available (or not, as the case realistically is) due to the coronavirus crisis. Alarmist headlines in the media, such as 'Britain's Job Market is the Worst in the Global Index" do not help create a jobs market in which we might get out of our predicament, while studies by job-boards themselves which show the number of applications by region and sector are, if the pattern shown in this study is accurate, potentially very misleading. I suspect that the fact that job-boards were more significant players six years ago, added to the fact that their role has diminished slightly as better tech allows targeted, 'inch-wide, mile-deep' searches by recruiters while social media have increasingly offered another (not necessarily better to be fair) channel to the market, all conspire to make the figures shown here likely to be replicated should a similar study be done today.