The latest LinkedIn Talent Trends 2014 report states that 75% of the global workforce are not actively looking for a new job; the so-called Passive Candidates. No doubt this and other figures will become as widely quoted as those were in the 2012 report but can we trust them?
This blog probes the statistics in a little more detail – including the UK dimension – and compares it with some other secondary research.
The LinkedIn survey was based on a sample of 18,219 fully employed workers globally with a UK subgroup size of 742. Given that the LinkedIn member populations are about 300 million and 15 million respectively, can this be representative of the whole? Only if the margin of error is taken into account which is ± 0.74% for the global sample and a larger ± 3.67% for the UK subgroup.
As the data was collected in two samples in June and December 2013, this is not really a 2014 report, and apart from a comparison with super-passive candidate proportions in 2012, the report does not present any trends either.
The report fails to clarify how survey respondents were selected so there is the possibility of sampling bias. If we assume that respondents were all LinkedIn members, then the sample will be naturally biased towards the membership demographic and their stated job-seeking behaviour.
The unemployed were excluded from the report. The unemployment rate in the UK for example is about 28% (ONS), which means that the study undercounts the true number of active job seekers and cannot be directly compared with research, which includes this group.
The only trend observation of the report is that globally the proportion of super-passive candidates has fallen from 20% to 15%. The report uses this to make the case that more people on LinkedIn are becoming ‘approachable’.
The UK picture is, however, quite different with 21% describing themselves as super-passive (with an upper bound of 24% at the 95% confidence level).
The sub-categories of active/passive candidates used by LinkedIn have their origins in work by
Lou Adler who contributes to their market reports. These classifications have been adopted in other research contexts.
Sending out resumes, actively looking for a position
- Hunter, Searcher, Networker Tiptoer
Thinking about changing jobs and reaching out to close associates
Not looking for a job but willing to discuss opportunities
- Super Passive
Happily employed and not interested in a new opportunity
Does secondary research out there agree with LinkedIn?
Perhaps one of the reasons why the LinkedIn reports are widely quoted is that there is not much of it. Jobvite’s US Social Sourcing surveys in 2011/12 contained some figures. A specific vertical role and market segment was also examined from Recruiting Physicians Today (April 2014, Massachusetts Medical Society). These sources reveal significantly higher proportions of passive and super-passive candidates as shown below both with 95% confidence ranges.
Whilst the time periods and populations may not be directly comparable, the difference in these results should lead us to question whether the LinkedIn report offers a truly unbiased picture upon which to make sourcing decisions. Or even whether investing time and money in LinkedIn really produces a good passive candidate ROI?
The Jobvite surveys included respondents with a wider social media footprint, which may explain some of the difference. LinkedIn-centric members could simply exhibit more active job-seeking behaviour than those frequenting non-business platforms. For every active candidate on LinkedIn there may be three passive ones, but there could be six to nine living primarily in other social networks, and those with more technical skills (such as physicians) who may be lurking on specialist sites.
So do I trust LinkedIn’s findings?
The answer is yes, but with the rider that we should be aware of the sampling bias, and provided we do not take the total figure as being the UK picture. Also we need to be aware that the differences in specific sectors are not represented; any survey of LinkedIn membership is not representative of workforce, or vertical sectors. Their sample is naturally biased towards its community, but I would question how much passive talent is on the platform really. It is a well-known fact that LinkedIn is the social site preferred by professionals when actively looking for opportunities to move their careers on, so it would have been more accurate to include passive voices.
So how then do you track down those latent candidates whose social footprint extends far beyond the LinkedIn boundary without time consuming manual methods? You might be prepared to pay a hefty agency fee on the promise of a fill months down the line. If you want to source in-house though, how can you invite enough passive applicants to produce a solid shortlist and still hit the hiring manager’s SLA?
The dawn of the big data age in recruitment means there are indeed much smarter ways to hunt down and attract talent before your competitors do. I forecast the way we recruit in the UK is beginning to change dramatically and the expensive recruitment agencies should make sure they are prepared for this.