• RSS Feed

The LinkUp Blog The Industry's Best-Kept Secret

October 9, 2008 / Toby Dayton

LinkUp Data Shows Surprisingly Positive Jobs Picture In September

Flying in the face of the U.S. Department of Labor September jobs data which showed a loss of 159,000 non-farm jobs during the month, LinkUp’s job data for September presents a far more positive outlook on the U.S. employment numbers. In fact, 45 states in the U.S. showed an increase in new jobs between August and September, while 48 states showed an increase in total job openings. New job listings on LinkUp rose from 643,175 in August to 726,472 in September, an increase of 83,297 new jobs. Total job listings on the site rose from 860,375 in August to 951,970 in September, an increase of 91,595 total jobs.

LinkUp is a job board that aggregates individual job postings pulled directly from company web sites such as Target, Apple, or Google. Because the job listings come directly and solely from the career sections of corporate and employer web sites, the jobs are real, current, and often unadvertised anywhere else on the web. LinkUp’s job postings also contain no duplicate listings for the same job because they are only sourced from a single site – the actual employer posting the job opening.

The LinkUp data above is not meant to present an accurate report on the U.S. economy as a whole. It is, rather, a relevant snapshot of the number of new and total job postings published on a huge number of company web sites from around the country (8,321 companies were included in the data set above). To the extent that the aggregate job growth or decline from a set of 8,000+ companies that post job openings on their company web site is statistically significant, the data may or may not be representative of what is happening in the broader economy. Given the disparity between the LinkUp data and the Department of Labor data, I am not sure that it is representative (though it’s tough to be certain given the constant revising that takes place with the government employment numbers). At any rate, the data above is certainly noteworthy.

As a side note, we significantly improved the methodology of how we obtain our jobs data each month by ‘normalizing’ the data set between months. Because we are adding new companies and employers to our site every month, the data in the past tended to show artificially inflated job numbers in a month to month comparison. Beginning this month, however, the jobs data for month-to-month comparisons will contain the exact same set of employers for the 2 months being reported.

For example, the data for August and September in the chart above is obtained from the exact same 8,321 companies across the U.S. included in the database on August 1st, 2008. In August and September, we added an additional 3,136 companies into the database, and the job postings for those companies and employers are not included anywhere in the chart above. The total 11,457 companies in the database at the end of September will be used to examine job growth or decline between October and November. The same companies in the database as of September 1st will be used to compare September and October job growth or decline. (As a result of this methodology, it is therefore not valid to compare October data or any month thereafter to August data due to the addition of those 3,136 new companies into the data set during August and September).