How the Online Labour Index is constructed
The Online Labour Index is an index that measures the utilization of online labour platforms over time and across countries and occupations.
Online labour platforms are here understood as platforms through which buyers and sellers of labour or services transact fully digitally. That is, we require that the worker and employer are matched digitally, the payment is conducted digitally via the platform, and that the result of the work is delivered digitally, excluding platforms for local services such as Uber and Airbnb. The index is based on tracking all projects and tasks posted into a selected sample of platforms, using API access and web scraping.
Which platforms are measured?
We currently limit our sample to the largest English language platforms as indicated by the unique monthly visitor estimate provided by Alexa.
To approximate the coverage of the OLI, we collected a list of 40 prominent English-language online labour platforms, retrieved their monthly estimated unique visitor counts from Alexa, and selected the top five. Using Alexa’s figures, we estimate that the platforms in our sample account for at least 60% of all traffic to English-language online labour platforms. They also represent a range of different market mechanisms and contracting styles, from online piecework to hourly freelancing.
The data from which the OLI is calculated is collected by periodically crawling the list of vacancies available on each of the sample platforms. As in conventional labour markets, a vacancy refers to a job, project, or task offered by a firm that wishes to hire a worker. For each crawl, we save the status of each vacancy: open, in progress, or completed. Comparing changes in statuses allows us to calculate the number of new and filled vacancies between two crawls.
The main shortcoming of this approach is that we do not observe vacancies which were either posted and completed between two crawls, or which were completed without a vacancy being posted. The latter might happen if a vacancy is filled without it being posted on a platform. These hidden vacancies exist, and remain unmeasured, in traditional vacancy statistics as well. Notwithstanding these caveats, we believe that our measure fulfils its purpose of tracking the volume of work transacted on the platforms.
Besides vacancy status, we also collect the occupation classification and employer country for each vacancy, when available.
The Online Labour Index is calculated from the number of new vacancies on each day.
As we are not tracking every single online labour market platform, and due to differences in details of data collection, it is not be possible to produce aggregates of absolute numbers; instead, the product is a set of indices that track changes over time in relation to the start of the data collection. In other words, we have normalised our time series so that the value in May of 2016 equals 100.
For more details on the insights provided by OLI, see here.
This post is a first post in a series of blog posts which discuss the methodology of the construction of the OLI. For details of how the OLI is constructed, see also How is online work classified on the OLI.
Note: This post was originally published on the OII's iLabour project blog on . It might have been updated since then in its original location. The post gives the views of the author(s), and not necessarily the position of the Oxford Internet Institute.