Scratching beneath the surface of data headlines
A press release landed on my desk on Tuesday with a bold claim. UK database analysts are earning less than half the salary of their peers across the pond.
Research from Redgate Software proclaimed that the average salary for a UK database administrator is $44,483, 47% of the average wage in the US of $94,176.
I thought, "that could be true. Everything is bigger is the States."
But aren’t we in the UK in the midst of a skills scarcity, with high demand and low supply across the data and analytics sector? With this being the case, one would assume that pay would be higher in the UK.
This data needed to be delved into.
Firstly, who collated this information? The survey was “carried out on the Redgate Software community site SQLServerCentral.” But what’s that when it’s at home?
Helpfully, a nice blurb at the end of the release explained. “Redgate makes ingeniously simple software used by 804,745 IT professionals and is the leading Microsoft SQL Server tools vendor. Redgate's philosophy is to design highly usable, reliable tools which elegantly solve the problems developers and DBAs face every day, and help them to adopt database DevOps. As a result, more than 100,000 companies use products in the Redgate SQL Toolbelt, including 91% of those in the Fortune 100.”
The release also said that survey took place and the end of 2016 with 432 respondents. The raw data could be downloaded from SQLServerCentral.
So I registered and downloaded. SQLServerCentral soon after sent me an email informing me that I am part of a community of 1.6 million data professionals.
The spreadsheet shows the records of 435 respondents. Assuming the larger figure is true, only a miniscule percentage of members took part in the survey - 0.027% to be precise.
Then I looked more closely at the spreadsheet. One thing about the data that stood out was that many of the respondents seemed to be from the USA. I did the maths and I was right. It turns out that 281 - or 65% - of the respondents are from the US.
The US is followed by India, the country in which 33 (7.5%) respondents are located, then the UK with 30 (6.9%) and Canada with 14 (3.2%). Total respondents from the remaining 36 countries are all in the single figures. Thirty of the 40 countries represented only had one, two or three respondents.
I confess that I am no data whiz, but I am sure that a bigger sample size is needed to give meaningful and representative results for data outside the US. When sample sizes are small, anomalous and outlying responses can skew the findings.
Italy is home to one of those outliers. Five respondents reported a mean average salary of $73,000, but the median salary is just $44,000. Italy’s results have been skewed by an outlier - one person with 19 years’ experience who takes home $200,000 a year.
From the findings, SQLServerCentral editor Steve Jones came to a conclusion and stated: “The message is clear - those in the US…have the highest average salaries, making them the top choices for experienced DBAs, while the UK lags behind the rest of Europe when it comes to take-home pay.”
It wouldn’t be so bad if the survey owners made it clear that they are just talking about a self-selected group of people within the SQLServerCentral community. But that is not the case.
I decided to look at other sources to see how they compare using Payscale, but I should note that I don’t know the methodology Payscale uses. Also, I didn’t know if I was comparing apples and oranges
Diod Steve and his team include the salaries of independent and agency contractors when working out average salaries? Were the hourly rates converted into annualised salaries? If so, what assumptions were made about the number of hours worked per week and weeks worked per year? Without guidance notes or appendices, there was no way to tell.
The results on Payscale are different what SQLServerCentral put forward, but also not a million miles away.
For all countries except for the UK, SQLServerCentral showed a higher salary than Payscale. What does this tell us? Perhaps people who are part of online communities earn more than people who don’t. It could be that high-earning database administrators are more likely to take part in surveys.
The moral of the story - if there is one - is to always go to the source, interrogate the data and look for other sources of the same information. You might still be none the wiser but at least you will have information from which to draw your own conclusions.